From Prejudice to Intergroup Emotions : Differentiated Reactions to Social Groups 9781317710219, 9781841690476

Citation preview

ZZZURXWOHGJHFRP

FROM PREJUDICE TO INTERGROUP EMOTIONS

Page Intentionally Left Blank

FROM PREJUDICE TO INTERGROUP EMOTIONS Differentiated Reactions to Social Groups

Edited by

Diane M. Mackie

University of California, Santa Barbara

and Eliot R. Smith Purdue University

First published 2002 by Psychology Press Published 2016 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, 0X14 4RN 711 Third Avenue, New York, NY 10017, USA Routledge is an imprint of the Taylor & Francis Group, an informa business Copyright © 2002 by Taylor & Francis, Inc.

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Notice^ Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. ISBN 13: 978-1-84169-047-6 (hbk) ISBN 13: 978-1-84169-048-3 (pbk) Cover Design: Pearl Chang Cover Image: Corbis

Library of Congress Cataloging-in-Publication Data Mackie, Diane M. From prejudice to intergroup emotions : differentiated reactions to social groups / Diane M. Mackie and Eliot R. Smith p. cm. Includes bibliographical references and index. ISBN 1-84169-047-3 1. Intergroup relations. 2. Social groups. 3. Prejudices. 4. Emotions— Social aspects. I. Smith, Eliot R. II. Title. HM716 .M33 2002 302.3— dc21

2002017835

Contents Contributors Preface 1 Beyond Prejudice: Moving from Positive and Negative Evaluations to Differentiated Reactions to Social Groups Diane M. Mackie and Eliot R. Smith

ix xi 1

2 Social Self-Discrepancies and Group-Based Emotional Distress Aharon Bizman and Yoel Yinon

13

3 Promotion and Prevention Forms of Ingroup Bias James Y. Shah, Paige C. Brazy, and E. Tory Higgins

31

4 Antecedents and Consequences of Collective Guilt Nyla R. Branscombe, Bertjan Doosje, and Craig McGarty

49

5 Intergroup Emotions and Self-Categorization: The Impact of Perspective-Taking on Reactions to Victims of Harmful Behavior Vincent Yzerbyt, Muriel Dumont, Ernestine Gordijn, and Daniel Wigboldus

67

6 Intergroup Encounters and Threat: A Multi-Method Approach 89 Jim Blascovich, Wendy Berry Mendes, and Mark D. Seery v

Vi

CONTENTS

7

Experiencing Intergroup Emotions

111

Thierry Devos, Lisa A. Silver, Diane M. Mackie, and Eliot R. Smith 8

Expressing Emotions and Decoding Them: Ingroups and Outgroups Do Not Share the Same Advantages

135

Jacques-Philippe Leyens, Stéphanie Demoulin, Michel Désert, Jeroen Vaes, and Pierre Philippot 9

The Role of Affect in Determining Intergroup Behavior: The Case of Willingness to Engage in Intergroup Contact

153

John F. Dovidio, Victoria M. Esses, Kelly R. Beach, and Samuel L. Gaertner 10 Close Encounters of the Suspicious Kind: Outgroup Paranoia in Hierarchical Trust Dilemmas

173

Roderick M. Kramer and John T. Jost

11

The Role of Threat in Intergroup Relations

191

Walter G. Stephan and C. Lausanne Renfro

12 Intergroup Emotions and Images

209

Marilynn B. Brewer and Michèle G. Alexander

13 The System Justification Motive in Intergroup Relations

227

John T. Jost and Roderick M. Kramer 14 Emotions Up and Down: Intergroup Emotions Result from Perceived Status and Competition Susan T. Fiske, Amy J. C. Cuddy, and Peter Glick

247

CONTENTS

15 Intergroup Emotions: A Biocultural Approach Steven L. Neuberg and Catherine A. Cottrell

265

16 Commentary Eliot R. Smith and Diane M. Mackie

285

Index

301

VU

Page Intentionally Left Blank

Contributors Michèle G. Alexander University of Maine Kelly R. Beach Syracuse University Aharon Bizman Bar-Ilan University Jim Blascovich University of California, Santa Barbara Nyla R. Branscombe University of Kansas Paige C. Brazy University of Wisconsin-Madison Marilynn B. Brewer Ohio State University Catherine A. Cottrell Arizona State University Amy J. C. Cuddy Princeton University Stephanie Demoulin University of Louvain at Louvain-laNeuve Michel Desert University of Louvain at Louvain-laNeuve

Muriel Dumont University of Louvain at Louvain-laNeuve Victoria M. Esses The University of Western Ontario Susan T. Fiske Princeton University Samuel L. Gaertner University of Delaware Peter Glick Lawrence University Ernestine Gordijn University of Amsterdam E. Tory Higgins Columbia University John T. Jost Stanford University Roderick M. Kramer Stanford University Jacques-Philippe Leyens Catholic University of Louvain at Louvain-la-Neuve Diane M. Mackie University of California, Santa Barbara

Thierry Devos San Diego State University

Craig McGarty Australian National University

Bertjan Doosje University of Amsterdam

Wendy Berry Mendes University of California, Santa Barbara

John F. Dovidio Colgate University

Steven L. Neuberg Arizona State University

ix

X

Pierre Philippot University of Louvain at Louvain-laNeuve

Walter G. Stephan New Mexico State University

C. Lausanne Renfro New Mexico State University

Jeroen Vaes University of Louvain at Louvain-laNeuve

Mark D. Seery University of California, Santa Barbara

Daniel Wigboldus University of Amsterdam

James Y. Shah University of Wisconsin-Madison

Yoel Yinon Bar-Ilan University

Lisa A. Silver University of California, Santa Barbara

Vincent Yzerbyt University of Louvain at Louvain-laNeuve

Eliot R. Smith Purdue University

Preface

O

ne of the main motives behind our continued collaboration is our mutual interest in theoretical integration in social psychology. This interest sparked our first co-authored project, an introductory textbook that presents all the different domains of our field as unified by a common set of integrative theoretical principles. The same interest was a major impetus for our 1998 Psychological Review paper advocating further integration across areas in the study of intergroup relations. In that article we explicitly urged social psychologists to integrate emotion theory into the study of prejudice and discrimination, an idea first suggested and developed by Smith (1993). We have now started to take our own advice, developing lines of research focusing on whether and how and why people might experience emotions not because of events that occur to them personally, but because of events that affect groups to which they belong. Obviously, when people experience emotion because of or on behalf of their group memberships, these emotions may play an important role in their reactions to outgroups, and so ultimately in patterns and outcomes of intergroup relations in society. As our research on these topics developed, we began to discover others engaged in similar attempts to bring emotion into the study of prejudice and intergroup relations. The specific topics being investigated by different researchers were varied, and the theoretical frameworks taken as starting points were diverse. Nevertheless, all these projects seemed to reflect a similar type of discomfort with the current state of knowledge in the field. Researchers asked questions like the following: What is the emotional significance of group membership, rather than merely its cognitive consequence? How can we understand the apparently context-sensitive nature of prejudice and discrimination, if prejudice is merely a positive or negative evaluation permanently and invariably affixed to an outgroup itself? Do people really react in similar negative ways to all kinds of negatively evaluated outgroups? And all these researchers seemed to converge on the incorporation of affective and emotional reactions as the best means of moving forward theory and research in this area. We brought together some of these approaches to the topic at a sympo-

xi

XÜ

PREFACE

sium we organized at the Society of Experimental Social Psychology annual meeting in St. Louis, Missouri, in October 1999. This symposium brought to our attention still other colleagues involved in similar endeavors. Together with the enthusiastic audience response to the symposium, this convinced us that a single volume bringing together these various research programs could be a timely and valuable impetus to further development in the area. As we talked to potential contributors, their enthusiasm matched our own. We were thus able to assemble an outstanding group of authors who brought to bear on these issues an incredible diversity of perspectives, richness of theoretical frameworks, and creativity of insight. The chapters that follow include descriptions of multiple research programs, all of which add to the general picture of the important and intricate role of emotion in intergroup relations. The chapters also include valuable and exciting speculation about what more needs to be done and in which new directions the field could profitably turn. Our contributors managed to maintain their enthusiasm for the project in the face of our draconian enforcement of deadlines and extensive editorial feedback, and the present volume—and the fact that it was submitted very close to the contract deadline—is the result of all our collective efforts. We have a very happy emotional reaction to this ingroup. Like so many intellectual pursuits, this volume reflects the participation, help, and contributions of many people. We wish to thank in particular our collaborators and colleagues at Santa Barbara and Purdue, all of whom have contributed in varying degrees to our discussions of these issues. Those who have been our closest collaborators and often co-authors on related research are Thierry Devos, Amber Garcia, Dan Miller, and Lisa Silver. Other valuable contributions have come from Jim Blascovich, Heather Claypool, Nancy Collins, Dave Hamilton, Stan Klein, Brian Lickel, Brenda Major, Jeremy Sack, Karen Schudel, Mike Silver, Abe Rutchick, Vinita Thakkar, and Crystal Wright. The project would have remained just an idea without the enthusiastic backing from Alison Mudditt at Psychology Press. Alison liked both the idea and the list of contributors, and continued to quietly encourage us through the development of the project and some important professional changes in her life. In her place, Caroline Osborne has more than capably seen the volume through to completion. Finally, we want to thank each other for the benefits of co-editorship. As usual, working together has not only been intellectually rewarding, but fun.

REFERENCES Mackie, D. M., & Smith, E. R. (1998). Intergroup relations: Insights from a theoretically integrative approach. Psychological Review, 105, 499-529. Smith, E. R. (1993). Social identity and social emotions: Toward new conceptualizations of

prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 297-315). San Diego, CA: Academic Press.

1

Beyond Prejudice

Moving from Positive and Negative Evaluations to Differentiated Reactions to Sociai Groups DIANE M. MACKIE ELIOT R. SMITH

T

f he annals of history and contemporary news sources bear overwhelming witness to the variety of ways in which outgroups are devalued, discriminated against, and sometimes decimated by the members of other groups. The human toll of such negative intergroup behavior, as well as its apparent near universality, and its seeming intransigence, have always loomed as much more than just theoretical and empirical challenges to social psychologists. Horrified at the scope and extremity of Nazi-exploited anti-Semitism, and frustrated at the obstinacy of international, religious, and ethnic strife, social psychologists have always recognized the urgent practical burden that their study of discrimination must bear. In attempting to understand, predict, and inhibit such negative intergroup behavior, social psychologists have typically focused on prejudice—a positive or negative evaluation of a group and its members—as the cause of discrimination (Allport, 1954; Crosby, Bromley, & Saxe, 1980; Hogg & Abrams, 1988; Mackie & Smith, 1998; Sigall & Page, 1971; Spencer & Fein, 1997; Tajfel, 1969; Zanna, 1994). This evaluative model of prejudice is almost as old as the discipline itself (Allport, 1946, 1954; Jones, 1998). The development of attitude theory made social psychology uniquely impermeable to behaviorist inroads, and evaluations with social groups as their object were central in the development of attitude theory. Theories of the antecedents of prejudice toward other groups have changed remarkably over the course of the decades, reflecting the discipline s changing preoccupations with personality, cognitive, and motivational explana-

1

2

FROM PREJUDICE TO INTERGROUP EMOTIONS

tions of phenomena. But the end product of such theories has typically been prejudice defined as a (usually) negative evaluation of, dislike for, or antipathy toward a group and its members. In this attitude-based model, discrimination— behavior directed toward a group or its members—can then be viewed as evaluation-consistent behavior. The fruitfulness of this approach is reflected in the many theoretical, empirical, and practical achievements obvious in any review of stereotyping, prejudice, or intergroup relations (Brewer & Brown, 1998; Brown & Gaertner, 2001; Fiske, 1998; Mackie & Smith, 1998; Sedikides, Schopler, & Insko, 1998). As with any maturing theory, however, the very success of the endeavor makes clear its limitations. In its most traditional form, and borrowing from its attitudinal ancestors, prejudice as a unidimensional liking or disliking associated with the mental representation of a social group has some significant difficulties in explaining intergroup behavior. Not coincidentally, recent developments in attitude theory have also started to acknowledge these same limitations. First, traditional models of prejudice provide no theoretical basis for considering differentiated reactions to social groups. By the attitude model, prejudice toward groups who have negative characteristics (and are thus negatively evaluated) or who engender negative feelings (and are thus negatively evaluated) is identical. So too is prejudice toward groups that elicit fear (and are thus negatively evaluated) and toward those that elicit anger (and are thus negatively evaluated). Dissatisfaction with this type of inflexibility in the attitude domain is reflected in the development of component models, in which attitudes based primarily on affective versus cognitive versus behavioral foundations are distinguished (Chaiken, Pomerantz, & Giner-Sorolla, 1995; Edwards & von Hippel, 1995; Zanna & Rempel, 1988). To our knowledge, no current model has gone further in looking at the different types of information within one of the component classes upon which evaluations are thought to be based. Second, in traditional attitude (and therefore prejudice) approaches, evaluation is seen as adhering to or being associated with the attitude object in an allor-none manner. As noted by Bodenhausen, Mussweiler, Gabriel, and Moreno (2001), this approach has generally assumed that "pervasive, culturally embedded forms of social conditioning tend to produce consistent patterns of affective reactions to certain social groups" (p. 321). In other words, negative affective responses become associated with groups through a conditioning mechanism ultimately leading to the responses becoming automatically activated as part of the mental representation of the group. This means that a negatively evaluated group—especially one whose negative evaluation is automatically activated as part of its mental representation—is negatively evaluated, regardless of circumstance. As Smith (1993) pointed out, such a view is particularly problematic for the highly nuanced and complex ways in which "model" minority groups, such as Asian Americans, are treated. Under certain conditions, such groups seem to elicit positive evaluation for their perceived characteristics (such as intelligence

BEYOND PREJUDICE 3 and ambition) that are valued by majority groups. Under other circumstances, however, those same qualities appear to threaten majority group dominance, provoking negative evaluations. Struck by the same difficulties, more recent approaches to attitude theory hold that multiple evaluations of objects are possible (Wilson & Hodges, 1992) or have focused on features of an attitude object that allow it to be categorized in multiple ways perhaps eliciting multiple evaluations (Smith, Fazio, & Cejka, 1996). Nevertheless, perhaps because such views undermine one of the classic functions of attitudes (allowing objects to be easily and uniformly evaluated across time and instance), they have not penetrated far into classical attitudinal theory. Attitude objects about which one is ambivalent raise a third problem for traditional models of attitude, perhaps especially in the intergroup domain. Reactions to many social groups appear to reflect ambivalence, whether this refers to the inability to form a clear evaluation or the presence of both positive and negative components of evaluations. Classic theories of ambivalence in prejudice (Katz, 1981; Katz & Glass, 1979; Katz, Wackenhut, & Hass, 1986; Kinder & Sears, 1981; McConahay, 1983; Sears, 1988) have focused more on the presence of incompatibility among attitude components, and this has also been the main focus in attempts to extend attitude theory by focusing on ambivalence (Fabrigar & Petty, 1999). Such approaches stop short, however, of considering the fact that multiple inconsistent evaluations (and not just components of evaluations) to an object might be possible. Recently, however, Cacioppo and Berntson (1994) have argued that evaluation is not a unidimensional construct. Positive and negative evaluations may be quite independent of one another, opening the door to the possibility that both positive and negative evaluations (and not just positive and negative antecedents of evaluation) might adhere to an attitude object.

CHAPTER PREVIEWS Perhaps because these issues seem even more pressing in the intergroup relations domain, intergroup relations researchers have also started pushing beyond theories that associate merely positive and negative evaluations to social groups. These researchers have started to explore the antecedents, nature, and consequences of much more complex, modulated, and differentiated reactions to social groups—both membership groups (ingroups) and groups considered psychological outgroups. These are the approaches we have drawn together in this volume. The theories or programs of research that are being developed contribute to a view of intergroup relations that is more group, situation, and context specific than that allowed by earlier approaches. The authors pursue this view by moving beyond the traditional attitude model of prejudice in important ways, drawing

4

FROM PREJUDICE TO INTERGROUP EMOTIONS

on a broad range of theoretical ancestry to develop models of why, when, and how differentiated reactions to groups arise, and what their consequences might be. One of the common avenues chosen by many intergroup relations researchers for moving beyond the attitude model is a refocusing of interest on emotion as a theoretical base for understanding prejudice. Such a focus is appealing for several reasons. Affect and emotion seem more intuitively to afford explanations of "hot" or virulent forms of discrimination. Affect seems to be more closely associated with behavior than some cognitive antecedents (Esses & Dovidio, 2000; Millar & Tesser, 1986a). And affective aspects of prejudice may apparently linger even after considerable change in its cognitive component. In G. W. AUport s classic aphorism: "Defeated intellectually, prejudice lingers emotionally" (Allport, 1954, p. 311). Perhaps of even greater importance, however, is that prevailing theories of emotion offer a conceptual structure paralleling that of attitude theory. Just as beliefs produce attitudes that produce behavior in traditional attitude models, appraisal theories of emotion suggest that cognitive assessment produces emotions that produce associated action patterns (Ellsworth, 1991; Frijda, 1986; Lazarus, 1991; Roseman, 1984; Scherer, 1988; Smith & Ellsworth, 1985). The conceptual gain from importing emotion theory into the intergroup domain, however, centers on the differentiation and specificity that it allows. Appraisals are context specific, and thus groups in different social contexts, or the same group in different social contexts, may elicit different emotions. These different emotions are different in important ways regardless of whether they share some component of, for instance, negative evaluation. So (negative) anger is quite distinct in antecedents and consequences from (negative) fear, which in turn is distinct from (negative) contempt or disgust. As emotions arise from appraisals of circumstances, multiple, different, and conflicting emotions may be experienced about the same group at different times, in different settings, or even simultaneously. A focus on the feelings or emotions aroused, perhaps transiently, perhaps chronically, by intergroup interactions, rather than the static evaluations inherently associated with particular groups, is a theme common to all contributions to the volume. Conceptualizing prejudice as an emotion almost inevitably entails other theoretical bootstrapping, and two such developments also feature as general streams of thought in most of the contributions to the volume. First, emotion has been conceptualized as an individual process and product. How then can group membership and intergroup interactions be the vehicle for emotional reactions? In fact, the development of social identity (Hogg & Abrams, 1988; Tajfel, 1982) and self-categorization (Turner 1985; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) theories made clear that when group members perceive themselves as interchangeable exemplars of the membership group, those ingroups and ingroup memberships become part of the self. Thus the psycho-

BEYOND PREJUDICE

logical self extends beyond the individual to include the group, so group membership and intergroup interactions become important sources of emotion (Smith, 1993). Several of the chapters in the volume develop this perspective, demonstrating the importance of categorization processes to the generation of emotions, even in situations in which events do not impinge on individuals at all. Second, emotions are inherently functional in that they guide self-regulation. Whereas the explanation of discrimination has always been important in the intergroup domain, adherence to the attitude model of prejudice has led researchers to conceptualize discrimination as action consistent with only either the perceived positive or negative evaluation of the target. Perhaps of most appeal in the conceptual leverage offered by considering prejudice as social emotion is that different emotions are associated with equally specific and differentiated tendencies to act—that is, to self-regulate. Anger is an approach emotion, whereas fear increases avoidance; sadness elicits withdrawal, disgust separation, and so forth. Thus with the self extended into the group, self and social regulatory processes could also be seen as the goal of intergroup emotions and behavior. The regulatory nature of emotional approaches to prejudice also features prominently in many of the contributing chapters. In moving beyond prejudice, researchers have looked to different aspects of the intergroup context as the source of differentiated reactions to the ingroup and to other groups, and we have used these predominant foci to organize the contributions to the volume. The first set of chapters (Chapters 2 through 8) in the volume all locate differentiated emotional reactions to groups in categorization processes and the appraisals that follow from them. These chapters tend also to have in common the demonstration of emotion as arising from group membership and being experienced on behalf of the group, regardless of whether events impinge personally on individuals or not. The first two chapters focus on self-regulatory emotions that can be aroused purely as a result of group membership. In Chapter 2, Bizman and Yinon extend Higgins' (1987, 1989) theory of self-discrepancy by assessing ideal and ought discrepancies from the actual self that result from membership in certain groups. Consistent with studies of discrepancies operating at the personal level, the authors find that actual-ideal group discrepancies uniquely produce dejection-related emotions, whereas actual-ought group discrepancies are linked to agitation-related emotions. Although Bizman and Yinon do not pursue the implications of these different emotional reactions to the ingroup as a basis for prejudice toward outgroups, the implication is clear—self-regulatory motives and processes powerfully affect emotional reactions to the ingroup. The implications of such self-regulatory processes for reactions to both ingroups and outgroups is the focus of Chapter 3, by Shah, Brazy, and Higgins. These authors extend into the intergroup domain the tenets of Higgins' (1997)

5

6

FROM PREJUDICE TO INTERGROUP EMOTIONS

regulatory focus theory, in which the goals of approaching gains (promotion) and avoiding losses (prevention) are seen as quite independent. Satisfaction or frustration of these different goals triggers different affective states even in regards to the same target. The authors demonstrate that individual differences in sensitivity to promotion or prevention goals as well as the manipulated salience of such goals dictate both emotional reactions to and behavior toward other ingroup and outgroup members. A concern with promotion versus prevention also seems to tune intergroup bias: whereas promotion concerns result in a positive focus on the ingroup, prevention concerns seem to result in a negative focus on the outgroup. Thus regulatory processes can affect both emotions experienced about and because of group membership and emotions experienced toward other groups because of that membership. The role of group membership—or the social extension of the self—in dictating emotional reactions to ingroups and outgroups, especially through the process of appraisal, is the common concern of the next three chapters in this set. In common with the first two chapters, these contributions treat categorization into a group as a prerequisite for an emotional experience, but those emotional reactions are also tuned by the particular circumstances of specific intergroup interactions. Categorization processes, for example, play a key antecedent role in Branscombe, Doosje, and McGartys treatment of collective guilt (Chapter 4). These authors draw a distinction between categorization as a group member, necessary in their view for individuals to feel collective guilt, and identification, which they define as a long-term commitment to the group. Whereas other approaches have assumed or found that the higher the identification with the group the more strongly an intergroup emotion might be felt, these authors point out that strong identification with the group as a source of pride inhibits acceptance of collective guilt. Thus the role of categorization and identification in intergroup emotions may be more complex than often assumed. Appraisal processes are also a source of differentiated reactions in this approach, however, because guilt occurs when the ingroup is seen as responsible for violating an important group norm or standard. One of the aspects that make this chapter somewhat unique is the role attributed also to intragroup processes. Specifically, Branscombe and colleagues suggest that collective guilt is an especially likely experience when there is dissent and discussion within the ingroup about the ingroup s behavior. This is a factor worth considering in other approaches as well—what kinds of intragroup processes might make the experience of shared group emotions more versus less likely? Continuing the idea that categorization influences appraisal, which in turn triggers specific emotion, Yzerbyt, Dumont, Gordijn, and Wigboldus (Chapter 5) demonstrate that the identity of interacting groups can profoundly affect appraisals of those interactions and thus of the emotion elicited by those events. Their research makes use of clever categorization manipulations that make tar-

BEYOND PREJUDICE

get groups either appear as an ingroup or an outgroup depending on the available comparison. Their results show that events seen as negatively impacting ingroups are reacted to with anger compared to when those same events impinge upon outgroup members. Thus, emotional reactions to a protagonist outgroup will differ quite markedly depending on whether the outgroup s action impacts the ingroup or not: "prejudice" can be quite specific not to the target group, but to the relationship between the perceiver and the target group. Appraisal processes are also the basis for differentiated reactions toward members of other groups in Chapter 6, by Blascovich, Mendes, and Smith. Drawing on a bio-social-psychological framework, these authors show that identical interactions may be either threatening or challenging depending on whether the resources they demand are appraised as exceeding those available. Interactions with stigmatized or low status outgroups are more likely to engender perceptions of threat (demands exceed resources), whereas identical interactions with ingroup or high status group members engender challenge (resources exceed demands). Thus, reactions to other groups are once again differentiated by the relationship between the interactants, which in turn influences appraisals to drive emotional reactions. One of the unique strengths of the chapter is the authors' assessment of non self-report measures of challenge and threat, and their analysis of differences between non-reactant and self-report measures of emotional response to members of other groups. The next two chapters take a slightly broader focus that encompasses several distinct emotions. In Chapter 7, Devos, Silver, Mackie, and Smith offer an explicit integration of appraisal theories of emotion into the intergroup domain. Building on Smiths (1993) suggestion that prejudice may be better thought of as group-based emotions elicited by intergroup interaction, the authors present a model by which appraisals, emotions, and action are all relevant to the social, rather than the personal, self. When social identity is salient, group members interpret events in terms of whether they hurt or help the ingroup, and these appraisals dictate emotional and behavioral reactions directed at the cause of these events. In this case, "prejudice" is thought of as social emotions, and these emotions are context and appraisal specific. Together with the two preceding chapters, this chapter makes the point that such processes may occur even when events do not have personal impact on the individual—that is, that emotional experience may in fact be an intergroup, as well as an interpersonal, process. Chapter 8, by Leyens, Demoulin, Désert, Vaes, and Philippot, tells a cautionary tale for researchers who focus on emotion in the intergroup context. Most such approaches treat the experience of emotion as arising from appraisals of environmental events; these events trigger emotions about one s own membership groups and emotion toward various outgroups. But of course one key determinant of the emotion felt toward one s own group is the emotions that other group members appear to be expressing: signs of anger in fellow group members following an intergroup interaction signal that anger is an ap-

7

8

FROM PREJUDICE TO 1NTERGROUP EMOTIONS

propriate intergroup response. And one key determinant of the emotion felt toward another group is no doubt the emotion that members of that group seem to be expressing toward your group: signs of anger in the outgroup might appropriately trigger fear, for example. As Chapter 8 makes clear, however, reading and interpreting these emotional signs is far from unproblematic. In fact, people believe they send different signals than they apparently do when trying to convey emotion to ingroup and outgroup members, and different (mainly stereotypic) emotions are attributed to the same neutral faces said to be ingroup or outgroup members. Thus once again, categorization processes play a clear, but complex, role in emotional reactions to both ingroups and outgroups. In a second set of chapters (Chapters 9 through 15), the focus shifts from categorization and appraisal processes as the source of differentiation in intergroup relations to the behavioral and structural relations among groups. The early chapters in this set focus on specific interactions or patterns of interaction between groups as triggering differentiated emotional reactions. Chapter 9, for example, by Dovidio, Esses, Beach, and Gaertner, focuses on the nature of specific interactions between groups as the source of differentiated affective reactions toward groups. Importantly, they find that interactions that generate positive affect (as opposed to positive cognitions or evaluations) are uniquely superior in increasing the desire for further interaction with outgroup members. In many ways, this approach is the most faithful in the volume to a revised attitudebehavior model of prejudice. Nevertheless, it underscores clearly the importance of emotional reactions (as compared to cognitive associations) in determining intergroup behavior. It is also one of very few but important chapters in the volume that focus on positive rather than negative intergroup emotions, confirming the idea that differentiated reactions can predict both positive as well as negative intergroup behavior. Chapter 10, by Kramer and Jost, also focuses on specific patterns of interaction between groups as the source of particular emotional reactions. The authors argue that a particular form of relationship between groups triggers what they term outgroup paranoia—the feeling that the ingroup is being wronged in word or deed by a malevolent outgroup. Such feelings are specifically expected to arise when groups are engaged in a hierarchical trust dilemma. A trust dilemma occurs whenever groups could both gain more and lose more by cooperating than by competing. For example, two research teams pursuing the same issue may solve a puzzle faster by working together. But such a situation also entails the possibility that one team may withhold information while benefiting from the other team s open exchange. Trust dilemmas are made more treacherous, especially for the underdog, when they are hierarchical—that is, when one group has more status, resources, or power than the other. Once again, then, it is the nature of the interaction or structural relation between groups, rather than the nature of the groups themselves, that dictates emotional reactions. The next two chapters continue to focus on the nature of interactions be-

BEYOND PREJUDICE

tween groups as a determinant of emotions. In Chapter 11, Stephan and Renfro locate differentiated reactions to groups in the very different types of threats that outgroup existence, outgroup action, and outgroup interaction can be seen as having for the ingroup. In this sense, integrated threat theory is another example of appraisal theories extended to the intergroup domain. Threat types are differentiated by whether they seem aimed at endangering the existence of the individual or the ingroup (realistic threats at the individual or group level), or at threatening the individuals or ingroup s value, identity, or way of life (symbolic threats at the individual or group level). To the extent the nature of the threat is differentiated in this way, specific emotional and behavioral reactions toward the outgroup will also be differentiated. Reactions toward groups are further differentiated because the degree to which each type of threat is salient in any given context depends at least in part on how the intergroup relations in that context are construed. Chapter 12, by Brewer and Alexander, develops and extends image theory, developed in the political psychology domain, into a more general theory of intergroup relations. By this account, the nature of the relationship between groups—whether dominant or subordinate, competitive or exploitative, threatening or secure, for example—produces appraisals that dictate both "images" of and emotional reactions to outgroups. While the research described in the chapter reinforces the view that specific group relationships can determine specific forms of prejudice, it also demonstrates that stereotypes (images) can also be the consequence rather than the antecedent of these reactions. Thus reactions are differentiated in this approach by the competitive and status relations between groups, but those interactions are also explicitly translated into appraisals that then serve as proximal determinants of emotion. The focus continues to shift to the societal level in Jost and Kramer s (Chapter 13) analysis of the system justification motive in producing differentiated reactions to groups. According to system justification theory, individuals need to accept the status quo, and this need largely dictates one s feelings about one s own and other groups. High status groups are reacted to favorably, even by non-members—their place in the social hierarchy ensures that they are. But members of socially devalued ingroups may feel ambivalence toward their groups, demonstrating again that "prejudice" is not only differentiated by group and context, but can also encompass affectively inconsistent reactions to the one target group. The reality of such intergroup ambivalence—that groups not only can but quite typically do elicit multiple affectively inconsistent reactions—is amply demonstrated in Chapter 14 by Fiske, Cuddy, and Glick. Ambivalence derives from the perception of many groups as both competent but cold, or incompetent but warm. Yet other groups may be seen as both warm and competent or cold and incompetent, eliciting affectively consistent reactions. In this approach, the modulation and nuancing of prejudice depend on the status and competi-

9

10

FROM PREJUDICE TO INTERGROUP EMOTIONS

tion relations among groups: Status predicts competence; competition negatively predicts warmth. And echoing the main point of Chapter 13, the authors note that society's dominant ideology is well served by matching particular kinds of prejudice to particular kinds of groups, grounding these differentiated reactions to groups in broader social processes as well. Finally, Chapter 15 by Neuberg and Cottrell takes an evolutionary perspective to make the point that more than one "type" of prejudice toward outgroups is likely given the naturally selected nature of social life. Given the variety of ways in which other groups contribute to or hinder the effective functioning of the ingroup, affective reactions to particular groups ranging from satisfaction to anger, from resentment to fear, from disgust to ambivalence, as well as specific behavioral strategies associated with those reactions, are to be expected. Given the priority of individual reproduction, these authors argue that emotional reactions to events that impact the individual will take priority over reactions to events that impact the group, and that survival threats will take priority over identity threats. Such a hierarchy of concerns is an interesting point of potential difference between the evolutionary perspective and, for example, the social identity perspective. While concurring with many other chapters that competition over resources and status plays a key role in reactions to outgroups, Neuberg and Cottrell attempt to provide a foundation for why such concerns should have arisen. Thus each chapter represents a contribution to the idea that reactions to groups are more differentiated, and particularly more emotionally differentiated, than anticipated by theories that equate prejudice with evaluation. The breadth of perspectives from which these contributions are derived and the international identity of the contributors speak to a widely seen need to develop new theory and research to advance the study of intergroup relations. We see these contributions as representing the beginnings of such a new movement. The final chapter, by Smith and Mackie, is explicitly theoretical and at times somewhat speculative. The goal of Chapter 16 is to present an overarching conceptual framework that integrates the research described in this volume and to suggest some links to other bodies of theory. We see the chapters in the volume as reflecting a theoretical shift of direction in intergroup relations from the earlier emphasis on knowledge about ingroups and outgroups (i.e., like/dislike for groups, group stereotypes) to an incorporation of group membership within a broadened regulatory framework. In this perspective the physical individual is no longer a boundary of self-regulatory processes, consistent with the idea that the psychological self extends beyond the individual to include fellow group members (social identity theory, self-categorization theory). In discussing all these points, the final chapter will describe the new perspective s similarities and differences from existing models of intergroup relations.

BEYOND PREJUDICE

REFERENCES Allport, G. W. (1946). Controlling group prejudice. Philadelphia: American Academy of Political and Social Science. Allport, G. W. (1954). The nature of prejudice. New York: Anchor Books. Bodenhausen, G. V, Mussweiler, T., Gabriel, S., & Moreno, K. N. (2001). Affective influences on stereotyping and intergroup relations. In J. P. Forgas (Ed.), Handbook of affect and social cognition (pp. 319-343). Mahwah, NJ: Lawrence Erlbaum. Brewer, M. B., & Brown, R. J. (1998). Intergroup relations. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (Vol. 2, 4th éd., pp. 554594). Boston: McGraw-Hill. Brown, R. J., & Gaertner, S. L. (Eds.). (2001). The Blackwell handbook of social psychology: Intergroup processes. Maiden, MA: Blackwell Publishers. Cacioppo, J. T., & Berntson, G. G. (1994). Relationships between attitudes and evaluative space: A critical review, with emphasis on the separability of positive and negative substrates. Psychological Bulletin, 115,401423. Chaiken, S., Pomerantz, E. M., & GinerSorolla, R. (1995). Structural consistency and attitude strength. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences. Hillsdale, NJ: Lawrence Erlbaum. Crosby, E, Bromley, S., & Saxe, L. (1980). Recent unobtrusive studies of Black and White discrimination and prejudice: A literature review. Psychological Bulletin, 87, 546-563. Edwards, K., & von Hippel, W. (1995). Hearts and minds: The priority of affective versus cognitive factors in person perception. Personality and Social Psychology Bulletin, 10, 996-1011. Ellsworth, P. C. (1991). Some implications of cognitive appraisal theories of emotion. In K. T. Strongman (Ed.), International review of studies on emotion (pp. 143-161). New York: Wiley. Esses, V. M., & Dovidio, J. F. (2000, June). The

role of affect and cognition in determining intergroup contact. Paper presented at the bi-annual convention of the Society for the Psychological Study of Social Issues, Minneapolis, MN. Fabrigar, L. R., & Petty, R. E. (1999). The role of the affective and cognitive bases of attitudes in susceptibility to affectively and cognitively based persuasion. Personality and Social Psychology Bulletin, 25(3), 363381. Fiske, S. T. (1998). Stereotyping, prejudice, and discrimination. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (Vol. 2,4th éd., pp. 357411). Boston: McGraw-Hill. Frijda, N. H. (1986). The emotions. Cambridge, England: Cambridge University Press. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340. Higgins, E. T. (1989). Self-discrepancy theory: What patterns of self-beliefs cause people to suffer? In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 22, pp. 93-136). San Diego, CA: Academic Press. Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52,1280-1300. Hogg, M. A., & Abrams, D. (1988). Socialidentifications. London: Routledge. Jones, E. E. (1998). Major developments in five decades of social psychology. In D. T Gilbert & S. T. Fiske (Eds.), The handbook of social psychology (Vol. 2,4th éd., pp. 3-57). New York: McGraw-Hill. Katz, J. (1981). Stigma: A social psychological analysis. Hillsdale, NJ: Lawrence Erlbaum. Katz, I., & Glass, D. C. (1979). An ambivalence-amplification theory of behavior toward the stigmatized. In W. Austin & S. Worchell (Eds.), The social psychology of intergroup relations (pp. 55-84). Monterey, CA: Brooks/Cole. Katz, I., Wackenhut, J., & Hass, R. G. (1986). Racial ambivalence, value duality, and behavior. In J. F. Dovidio & S. L. Gaertner

i t

12

FROM PREJUDICE TO INTERGROUP EMOTIONS

(Eds.), Prejudice, discrimination, and racism (pp. 35-59). Orlando, FL: Academic Press. Kinder, D. R., & Sears, D. O. (1981). Prejudice and politics: Symbolic racism versus racial threats to the good life. Journal of Personality and Social Psychology, 40(3), 414431. Lazarus, R. S. (1991). Progress on a cognitivemotivational-relational theory of emotion. American Psychologist, 46, 819-834. Mackie, D. M., & Smith, E. R. (1998). Intergroup relations: Insights from a theoretically integrative approach. Psychological Review, 105, 499-529. McConahay, J. B. (1983). Modern racism and modern discrimination: The effects of race, racial attitudes, and context on simulated hiring decisions. Personality and Social Psychology Bulletin, 9(4), 551-558. Millar, M. G., & Tesser, A. (1986a). Thoughtinduced attitude change: The effects of schema structure and commitment. Journal of Personality and Social Psychology, 51(2), 259-269. Millar, M. G., & Tesser, A. (1986b). Effects of affective and cognitive focus on the attitudebehavior relation. Journal of Personality and Social Psychology, 51(2), 270-276. Roseman, I. J. (1984). Cognitive determinants of emotion: A structural theory. Review of Personality ¿r Social Psychology, 5, 11-36. Scherer, K. R. (1988). Criteria for emotionantecedent appraisal: A review. In V. Hamilton, G. H. Bower, & N. Frijda (Eds.), Cognitive perspectives on emotion and motivation (pp. 89-126). Dordrecht, The Netherlands: Kluwer Academic Publishers. Sears, D. O. (1988). Symbolic racism. In P. A. Katz & D. A. Taylor (Eds.), Eliminating racism: Profiles in controversy (pp. 53-84). New York: Plenum Press. Sedikides, C , Schopler, J., & Insko, C. A. (Eds.). (1998). Intergroup cognition and intergroup behavior. Mahwah, NJ: Lawrence Erlbaum. Sigall, H., & Page, R. (1971). Current stereotypes: A little fading, a little faking. Journal

of Personality and Social Psychology, 18, 247-255. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and Social Psychology, 48, 813-838. Smith, E. R. (1993). Social identity and social emotions: Toward new conceptualizations of prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 297-315). San Diego, CA: Academic Press. Smith, E. R., Fazio, R. H., & Cejka, M. A. (1996). Accessible attitudes influence categorization of multiply categorizable objects. Journal of Personality and Social Psychology, 71, 888-898. Spencer, S. J., & Fein, S. (1997). Prejudice as self-image maintenance: Affirming the self through derogating others. Journal of Personality and Social Psychology, 73, 31-44. Tajfel, H. (1969). Cognitive aspects of prejudice. Journal of Social Issues, 25, 79-97. Tajfel, H. (1982). Social psychology of intergroup relations. Annual Review of Psychology, 33, 1-39. Turner, J. C. (1985). Social categorization and the self-concept: A social cognitive theory of group behaviour. In E. J. Lawler (Ed.), Advances in group processes: Theory and research (Vol. 2, pp. 77-122). Greenwich, England: JAI Press. Turner, J. C , Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Oxford, England: Blackwell. Wilson, T. D., & Hodges, S. D. (1992). Attitudes as temporary constructions. In L. L. Martin & A. Tesser (Eds.), The constructions of social judgments (pp. 37-65). Zanna, M. P. (1994). On the nature of prejudice. Canadian Psychology, 35, 11-23. Zanna, M. P., & Rempel, J. K. (1988). Attitudes: A new look at an old concept. In D. Bar-Tal & A. W. Kruglanski (Eds.), The social psychology of knowledge (pp. 315-334). New York: Cambridge University Press.

2

Social Self-Discrepancies and Group-Based Emotional Distress

AHARON BIZMAN YOEL YINON

ñ

ecently, in an article published in a highly influential newspaper in Israel, in the context of the responses of Israeli government to the Pales, tinian uprising, a well-known columnist observed that "in Israel today everything is measured in terms of utility or gain and loss. Values such as loyalty and responsibility for people who risked their life on behalf of Israel, are not anymore codes of normative behavior. . . . Israel today is a state without a sense of loyalty, responsibility, obligation and honor" (Harel, 2001). This is an example of an Israeli commentator who points at the wide gap between what he would like his ingroup (the state of Israel) to be and what he observes his ingroup actually is. In other words, the gap described here is between the perceived actual and ideal image of ones ingroup. Although the emotional effects of discrepancies between what individuals perceive themselves to be and what they would like to be have been discussed by social psychologists, to our knowledge no effort has been directed at studying the emotional effects of the corresponding discrepancies associated with one s ingroup. The present chapter is an attempt to extend Higgins' self-discrepancy theory to discrepancies involving the social component of the self.

13

14

FROM PREJUDICE TO INTERGROUP EMOTIONS

SELF-CATEGORIZATION THEORY AND GROUP-BASED EMOTIONS Many theorists agree that the self contains two distinct components: the individual self and the social (or collective) self (Brewer & Gardner, 1996). Social identity theory (Tajfel & Turner, 1986) argues that an important aspect of an individuals self-image is derived from memberships in social groups and social categories. According to self-categorization theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987), self-definition is dependent on the social context, and people can define themselves in personal terms or in terms of particular group memberships. When the social context leads to self-categorization at the personal level, the person defines himself/herself by attributes and characteristics that differentiate one from others. When the social context leads to selfcategorization as a member of a particular social group, the person defines himself/herself in terms of group-typical attributes and characteristics and differentiates his/her group from other groups. In other words, group membership becomes part of the self-concept when people identify themselves as group members. The stronger the group identification, the more the group membership is incorporated into the self-concept. The attributes characterizing the personal aspect of the self may differ from the attributes characterizing the social aspect of the self (Crocker, Luhtanen, Blaine, & Broadnax, 1994). The above conceptualization of personal and social levels of the self implies that the basis for self-evaluation varies according to which level of the self is in focus. Self-evaluation at the personal level is related to one s personal attributes and characteristics relative to those of relevant other individuals, whereas self-evaluation at the group level is related to the status of one s ingroup relative to that of relevant outgroups (Brewer & Gardner, 1996; Luhtanen & Crocker, 1992). Thus for instance, Rosenbergs common measure for personal self-esteem asks people to respond to items such as: "I am able to do things as well as most other people" (Rosenberg, 1965). In contrast, Luhtanen and Crocker's Collective Self-Esteem Scale asks people to respond to items such as: "I often regret that I belong to some of the social groups I do." Depending on its nature, information pertaining to the self may produce positive or negative feelings. As the social component is an integral part of the self, it follows that information relevant to the social self may also evoke emotional responses. Thus, when an individual self-categorizes himself/herself as a member in a particular social group, cognitions about this group may arouse emotional responses based on his/her group membership. That is, individuals may experience not only personal-based emotions but also group-based emotions. Indeed, studies investigating the BIRG (Basking in Reflected Glory) phenomenon have demonstrated that sport fans may experience positive as well as negative group-based emotions (e.g., Cialdini et al., 1976; Hirt, Zillmann, Erickson, & Kennedy, 1992). Thus, in Hirt et al. s (1992) study, team success

SOCIAL SELF-DISCREPANCIES

induced positive mood and elevated self-esteem, whereas team failure induced negative mood and decreased self-esteem, among fans who strongly identify with their team. Furthermore, Sloan (1979) has found that fans experience specific emotions depending on game outcome: anger, discouragement, irritation, frustration, sadness, feeling upset, and hostility increased following team failure, while happiness, satisfaction, and feeling pleased decreased after team loss and increased after a tough victory. Group-based emotions were found not only among sport fans. For instance, Doosje, Branscombe, Spears, and Manstead (1998) found that when Dutch participants read about unfavorable past behavior of their national group (with which they personally were not involved) they experienced group-based guilt feelings. The above mentioned studies indicate that emotional experience can be personal- or group-based. However, not much theoretical work has been directed at describing when and why people may experience different types of groupbased emotions. Given that group membership may become part of the self, one can assume that the psychological mechanisms underlying group-based emotions are the same as those underlying personal-based emotions. Indeed, Smith (1993) integrated appraisal theories of emotions (e.g., Scherer, 1988; Smith & Ellsworth, 1985) with self-categorization theory suggesting that ". . . to the extent a self-categorization functions as a self aspect, appraisal of events or situations with respect to that social aspect of identity will also trigger emotions" (p. 303). However, Smiths main concern was with ingroup members' emotions directed toward the outgroup, such as, fear, disgust, contempt, anger, and jealousy (see also Mackie, Devos, & Smith, 2000). We are interested in the emotions associated with the social self that are not directed toward the outgroup. Specifically, the present chapter is an attempt to employ self-discrepancy theory (Higgins, 1987,1989) in order to predict which types of emotional distress people may experience as a result of their group memberships.

SELF-DISCREPANCY THEORY (SDT)-EXTENDED Self-discrepancy theory delineates the relation between different types of selfstate representations and distinct kinds of emotional distress. The theory identifies three domains of self-representation: (a) the actual self, which includes the traits or characteristics that people believe they actually possess; (b) the ideal self, which includes the attributes that people wish or hope to possess; and (c) the ought self, which contains the attributes that people believe they have the duty, responsibility, or obligation to possess. Each of the three domains of the self can be viewed from ones own standpoint (i.e., the attributes one believes one does possess, the attributes one wishes to possess, and the attributes one believes one should possess) or from the perceived standpoints of one or more important others (i.e., the attributes one assumes one s mother, father, or

15

16

FROM PREJUDICE TO INTERGROUP EMOTIONS

some other individual believes one actually possesses, wishes one to possess, or believes one should possess). The actual self from one's own standpoint is identical to what is typically named as the self-concept. The ideal and the ought self (from own or other perspective) function as self-guides against which individuals compare their actual self. According to the theory, people strive to ensure that their actual self matches their ideal and ought self. The greater the discrepancy between the actual self and a self guide, the greater the psychological discomfort. The major proposition of the theory is that the actual-ideal and the actual-ought self-discrepancies are differentially associated with distinct types of emotional distress. The actual-ideal discrepancy reflects the absence of positive outcomes (resulting from failure to possess desired attributes) and is linked to dejection-related emotions (e.g., disappointment, sadness, dissatisfaction). The actual-ought discrepancy reflects the presence of negative outcomes (e.g., sanctions, resulting from failure to fulfill prescribed duties and obligations) and is linked to agitation-related emotions (e.g., apprehension, nervousness, tension). This proposition has been supported in several studies. The actual-ideal discrepancy has been uniquely associated with depression and dejection, while the actual-ought discrepancy has been uniquely related to anxiety and agitation (e.g., Carver, Lawrence, & Scheier, 1999; Higgins, Klein, & Strauman, 1985). The actual (from own standpoint)-ought (from other standpoint) discrepancy, but not the actual-ideal discrepancy, was found to be related to two measures of social anxiety (Strauman & Higgins, 1988): Fear of Negative Evaluation and Social Avoidance and Distress Scales (Watson & Friend, 1969). In addition, the actual-ideal discrepancy has been found to be negatively related to self-esteem (Moretti & Higgins, 1990). Higgins' self-discrepancy theory does not relate to the distinction between the individual- and group-based aspects of the self. However, review of the studies examining Higgins' theory indicates that the specific procedures and measures used to assess self-discrepancies and emotions engage mainly in the individual component of the self. First, participants are requested to concentrate on the attributes of their individual self (be it actual, ideal, or ought). Thus, they seem to assess discrepancies at the individual-self level. Second, self-discrepancies are related to personal-based emotions and to personal selfesteem. As described above, through the process of group identification, group membership becomes a part of the self. Thus, it can be argued that the notion of discrepancies between domains of self-representations and its motivational and emotional consequences may be applicable to the social-self level as well. At the individual-self level, discrepancies refer to one's personal attributes. At the social-self level, the discrepancies refer to the attributes of one's ingroup. If indeed the psychological mechanism underlying group-based emotional distress is the same as the one underlying personal-based emotional distress, then,

SOCIAL SELF-DISCREPANCIES

when people define themselves in terms of their group memberships, the distinct negative emotional outcomes of the different types of self-discrepancies involving the individual self would be also found with regard to discrepancies involving the social self. That is, different types of discrepancies referring to the attributes of ones ingroup (group discrepancies) would be associated with distinct group-based emotional distress (dejection versus agitation) and to the collective self-esteem. The first step in extending SDT to the social self is, naturally, to take the elements composing the SDT and identify their parallels in the social component of the self. Correspondent to the three domains of self-representation regarding the personal self, one can identify three domains of self-representation regarding the social self: the actual group, which contains the attributes that individuals believe their ingroup members actually possess; the ideal group, which contains the attributes that individuals wish or hope their ingroup members would possess; and the ought group, which is comprised of the attributes that individuals believe their ingroup members have the duty, responsibility, or obligation to possess. Each of these domains of the social self can be viewed from several standpoints: the individuals own perspective on his/her ingroup members (i.e., the attributes the individual believes they do possess, the attributes the individual wishes them to possess, and the attributes the individual believes they should possess), or the perceived standpoint of significant outgroups (i.e., the attributes the individual assumes a given relevant outgroup believes his ingroup members actually possess, wishes them to possess, or believes they should possess). We will return to the issue of the perceived standpoint of a significant outgroup later. The ideal group and the ought group constitute the social-self guides against which individuals compare their actual group. Corresponding to personal self-discrepancies, group-discrepancies can be defined as follows: actual-ideal group—the extent to which perceived actual group members' attributes differ from the attributes someone (self or an outgroup) wishes or hopes group members to possess; and actual-ought group—the extent to which perceived actual group members' attributes differ from the attributes someone (self or an outgroup) believes group members are obligated or have the duty to possess.

ATTRIBUTES CONTAINED IN THE INDIVIDUAL VERSUS THE SOCIAL SELF Do the attributes and characteristics contained in the actual, ideal, and ought social self differ from those comprising the actual, ideal, and ought personal self, respectively? In a recent study (Bizman, Yinon, & Krotman, 2001), we adapted Higgins' Selves Questionnaire (Higgins et al., 1985), asking 118 native Israelis to list three sets of 6 to 10 attributes referring to the actual, ideal, and

17

18

FROM PREJUDICE TO INTERGROUP EMOTIONS

ought Israelis: for the actual Israelis to "think about us as Israelis. In your opinion, which attributes Israelis actually possess"; for the ideal Israelis to "think about the ideal Israeli, that is, the attributes you wish and hope the ideal Israeli would possess"; for the ought Israeli to "think about the ought Israeli. In your opinion, which attributes Israelis should possess, considering Israelis' responsibilities and duties." Following each set of attributes, participants rated the extent to which each attribute is possessed by Israelis (actual group), they wish it would be possessed by the Israelis (ideal group), and they believe it should be possessed by the Israelis (ought group). In addition, one week later they were asked to list 6 to 10 attributes representing their actual, ideal, and ought personal self, and to rate the extent to which they possess each attribute (actual self), wish they would possess each attribute (ideal self), and believe they should possess each attribute (ought self). We compared the lists of attributes, contained in the three domains of personal and social self-representations, considering only attributes mentioned by at least 10% of the respondents. The lists of the attributes appearing in the three personal self-representations were found to be similar to each other and contained mostly positive attributes (e.g., friendliness, involvement, ambitiousness). In contrast, the lists of attributes appearing in the three social self-representations had much less in common. Whereas 39% of the attributes of the actual Israeli seemed to be negative (e.g., impudent, impatient, aggressive, conceited, shallow), only two of the attributes in the actual personal self seemed to be negative (nervous, introverted). For each self-representation (actual, ideal, ought), we compared the attributes appearing in the personal and Israelis lists. For each list we counted the number of overlapping (attributes appearing in both lists), unique (attributes appearing in the Israelis list but not in the personal list, and vice versa), and opposite attributes (attributes appearing in one list and their opposite in the other list). The results of this analysis are presented in Table 2.1. As can be seen from the table, concerning the actual self-representation, 39% of the actual Israeli attributes appeared also in the actual personal self, and 39% of them (e.g., patriotism, comradeship, conceit) were unique to the actual Israeli. Thirty-five percent of the actual personal-self attributes appeared also in the actual Israeli, and an additional 45% of these attributes (e.g., sensitivity, TABLE 2.1. Percentages of Overlapping, Unique, and Opposite Attributes According to Israeli and Personal Self-Representations Actual Attributes Overlapping Unique Opposites Total

Ideal

Ought

Israelis

Personal

Israelis

Personal

Israelis

Personal

39 39 22 100

35 45 20 100

89 11

67 33

82 18

61 39

100

100

100

100

SOCIAL SELF-DISCREPANCIES

responsibility, sense of humor) were unique to the actual personal self. Also, four attributes (22% of the actual Israeli attributes, and 20% of the actual personal-self attributes) were opposites (negative in the actual Israeli and positive in the actual personal self: impudence versus politeness, intolerance versus tolerance, aggressiveness versus moderation, impatience versus patience). Concerning the ideal self-representation, 89% of the ideal Israeli attributes appeared also in the ideal personal self, and only 11% of them (patriotism and comradeship) were unique to the ideal Israeli. Sixty-seven percent of the ideal personal-self attributes appeared also in the ideal Israeli, and 33% of them were unique to the ideal personal self (e.g., courage, sense of humor, responsibility). In regard to the ought self-representation, 82% of the ought Israeli attributes appeared also in the ought personal self, and only 18% of them (patriotism, courage, and comradeship) were unique to the ought Israeli. Sixty-one percent of the ought personal-self attributes appeared also in the ought Israeli, and 39% of them were unique to the ought personal self (e.g., responsibility, sensitivity, optimism). These findings indicate that the attributes and the characteristics contained in the actual, ideal, and ought social self overlap only partially with those comprising the actual, ideal, and ought personal self, respectively. Thus, they suggest that group- and personal self-discrepancies might be distinct. However, it might still be possible that the attributes contributing to group-discrepancies are similar to those contributing to personal self-discrepancies. To address this issue, we first calculated for each participant discrepancies scores, both for the group- and personal self-representations (see Higgins, Bond, Klein, & Strauman, 1986). For the group-discrepancies, the list of attributes of the actual Israeli was compared to the list of attributes of the ideal and the ought Israeli. The number of matches (synonymous attributes appearing in both lists with extent rating that did not differ by more than 1 point), true mismatches (antonymous attributes appearing in both lists), mismatches of extent (synonymous attributes appearing in both lists with extent rating differing by 2 or more points), and nonmatches (attributes appearing only in one list but not in the other) were counted. Actual-ideal and actual-ought Israelis discrepancy scores were calculated by subtracting the number of matches from the total number of mismatches, with true mismatches given a weight of 2 and mismatches of extent, and matches, a weight of 1. The procedure for calculating the actual-ideal and actual-ought personal self-discrepancies scores was similar to that detailed with regard to group-discrepancies. This time, the list of attributes of the actual personal self was compared with the list of attributes of the ideal and ought personal self. Finally we calculated, for each participant, the percentage of attributes contributing to the actual-ideal and actual-ought group-discrepancies that were the same as those contributing to the actual-ideal and actual-ought personal self-discrepancies, respectively. The mean percentage of matching attributes for the actual-ideal group-discrepancy and actual-ideal personal self-discrep-

19

20

FROM PREJUDICE TO INTERGROUP EMOTIONS

ancy pair was 18.08, and for the actual-ought group-discrepancy and actualought personal self-discrepancy pair 16.73. Only among 1.7% of the participants, all the attributes contributing to the actual-ideal group-discrepancy contributed also to the actual-ideal personal self-discrepancy In contrast, among 50% of the participants, none of the attributes contributing to the actual-ideal group-discrepancy contributed also to the actual-ideal personal self-discrepancy. Only among 5.1% of the participants all the attributes contributing to the actual-ought group-discrepancy contributed also to the actual-ought personal self-discrepancy. In contrast, among 62.7% of the participants none of the attributes contributing to the actual-ought group-discrepancy contributed also to the actual-ought personal self-discrepancy. These findings show that the content of group- and personal self-discrepancies is quite different.

ATTRIBUTES OF THE GROUP AS A WHOLE It should be noted that upon applying SDT to the social self, we conceptualized group representations in terms of the attributes and characteristics of the ingroup members. This notion is in line with the self-categorization theory proposition that when a given group membership is salient, people self-stereotype and perceive similarity between themselves and other group members. However, one can conceptualize group representations in terms of the attributes of the group as an entity, which may be distinct from those characterizing the group members. In our case, we asked the participants to list the attributes of actual, ideal, and ought Israelis. However, it might be that asking them to indicate the actual, ideal, and ought attributes of the state of Israel would probably have yielded a different list of attributes. For example, people may perceive their ingroup as, wish their ingroup to be, or believe that their ingroup should be democratic, capitalistic, religious, cohesive, strong, and the like. Thus, recently the former chief of the Israeli Internal Security Agency, when addressing a group of governmental economists, raised the issue of "what kind of Israel do we want? If we want it to be Jewish and democratic it cannot employ a policy of economic apartheid" ("Ayalon Observation," 2000). In the same vein, one can conceive of the controversy in Israel regarding the state-religion relationship as an example of a conflict around the ideal and ought group. The religious people abide by biblical commands and therefore would like Israel to adopt religious laws (ideal group). They also believe that Israel ought to adopt religious laws (ought group). In contrast, the secular people would like Israel to adopt modern laws (ideal group) and also believe that Israel ought to adopt modern laws (ought group). The actual-ideal and actual-ought group-discrepancies based on the attributes of the ingroup as a whole may not necessarily be correlated with actual-ideal and actual-ought group-discrepancies based on the attributes of the ingroup members. It seems to us that both types of group-discrepancies maybe

SOCIAL SELF-DISCREPANCIES

associated with group-based psychological distress, depending on which type of group-discrepancies is more salient.

GROUP-DISCREPANCIES AND GROUP-BASED AGITATION AND DEJECTION Our central proposition claims that when people are operating on the basis of their group membership, the actual-ideal discrepancy would be uniquely related to group-based dejection-related emotions, but not to group-based agitation-related emotions. In contrast, the actual-ought discrepancy would be uniquely related to group-based agitation-related emotions, but not to groupbased dejection-related emotions. In the above-mentioned study, group-based emotions were assessed by asking participants: "As an Israeli, when you think about the Israeli group, to what extent do you experience each one of the following emotions." Six emotions related to dejection (four negative—sadness, disappointment, dissatisfaction, hopelessness, and two positive—happiness, joy), and seven emotions related to agitation (five negative—apprehension, threatened, uneasiness, tension, nervousness, and two positive—relaxation, calmness). These emotions were employed in previous SDT studies (Alexander & Higgins, 1993; Higgins, 1987; Higgins et al., 1985; Higgins, Shah, & Friedman, 1997; Roese, Hur, & Pennington, 1999) and seemed to us as frequently indicated by Israelis when thinking about their fellow countrymen. To examine the association between group-discrepancies and group-based psychological distress, we adopted procedures previously used in self-discrepancy research. First, since group-based dejection- and agitation-related emotions measures were highly correlated (r = .61, p < .001), we computed residual scores by regressing participants' dejection scores on agitation scores, and vice versa. Next, partial correlations were computed between the two types of group discrepancies and the residualized scores of the two measures of group-based emotions. Consistent with expectations, after controlling for actual-ought discrepancy, actual-ideal discrepancy was significantly correlated with group-based dejection-related emotions (pr = .22, p < .05) but not to group-based agitationrelated emotions (pr = -.08). In contrast, after controlling for actual-ideal discrepancy, actual-ought discrepancy was significantly correlated with group-based agitation-related emotions (pr = .23, p < .05), but not to group dejection-related emotions (pr = .04). As group- and personal self-discrepancies were found to be moderately correlated (r = .37, between actual-ideal group- and personal-discrepancies; r = .26, between actual-ought group- and personal-discrepancies), one can ask whether group-related discrepancies are related to group-based psychological discomfort over and above the contribution of personal self-discrepancies. As

21

22

FROM PREJUDICE TO INTERGROUP EMOTIONS

an answer, we computed partial correlations between the two types of groupdiscrepancies and the two measures of group-based emotions, controlling also for personal self-discrepancies. The findings indicated that the pattern of associations between group-discrepancies and group-based emotions was not changed. Each of the six dejection- and seven agitation-related emotions was similarly analyzed. It was found that after controlling for actual-ought discrepancy, actual-ideal discrepancy was significantly correlated with five out of the six groupbased dejection-related emotions (sadness, disappointment, dissatisfaction, hopelessness, happiness, and joy) but with none of the seven group-based agitation-related emotions. In contrast, after controlling for actual-ideal discrepancy, actual-ought discrepancy was significantly correlated with all seven group-based agitation-related emotions (apprehension, threatened, uneasiness, tension, nervousness, relaxation, and calmness). Actual-ought discrepancy was significantly associated with only three out of the six group-based dejectionrelated emotions (sadness, dissatisfaction, hopelessness) after controlling for actual-ideal discrepancy. The pattern of the relationship we found between group-discrepancies and group-based emotional distress is similar to those found by Higgins (1987, 1989) in regard to discrepancies and negative emotional experience at the personal level. As such, the findings suggest that similar psychological processes underlie the various types of emotional discomfort that a person may experience as an individual and as a result of being a group member. At the individual level, the distinct experiences of dejection- and agitation-related emotions are aroused when one perceives that his/her actual self does not match his/her ideal or ought self, respectively. Similarly, at the social-self level, these distinct types of negative emotions seem to be aroused when individuals perceive that their ingroup s attributes do not match those they wish or believe their ingroup ought to possess, respectively. Moreover, group-discrepancies associations with groupbased emotions remained significant even after controlling for personal selfdiscrepancies. In addition, the attributes associated with group-discrepancies were found to be different from those associated with personal self-discrepancies. Taken together, these findings indicate that group membership may produce negative emotional experiences although the individual may contribute directly very little to the group discrepancies. Group- and personal self-discrepancies were found, in our study, to be only moderately correlated, indicating that one can have large personal selfdiscrepancies and concurrently small social self-discrepancies, and vice versa. Similarly, we can expect that the magnitude of discrepancies related to one s particular social identity is not necessarily correlated with the magnitude of discrepancies involving ones other social identities. The consequent experienced emotional distress would probably be a function of the relative salience of ones various social identities, as well as the salience of the personal identity.

SOCIAL SELF-DISCREPANCIES

Recently, Mussweiler, Gabriel, and Bodenhausen (2000) demonstrated that people may emphasize or deemphasize their social identities in order to protect themselves from unpleasant outcomes of interpersonal comparison. These findings suggest that people may cope with emotional distress associated with selfdiscrepancies related to their being members of a particular group by defining themselves either as a part of some other group, associated with a small magnitude of discrepancies, or in individual terms, when the magnitude of personal self-discrepancies is small. Thus, for example, if thinking of themselves as a part of their national group evokes emotional distress among Israelis (due to either actual-ideal or actual-ought group-discrepancies), they may actively shift their focus to a different group-based identity (e.g., gender, religious). Having demonstrated that emotional distress can also be group-based has a methodological implication for studies relating types of psychological distress to self-discrepancies. Some of these studies measure personal self-discrepancies at time 1 and ask participants to indicate the emotions they experienced during the past week, at time 2, without referring to whether the emotions were personal- or group-based (e.g., Carver et al., 1999). To the extent that the emotions experienced were group-based, they would not be associated with personal self-discrepancies but rather with group-discrepancies. Thus, measuring discrepancies both in the social and personal self may provide a more complete picture concerning the relation between types of emotional distress and selfdiscrepancies. Alternatively, one can develop methods of measuring the same emotions experienced as originating from the personal and the social components of the self, separately. Considering both personal- and group-based emotions is especially needed in periods when people tend to think of themselves as group members, such as when the ingroup is threatened. For example, in Israel, during periods of unrest among the Palestinians, asking Israelis to report about their current emotional experience, one may get an answer such as "personally I am fine, but I am upset because our government has a too soft policy toward the Palestinians/'

GROUP-DISCREPANCIES AND COLLECTIVE SELF-ESTEEM Reviewing frequently used measures of self-esteem (e.g., Rosenberg, 1965), Moretti and Higgins (1990) concluded that these measures are based on models looking at the self-esteem as reflecting the perceived discrepancy between one's actual and desired characteristics. Accordingly, they hypothesized and found the actual-ideal discrepancy (but not the actual-ought discrepancy) to be negatively related to self-esteem. Extrapolating from the personal level to the social level of the self, we expected actual-ideal group-discrepancies to be related to collective self-esteem. Luhtanen and Crocker (1992) construed the collective

23

24

FROM PREJUDICE TO INTERGROUP EMOTIONS

self-esteem as containing four components: private CSE (the individuals' own evaluation of their social group); public CSE (the individuals' judgment of how other people evaluate their social group); membership CSE (self-evaluation of the extent to which one is a good member of one's social group); and importance to identity (perceived importance of one's group membership to one's self-concept). The private component of CSE is considered to be the grouplevel equivalent of personal self-esteem (Crocker & Luhtanen, 1990). Hence, we expected only the actual-ideal group discrepancy to be uniquely related to the private CSE. In the aforementioned study, the four subscales of Collective Self-Esteem Scale (Luhtanen & Crocker, 1992) were adapted to assess CSE with regard to the participants' being Israelis: (a) private CSE assessed participants' evaluation of the Israeli group (e.g., "I often regret that I am an Israeli"); (b) public CSE assessed participants' judgment of how positively other people evaluate the Israeli group (e.g., "In general, other groups respect the Israeli group"); (c) membership esteem assessed participants' judgment of how worthy each is as a member of the Israeli group (e.g., "I am a worthy member of the Israeli group"); and (d) importance to identity assessed the importance of being an Israeli to each participant's self-concept (e.g., "Being a native Israeli is an important part of my self-image"). We computed partial correlations between the two types of group-discrepancies and the four collective self-esteem subscales. Consistent with expectation, after controlling for actual-ought discrepancy, actual-ideal discrepancy was significantly related to the private subscale (pr = -.29, p < .01), but not to the public and importance to identity subscales. In contrast, after controlling for actual-ideal discrepancy, actual-ought discrepancy was not significantly related to any of the subscales. Unexpectedly, actual-ideal discrepancy was also uniquely related to the membership subscale. To examine whether group-related discrepancies are related to CSE over and above the contribution of personal self-discrepancies, we computed partial correlations between the two types of group-discrepancies and the four CSE subscales controlling also for personal self-discrepancies. The findings indicated that the pattern of associations between group-related discrepancies and the CSE subscales was not changed. Thus, it seems that the relations between groupdiscrepancies and CSE cannot be explained only by their association with personal self-discrepancies. Moretti and Higgins' (1990) finding indicating that only the actual-ideal self-discrepancy is negatively related to self-esteem was interpreted as suggesting that the commonly used measures of personal self-esteem (e.g., Rosenberg, 1965) reflect mainly the dejection-related rather than the agitation-related outcomes of low self-esteem. In the same vein, our findings suggest that the private subscale, as a group-level equivalent of personal self-esteem (Crocker & Luhtanen, 1990), captures the dejection-related, rather than the agitation-

SOCIAL SELF-DISCREPANCIES

related, outcomes of ones low evaluation of ones social group. In line with this suggestion, the correlation between the private CSE with dejection-related emotions was higher than with agitation-related emotions (r = -.70 and r = -.36, respectively, t [115] = 5.82, p < .001).

GROUP-DISCREPANCIES AND FEAR OF ONE'S INGROUP BEING NEGATIVELY EVALUATED One component of social anxiety was defined by Watson & Friend (1969) as "apprehension about others' evaluations, distress over their negative evaluations, avoidance of evaluative situations, and the expectations that others would evaluate oneself negatively" (p. 449). They also developed a scale measuring the Fear of Negative Evaluation (FNE). Strauman & Higgins (1988) found the actual (from own standpoint)-ought (from other standpoint) discrepancy, but not the actual-ideal discrepancy, to be related to FNE. We thought that the equivalent of fear of negative evaluation at the group level may be construed as one s fear that one s ingroup members would be evaluated negatively by other groups (FNEG). Hence, we adapted 11 items from Watson and Friends FNE scale to measure participants' fear that Israelis would be evaluated negatively by other groups (e.g., "I am concerned with what other groups are thinking about the Israelis"), adding the item "When Israelis travel abroad, I worry about the impression they make on others." Our notion of FNEG is somewhat similar to the concept of stereotype threat described by Steele (1997) as the threat that others may employ a negative stereotype about one's group in order to interpret one's behavior. However, from our point of view, when group membership is salient, the threat may arise even if the negative stereotype may be utilized as an interpretation of the behavior of other ingroup members rather than one's own behavior. Moving from the personal level to the social level of the self, we expected actual-ought discrepancy, but not the actual-ideal discrepancy, to be uniquely related to FNEG. Using the above-mentioned statistical procedures, we indeed found that after controlling for actual-ideal discrepancy, actual-ought discrepancy was significantly correlated with FNEG (pr = .25, p < .01). In contrast, after controlling for actual-ought discrepancy, actual-ideal discrepancy was not related to FNEG (pr = .09, p < .01). Additional analyses revealed that the latter findings remained intact after controlling also for personal self-discrepancies. Again, this suggested that the relations between group-discrepancies and FNEG cannot be explained only by their association with personal self-discrepancies. According to SDT (Strauman & Higgins, 1988), FNEG is expected to be uniquely predicted by actual-ought discrepancies from the other standpoint rather than the own standpoint. Yet, we found it to be uniquely related to actual-ought discrepancies from the own standpoint. This finding may be due to

25

26

FROM PREJUDICE TO INTERGROUP EMOTIONS

the correlation between discrepancies from the other and own standpoints (Higgins et al., 1985). Thus, it still may be that the actual-ought discrepancies from the other perspective only, but not from the own perspective, would be uniquely related to FNEG.

WHO IS THE SIGNIFICANT "OTHER"? In SDT studies at the personal level, the participant is requested to choose among specific significant others (e.g., father, mother, best friend), but when moving to group-discrepancies (national groups in particular), it is unclear who the significant "other" might be. It seems that there is no equivalent figure to "parents" when turning to the group level. However, a significant friendly outgroup may be considered as an equivalent to a friend at the personal level, for example, an outgroup whose attitudes toward one's ingroup are important. Specifically, for Israelis, national groups such as the United States or other Western countries, whose attitudes toward Israel and Israelis are probably important to most Israelis, may serve as significant outgroups. The significance of a friendly outgroup may be based on a process of identification, similar to the same process at the interpersonal level (Kelman, 1958). In addition, an outgroup may become significant when one perceives the ingroup as dependent on the outgroup for positive outcomes, such as in the case of a minority group being dependent on the majority group. For example, immigrants may perceive the absorbing group as a significant outgroup. In an ongoing study we asked young Israelis to name an important state friendly toward Israel, and to list the actual, ideal, and ought attributes of the Israelis from that state standpoint, as well as from their own standpoint. Participants found it easy to carry out this task (incidentally, most of the participants indicated the United States). Comparing the lists of attributes appearing under the ideal and ought Israelis from own standpoint to those appearing under the ideal and ought Israelis from the friendly state standpoint, respectively, showed that the pairs of lists of attributes were mostly different. The mean percentage of matching attributes for the actualideal Israelis from own and the friendly state standpoints pair was about 39%, and the mean percentage of matching attributes for the actual-ought Israelis from own and the friendly state standpoints pair was about 43%. At the group level one may also consider a hostile outgroup to be a relevant significant other. That is, perceiving which attributes a hostile outgroup wishes one's ingroup to possess may serve as a social self-representation one would like his/her group to avoid. For example, if your enemy would like your ingroup to be weak, it may drive you to believe that your ingroup should be stronger than it is now. A similar concept, at the personal level, was developed by Markus and Nurius (1986) in their model of possible selves. There, the feared self is defined as the qualities one does not want to become but fears that he

SOCIAL SELF-DISCREPANCIES

would become. Thus, in the case of a hostile outgroup, it is expected that the larger the discrepancy between the actual group and the "feared" group, the less the psychological distress.

CONCLUDING REMARKS Recently, Higgins (1998) developed a Regulatory Focus Theory in which a distinction is proposed between promotion and prevention focus. A person whose focus is on promotion is oriented toward accomplishments, hopes, and aspirations. A person whose focus is on prevention is oriented toward safety, responsibilities, and obligations. The concepts of ideal and ought self-guides are associated with the promotion and the prevention focuses, respectively. That is, the focus is on promotion when people regulate their self toward ideals. The focus is on prevention when people regulate themselves toward oughts. Accordingly, the strength of the ideal self-guide reflects the strength of the promotion focus, and the strength of the ought self-guide reflects the strength of the prevention focus. Based on the suggestion that the relation between goal attainment and emotional responses is moderated by goal strength (Clore, 1994), Higgins et al. (1997) hypothesized and found that the unique association between actualideal personal self-discrepancy and dejection-related emotions became stronger as the strength of the promotion focus (conceptualized as ideal self-guide accessibility) increased. Similarly, the unique relation between actual-ought personal self-discrepancy and dejection-related emotions became stronger as the strength of the prevention focus (i.e., ought self-guide accessibility) increased. Extrapolating from the personal level to the social level of the self, one can expect that the ideal and ought group strength would have the same moderating roles. In other words, the effect of actual-ideal group-discrepancies on groupbased dejection-related emotions would be more pronounced among persons who are mainly concerned with their group accomplishments and aspirations. Similarly, the effect of actual-ought group-discrepancies on group-based agitation-related emotions would be more pronounced among persons who are mainly concerned with their group duties and obligations. A related interesting question concerns the extent to which chronic individual differences in the strength of promotion and prevention focuses at the personal level of the self are correlated with those at the social level of the self. It is not inconceivable to think that a person may be concerned about his/her personal accomplishments, hopes, and aspirations (ideal personal self), and simultaneously concerned about his/her ingroup responsibilities and obligations (ought group), and vice versa. Based on Shah and Higgins' (1998) study, Higgins (1998) suggested that the emotional responses toward objects and events of individuals with a promotion focus will be along the dejection/cheerfulness di-

27

2 8 FROM PREJUDICE TO INTERGROUP EMOTIONS mension. In contrast, the emotional responses toward objects and events of individuals with a prevention focus will be along the agitation/quiescence dimension. Correspondingly, for persons who are mainly concerned about their ingroup responsibilities and obligations, being a group member will arouse emotional experiences along the agitation/quiescence dimension. Whereas for persons who are mainly concerned about their ingroup accomplishments and hopes, being a group member will arouse emotional experiences along the dejection/cheerfulness dimension. If indeed the strength of the regulatory focuses at the individual level of the self are not highly correlated with their parallels at the social level of the self, then people may experience different types of emotions (dejection/cheerfulness versus agitation/quiescence) as individuals and as group members. Based on the proposition that salient group memberships are embedded in the self, the present chapter adapted SDT concepts to delineate the relation between social-self discrepancies and group-based emotional distress. The findings presented in this work demonstrate that the distinct negative emotional outcomes of different types of self-discrepancies involving the individual self appear also in regard to discrepancies involving the social self. The actual-ideal and actual-ought discrepancies were found to be related (in the pattern predicted) to group-based dejection/agitation-related emotions, collective self-esteem, and fear that one s ingroup would be negatively evaluated. According to self-categorization theory, when people categorize themselves as members of a particular social group, they define themselves in terms of group-prototypical attributes. The present work suggests that this self-definition pertains not only to one's actual attributes, but to one's ideal and ought attributes as well. From a broader perspective, the present work suggests that concepts appearing in self-regulation models maybe extrapolated to the social aspect of the self. For example, adopting SDT we focused on the emotional outcomes of group-discrepancies. However, Carver and Scheier (1998) proposed that discrepancy per se is not the critical variable affecting emotional outcomes, but rather the sensed rate of discrepancy reduction. From this point of view, perceiving one's actual group as discrepant from the ideal or ought group but moving toward it at a more or less than adequate rate, would produce positive or negative affect, respectively. In conclusion, extending models describing selfregulation processes related to the personal self to those involving the social self may help us to better understand how behavior is regulated when people define themselves in terms of their group memberships.

REFERENCES Alexander, M. J., & Higgins, E. T. (1993). Emotional trade-offs of becoming a parent: How social roles influence self-discrepancy effects. Journal of Personality and Social

Psychology, 65, 1259-1269. Ayalon observation. (2000, December 6). Haaretz, p. Bl. Bizman, A., Yinon, Y., & Krotman, S. (2001).

SOCIAL SELF-DISCREPANCIES

Group-based emotional distress: An extension of self-discrepancy theory. Personality and Social Psychology Bulletin, 27, 1291— 1300. Brewer, M. B., & Gardner, W. (1996). Who is this "we"? Levels of collective identity and self representations. Journal of Personality and Social Psychology, 71, 83-93. Carver, C. S., Lawrence, J. W., & Scheier, M. F. (1999). Self-discrepancies and affect: Incorporating the role of feared selves. Personality and Social Psychology Bulletin, 25,783792. Carver, C. S., & Scheier, M. F. (1998). On the self-regulation of behavior. Cambridge, England: Cambridge University Press. Cialdini, R. B., Borden, R. J., Thorne, A., Walker, M. R., Freeman, S., & Sloan, L. R. (1976). Basking in reflected glory: Three (football) field studies. Journal of Personality and Social Psychology, 34, 366-375. Clore, G. L. (1994). Why emotions vary in intensity. In P. Ekman & R. J. Davidson (Eds.), The nature of emotion: Fundamental questions (pp. 386-393). New York: Oxford University Press. Crocker, J., & Luhtanen, R. (1990). Collective self-esteem and ingroup bias. Journal of Personality and Social Psychology, 58, 60-67. Crocker, J., Luhtanen, R., Blaine, B., & Broadnax, S. (1994). Collective self-esteem and psychological well-being among White, Black, and Asian college students. Personality and Social Psychology Bulletin, 20, 503513. Doosje, B., Branscombe, N. R., Spears, R., & Manstead, A. S. R. (1998). Guilty by association: When one's group has a negative history. Journal of Personality and Social Psychology, 75, 872-886. Harel, I. (2001, January 18). Without loyalty, without honor. Haaretz, p. Bl. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340. Higgins, E. T. (1989). Self-discrepancy theory: What patterns of self-beliefs cause people to suffer? In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 22, pp. 93-136). San Diego, CA: Academic Press. Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational

principle. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp. 1-46). New York: Academic Press. Higgins, E. T., Bond, R. N., Klein, R., & Strauman, T. (1986). Self-discrepancies and emotional vulnerability: How magnitude, accessibility and type of discrepancy influence affect. Journal of Personality and Social Psychology, 51, 1-15. Higgins, E. T., Klein, R., & Strauman, T. (1985). Self-concept discrepancy theory: A psychological model for distinguishing among different aspects of depression and anxiety. Social Cognition, 3, 51-76. Higgins, E. T., Shah, J., & Friedman, R. (1997). Emotional responses to goal attainment: Strength of regulatory focus as moderator. Journal of Personality and Social Psychology, 72, 515-525. Hirt, E. R., Zillmann, D., Erickson, G. A., & Kennedy, C. (1992). Cost and benefits of allegiance: Changes in fans' self-ascribed competencies after team victory versus defeat. Journal of Personality and Social Psychology, 63, 724-738. Kelman, H. C. (1958). Compliance, identification and internalization: Three processes of attitude change. Journal of Conflict Resolution, 2, 51-60. Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: Self-evaluation of one's social identity. Personality and Social Psychology Bulletin, 18, 302-318. Mackie, D. M., Devos, T., & Smith, E. R. (2000). Intergroup emotions: Explaining offensive action tendencies in an intergroup context. Journal of Personality and Social Psychology, 79, 602-616. Markus, H., & Nurius, P. (1986). Possible selves. American Psychologist, 41, 954969. Moretti, M. M., & Higgins, E. T. (1990). Relating self-discrepancy to self-esteem: The contribution of discrepancy beyond actualself rating. Journal of Experimental Social Psychology, 26, 108-123. Mussweiler, T., Gabriel, S., & Bodenhausen, G. V (2000). Shifting social identities as a strategy for deflecting threatening social comparisons. Journal of Personality and Social Psychology, 79, 398-409. Roese, N. J., Hur, T, & Pennington, G. L. (1999). Counterfactual thinking and regu-

29

30

FROM PREJUDICE TO INTERGROUP EMOTIONS

larity focus: Implications for action versus inaction and sufficiency versus necessity. Journal of Personality and Social Psychology, 77, 1109-1120. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press. Scherer, K. R. (1988). Cognitive antecedents of emotion. In V. Hamilton, G. H. Bower, & N. H. Frijda (Eds.), Cognitive perspectives on emotion and motivation (pp. 89-126). Dordrecht, The Netherlands: Kluwer Academic Publishers. Shah, J., & Higgins, E. T. (1998). Emotional evaluations of self and other attitude objects: Distinct sensitivities from regulatory focus. Unpublished manuscript, Columbia University, New York. Sloan, L. R. (1979). The function and impact of sports for fans: A review of theory and contemporary research. In J. H. Goldstein (Ed.), Sports, games, and play (pp. 219-262). Hillsdale, NJ: Lawrence Erlbaum. Smith, E. R. (1993). Social identity and social emotions: Toward new conceptualizations of prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group

perception (pp. 297-315). San Diego, CA: Academic Press. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and Social Psychology, 48, 813-838. Steele, C. N. (1997). A threat in the air: How stereotypes shape intellectual identity and performance. American Psychologist, 52, 613-629. Strauman, T. J., & Higgins, E. T. (1988). Selfdiscrepancies as predictors of vulnerability to distinct syndromes of chronic emotional disease. Journal of Personality, 56,685-707. Tajfel, H., & Turner, J. C. (1986). Social identity theory of intergroup conflict. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (pp. 7-24). Chicago: Nelson-Hall. Turner, J. C , Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Oxford, England: Basil Blackwell. Watson, D., & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33, 448457.

3

Promotion and Prevention Forms of Ingroup Bias JAMES Y. SHAH PAIGE C. BRAZY E. TORY HIGGINS

O

ur propensity to favor those groups to which we belong has long been a central concern of social psychology (Sumner, 1906). Such concern is not without justification, as ingroup bias has proven to be as pervasive as it is varied. Indeed, ingroup bias has been observed not only in enduring real-world social groups that carry profound significance for their members (e.g., the various ethnic or national groups), but also in the more ephemeral and relatively insignificant groups that constantly "form" and "disband" in our everyday lives. (Allen & Wilder, 1975; Tajfel, Billig, Bundy, & Flament, 1971). Given the pervasiveness and significance of this bias, it is not surprising that considerable research efforts have been directed toward understanding its antecedents and consequences. Traditionally, this bias has been conceptualized as general liking for one group over another, and this approach has certainly provided important insights into the characteristics of this bias and to the motivations that drive it (for a review, see Brewer & Brown, 1998; Fiske, 1998). Yet, as recently noted by Mackie, Devos, and Smith (2000), such a conceptualization fails to consider the diverse nature of this phenomenon. To echo questions raised by these researchers, why do some outgroups invoke specific emotions, such as fear, while others simply "bring us down"? When do we show such bias by attending to our own groups, and when do we actually avoid outgroups? The purpose of the present chapter is to build on past research suggesting that the key to answering such questions lies in the consideration of the regulatory and affective needs that may be fulfilled through ingroup bias. After briefly reviewing different ways in which ingroup bias may be manifested evaluatively,

31

32

FROM PREJUDICE TO INTERGROUP EMOTIONS

emotionally, and behaviorally, we propose that a consideration of the fundamental regulatory needs that groups may fulfill would lend insight into how such bias is manifested in thought, action, and feeling.

MANIFESTATIONS OF INTERGROUP BIAS Evaluative Manifestations Early functional approaches to understanding ingroup bias focused on the evaluative component of bias and assumed that positive ingroup evaluation was inextricably linked to a negative view of outgroups, since often they were the "common products of the same situation" (see Sumner, 1906). Indeed, many influential theories of intergroup relations, most notably Social Identity Theory (SIT; Tajfel & Turner, 1979) and Realistic Conflict Theory (RCT; LeVine & Campbell, 1972; Sherif & Sherif, 1953), have, for differing reasons, assumed that positive evaluations of ingroups and outgroups is a "zero-sum" phenomenon in which increases in one inevitably decreases the other. RCT, for instance, proposes that ingroup bias results from hostility in response to a competitive (threatening) outgroup. SIT theory, on the other hand, would also suggest that this positive relation results from the fact that increasing the positivity of ingroups and the negativity of outgroups both serve to enhance or maintain self-esteem. Allport (1954), however, was critical of this underlying assumption for failing to consider the "psychological primacy" of the ingroup and that positive feelings toward this group may often precede the development of any outgroup negativity. In his view, then, liking the ingroup need not necessarily relate to disliking the outgroup because independent factors influence these different evaluations. Indeed, a number of laboratory and field studies have demonstrated that ingroup positivity need not relate to outgroup negativity (for review see Brewer, 1999). For instance, Brewer and Campbell (1976) found that while 30 different ethnic groups in East Africa all showed evidence of ingroup identification, there was no significant relation of such identification to outgroup distancing. Of course, under some circumstances these two evaluations maybe linked. A more recent study by Duckett and Mphuthing (1998), examining group identification and intergroup attitudes, did find that attitudes toward outgroups influenced ingroup identification, but only when these outgroups posed a threat. Thus, the relation of ingroup liking and outgroup derogation varied as a function of the perceived threat of the outgroup. One implication of this dissociation between ingroup positivity and outgroup negativity is that different types of group evaluations may be made in relative independence from each other. This, in turn, suggests that ingroup bias may be defined separately in terms of the different ways in which groups can be evaluated. A similar argument has been made with respect to positive and negative

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

evaluations of attitude objects more generally (see Cacioppo, Gardner, & Berntson, 1997).

Emotional and Behavioral Manifestations Just as groups may be evaluated differently, so might they be experienced and acted on differently. Mackie et al. (2000) have recently sought to go beyond a consideration of ingroup bias simply in terms of the positivity and negativity of group evaluations, and have attempted to specify a wider range of emotional and behavioral reactions one may have toward ingroups and outgroups. They suggest that, similar to appraisal theories of emotion, models of ingroup bias should specify both the emotional reactions one can have toward ingroups and outgroups, and the behaviors that can arise from, or coincide with, these feelings (see also Mackie & Smith, 1998). Indeed, by manipulating the salience and social support of individuals' ingroups, Mackie et al. (2000) were able to demonstrate that intergroup anger can be distinguished from intergroup fear, and that individuals' tendency to approach (and act against) an outgroup can be distinguished from their tendency to avoid this group. Further support for this perspective comes from research on behavioral discrimination in the minimal group paradigm, which has consistently found that the allocations of rewards to ingroup and outgroup members does not correspond to the allocation of costs or penalties (see Tajfel et al., 1971). Thus, although studies examining the allocation of rewards to minimally defined ingroup and outgroup members typically find that participants favor ingroups in allocating rewards, this effect was significantly decreased when distributing penalties. Hewstone, Fincham, and Jaspars (1981), for instance, found significantly less ingroup bias when they altered the standard paradigm by asking participants to subtract reward from ingroup and outgroup members who had previously been given an initial sum. Similarly, Mummendey et al. (1992) found no evidence of ingroup bias when participants were asked to administer unpleasant high-pitched tones to ingroup and outgroup members, again suggesting the relative independence of these two forms of behavioral bias. Given that ingroup bias may involve different actions and emotions that do not necessarily correspond with each other, how might these various manifestations of bias be understood? In addressing this question, researchers have examined the motivations underlying such bias. Thus, before proposing our own self-regulatory perspective on ingroup bias, let us first briefly review other such perspectives.

MOTIVATIONAL DETERMINANTS OF INGROUP BIAS Social identity theory (Tajfel & Turner, 1986) has proposed a very influential self-regulatory perspective on ingroup bias, in assuming that group member-

33

34

FROM PREJUDICE TO INTERGROUP EMOTIONS

ship represents an important component of self-identity, and thus constitutes a significant source of self-esteem. Individuals can fulfill their esteem needs by increasing the status of groups to which they belong relative to various outgroups. Individuals, then, may favor ingroups over outgroups in order to boost, by association or "reflected glory," their own personal status and self-esteem. One implication of this widely examined self-enhancement hypothesis is that individuals with a particularly strong need for self-esteem enhancement will manifest a correspondingly elevated degree of ingroup bias. Consistent with the self-enhancement hypothesis, Fein and Spencer (1997) found that in those individuals whose self-esteem had been threatened by negative feedback, derogating a stereotyped target led to an increase in self-esteem. Abrams and Hogg (1988) described two studies revealing a significant negative correlation between personal self-esteem and ingroup bias, and Finchilescu (1986) found that groups experimentally assigned low status demonstrated greater ingroup bias. Not all evidence, however, has proven consistent with this theory. Crocker and Schwartz (1985), for instance, found that individuals low in self-esteem denigrate both outgroup and ingroup members (see also Crocker, Thompson, McGraw, & Ingerman, 1987). Sachdev and Bourhis (1987) found that, contrary to what one might expect, high status groups showed an even more pronounced ingroup bias than did low status groups. Recently, Aberson, Healy, and Romero (2000) attempted to reconcile these conflicting findings, and in the process suggested how this regulatory need may come to affect the type of bias one shows toward ingroups and against outgroups. Interpreting the results of a meta-analysis, these researchers concluded that both high- and low-self-esteem individuals demonstrate ingroup bias, but do so in different ways. High-self-esteem individuals showed more direct bias (i.e., a bias when they were actively participating in the ingroup). Low-self-esteem individuals, on the other hand, tended to show more indirect bias, defined as bias exhibited when not actively participating in the ingroup (i.e., an observer; see Brown, Collins, & Schmidt, 1988). Yet self-esteem need not be the only need satisfied by group membership. Indeed, classic social psychological conceptions of group function have stressed how members' personal attitudes, experiences, and behavior are anchored in group norms, suggesting that ingroups play a critical epistemic role as a primary source of social comparison and verification (Asch, 1952; Festinger, 1954; Kelley, 1955; Lewin, 1947; Sherif, 1936). Because a meaningful validation of beliefs or values is more likely to be sought from individuals with whom one shares relevant social identities, ingroups may play a uniquely important role in defining one's social reality and in bringing one certainty or closure (Hogg & Abrams, 1993; Hardin & Higgins, 1996; Shah, Kruglanski, & Thompson, 1998). Although the previous review is far from comprehensive, it does allow us to draw two fundamental conclusions about ingroup bias: (a) such bias may serve various self-regulatory functions, and (b) an understanding of these functions may reveal something about how this bias is expressed. With this in mind,

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

a primary purpose of the present chapter is to propose a new regulatory distinction as underlying distinct forms of the ingroup bias: expressions of this bias that arise from our fundamental needs for either nurturance through promotion or security through prevention. Below, we elaborate on this regulatory distinction under the assumption that its application to the phenomena of ingroup bias may yield important insights into how this bias unfolds.

THE REGULATORY FOCUS OF INGROUP BIAS Among the most primary of human needs are the needs for nurturance and security. Regulatory focus theory (Higgins, 1997) proposes that there are two distinct types of regulatory systems concerned with meeting the basic needs for nurturance and gain, and for security and safety. The two self-regulation systems (labeled promotion and prevention, respectively) fulfill these needs through the pursuit of different types of goals (or desired end-states) and the use of different behavioral means, and are also characterized by differing evaluations and emotional experiences. In addressing our needs for nurturance and accomplishment, a promotion focus involves the pursuit of hopes, wishes, and aspirations (ideals). It also involves a sensitivity to the presence or absence of gain, and employs behavioral approach strategies. Thus, individuals show a tendency to pursue ideals, such as doing well in school, by actively pursing opportunities for advancement toward this goal (e.g., seeking out time to study for exams). In addressing our needs for protection and safety, a prevention focus, in contrast, involves the pursuit of duties, obligations, or responsibilities (oughts), a concern with the presence or absence of loss, and behavioral avoidance strategies. Thus, if individuals view doing well at school as an obligation, they may consequently pursue this obligation by avoiding situations that would set them back (e.g., avoiding parties the week of an exam). A study on friendship by Higgins, Roney, Crowe, and Hymes (1994) lent support to this regulatory focus distinction by finding that participants whose promotion focus was primed using their ideals showed a preference for friendship strategies that exemplified approaching a match to a desired goal (i.e., supporting a friend). Alternatively, individuals whose prevention focus was primed using their oughts showed a preference for friendship strategies that exemplified avoiding a mismatch to a desired goal (i.e., avoid losing touch with a friend). Similarly, Shah, Higgins, and Friedman (1998) found that individuals with a promotion focus who pursued a task goal of solving a series of anagrams tended to dwell on those trials in which successful completion would move them closer to the overall task goal. Individuals with a prevention focus, on the other hand, tended to dwell on those trials in which failure would set them back from goal attainment.

35

36

FROM PREJUDICE TO INTERGROUP EMOTIONS

The pursuit of promotion and prevention has also been linked to distinct emotional experiences. Because promotion involves the absence or presence of nurturance and achievement, its pursuit has been shown to be experienced emotionally in terms of cheerfulness- and dejection-related emotions. In contrast, because prevention involves the presence or absence of security, its pursuit has been shown to be experienced emotionally in terms of quiescence- and agitation-related emotions (Higgins, 1987, 1998). These distinct emotional experiences have recently been shown to apply not only to how individuals experience success and failure, but also to how they come to emotionally appraise the world around them. Shah and Higgins (2001), for instance, demonstrated that promotion-focused individuals appraised objects in their environment in terms of how cheerful or dejected they made them feel. Alternatively, individuals' prevention focus appraised the same objects in terms of quiescence or agitation. Given that regulatory focus has been shown to differentially influence the ways in which individuals respond toward goals and objects in their environment, and the well-examined psychological significance of ingroups and outgroups (for a review see Brewer & Brown, 1998), it seems likely that these regulatory distinctions are also relevant to understanding individuals' evaluations, feelings, and interactions with ingroups and outgroups. Indeed, because regulatory focus offers a motivational distinction with specific evaluative, emotional, and behavioral consequences, it maybe particularly useful for specifying distinct manifestations of ingroup bias. We have begun to explore the influence of regulatory focus on ingroup evaluations, emotions, and behavior. Following the example of Mackie and Smith (1998), we propose that regulatory focus may help to provide a better understanding of how we come to evaluate, feel about, and behave toward different groups. Specifically, we propose that while both a promotion and prevention focus may relate to ingroup bias, they will do so in different ways. Thus, individuals' momentary or chronic regulatory focus may determine the manner in which individuals favor their ingroup over other groups evaluatively, emotionally, and behaviorally.

REGULATORY FOCUS AND BEHAVIORAL MANIFESTATIONS OF INGROUP BIAS In a recent study, Shah, Brazy, and Higgins (2001) measured the strength of participants' promotion and prevention focus by employing a measure conceptually similar to that used by Higgins, Shah, and Friedman (1997). Faster lexical decisions involving attributes the participant had previously indicated that he/she ideally wanted to possess was assumed to indicate greater promotion focus (ideal strength), whereas faster lexical decisions involving attributes that the person felt he/she ought to be was assumed to represent a greater preven-

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

tion focus (ought strength). It is important to note that this measure of regulatory focus strength was in no way related to an individuals group memberships or intergroup relations; it was simply a measure of an individual's chronic promotion or prevention orientation. After completing this measure the participants were informed that the session would involve them being paired with a teammate in competition against another pair in a team task. The experimenter explained that some of the other players in the participant s game were assigned to a different waiting room, so that they would not meet until the team task. However, the experimenter did want the players to have the opportunity to learn a few things about each other before meeting. Thus, each participant had his/her picture taken using a digital camera and was asked to fill out an information sheet, which would presumably be given to the other participants in the game. In return, each participant received the same personal information from the other participants as well as printed copies of their pictures. After viewing the pictures and reading the provided information, participants were then brought to another room and told to take a seat in one of the available chairs. Participants were told that one of the chairs was already "taken" (as indicated by the presence of a backpack) by either their teammate or one of their competitors. This manipulation was a modification of one employed by Macrae, Bodenhausen, Milne, and Jetten (1994). With this manipulation, we were able to examine, independently, the extent to which participants chose to approach (sit closer to) an ingroup member or to avoid (sit further away from) an outgroup member. As illustrated in Table 3.1, the strength of participants' promotion focus predicted the degree to which participants sat closer to the chair identified as a teammate's, whereas the strength of their prevention focus predicted the degree to which participants sat further from a chair identified as a competitor's. A consideration of regulatory focus may increase our understanding of the negative outcome allocation effect described earlier. It suggests, for instance that while individuals may be less likely overall to express ingroup bias in terms of punishment than in terms of rewards, this behavioral distinction may be moderated by differences in regulatory focus. Indeed, while Otten, Mummendey, and Blanz (1996) also found evidence suggesting that individuals are less likely TABLE 3.1. Behavioral Manifestation of Intergroup Bias by Regulatory Focus Strength Regulatory Focus Strength Ideal Strength Ought Strength 0

p < .05

Type of Intergroup Bias Ingroup Approach

Outgroup Distancing

.31* -.14

-.13 .28°

37

38

FROM PREJUDICE TO INTERGROUP EMOTIONS

to demonstrate ingroup bias with respect to the allocation of negative stimuli, this effect was significantly lessened under conditions in which one s ingroup felt threatened (as when given inferior status). The present perspective suggests that the manipulated threat posed by the outgroup may have invoked a prevention focus that, in turn, would encourage active distancing from outgroups through their derogation or punishment.

REGULATORY FOCUS AND EMOTIONAL MANIFESTATIONS OF INGROUP BIAS Individual differences in regulatory focus may not only affect the manner in which one behaves toward ingroup and outgroup members, but also how one comes to feel about them emotionally. Specifically, regulatory focus theory suggests that this bias will occur along the specific emotional dimensions associated with promotion and prevention (cheerfulness- and dejection-related and relaxation- and agitation-related, respectively). To test this idea, we (Shah et al., 2001) recently assessed individual differences in regulatory focus strength using the same lexical decision task described earlier. After assessing these differences in University of Wisconsin students, we then asked them to describe how they felt emotionally about other University of Wisconsin students and how they felt emotionally about students from a rival university (Michigan State). Specifically, students were asked to determine the degree to which they thought they would feel "cheerful," "dejected," "relaxed," and "agitated" around University of Wisconsin students and Michigan State students generally. Separate promotion and prevention manifestations of emotional ingroup bias were calculated by summing the dejection-related and agitation-related emotion ratings. The positive emotions were first reverse scored, and then summed with the negative emotions to form the dejection-related emotions (i.e., reverse of cheerful scores added to dejected scores) and, separately, the agitation-related emotions (i.e., reverse of relaxed scores added to agitated scores). We then compared how participants felt emotionally about the two groups. As indicated in Table 3.2, participants' reported level of ingroup bias in terms of cheerfulness- and dejection-related emotions was uniquely reTABLE 3.2. Emotional Manifestation of Intergroup Bias by Regulatory Focus Strength. Regulatory Focus Strength Ideal Strength Ought Strength * p < .05

Type of Intergroup Bias Cheerfulness-Dejection

Relaxation-Agitation

.17° -.13

-.09 .18*

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

lated to the strength of their promotion focus. Participants' reported level of bias in terms of relaxation- and agitation-related emotions was uniquely predicted by the strength of their prevention focus.

SITUATIONAL DIFFERENCES IN REGULATORY BIAS To this point, regulatory focus has been conceptualized as a difference that varies individually, affecting how one chronically feels about ingroups and outgroups and how one chronically acts toward or against them. Yet, as discussed earlier, regulatory focus may also differ from moment to moment, and such situational differences may similarly influence how one comes to presently view ingroups and outgroups. With this in mind, Shah et al. (2001) attempted to manipulate individuals' regulatory focus while engaged in a competition with a teammate against another pair. This was done by manipulating their incentive to perform well in the competition. Participants were told that if their team won the competition they would receive two extra credits for their participation (versus only one if they lost), but the manner in which they were given this information differed in each experimental condition. To induce an orientation toward gain and achievement, participants in the promotion condition were told that they would be receiving one extra credit for their participation, but if their team won the competition, they would receive an additional extra credit. To induce an orientation toward security and the prevention of loss, participants in the prevention condition were told that they would be receiving two extra credits for their participation, but if their team lost, they would lose one of these credits. After receiving either the promotionor prevention-framed contingency, the participants were given information about their teammate and one of the members of the competing team in much the same fashion as was described earlier. After reviewing the provided information, participants were then asked to indicate how "cheerful," "dejected," "relaxed," and "anxious" they thought these individuals (teammate and competitor) would make them feel, and how much they thought they would want to get to know these individuals or avoid them in a different context. From these measures, ingroup bias scores in terms of behavioral approach and avoidance, and in terms of cheerfulness-, dejection-, relaxation-, and agitation-related emotions, were calculated. Differences in the extent of these particular forms of ingroup bias as a function of regulatory focus framing are presented in Figure 3.1. As is evident, individuals in the promotion condition reported more ingroup bias in terms of getting to know their teammate during the competition, and in terms of cheerfulness- and dejection-related emotions. Participants in the prevention-framed condition, on the other hand, showed more ingroup bias in terms of avoiding their competitor during the competition, and in terms of relaxationand agitation-related emotions.

39

40

FROM PREJUDICE TO INTERGROUP EMOTIONS

FIGURE 3.1. Emotional and behavioral manifestation of intergroup bias as a function of regulatory focus manipulation.

Regulatory Focus and the Nature of Implicit Associations The regulatory focus perspective on group identification has implications not only for t h e deliberate emotional and behavioral bias one shows against outgroups, but also for the more implicit negative associations one may make to these groups. Recent work by Lepore and Brown (1997), for instance, suggests that the nature of implicit associations one makes with ethnic categories can vary as a function of prejudice level. This research found that individuals high

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

in prejudice were found to show more automatic activation of negative stereotypic information when primed with labels for the general category of "Black person" (e.g., "Blacks" or "Afro-Caribbean"). Using a sequential priming procedure similar to that proposed by Bargh, Bond, Lombardi, and Tota (1986), Lepore and Brown (1997) found that high prejudice individuals were quicker to recognize negative stereotypic words when first primed with category labels representing the general "Black people" category. Devine, Plant, Amodio, Harmon-Jones, and Vance (2002) extended this research by showing that individuals' internal motivation to respond without prejudice is linked to their implicit bias, as assessed by the degree to which ethnic category labels facilitate recognition of positive and negative targets. As Devine and colleagues have demonstrated (Devine, Plant, & Brazy, 2002; Plant & Devine, 1998), high internally motivated individuals respond with less prejudice on implicit measures, whereas low internally motivated individuals show greater implicit prejudice. Recently, we (Brazy & Shah, 2002) attempted to show that regulatory focus may further specify the types of implicit associations made with the general "Black person" category in terms of the specific emotions relating to promotion and prevention. After measuring participants' internal motivation to respond without prejudice (see Plant & Devine, 1998), and participants' regulatory focus strengths (using the aforementioned lexical decision task), we had participants complete a sequential priming procedure (similar to Lepore and Brown, 1997) using emotion words relating to agitation and dejection. We predicted that differences in the strength of participants' focus on promotion and prevention, and their prejudice level would differentially influence the degree to which they implicitly associated the "Black person" category with either dejectionrelated emotions or agitation-related emotions, as seen by how quickly participants responded to these words when first primed with the "Black person" category. Indeed, participants' promotion focus strength was generally found to relate to the degree to which they associated dejection-related emotions with the "Black person" category. These main effects, however, were significantly moderated by participants' internal motivation to respond without prejudice. Specifically, the effects of promotion and prevention focus strength on emotional responses were significantly more pronounced in those individuals with a relatively low internal motivation to respond without prejudice than those whose motivation was relatively high. To better illustrate these significant interactions of continuous variables, we first performed a median split on participants' internal motivation to respond without prejudice and examined the effects of regulatory focus on these two different types of emotional associations (dejection-related and agitation-related) for individuals relatively "high" and "low" in their internal motivation to respond without prejudice, as detailed in Table 3.3. In sum, though a promotion focus and prevention focus were both linked to the expression of ingroup bias, the manner in which it was expressed differed in accordance with the regulatory need being pursued. Promotion-focused in-

41

42

FROM PREJUDICE TO INTERGROUP EMOTIONS

TABLE 3.3. Implicit Associations as a Function of Internal Motivation to Respond without Prejudice and Regulatory Focus High Internal Motivation to Respond without Prejudice Strength of Implicit Association Regulatory Focus Strength Promotion (ideal) Prevention (ought)

Black-Agitation

Black-Dejection

.06 .17*

.11 .03

° p < .05 Low Internal Motivation to Respond without Prejudice Strength of Implicit Association Regulatory Focus Strength Promotion (ideal) Prevention (ought)

Black-Agitation

Black-Dejection

-.18* .34°'°

.25**° -.09

°p< .05, ***p< .001

dividuals expressed ingroup bias in terms of approach-related evaluations and behaviors (e.g., approaching ingroup members over outgroup members) and the specific emotions associated with promotion (cheerfulness and dejection). Prevention-focused individuals, on the other hand, expressed ingroup bias in terms of avoidance-related evaluations and behaviors (avoiding outgroup members more than ingroup members), and in terms of prevention-related emotions (quiescence and agitation). Strikingly, the differential effects of regulatory focus on ingroup bias have been shown to be evident not only in enduring social groups of significant personal importance (such as one's university affiliation), but even with what might seem highly ephemeral and temporary group memberships of patently short duration (such as the "minimal groups" described earlier). Indeed, common experience confirms that we often react to a complete stranger (e.g., a fellow fan at a sporting event) as if he/she were a member of a highly significant ingroup with the potential to bring us momentary reward or peace of mind. Yet, often only minutes later, once the context has shifted and the momentary "ingroup feeling" has dissolved, this strangers "regulatory significance" virtually ceases to matter, and his/her influence over our thoughts, feelings, and actions all but disappears. The distinct effects of ingroup bias when types of regulatory focus are situationally induced reveals how this can happen.

FUTURE DIRECTIONS The research described here offers compelling initial support for the hypothesis that group members' regulatory focus generally influences the manner in

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

which they favor their ingroups and derogate outgroups. Our theoretical framework also suggests a number of additional directions to explore in the future. Four such possibilities are detailed below.

Regulatory Focus and the Attractiveness of Croup Qualities Although our present analysis has focused on how we feel about and behave toward groups we already belong to or are excluded from, it may also have implications for how and when one joins and leaves groups (see also Smith, Murphy, & Coats, 1999). Specifically, although we may be "born into" some ingroups, such as those defining our ethnicity and gender, many others we may join by choice, and our analysis suggests that differences in regulatory focus may determine the types of group characteristics that encourage or discourage group membership. Group attraction may be determined by the potential for groups to fulfill promotion and prevention needs. Thus, participants' regulatory focus leads them to seek out and favor those groups that, because of their particular characteristics, are perceived to have the potential to address promotion and prevention concerns. Interestingly, different group characteristics may be associated with the potential to provide nurturance versus security, raising the possibility that promotion and prevention individuals are attracted to different characteristics of ingroups. Prevention-oriented individuals, for instance, may be attracted to large ingroups of high status because such groups may be in a better position to provide the security they seek. Promotion individuals, on the other hand, may be attracted to smaller ingroups of lower status because such groups may offer them more attention and allow them more control. Thus, regulatory focus may ultimately be shown to affect not only the manner in which individuals favor ingroups over outgroups, but also the types of ingroups one prefers. In specifying how regulatory needs affect the attractiveness of certain group qualities, this possibility is reminiscent of Brewers (1993) model of optimal distinctiveness, which describes how individuals' distinct need for relatedness and distinctiveness combine to influence their group identifications.

Experiencing Ingroup and Outgroup Members Indioidually Regulatory focus may also yield important insights into how we come to view the individual members of ingroups and outgroups. Recent research on stereotyping and prejudice, most notably by Devine and her colleagues, has paid particular attention to the motivations underlying this process and the differing implications they have for the expression of prejudice and discrimination toward members of outgroups (see Devine et al., 2002; Plant, 2000; Plant & Devine, 1998). Similarly, regulatory focus may be shown to influence not only how one evaluates and emotionally appraises individual outgroup members (such as whether they are viewed in terms of their potential for nurturance vs. security),

43

44

FROM PREJUDICE TO INTERGROUP EMOTIONS

but how one acts toward or against these individuals. Thus, whereas a prevention focus may induce one to more actively punish a threatening outgroup member, a promotion focus may cause one to withhold rewards, or advancement, from the same individual. This influence may also extend to individuals' experiences with fellow ingroup members, affecting, for instance, the nature of their friendships within these groups. Just as personal relationships may be characterized in terms of their "communal" or "exchange" qualities (see Clark & Mills, 1993), regulatory focus theory suggests that personal relationships may be characterized with respect to other qualities. For instance, individuals may characterize their friendships by the degree to which they are secure or exciting, or the degree to which they invoke happiness versus quiescence. Regulatory focus may specify one's sensitivity to bias from outgroup members. Steele and his colleagues (Steele, 1997; Steele & Aronson, 1995) have suggested that members of discriminated groups often feel threatened by situations in which their own actions may validate the stereotype others have of them. The anxiety that accompanies such perceived threat has been shown to negatively impact individuals' performance and subsequent motivation. Although little research to date has explored how such stereotype threat may vary individually, the present analysis suggests that one's reaction to such threat may be significantly moderated by individual or situational differences in concern with security and prevention. Specifically, individuals with a high need for security through prevention may be particularly vulnerable to the threat posed by the psychological "presence" of negative stereotypes. Moreover, a focus on prevention may also dictate how such threat is handled in terms of approach and avoidance behavior. Thus, individuals with a strong prevention orientation may be particularly sensitive to the threat posed by salient stereotypes relevant to their own performance, and particularly likely to react to this threat through avoidance means. Such intriguing possibilities await future study.

Ingroup Bias as Mood Regulation If ingroup bias has emotional consequences, it raises the possibility that such bias may also serve to regulate mood. The present approach suggests that specific attempts at mood regulation may also lead to specific types of intergroup bias, as specified by regulatory focus theory. Thus, just as Fein and Spencer (1997) found that individuals may be motivated to derogate others as a means for raising self-esteem, regulatory focus theory suggests that attempts to reduce specific types of negative moods may lead individuals to specific types of intergroup bias. Whereas agitated individuals may attempt to feel more relaxed through intergroup bias that involves the avoidance of outgroups, depressed individuals may attempt to feel happier through intergroup bias that involves approaching their ingroup.

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

Collective Regulatory Focus Finally, although the present analysis has considered how ingroup bias may serve to fulfill individual promotion and prevention needs, individuals' self-concept may also include their important relationships and group memberships (see, for instance, Markus & Kitayama, 1991). Ingroup bias, then, may arise to address not only the promotion and prevention concerns of the group members but the regulatory concerns of the group as a whole. Thus, just as individuals may pursue individual and collective self-esteem (see Crocker & Luhtanen, 1990), so may they pursue promotion and prevention for themselves and for those groups incorporated into their self-concept (for a recent discussion on role of interdependence in regulatory focus, see Lee, Aaker, & Gardner, 2000). Such a possibility suggests, among other things, that individuals who have satisfied their own personal needs for promotion and prevention may nevertheless exhibit bias against outgroups if such favoritism would bring about security or achievement for the group as a whole.

CONCLUSIONS In his seminal work on prejudice, Allport (1954) recognized the "shifting nature of ingroups," suggesting that while each individual may value membership to specific groups, his/her conception of these groups is not "unaffected by the tempers of the time." In the present chapter we have suggested that conceptions of ingroups and outgroups are not unaffected by the regulatory focus of the moment. Whether such focus arises from situational or dispositional influences, it may come to specify the evaluative, emotional, and behavioral nature of the bias we show toward our own groups and against others. In examining the complexity of individuals' behavior and emotions in in^ tergroup interactions, the present analysis suggests that individuals' focus on promotion and prevention may offer crucial insight into how ingroup bias unfolds. Moreover, this fundamental regulatory distinction may ultimately help us understand of not only how we perceive, feel, and act toward certain groups, but also toward the individuals who comprise them.

REFERENCES Aberson, C. L., Healy, M., & Romero, V. (2000). Ingroup bias and self-esteem: A meta-analysis. Personality and Social Psychology Review, 4, 157-173. Abrams, D., & Hogg, M. A. (1988). Comments on the motivational status of self-esteem in social identity and inter-group discrimina-

tion. European Journal of Social Psychology, 18, 317-334. Allen, V. L., & Wilder, D. A. (1975). Categorization, belief summary, and inter-group discrimination. Journal of Personality and Social Psychology, 32, 971-977. Allport, G. W. (1954). The nature of prejudice.

45

46

FROM PREJUDICE TO INTERGROUP EMOTIONS

Garden City, NY: Doubleday Anchor. Asch, S. E. (1952). Social psychology. Englewood Cliffs, NJ: Prentice Hall. Bargh, }. A., Bond, R. N., Lombardi, W. J., & Tota, M. E. (1986). The additive nature of chronic and temporary sources of construct accessibility Journal of Personality and Social Psychology, 50, 869-878. Brazy, P. C , & Shah, J. Y. (2002, February). Regulatory focus and the nature of implicit associations. Poster session presented at the annual meeting of the Society for Personality and Social Psychology, Savannah, GA. Brewer, M. B. (1979). Ingroup bias and the minimal intergroup situation: A cognitivemotivational analysis. Psychological Bulletin, 86, 307-324. Brewer, M. B. (1993). Social identity, distinctiveness, and ingroup homogeneity. Social Cognition, 11, 150-164. Brewer, M. B. (1999). The psychology of prejudice: Ingroup love or outgroup hateP/owrnal of Social Issues, 55, 429-444. Brewer, M. B., & Brown , R. (1998). Intergroup relations. In D. T Gilbert, S. T Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th éd., Vol. 2, pp. 554-594). Boston: McGraw-Hill. Brewer, M. B., & Campbell, D. T. (1976). Ethnocentrism and intergroup attitudes: East African evidence. Beverly Hills, CA: Sage. Brown, J. D., Collins, R. L., & Schmidt, G. W. (1988). Self-esteem and direct versus indirect forms of self-enhancement. Journal of Personality and Social Psychology, 55, 445453 Cacioppo, J. T, Gardner, W. L., & Berntson, G. G. (1997). Beyond bipolar conceptualizations and measures: The case of attitudes and evaluative space. Personality and Social Psychology Review, 1(1), 3-25 Clark, M. S., & Mills, J. (1993). The differences between communal and exchange relationships: What it is and what it is not. Personality and Social Psychology Bulletin, 19, 684691. Crocker, }., & Luhtanen, R. (1990). Collective self-esteem and ingroup bias. Journal of Personality and Social Psychology, 58, 60-67. Crocker, J., & Schwartz, I. (1985). Prejudice and ingroup favoritism in a minimal inter-

group situation: Effects of self-esteem. Personality and Social Psychology Bulletin, 11, 379-386. Crocker, J., Thompson, L. L., McGraw, K. M., & Ingerman, C. (1987). Downward comparison, prejudice, and evaluations of others: Effects of self-esteem and threat. Journal of Personality and Social Psychology, 52, 907916. Devine, P. G., Plant, E. A., Amodio, D., Harmon-Jones, E., & Vance, S. (2002). Exploring the relationship between implicit and explicit prejudice: The role of motivations to respond without prejudice. Journal of Personality and Social Psychology, 82, 835-848. Devine, P. G., Plant, E. A., & Brazy, P. C. (2002). The bogus pipeline and motivations to respond without prejudice: Revisiting the fading andfaking of racial prejudice. Manuscript submitted for publication. Duckett, J., & Mphuthing, T. (1998). Group identification and intergroup attitudes: A longitudinal analysis in South Africa. Journal of Personality and Social Psychology, 74(1), 80-85. Fein, S., & Spencer, S. }. (1997). Prejudice as self-image maintenance: Affirming the self through derogating others. Journal of Personality and Social Psychology, 73(1), 3 1 44. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117140. Finchilescu, G. (1986). Effects of incompatability between internal and external group membership criteria on inter-group behavior. European Journal of Social Psychology, 16, 83-87. Fiske, S. T (1998). Stereotyping, prejudice, and discrimination. In D. T. Gilbert, S. T Fiske, & G. Lindsey (Eds.), The handbook of social psychology (Vol. 2, 4th éd., pp. 357-411).. New York: McGraw-Hill. Hardin, C. D., & Higgins, E. T. (1996). Shared reality: How social verification makes the subjective objective. In R. M. Sorrentino & E. T Higgins (Eds.), The handbook of motivation and social cognition (Vol. Ill, pp. 2 8 84). New York: Guilford Press. Hewstone, M., Fincham, F., & Jaspars, J. (1981). Social categorization and similarity

PROMOTION AND PREVENTION FORMS OF INGROUP BIAS

in intergroup behaviour: A replication with "penalties." European Journal of Social Psychology, il,101-107. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340. Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52,1280-1300. Higgins, E. T. (1998). Emotional experiences: The pains and pleasures of distinct regulatory systems. In R. Kavanaugh, & B. Zimmerberg (Eds.), Emotion: Interdisciplinary perspectives (pp. 203-241). Hillsdale, NJ: Lawrence Erlbaum. Higgins, E. T., Roney, C , Crowe, E., & Hymes, C. (1994). Ideal versus ought predilections for approach and avoidance: Distinct selfregulatory systems. Journal of Personality and Social Psychology, 66, 276-286. Higgins, E. T., Shah, J. Y., & Friedman, R. (1997). Emotional responses to goal attainment: Strength of regulatory focus as moderator. Journal of Personality and Social Psychology, 72, 515-525. Hogg, M. A., & Abrams, D. (1993). Towards a single process uncertainty-reduction model of social motivation in groups. In M. A. Hogg & D. Abrams (Eds.), Group motivation: Social psychological perspectives (pp. 173190). Hertfordshire, England: Harvester Wheatsheaf. Kelley, H. H. (1955). Salience of membership and resistance to change of group-anchored attitudes. Human Relations, 8, 275-289. Lee, A. Y., Aaker, J. L., & Gardner, W. L. (2000). The pleasures and pains of distinct self-construals: The role of interdependence in regulatory focus. Journal of Personality and Social Psychology, 78, 1122-1134 Lepore, L., & Brown, R. (1997). Category and stereotype activation: Is prejudice inevitable? Journal of Personality and Social Psychology, 72, 275-287. LeVine, R., & Campbell, D. (1972). Ethnocentrism: Theories of conflict, ethnic attitudes, and group behavior. New York: Wiley. Lewin, K. (1947). Group decision and social change. In T. M. Newcomb & E. L. Hartley (Eds.), Readings in social psychology (pp. 211-234). New York: Holt. Mackie, D. M., Devos, T., & Smith, E. R. (2000). Intergroup emotions: Explaining

offensive action tendencies in an intergroup context. Journal of Personality and Social Psychology, 79, 602-616. Mackie, D. M., & Smith, E. R. (1998). Intergroup relations: Insights from a theoretically integrative approach. Psychological Review, 105(3), 499-529. Macrae, C. N., Bodenhausen, G. V, Milne, A. B., & Jetten, J. (1994). Out of mind but back in sight: Stereotypes on the rebound. Journal of Personality and Social Psychology, 67, 808-817. Markus, H., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion and involvement. Psychological Review, 98, 224-253. Mummendey, A., Simon, B., Dietze, C , Gruenert, M., Haeger, G., Kessler, S., Lettgen, S., & Schaferhoff, S. (1992). Categorization is not enough: Intergroup discrimination in negative outcome allocations. Journal of Experimental Social Psychology, 28, 1-16. Otten, S., Mummendey, A., & Blanz, M. (1996). Intergroup discrimination in positive and negative outcome allocations: Impact of stimulus valence, relative group status, and relative group size. Personality and Social Psychology Bulletin, 22, 568-581. Plant, E. A. (2000). Approach and avoidance regulatory concerns for interracial interactions. Unpublished manuscript, University of Wisconsin-Madison. Plant, E. A., & Devine, P. G. (1998). Internal and external motivation to respond without prejudice. Journal of Personality and Social Psychology, 75,811-832. Sachdev, I., & Bourhis, R. Y (1987). Status differentials and inter-group behavior. European Journal of Social Psychology, 17, 277-293. Shah, J. Y, Brazy, P. C , & Higgins, E. T. (2001). Promoting us vs. preventing them: The influence of regulatory focus on the manifestations of intergroup bias. Manuscript submitted for publication. Shah, J. Y, & Higgins, E. T. (2001). Regulatory concerns and appraisal efficiency: The general impact of promotion and prevention. Journal of Personality and Social Psychology, 80, 693-705. Shah, J. Y, Higgins, E. T., & Friedman, R.

47

48

FROM PREJUDICE TO INTERGROUP EMOTIONS

(1998). Performance incentives and means: How regulatory focus influences goal attainment. Journal of Personality and Social Psychology, 74, 285-293. Shah, J. Y., Kruglanski, A. W, & Thompson, E. P. (1998). Membership has its (epistemic) rewards: Need for closure effects on intergroup bias. Journal of Personality and Social Psychology, 75, 383-393. Sherif, M. (1936). The psychology of social norms. New York: Harper. Sherif, M., & Sherif, C. (1953). Groups in harmony and tension. New York: Harper. Smith, E. R, Murphy, J., & Coats, S. (1999). Attachment to groups: Theory and management. Journal of Personality and Social Psychology, 77, 94-110. Steele, C. M. (1997). A threat in the air: How stereotypes shape intellectual identity and performance. American Psychologist, 52, 613-629

Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of African Americans. Journal of Personality and Social Psychology, 69,797-811. Sumner, W. G. (1906). Folkways. New York: Ginn. Tajfel, H., Billig, M. G., Bundy, R. F., & Flament, C. (1971). Social categorization and intergroup behavior. European Journal of Social Psychology, 1, 149-177. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks/Cole. Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (2nd éd., pp. 7-24). Chicago: Nelson-Hall.

4

Antecedents and Consequences of Collective Guilt

NYLA R. BRANSCOMBE BERTJAN DOOSJE CRAIG McGARTY

E

motions and their antecedents have been primarily conceptualized in individualistic terms. We argue, however, that some important social emotions are dependent on the nature of salient intergroup relationships. Precisely because emotions are central to human psychological experience, understanding how they can come to be shared among group members at a particular historical juncture is of considerable theoretical importance. While other chapters in this volume assess various emotional responses toward different outgroups, we will explore emotional responses that stem from a focus on the ingroups own negative history and its treatment of an outgroup. The particular emotion we examine in-depth—collective guilt—is by no means universally experienced, although it is presently a prominent emotional response in a number of social groups around the world. Our goal for this chapter is to establish (a) the critical conditions for collective guilt to be experienced, (b) what factors contribute to the intensity of this group-based emotion, and (c) the social consequences that can flow from collective guilt.

Preparation of this chapter was facilitated by a NWO (the Dutch Organization for Scientific Research) Visiting Professorship grant and an Australian National University Visiting Scholar award to Nyla R. Branscombe, as well as an NWO award to organize an "Expert Meeting on Collective Guilt" in Amsterdam, July 2000. We thank Diane Mackie, Michael Schmitt, Eliot Smith, and Michael Wohl for their helpful comments on an earlier draft of the manuscript.

49

50

FROM PREJUDICE TO INTERGROUP EMOTIONS

Social Identity Theory (Tajfel & Turner, 1979) provides the starting point for our doubts about whether social emotions are solely a function of individualistic antecedents. To the extent that group members share goals and vested interests, emotional states that stem from such motivations are likely to be shared. One general goal that group members are likely to share is the desire to maintain a positive image of the group. Consistent with this assumption, numerous researchers have reported that people take pride in the accomplishments of their group even though they personally have done nothing to contribute to its glory (see Cialdini, Borden, Thorne, Walker, Freeman, & Sloan, 1976; Wann & Branscombe, 1990). Such emotional responses to positive, as well as negative or threatening, group outcomes are moderated by level of group identification. Depending on how identified people are with their group, they will vary in the extent to which group-level pride is expressed and in whether they attempt to distance from the group when it might be seen as responsible for a negative outcome (Branscombe, Ellemers, Spears, & Doosje, 1999; Doosje & Ellemers, 1997). Self-categorization theory helps explain the experience of group-based emotion by rejecting the assumption that the self can be equated with the individual (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; Turner, Oakes, Haslam, & McGarty, 1994; Turner & Onorato, 1999). The scope of the self can measurably extend beyond the individual psychological entity to include group memberships (Smith & Henry, 1996; Tropp & Wright, 2001). In short, emotional responses to social events depend on how the self is categorized. Individualistic responses to event outcomes are expected when the person categorizes at the personal level, whereas collective or group-based responses result when the self is categorized as a member of a group (Oakes, Haslam, & Turner, 1994; Schmitt, Silvia, & Branscombe, 2000; Turner et al., 1994). Importantly, grouplevel emotions cannot occur without first categorizing the self as a group member. We cannot feel pride in our group s positive accomplishments, or guilt based on our group's harmful actions, unless we have first categorized ourselves as members of that group. Theories aimed at illuminating the determinants of specific emotions such as guilt have assumed that guilt results from a perceived violation of an important standard or expectation (Frijda, Kuipers, & ter Schure, 1989; Weiner, 1995). While personal guilt may indeed be due to discrepancies between the self's actions and various possible standards (see Devine & Zuwerink, 1994; Fazio & Hilden, 2001; Higgins, 1987), we propose that collective guilt can occur when the ingroup is perceived as having treated an outgroup unjustly or as having violated moral standards. Although we claim that collective guilt can occur even when there is no personal responsibility for a negative outcome, there are several points of agreement between our view and that of personal guilt researchers (Devine, Monteith, Zuwerink, & Elliot, 1991; Lewis, 2000; Tangney & Fischer, 1995). We concur that the prerequisites for experiencing guilt include

COLLECTIVE GUILT

an awareness that some action or inaction brought harm to others, and an acknowledgment that the action or inaction represents a violation of an important standard. While there is considerable public debate about whether people should feel collective guilt for historical events that they are not personally responsible for (Barkan, 2000; Goldhagen, 1996; Gross, 2001), such questions do not address the issue of whether people can and do experience such guilt. We begin by examining the empirical evidence concerning our assumption that collective guilt can be experienced, even though personal responsibility for the outcome is absent.

DISTINGUISHING COLLECTIVE AND PERSONAL BASES OF GUILT In order to assess collective guilt when personal responsibility for a harmful outcome received by an outgroup had been eliminated, Doosje, Branscombe, Spears, and Manstead (1998, Study 1) first lead participants to categorize themselves as members of a group, "the deductive thinkers." After ostensibly interacting via their computers with other people who were members of either their ingroup or the outgroup, we provided participants with two forms of false feedback—whether they personally had been biased in interacting with the outgroup and whether they belonged to a group that had been. We reasoned that if participants believed they personally had mistreated the outgroup, then they would have a personal basis for experiencing guilt about the harm the outgroup experienced. If, however, participants believed they personally had been fair in their treatment of the outgroup, there would be no personal basis for experiencing guilt about the treatment the outgroup had received. Accordingly, half of the participants were told that their personal behavior toward members of the outgroup during a series of group tasks had been fair and unbiased, while the other half were told that their personal behavior toward the outgroup had been biased and unfair. Because it was crucial for testing our hypothesis that people can feel collective guilt even when any personal basis for guilt has been removed, we also administered a group history false feedback manipulation. This form of feedback was intended to vary whether other members of the participant's ingroup had been prejudiced in their treatment of the outgroup or not. Accordingly, participants were told either that other members of their ingroup had been unfairly biased and had displayed a consistent tendency to be prejudiced against the outgroup, or that the other members of their ingroup had not behaved in a prejudiced manner toward the outgroup. With this procedure, we were able to vary whether the individual had a personal basis for guilt concerning the treatment members of the outgroup received or not, in addition to whether their

51

52

FROM PREJUDICE TO INTERGROUP EMOTIONS

ingroup had harmed the outgroup or not. The degree to which collective guilt was experienced across these four conditions was assessed with Branscombe, Slugoski, and Kappen's (2001) 5-item measure of collective guilt acceptance (e.g., "I feel guilty about the negative things inductive thinkers have done to deductive thinkers"). Doosje et al. (1998) found significantly greater collective guilt when the person belonged to a group that had behaved in a prejudiced fashion compared to when their ingroup had been fair toward the outgroup, even though the individual had not personally been prejudiced against the outgroup. Thus, the experience of collective guilt is not simply due to the individual's own actions or perceived personal responsibility for the harm experienced by the outgroup. Collective guilt can be experienced even when the individual has clearly played no role in the mistreatment of another group and cannot be seen as personally responsible for the harm done. When people belong to a group that has acted toward an outgroup in a way that is seen as unjust, under the psychological conditions that we elaborate on below, then people can indeed experience collective guilt. Collective guilt can be empirically distinguished from personal guilt and numerous other personal beliefs. Branscombe et al. (2001) found that the acceptance of collective guilt is weakly correlated (rs < .3) with both trait and state measures of willingness to accept personal guilt for negative outcomes (Kugler & Jones, 1992). Furthermore, the degree to which collective guilt is reported concerning the ingroup s history is not significantly correlated with measures assessing adherence to high personal moral standards, personal selfesteem, social desirability, or beliefs about the world in general (e.g., just-world beliefs and equal opportunity beliefs). Collective guilt is, however, significantly correlated with a number of theoretically relevant group-level measures. For example, White Americans who score higher on a measure of social dominance (e.g., are more racist) are also less likely to accept collective guilt for their group s racist past. In contrast, those who report more collective guilt are more likely to favor affirmative action for minority groups who have been harmed by Whites in the past (Branscombe et al., 2001; Swim & Miller, 1999). Acceptance of collective guilt has been shown to be consistently associated with a more liberal political orientation (rs < .3) in American (Branscombe et al., 2001), Australian (McGarty, Anutei, & Hendres, 2001), and Dutch (Doosje et al., 1998, 2001) samples. However, despite the correlation between political orientation and acceptance of collective guilt in dominant group members, responses toward the historically harmed outgroup (e.g., willingness to make reparations) have been consistently mediated by collective guilt rather than political orientation. There are also group differences in willingness to accept collective guilt reflecting the stable intergroup differences in power that exist within a given social context. In Australia, there is evidence of collective guilt concerning the historically harsh treatment of the Indigenous population (McGarty et al., 2001).

COLLECTIVE GUILT

In the Netherlands, collective guilt in relation to the colonial occupation of Indonesia (Doosje et al., 1998; Zebel, Doosje, & Spears, 2001) and in relation to the Dutch involvement in slavery (Zebel et al., 2001) has been documented. In Canada, English speakers, compared to French speakers, are more likely to accept collective guilt for the past harmful actions of their group toward the other linguistic group (Branscombe et al., 2001). In the United States, there is evidence concerning the existence of White guilt in relation to the treatment of African Americans (Branscombe, 2001; Branscombe et al., 2001; Swim & Miller, 1999). However, all members of these national groups do not report feeling collective guilt for the harm their group has done, nor is collective guilt experienced for all groups that have been harmed by the ingroup. Although Americans might be expected to also feel guilty about the treatment of their Native population or other ethnic minority groups within the United States such as Mexican Americans, they do so to a lesser extent compared to the guilt they report concerning slavery and African Americans (Barkan, 2000; Branscombe, 2001). In addition, harmful actions by the U.S. military toward groups outside the nation (e.g., the bombing of Cambodia during the Vietnam War, or the occupation of the Philippines in 1898) generate relatively little collective guilt (Branscombe, 2001). Such findings imply that collective guilt is not simply a function of the length of time since the historical events occurred, whether the harmed group is an internal or external one, or even the degree of the moral violation per se. Suffice it to say, collective guilt does not appear to be universally experienced for all harm done to outgroups. Willingness to accept collective guilt depends on aspects of the ongoing relations between the groups (see Barkan, 2000). We turn now to the minimally necessary conditions for the experience of collective guilt, and then we consider factors that can augment or attenuate the level of collective guilt experienced. The consequences of collective guilt for shaping intergroup relations and motivating political action are addressed in the final section of the chapter.

KEY ANTECEDENTS FOR THE ACCEPTANCE OF COLLECTIVE GUILT We focus here on the processes operating in members of dominant groups to produce collective guilt. By dominant groups we mean social categories that are perceived to be of higher status than some comparison category, in a given social context. We include those groups that have been systematically privileged compared to members of another category. Importantly, for collective guilt to be experienced, the higher status position of the dominant group must be at least open to the interpretation that it resulted from the exploitation of the disadvantaged group (see Branscombe, 1998; Montada & Schneider, 1989). If

53

54

FROM PREJUDICE TO INTERGROUP EMOTIONS

it is not seen as having resulted from unfairness or some form of exploitation in the past, and instead the status difference is seen as highly legitimate, then little collective guilt can be expected. We propose that there are two necessary conditions for members of a dominant group to feel collective guilt. The first is that they must self-categorize as a member of that group, although they need not highly identify with that group. The second is that they must perceive their group to be responsible for a salient illegitimate action or set of conditions that violates a moral value that the group currently subscribes to. Where these conditions hold, collective guilt will be augmented to the extent that there is a continuing status advantage that has not been repaired, and there is a conflict among members of the dominant group about the nature of the past or ongoing relations between the groups. Where dominant group members believe that the sins of the past have been corrected, or opinion about the past is not polarized and made a matter of public debate, there may be relatively little collective guilt. In order to experience collective guilt, dominant group members must first categorize themselves as members of the accused group. If people exclude themselves from the category, they can potentially escape any taint that would accrue to them and avoid accepting guilt for what the group has done. Thus, we argue that it is crucial for people to feel that they belong to a certain category in order to accept collective guilt. While this may seem an obvious precondition, it becomes less obvious when taking into account the multiple ways in which people can categorize themselves. For example, U.S. citizens whose parents were Italian can choose to perceive themselves in terms of the subordinate identity (Italian American), rather than in terms of the superordinate identity (American), and thereby avoid the guilt that would apply exclusively to the latter. People s ability to categorize themselves in terms of alternative identities is even more obvious when they are literally members of multiple groups (e.g., if they hold both Italian and American citizenship). By shifting self-categorizations, people in this position may be able to avoid the guilt by association that would follow from making salient the negative history of one of those category memberships. Collective guilt should be distinguished from moral outrage, which, we argue, can occur without self-categorization as a member of the perpetrator group. Members of groups not directly involved in an intergroup conflict can feel moral outrage at the harm done to a disadvantaged group, but guilt is unlikely. Montada and Schneider (1989) similarly suggested that the primary difference between moral outrage and guilt involves who is seen as the source of the harm. Specifically, they suggested that guilt occurs when the self is advantaged at the expense of the other. For collective guilt to occur, the focus must be on the actions or inactions of one s own ingroup that resulted in harm to the outgroup. Moral outrage, in contrast, is expected to occur when injustice or disadvantage is perceived to have occurred to others, without the ingroup necessarily being implicated. Thus, moral outrage, but not guilt, should occur when

COLLECTIVE GUILT

one outgroup is mistreated or harmed by another outgroup (e.g., Americans can be outraged by the harm done by one ethnic group to another ethnic group in Rwanda, but they will have little reason to feel guilt). Identification with a group, which we equate with long-term commitment to the group (see Ellemers, Doosje, & Spears, in press; Ellemers, Spears, & Doosje, 1999), is not the same thing as self-categorization as a member of the group. According to self-categorization theory (Turner et al., 1987), level of identification is one determinant of relative accessibility or perceiver readiness (Oakes et al., 1994) to self-categorize as a member of a group. That is, high identifiers will be more likely to self-categorize in terms of their group membership compared to low identifiers. While it is necessary for a person to have some sense of belonging to a dominant group (i.e., self-categorize) in order for collective guilt acceptance to occur, high identification or involvement with the group is not necessary. In fact, high identification with the ingroup may lead to justification processes and, at least under some circumstances, lower the level of collective guilt that is experienced. Support for this hypothesis was obtained by Doosje et al. (1998, Study 2). In that research, Dutch participants who were either high or low in identification with their national group were presented with one of three different versions of the history of the Dutch colonial occupation of Indonesia. When the ingroup s history was presented as thoroughly negative, with the Dutch being said to be exploitive and responsible for the death of many Indonesians, both high and low Dutch identifiers similarly reported feeling collective guilt. When the history was presented as thoroughly positive, where the Dutch were said to have created excellent services and institutions in Indonesia, collective guilt was significantly lower among both high and low identifiers. An important difference between high and low Dutch identifiers emerged, however, under conditions where the ingroup's history was presented in an ambiguous manner—where the ingroup was portrayed as being responsible for both negative and positive outcomes. In this case, high Dutch identifiers experienced significantly less collective guilt than did low Dutch identifiers. This pattern of results might seem paradoxical: the people who are most likely to experience collective guilt are those who report lower commitment to the collective that the group represents. It is precisely when people have little attachment to a group identity, while simultaneously being unable to avoid the categorization, that they will be most likely to acknowledge the negative aspects of their group, evoking greater collective guilt. In contrast, those who identify strongly with their group are more likely to actively defend their group, which will make them less susceptible to collective guilt. High identifiers can defend the group in several ways including claiming the current ingroup is not the "same" as it was during a prior historical period, or claiming greater ingroup variability (with the historical "black sheep" being separated psychologically from the current ingroup; see Marques & Paez, 1994).

55

56

FROM PREJUDICE TO INTERGROUP EMOTIONS

The Doosje et al. (1998, Study 2) research found support for the latter possibility with a measure of perceived ingroup variability. In the ambiguous condition, where collective guilt can be potentially avoided, high Dutch identifiers reported greater perceived ingroup variability than did low identifiers. In addition, the more intragroup variability the high identifiers perceived in this condition, the less collective guilt they reported. The second necessary condition for the experience of collective guilt in dominant groups is that they must perceive that a moral value violation has occurred. That is, they must see their group s actions as illegitimate and as inconsistent with the current humanitarian norms of the ingroup. Thus, Nazis will experience no guilt in relation to the Holocaust (racial genocide being consistent with the norms of Nazism), but contemporary Germans may experience collective guilt if they see the actions of the Nazis and their supporters as inconsistent with what it now means to be German. As is the case with guilt at the individual level (Lewis, 2000), in order to experience guilt as a group member, one has to experience a discrepancy between past behavior and some salient moral standard. In other words, one has to interpret the history of the dominant ingroup as morally wrong in some fundamental ways.

FACTORS THAT CAN AUGMENT OR ATTENUATE COLLECTIVE GUILT Groups have a variety of means available to them for avoiding the conclusion that their group s history is immoral. One possibility involves the perpetrator group claiming that it was more victimized than the other group (Buruma, 1999). With such a relative comparison, the ingroup s behavior can be considered to be relatively less of a moral violation. For example, some Japanese did this after World War II by comparing the Nagasaki and Hiroshima bombings to Pearl Harbor, and thereby lowering the severity of the ingroup s acts. Likewise, some Israelis currently employ the Holocaust as a more severe comparison as a means of lowering the gravity of the moral violation involved in the ingroup s treatment of the Palestinians. Such shifts in the standard that is used to compare the ingroup s actions against allows people to strategically avoid experiencing collective guilt. Likewise, including other groups who have committed similar moral violations in the frame of reference may alter the standard employed (e.g., "we were no worse than other nations in the Colonial period"). Another means that perpetrator groups can employ to lessen collective guilt involves attempts to increase the legitimacy of the ingroup s actions (which would, in effect, lower the severity of the moral violation). This can be accomplished in a variety of ways. While admitting the harmful actions, perpetrator group can attempt to justify its actions as having been provoked by the victimized group (e.g., claims by some Turkish people that the genocide of the Arme-

COLLECTIVE GUILT

nians was provoked by their traitorous behavior; see Staub, 1989). In general, increasing the perceived threat from the outgroup, as a justification for the ingroup's harmful actions, is a likely means of reducing ingroup collective guilt (Baumeister, 1999). Dehumanization of the outgroup (e.g., seeing them as unworthy of moral or human treatment at all; Bar-Tal, 1989) may be a vital mechanism for avoiding collective guilt by decreasing the extent to which a moral violation is perceived as having been committed. Another important means of avoiding collective guilt would involve denying the ingroup s involvement in the moral violation. McGarty et al. (2001) found that Romanian participants contested the historical involvement of Romanians in anti-Jewish actions during World War II, and consequently reported little collective guilt about the harm that occurred. Without a challenge to the legitimacy of the status differential between the dominant and disadvantaged groups, collective guilt among members of the dominant group may be unlikely. Only illegitimately perceived forms of status inequality will be capable of evoking collective guilt. Consider the inequality that currently exists between members of different age groups. The middleaged enjoy a stable status advantage over the elderly and young, but this does not translate into collective guilt among the middle-aged concerning their group s position and benefits (Garstka, Branscombe, & Hummert, 2001). Where the status differential is seen as highly legitimate, little collective guilt is likely among members of the advantaged group. Indeed, when members of these three age groups were first reminded of the economic advantages they receive based on their age group membership, the middle-aged reported significantly less collective guilt than did either the young or the elderly (Garstka et al., 2001). Furthermore, the middle-aged reported significantly greater perceived legitimacy of their age group advantages than did either the young or the elderly. Unless lower status groups such as the young or the elderly are able to create a divergence of opinion amongst the middle-aged concerning the illegitimacy of their ongoing group-based advantage, collective guilt about the inequality is unlikely in the privileged. The importance of illegitimacy for emotional responses to intergroup inequality has been also explored among gender groups. Branscombe (1998) asked men and women to think about the ways in which they receive privileges or benefits based on their gender group membership. Men consistently reported more wide-ranging and impactful privileges (e.g., having greater freedom of movement without fear of sexual assault and expecting to earn greater incomes) relative to women (e.g., having doors opened for them and receiving flowers from dates). Following this privilege listing task, men reported lower collective self-esteem and greater distancing from their gender group membership compared to women. It was argued that because these male participants perceived the kinds of group-based privileges that they were forced to think about as illegitimate, they tried to reduce their link to the group as a means of avoiding the

57

58

FROM PREJUDICE TO INTERGROUP EMOTIONS

collective guilt that they might otherwise experience. Consistent with this reasoning, Montada and Schneider (1989) found that the more people perceived a causal connection between their own fortunate position and the position of a disadvantaged group, the more likely they were to report feeling guilt. Indeed, gaining benefits at the expense of the other group is especially applicable to gender groups because the ways one group is privileged are precisely the ways in which the other group is disadvantaged (Branscombe, 1998). Thus, the major factor we see as likely to augment the experience of collective guilt involves ongoing status differentials between the groups that are seen as illegitimate. Reminders of such ongoing advantage among dominant group members are likely to elicit guilt only to the extent that the advantage is seen as illegitimate (see also Leach, Iyer, & Snider, 2002; Steele, 1990). If the differential positions of the groups can be seen as legitimate, there is no basis for collective guilt. To the extent that differentials between the groups have been substantially reduced, reparations have been made, or reconciliation has already occurred, then the likelihood of collective guilt should be lower. Doosje et al. (2001) showed that the type of reparation that is believed to have already occurred can influence the extent to which collective guilt is experienced depending on how committed people are to their national group. Interestingly, low identifiers experienced less guilt compared to high identifiers when told that their government had already apologized for the harm done to the victimized group. However, when high identifiers were told that their government had already dealt with the legacy of the colonial past in financial terms (without apologizing), they experienced less collective guilt than did low identifiers. These results suggest that collective guilt may be reduced via very different sorts of group reparations among low and high identifiers. High identifiers may prefer to avoid the identity-threatening approach of apologizing for the ingroup s past because that involves an admission of the illegitimate harm done. In contrast, low identifiers may experience less guilt when they acknowledge the negative aspects of their nation s history and, importantly, appear to see monetary reparations as insufficient to make up for the past. Schmitt, Branscombe, and Brehm (2000) tested the idea that the costs of corrective action for harm done could affect the intensity of collective guilt. They hypothesized that collective guilt would increase as the costs of corrective action increased, up to a point where corrective action is perceived as subjectively too costly, at which point collective guilt would drop. In this research, all of the male participants were presented with information about the injustice of male privilege, in particular the advantage of not having to limit one's freedom of movement because of fear of sexual assault. The perceived cost of correcting this specific form of intergroup injustice which had been made salient was then varied. In three experimental conditions, Schmitt, Branscombe, and Brehm (2000) specified what would be required to reduce women's disadvantage relative to men in this domain. In the low cost condition, the participants were

COLLECTIVE GUILT

required to get 5 signatures for a petition to get better lighting on campus for women. In the moderate cost condition, they were told that they needed to get 50 people to sign the petition. In the high cost condition, they believed that they needed to each get 100 people to sign the petition to correct this social injustice. Collective guilt increased from the low to moderate cost condition, but dropped again when reparations were perceived as very costly (e.g., the 100 signatures condition). These results are consistent with other studies (see Brehm, 1999) showing increases in the target emotion (anger, fear, happiness) as a function of action difficulty, until the required action is seen as more costly than the goal is subjectively valued. We suggest therefore that collective guilt can be attenuated, when making reparations for the past is perceived as sufficiently costly that motivation to do so declines. One factor that is likely to augment collective guilt is the presence of ongoing conflict within the dominant group about the nature of its own history (e.g., whether its actions were indeed immoral or not). Collective guilt may be especially likely to be aroused when there is debate about the ingroup s history and its consequences for current intergroup status differentials. Barkan (2000) has argued that while it is generally members of the disadvantaged group who first seek to challenge the legitimacy of the dominant group s past actions, unless the disadvantaged group can convince at least some members of the dominant group that they have legitimate claims, disadvantaged groups alone are unlikely to be successful at inducing collective guilt. To the extent that members of a disadvantaged group convey a unified view concerning the past injustices they have experienced (as is often important for a minority to influence a majority, see Moscovici, 1980), they are likely to be more successful in inducing guilt among members of the dominant group. Thus, we suggest that such sub-group formation within the dominant group itself may be of considerable importance for collective guilt to be widely experienced. If the dominant group remains reasonably cohesive in its original perspective on the ingroup s history and continues to justify it, relatively little collective guilt about the intergroup status differential will be experienced. Consistent with this argument, Zebel et al. (2001) have shown that members of a dominant group are more likely to perceive the status differential as illegitimate and experience collective guilt when they accept the disadvantaged group s perspective on the nature of the intergroup relationship as valid. This finding is comparable to the relations between perspective taking and guilt for not helping that have been observed at the inter-individual level (Batson, 1998). Based on the necessary conditions and the factors that can attenuate collective guilt that we have outlined, it might be fair to expect that collective guilt will be, on the whole, relatively rare. Indeed, people appear to have a host of defenses that protect them from experiencing collective guilt. To experience collective guilt, people must be unable to escape from categorizing themselves as a member of a dominant group that has exploited or mistreated another group.

59

60

FROM PREJUDICE TO INTERGROUP EMOTIONS

They must also perceive the harm to the other group as illegitimate. Collective guilt does tend to be higher, though, when there is continuing advantage on the part of dominant group members, when the efforts to make reparations to the harmed group have been inadequate but are perceived as possible, and when some dominant group members have been convinced that the moral violation that occurred is indeed illegitimate.

CONSEQUENCES: POLITICAL RAMIFICATIONS OF COLLECTIVE GUILT We argue that collective guilt can motivate people to reduce inequality, but is not guaranteed to do so. This could take the form of support for affirmative action for the historically harmed group or other mechanisms of reparation. Several studies have now demonstrated that collective guilt acceptance mediates the relationship between perceptions of illegitimate group history and outgroup compensation (Doosje et al., 1998, 2001; Swim & Miller, 1999). Collective guilt can also encourage commitment to apologize to the harmed group, especially among low identifiers (Doosje et al., 2001). There are two different forms of group-based apologies, with each having rather different political consequences. The first involves collective atonement by a group s leader on behalf of the group as a whole. For example, Pope John Paul II, in his role as leader of the Catholic Church, recently expressed guilt on behalf of Catholics of the world for their historical treatment of the Jews. In 1998, former U.S. President Clinton, on a trip to Africa, apologized for the role that White Americans played in slavery. But, there are also interesting cases of "Sorry" movements, where individuals are encouraged to express their personal sentiments to members of the harmed group, without an official group-based apology. In Australia this has taken the form of non-Indigenous Australians signing "Sorry books" for Indigenous Australians (especially on "Sorry Day"). McGarty et al. (2001) showed that feeling collective guilt in Australia predicts support for making this sort of personal apology better than it predicts support for an official apology by the government. Such findings create a paradox: Why should a collective emotion be a better predictor of a personal action than support for a collective action? The first possibile explanation focuses on perceptions of costs of different forms of apology. As the Schmitt, Branscombe, and Brehm (2000) study showed, when costs are perceived to be very high, collective guilt can be lower than when the costs are seen as more moderate. Official apologies can have real political and financial consequences. As a result, collective apologies may be perceived as too costly a response to collective guilt. The second possibile explanation is that features of the political context can force an individual response to collective guilt (McGarty, 2000). In Australia, collective guilt and

COLLECTIVE GUILT

government apology have both been portrayed in the media as unhealthy, negative, inappropriate, and as backward-looking responses. The governing parties have therefore argued against the issuing of an official apology to Indigenous Australians in order to avoid these potential costs. In the face of such pressures, some people who feel collective guilt appear to have developed an argument of the form "because we are guilty, I should apologize," producing greater support for a personal than a group-based response. In contrast to the Australian case, McGarty et al. (2001) found that collective guilt was an equally strong predictor of support for personal and official apology among Romanians in relation to their wartime treatment of the Jews. Without the same political constraints as those present in Australia, the critical variable in predicting apology in that Eastern European context was whether Romanians believed they were responsible for the harsh treatment experienced by the Jews. Because, in part, an apology from one group to another involves seeking forgiveness, an unwillingness to do so may stymie reconciliation between groups. As Barkan (2000) suggested, the unwillingness of the Japanese government to officially apologize for war crimes has resulted in an inability for them to reconcile with the harmed groups (e.g., Korean "Comfort Women"). When, however, the political context supports a group-based apology for the past, successful reconciliation between the groups may be more likely to occur. In sharp contrast to the lack of successful resolution with the past in cases where apologies have not been forthcoming, the South African Truth and Reconciliation Commission following the Apartheid era appears to have been rather successful. Perhaps by encouraging public apology for the harmful and unjust actions of past White governments, it has helped to facilitate unification and create a new post-Apartheid South African identity. Likewise, in Argentina and Chile during the past decade public expressions of guilt by members of the perpetrator group have been undertaken in the hope that this will lead to reconciliation between the groups. However, it is important to keep in mind that, in these latter cases, the admission of guilt has typically come from the actual harm perpetrators rather than from their successors. More historically distant cases, where the direct perpetrators are no longer available for personal acceptance of guilt, may require public group-based apologies for reconciliation between the groups to occur. Such differences in possible strategies for dealing with a negative group history provoke an important question: Does making collective apologies enable reparations to the harmed group, or might they even preclude the making of reparations under some circumstances? Certainly, an apology can be construed as a form of reparation in its own right. Because it can be identity-affirming for the disadvantaged group it may be seen as highly desirable. It could be desirable also for the dominant group, especially if it is seen as less costly than continuing animosity. In order to understand what role apology serves for ei-

61

62

FROM PREJUDICE TO INTERGROUP EMOTIONS

ther group, we must consider who the "target audience" is for the apology. The apology may be aimed squarely at the harmed group, but it may be also designed for the consumption of other members of the dominant group, or for the court of international public opinion. In the latter cases, apology might be construed as an attempt to buy a clean conscience fairly cheaply. It is reasonable to suppose that apologies that are not accompanied by social change might be perceived negatively by the disadvantaged group, as a demand by the dominant group for forgiveness of its past without incurring the costs required to improve ongoing conditions. However, members of some historically harmed groups may be sensitive about receiving financial compensation for deeds that could be seen as ill-suited to framing in purely financial terms. Is financial compensation appropriate for the loss of loved ones? German reparations to Israel have been intentionally framed as compensation to Holocaust survivors, rather than as reparations for the lives lost (see Barkan, 2000). In the Netherlands, after much debate, the "Jewish community" accepted 200 million dollars from the Dutch government as compensation for the mistreatment that Dutch Jews received upon returning from Nazi concentration camps at the end of World War II. Indeed, some Jewish people expressed disapproval of this or any other form of financial settlement. They did not agree with construing the issue in purely financial terms. They preferred an official apology, in addition to greater national recognition of their history (e.g., in the form of a museum) as a means of keeping the memory of their groups victimization alive. With these historical examples in mind, we now briefly outline two distinct pathways through which collective guilt could influence ongoing political processes. First, collective guilt could motivate both an official apology and concrete reparations to the disadvantaged group. Alternatively, collective guilt could motivate an apology as a means of avoiding more costly forms of reparations. According to the first sequence, apology is the vehicle by which genuine social change can be achieved. Collective guilt motivates the apology as a symbol of the desire of the dominant group to make reparations for the past. The formal step of making an apology is simply a recognition that wrongs were committed and must be set right. In this case, the apology is important for its symbolic value in achieving reconciliation. The offer of an apology means that past wrongs are acknowledged, and implies a desire for the apology to be accepted by the disadvantaged group. In other words, the dominant group is communicating, by its act of apologizing, that it considers the judgment of the disadvantaged group to be important and valid. Barkan (2000) argued that this process offers much to both the dominant and disadvantaged group. According to this view, what is important is that both dominant and disadvantaged group members engage in a process of rewriting the historical relation between the groups so that both groups consider the revision to be a fair depiction of the past. The

COLLECTIVE GUILT 6 3 disadvantaged group can gain from this process by receiving acknowledgment of their historical and present situation. The dominant group can gain from this process by coming to terms with the legacy of its past. According to the second pathway, collective guilt can lead to apology, but this is negatively linked to willingness to make subsequent reparations. In other words, the symbolic statement may serve to undermine the willingness to go the extra mile for social change. In effect, a public apology may be able to act as a symbol of finishing with the past among dominant group members. Consistent with this possibility, Doosje et al. (2001) found that high identifiers' support for reparations were lower once an apology had been made. However, this was not the case with low identifiers; for them, financial reparations were still seen as necessary. Thus, there is some reason to suppose that level of identification with the group may make one of these pathways more appealing than the other.

CONCLUSIONS As we have illustrated, the antecedents of some emotions—those stemming from a focus on the actions of the ingroup—are not simply a function of processes operating at the level of the personal self. The experience of guilt and other affective responses depends on how the self is categorized in a given context (see Schmitt, Silvia, & Branscombe, 2000; Smith, 1993; Turner & Onorato, 1999; Yzerbyt, Dumont, Gordijn, & Wigboldus, this volume). The same emotion—guilt—can result from violations of standards that the self is responsible for upholding when personal identity is salient, or when one's ingroup is seen as being responsible for violating standards and social identity is salient. In one case, personal guilt would be expected, and in the other collective guilt is likely to be experienced. While we expect that the phenomenology of both forms of guilt would be similar, the two should drive different forms of behavior—personal or group-based (see also Devos, Silver, Mackie, & Smith, this volume). The experience and consequences of collective guilt are, like other intergroup responses, moderated by social identification. Those who are relatively low in identification with a group they cannot avoid categorizing the self in terms of appear to be more susceptible to collective guilt, whereas high identifiers have a number of defenses that enable them to avoid this emotion to a greater degree. Most importantly, collective guilt can, but need not, have consequences for collective action which can serve to change existing intergroup relations, correct past injustices, and reduce ongoing inequality. Inducing collective guilt is unlikely to be a social panacea; it must be coupled with actual change in the existing relations between the groups.

64

FROM PREJUDICE TO INTERGROUP EMOTIONS

REFERENCES Barkan, E. (2000). The guilt of nations: Restitution and negotiating historical injustices. New York: W. W. Norton. Bar-Tal, D. (1989). Delegitimization: The extreme case of stereotyping and prejudice. In D. Bar-Tal, C. F. Graumann, A. W. Kruglanski, & W. Stroebe (Eds.), Stereotyping and prejudice: Changing conceptions (pp. 169182). New York: Springer-Verlag. Batson, C. D. (1998). Altruism and prosocial behavior. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th éd., Vol. 2, pp. 282-316). Boston: McGraw-Hill. Baumeister, R. F. (1999). Evil: Inside human violence and cruelty. New York: W. H. Freeman. Branscombe, N. R. (1998). Thinking about one's gender group s privileges or disadvantages: Consequences for well-being in women and men. British Journal of Social Psychology, 37, 167-184. Branscombe, N. R. (2001). For what historical group actions do White Americans feel collective guilt? Unpublished data, University of Kansas, Lawrence, KS. Branscombe, N. R., Ellemers, N., Spears, R., & Doosje, B. (1999). The context and content of social identity threat. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social identity: Context, commitment, content (pp. 3 5 58). Oxford, England: Blackwell. Branscombe, N. R., Slugoski, B., & Kappen, D. M. (2001). Collective guilt: Its components and consequences. Unpublished manuscript, University of Kansas, Lawrence, KS. Brehm, J. W. (1999). The intensity of emotion. Personality and Social Psychology Review, 3, 2-22. Buruma, I. (1999). The joys and perils of victimhood. New York Review of Books, April 8. Cialdini, R. B., Borden, R. J., Thorne, A., Walker, M., Freeman, S., & Sloan, L. R. (1976). Basking in reflected glory: Three (football) field studies. Journal of Personality and Social Psychology, 34, 366-375. Devine, P. G., Monteith, M. J., Zuwerink, J. R., & Elliot, A. J. (1991). Prejudice with and

without compunction. Journal of Personality and Social Psychology, 60, 817-830. Devine, P. G., & Zuwerink, J. R. (1994). Prejudice and guilt: The internal struggle to overcome prejudice. In W. J. Lonner & R. S. Malpass (Eds.), Psychology and culture (pp. 203-207). Boston, MA: Allyn & Bacon. Doosje, B., Branscombe, N. R., Spears, R., & Manstead, A. S. R. (1998). Guilty by association: When one's group has a negative history. Journal of Personality and Social Psychology, 75, 872-886. Doosje, B., Branscombe, N. R., Spears, R., & Manstead, A. S. R. (2001). Effects ofingroup identification on feelings of group-based guilt. Unpublished manuscript, University of Amsterdam. Doosje, B., & Ellemers, N. (1997). Stereotyping under threat: The role of group identification. In R. Spears, P. J. Oakes, N. Ellemers & S. A. Haslam (Eds.), The social psychology of stereotyping and group life (pp. 257272). Oxford, England: Blackwell. Ellemers, N., Doosje, B., & Spears, R. (in press). Self and social identity. Annual Review of Psychology, 53. Ellemers, N., Spears, R., & Doosje, B. (Eds.). (1999). Social identity: Context, commitment, content. Oxford, England: Blackwell. Fazio, R. H., & Hilden, L. E. (2001). Emotional reactions to a seemingly prejudiced response: The role of automatically activated racial attitudes and motivation to control prejudiced reactions. Personality and Social Psychology Bulletin, 27, 538-549. Frijda, N. H., Kuipers, P., & ter Schure, E. (1989). Relations among emotion, appraisal, and emotional action readiness. Journal of Personality and Social Psychology, 57, 212228. Garstka, T. A., Branscombe, N. R., & Hummert, M. L. (2001). Legitimacy of agebased advantage in young, middle-aged, and elderly adults: Implications for collective guilt. Unpublished manuscript, University of Kansas, Lawrence, KS. Goldhagen, D. J. (1996). Hitler's willing executioners: Ordinary Germans and the holocaust. New York: Knopf.

COLLECTIVE GUILT Gross, J. T. (2001). Neighbours: The destruction of the Jewish Community in Jedwabne, Poland. Princeton, NJ: Princeton University Press. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319-340. Kugler, K. E., & Jones, W. H. (1992). On conceptualizing and assessing guilt. Journal of Personality and Social Psychology, 62,318327. Leach, C. W., Iyer, A., & Snider, N. (2002). Poisoning the consciences of the fortunate: The experience of relative advantage and support for social equality. In I. Walker & H. J. Smith (Eds.), Relative deprivation: Specification, development, and integration. New York: Cambridge University Press. Lewis, M. (2000). Self-conscious emotions: Embarrassment, pride, shame, and guilt. In M. Lewis & J. M. Haviland (Eds.), Handbook of emotions (2nd éd., pp. 623-636). New York: Guilford Press. Marques, J. M., & Paez, D. (1994). The "black sheep effect": Social categorization, rejection of ingroup deviates and perception of group variability. In W. Stroebe & M. Hewstone (Eds.), European review of social psychology (Vol. 5, pp. 37-68). Chichester, England: Wiley. McGarty, C. (1999). Categorization in social psychology. London: Sage. McGarty, C. (2000, July 12-14). Collective guilt in Australia. Paper presented at the Expert Meeting on Collective Guilt, Amsterdam. McGarty, C , Anutei, A.-M., and Hendres, D. (2001). Collective guilt and atoning for the past in Australia and Romania: Separating personal from official apologies. Unpublished manuscript, The Australian National University, Canberra, Australia. Montada, L., & Schneider, A. (1989). Justice and emotional reactions to the disadvantaged. Social Justice Research, 3, 313-344. Moscovici, S. (1980). Towards a theory of conversion behavior. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 13, pp. 209-239 ). San Diego, CA: Academic Press. Oakes, P. J., Haslam, S. A., & Turner, J. C. (1994). Stereotyping and social reality. Oxford, England: Blackwell.

Roseman, I., Wiest, C , & Swartz, T. (1994). Phenomenology, behaviors, and goals differentiate discrete emotions. Journal of Personality and Social Psychology, 67, 206-211. Schmitt, M. T., Branscombe, N. R., & Brehm, J. W (2000, July 12-14). Determinants of the intensity of collective guilt. Paper presented at the Expert Meeting on Collective Guilt, Amsterdam. Schmitt, M. T., Silvia, P., & Branscombe, N. R. (2000). The intersection of self-evaluation maintenance and social identity theories: Intragroup judgment in interpersonal and intergroup contexts. Personality and Social Psychology Rulletin, 26, 1598-1606. Smith, E. R. (1993). Social identity and social emotions: Toward new conceptualizations of prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 297-315). San Diego, CA: Academic Press. Smith, E. R. & Henry, S. (1996). An in-group becomes part of the self: Response time evidence. Personality and Social Psychology Bulletin, 22, 635-642. Staub, E. (1989). The roots of evil: The origins of genocide and other group violence. Cambridge, MA: Cambridge University Press. Steele, S. (1990). White guilt. In S. Steele (Ed.), The content of our character: A new vision of race in America (pp. 77-92). New York: Harper Collins. Swim, J. K., & Miller, D. L. (1999). White guilt: Its antecedents and consequences for attitudes towards affirmative action. Personality and Social Psychology Bulletin, 25, 500514. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks/Cole. Tangney, J. P., & Fischer, K. W (Eds.). (1995). Self-conscious emotions: The psychology of shame, guilt, embarrassment, and pride. New York: Guilford Press. Tropp, L. R., & Wright, S. C. (2001). Ingroup identification as the inclusion of the ingroup in the self. Personality and Social Psychology Bulletin, 27, 585-600. Turner, J. C , Hogg, M. A., Oakes, P. J., Reicher,

65

66

FROM PREJUDICE TO INTERGROUP EMOTIONS

S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Oxford, England: Basil Blackwell. Turner, J. C , Oakes, P. J., Haslam, S. A., & McGarty, C. (1994). Self and collective: Cognition and social context. Personality and Social Psychology Bulletin, 20,454-463. Turner, J. C , & Onorato, R. S. (1999). Social identity, personality, and the self-concept: A self-categorization perspective. In T. R. Tyler, R. M. Kramer & O. P. John (Eds.), The social psychology of the social self (pp. 11-46). Mahwah, NJ: Lawrence Erlbaum.

Wann, D. L., & Branscombe, N. R. (1990). Die hard and fair-weather fans: Effects of identification on BIRGing and CORFing tendencies. Journal of Sport and Social Issues, 14, 103-117. Weiner, B. (1995). Judgments of responsibility: A foundation for a theory of social conduct. New York: Guilford Press. Zebel, S., Doosje, B., & Spears, R. (2001). Taking the outgroup's perspective and collective guilt. Unpublished manuscript, University of Amsterdam.

5

intergroup Emotions and Self-Categorization The Impact of Perspective-Taking on Reactions to Victims of Harmful Behavior

VINCENT YZERBYT MURIEL DUMONT ERNESTINE GORDIJN DANIEL WIGBOLDUS

D

uring the Kosovo war, thousands of civilians were thrown on the roads of their province in the middle of the winter in an attempt to escape the troops of Serbian ultra-nationalists. Most ended up as refugees in Albania. Many were reported to have died on their way to Albania or Montenegro. To be sure, the rest of the world did not remain insensitive, and international reactions took many forms. In Belgium, the dreadful situation of the Kosovars encouraged the Dutch- and French-speaking communities to unite their efforts to collect money. In a telethon that made history, some 1.3 billion Belgian francs were gathered—an impressive average of three U.S. dollars per capita. In other words, had the equivalent program and public reaction taken place in the United States, some 750 million dollars would have been raised! In sharp contrast to what happened during the Bosnian war, the Belgian media never referred to religious issues. In the telethon program itself, journalists and entertainers alike did not make the slightest reference to religion. For political observers, this omission came as somewhat of a puzzle. For social psychologists interested in intergroup relations, the suspicion was even stronger. Would the benefit of the telethon have reached such remarkable figures had

67

68

FROM PREJUDICE TO INTERGROUP EMOTIONS

the Belgian population, the vast majority of which is of Christian background, been explicitly reminded that most Kosovars are Muslims and that Serbs are Christians? The research presented in this chapter indicates that the answer to this question is a definite no. Indeed, we suggest that the generosity of the Belgian viewers may have been far less spectacular had the perpetrators and the victims in the Kosovo war been presented under a different light. The key factor, we argue, is that the potential donors, although not directly affected by the perpetrators' negative behavior, likely saw the victims as belonging to the same category as themselves. As a result they tended to appraise the situation in very much the same way as if they themselves had been involved in the situation. As it happens, various strands of research produced results compatible with our intuition that inclusion of ourselves and others in the same category or, for that matter, that a strong overlap between the self and other people will be of critical importance in shaping our emotions and our actions. In a first section, we evaluate a number of research efforts that looked at the impact of merging other people and the self in the same unit (Heider, 1958). Starting with the work on self-expansion and on helping, we then examine related findings within the intergroup relations literature. We review evidence that people tend to equate self and others as long as they belong to the same category and explore recent empirical efforts showing the benefits of perspective-taking on stereotype control and ingroup bias. In the second section, we turn to the consequences of self-other overlap on the emergence of intergroup emotions and action tendencies. We quickly present Smiths (1993, 1999) appraisal model of intergroup emotions and concentrate on recent work devoted to the extended contact hypothesis and to the issue of collective guilt. In our third section, we provide an overview of our own research program on social emotions and behaviors. In several studies, we accumulated a very consistent set of findings showing the relevance of a self-categorization approach of emotional reactions and action tendencies. We conclude by stressing the relevance of self-other overlap and depersonalization in any attempt to understand and orient intergroup relations.

PERSPECTIVE-TAKING AND SELF-CATEGORIZATION: WALKING THROUGH THE WORLD IN OTHER PEOPLE'S SHOES As our introductory example suggests, the connection between the victims of dramatic events and distant observers is likely to be of crucial importance in shaping the observers' reactions. This very question has been at the heart of several lines of research. Indeed, the work on self-expansion, empathy and altruism, self-categorization, and perspective-taking all convey the message that

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

the degree of overlap between the others and the self has important affective, cognitive, and behavioral consequences. In a most innovative research program, Aron and colleagues (for a recent review, see Aron, Aron, & Norman, 2000) proposed that human beings have a core motivation to expand the self. People are striving to acquire social and material resources, perspectives, and identities that enhance their ability to accomplish goals. By entering new relationships, the self presumably increases in diversity and in perceived self-efficacy. Aron and collaborators mounted an impressive set of data confirming that (rapid) expansion of the self indeed produces (strong) positive affective reactions (Aron, Norman, Aron, McKenna, & Heyman, 2000). These authors also report on experiments demonstrating that, in the context of close relationships, the other is literally included in the cognitive representation of the self (see also, Davis, Conklin, Smith, & Luce, 1996). For example, Aron, Aron, Tudor, & Nelson (1991) found that, although participants were most likely to attribute both poles of pairs of opposite traits to self (for instance, they would indicate that the self is both extroverted and introverted), they also tended to choose both traits more often for a best friend than for a friendly acquaintance (Sande, Goethals, & Radloff, 1988). This pattern suggests that the knowledge structure representing the best friend resembles that of the self more than the knowledge structure of the friendly acquaintance. Addressing the issue of cognitive representations more directly, Aron et al. (1991) asked married participants to rate 90 trait adjectives for their descriptiveness of themselves and their spouse. After a filler task, they then made a "me-not me" reaction time choice for each one of these traits. To the extent that the spouse is part of the self, any trait on which the two people differ should render the self-description decision more difficult to make, thereby increasing the reaction time. The data confirmed that there is indeed an overlap in the cognitive structures representing the self and close others. Another domain obviously related to the issue of self-other overlap is prosocial behavior (for a review, see Dovidio & Penner, 2000). Within the context of the research on helping, the idea of seeing the other as part of the self is directly reminiscent of the notion of empathy. To the extent that empathy can be seen as some sort of confusion between the self and the other, we would expect self-other overlap to play a key role in the provision of help upon encountering a person in need. A number of studies support the idea that selfother overlap increases the likelihood that an observer will intervene and provide help. For instance, Cialdini, Brown, Lewis, Luce, and Neuberg (1997) found that participants' intention to help another person was mediated by both empathie feeling and inclusion of the other in the self. Interestingly, when both variables were entered in the equation, the impact of empathy was no longer significant and self-other overlap came out as the key mediator. Although Batson, Sager, Garst, Kang, Rubchinsky, & Dawson (1997; Batson, 1997,1998) disputed

69

7 0 FROM PREJUDICE TO INTERGROUP EMOTIONS the strong conclusion that empathy is nothing but self-other merging, it seems clear that including the other in the self is a most efficient strategy to increase empathy among observers (Neuberg, Cialdini, Brown, Luce, Sagarin, & Lewis, 1997). The impact of self-other overlap has also been shown in the context of intergroup relations. In an intriguing set of studies, Cadinu and Rothbart (1996) relied on minimal group settings (i.e., a context in which information about the other members of the group is virtually absent) to provide evidence that ingroup favoritism is essentially a self-anchoring effect. Participants in their study displayed a greater correlation between the self and ingroup ratings when the selfratings were made before the ingroup ratings. Also, participants were more likely to generalize from the self to the ingroup than from the ingroup to the self. To be sure, we would argue that knowledge about the other(s) needs not be extensive for the self to start incorporating other people s characteristics. Moreover, people s familiarity with a great many groups in their social environment offers room for much flexibility in the merging process. Stressing the changing nature of self-other overlap like no other approach, self-categorization theory (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) suggests that the contextual salience of a given social identity leads group members to perceive themselves as interchangeable exemplars of the group rather than as unique individuals (Haslam, Turner, Oakes, McGarty, & Reynolds, 1998; Oakes, Haslam, & Turner, 1994; Turner & Onorato, 1999). This focus on the similarities between people and other members of the group has come to be known as the depersonalization process. The internalization of the same group concept and categorical attributes by all group members typically gives way to uniform behavior and has been used to account for such diverse phenomena as stereotyping, group polarization, social influence, and leadership (for a review, see Turner & Reynolds, 2000). Although the depersonalization process is a key assumption of self-categorization theory, the empirical evidence demonstrating the confusion between the characteristics typical of the self and those typical of the ingroup long remained mostly indirect (but see Cadinu & Rothbart, 1996). In a study borrowing the paradigm used by Aron et al. (1991), Smith and Henry (1996) provided direct proof that the representations of the self and the ingroup overlapped and that ingroup members and ingroup features become part of the self. Indeed, these authors found that ingroup traits that matched the self gave rise to faster self-descriptive responses than ingroup traits that mismatched the self. Consistent with a large body of findings showing that the action generally takes place for the ingroup rather than for the outgroup (Brewer, 1999; Otten & Wentura, 1999; Yzerbyt, Castaño, Leyens, & Paladino, 2000), no such difference was observed for the traits describing the outgroup. In a follow-up study, Smith, Coats, and Walling (1999) further showed that participants' perception of their standing on a given trait influenced their response time on the ingroup description

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

task. People were slower to indicate that a self-descriptive versus non-self-descriptive trait described the ingroup. Such an inhibition pattern is precisely what would be expected if the ingroup is defined in terms that match the ones used to describe the self. As a set, these studies provide strong evidence that the mental representations of the self and the ingroup indeed overlap. Although the above data indicate that people see themselves as possessing characteristics that are similar to the ones possessed by their fellow group members, they hardly provide any hint that such overlap influences the way people interact with others, especially when these targets are members of negatively stereotyped social groups. A series of experiments by Galinsky and Moskowitz (2000) on perspective-taking does just that. In one study, Galinsky and Moskowitz (2000, Exp. 1) borrowed a paradigm used by Macrae, Bodenhausen, Milne, and Jetten (1994) to explore the consequences of stereotype suppression (for a review, see Wenzlaff &Wegner, 2000). Galinsky and Moskowitz (2000) gave participants a photograph of an elderly man and asked them to write an essay describing a typical day of the man either with suppression instructions ("previous research has demonstrated that thoughts and impressions are consistently influenced by stereotypic preconceptions, and therefore you should actively try to avoid thinking about the photographed target in such a manner"), perspective-taking instructions ("imagine a day in the life of this individual as if you were that person, looking at the world through his eyes and walking through the world in his shoes"), or no specific instructions. Participants then performed a lexical decision task before they wrote a second essay about a different elderly man. As expected, both suppression and perspective-taking instructions reduced the stereotypicality of the first description compared to the control condition. Also, perspective-taking participants wrote a more positive essay than participants in the two other conditions. The key prediction concerned participants' performance on the lexical decision task. In line with earlier work showing rebound effects (i.e., the hyperaccessibility of the stereotype after a suppression episode), suppression participants were quicker to make a decision for stereotype-consistent traits than stereotype-inconsistent traits. In line with predictions, no such effect was obtained for perspective-taking and control participants. Clearly, thus, perspective-taking restricted the expression of stereotypical beliefs without provoking the usual facilitation for stereotype-consistent traits on the lexical decision task. In a follow-up study, Galinsky and Moskowitz (2000, Exp. 2) adapted the paradigm used by Smith and Henry (1996) and further showed that the degree to which people see the self and the target of judgment as sharing a high number of characteristics is a direct predictor of the propensity to neglect stereotypic expectations. In a final study, Galinsky and Moskowitz (2000) reasoned that, to the extent that ingroup favoritism can be seen as a direct consequence of the association between the ingroup and the self (Cadinu & Rothbart, 1996; Otten &

71

7 2 FROM PREJUDICE TO INTERGROUP EMOTIONS Moskowitz, 2000; Smith & Henry, 1996), any increase of the overlap between the self and the outgroup, byway of taking the perspective of an outgroup member, would reduce the strength and ubiquity of the bias. This is exactly what the data revealed. As it happens, the evaluation of the outgroup was raised to a level equivalent to that of the ingroup, thereby showing that it was not so much the ingroup that was seen as distinct from the self than the outgroup that was now merging with the self (Galinsky, in press). In our view, these data stress in a most unambiguous way the role of the links between the self and the target group in the emergence of intergroup phenomena. In summary, the tendency to merge the self with the other, either by applying the characteristics of the self to the other or the characteristics of the other to the self, has been found to exert a profound influence on the way people take pleasure and satisfaction from a relationship or even provide help. At the intergroup level, there is also strong evidence that such phenomena as stereotyping or ingroup favoritism are shaped by a tendency to see overlap between the others and the self. In the next section, we turn to the effect of self-group overlap on emotional experiences.

FROM SELF-OTHER OVERLAP TO INTERGROUP EMOTIONS Remember the emerging impact of the telethon program designed to aid the refugees in Kosovo—with its emphasis on a somewhat indirect path to the emotional experience, this example stands in sharp contrast to what can generally be found in current work on emotion. Indeed, contemporary appraisal theories mainly address the issue of individual emotional experiences (Frijda, Kuipers & ter Schure, 1989; Roseman, 1984; Scherer, 1988; Smith & Ellsworth, 1985). They stage personal emotions as complex reactions to specific situations or events that include differentiated cognitions, feelings, and action tendencies. That is, an individual is believed to interpret a specific event mainly in terms of whether the event harms or favors the individuals goals and desires and whether the individual possesses or lacks the resources to cope with the event. This cognitive appraisal then triggers a specific emotional experience which, in turn, promotes particular behavioral reactions. Clearly, thus, little is said about the experience of emotions on behalf of other people, not to mention entire social groups. In an attempt to move beyond such a highly individualized context, and building upon self-categorization theory (Turner et al., 1987), Smith (1993,1999) proposed an extension of the appraisal models in which people were thought to be able to experience emotions on behalf of their group. In that model, individuals are not necessarily personally concerned with the event but they experience emotions because their group may be helped or hurt by it. The critical

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

factor here is the focus on individuals' identity. Importantly, Smiths (1993) model of social emotions insists on the idea that people may experience specific intergroup emotions instead of just a global and undifferentiated (and generally negative) affective reaction (Dijker, 1987). In turn, each specific emotion is thought to lead to a unique pattern of intergroup behaviors. Several lines of research are consistent with the idea that people s situation as group members would lead them to experience emotions that would otherwise be absent if they persisted in seeing themselves as individuals. As a case in point, the work on fraternal relative deprivation (for recent examples, see Smith, Spears, & Hamstra, 1999; Tropp & Wright, 1999) holds that only when people embrace the social identity of a deprived group will they be in a position to experience a negative affective reaction to its full extent and engage in collective action in an attempt to set the record straight. There is also ample work to indicate that people feel happy or sad depending on the success or failure of their group even if they did not personally contribute to the outcome. Indeed, in a series of field studies, Cialdini, Borden, Thorne, Walker, Freeman, and Sloan (1976) observed that students wore more apparel displaying the logo of their university on Mondays following a victory of the local football team than on Mondays following a defeat or a draw. Moreover, students used the pronoun "we" significantly more to designate their team when they were asked to describe a victory of the university team than when they were asked to describe a defeat or a draw. This tendency was more pronounced when the student s public self had been threatened, for example, when they had personally failed a knowledge test. In order to account for this so-called BIRG (Basking in Reflected Glory) effect, Cialdini and colleagues (1976) relied on Heiders (1958) balance concept. According to Heider, cognitive unit relations ("I am associated with the team") are accompanied by affective relations ("I have the same evaluative meaning as the team"). Framing the explanation as a self-presentation device, Cialdini et al. (1976) suggested that students are aware that this strive for consistency comes into play in other people s eyes. They thus very much want to underscore their association with the winning team. We doubt somewhat the sufficiency of the self-presentational account. Instead, we would claim that the association between the self and the ingroup is a consequence of a wholly intrapersonal process (i.e., an authentic emotional reaction to the success of their group). A recent study by Boen, Vanbeselaere, and Feys (2000) conducted on the Internet lends some credence to our intuition. These authors monitored the number of unique visitors on the most popular website attached to each one of the teams participating in the Belgian and Dutch soccer championship. They did this on the first working day after each game for half of the 1999-2000 soccer season. Boen et al. (2000) found that the number of visitors was significantly higher after a win than after a loss. Clearly, the largely private nature of

73

74

FROM PREJUDICE TO INTERGROUP EMOTIONS

the behavior observed by Boen and colleagues militates against a pure selfpresentational account and sits comfortably within our social emotional framework. As it turns out, the behavior of even a single ingroup member may also have a dramatic impact on the way people react. Imagine that you are informed that one of your friends entertains a pleasant relationship with an outgroup member. Assuming that this ingroup member is seen as part of your self and that the outgroup member is part of that ingroup members self, it is reasonable to expect that the outgroup member will be seen as part of your self. Because the outgroup is part of that outgroup member, one would anticipate some increase in the overlap between your self and the outgroup as a whole. This idea of transitivity in the self-other merging was tested in what has come to be known as the extended contact hypothesis (Wright, Aron, McLaughlin-Volpe, & Ropp, 1997). In one intriguing experiment, Wright et al. (1997, Exp. 4) used a modified minimal group paradigm and had participants believe that they were into one of two groups on the basis of their performance on an object estimation task. Participants then observed an ingroup and an outgroup member (both confederates) interacting. Byway of verbal and nonverbal cues, the two confederates represented their relationship as that of close friends, unacquainted strangers, or disliked acquaintances. As expected, observation of an interaction between cross-group friends produced more positive evaluations of the outgroup than did observation of an interaction between an ingroup and an outgroup member who were either strangers or disliked acquaintances. Last but not least, Wright and colleagues (1997) found no evidence for an ingroup bias among those participants who had witnessed a cross-group friendship, whereas a clear ingroup bias emerged in the two other conditions. The message here can be seen as one more illustration of Heider's (1958) balance theory. Instead of the usual "my friends friend is my friend," Wright et al.'s (1997) work illustrates the fact that "my group member s friend s group is my group." From the perspective of the social emotions model, a major limitation of the above empirical efforts is that the emphasis remains on very broad evaluative reactions regarding the ingroup or the outgroup with limited attention devoted to specific emotional reactions. A notable exception is the work on collective guilt conducted by Doosje, Branscombe, Spears, and Manstead (1998). Using a minimal group paradigm study, Doosje et al. (1998, Exp. 1) informed participants about their alleged group membership (inductive thinkers) and asked them to evaluate ingroup and outgroup products. Depending on conditions, participants then learned that their fellow ingroup members had displayed a high versus low level of bias in their evaluations. Also, they were told about their high or low level of personal bias. Participants' reactions indicated that they felt more collective guilt when their group had been biased, especially when they themselves believed that they had not systematically undervalued the outgroup. Moreover, these feelings were linked to participants' compensa-

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

tory behavior. In a follow-up study focusing on the Dutch as a group, Doosje et al. (1998, Exp. 2) not only measured participants' identification with their national group but they also manipulated the historical treatment of Indonesia, a former Dutch colony, by the Dutch. Participants learned that their ancestors had done a lot of good things, a lot of negative things, or both good and bad things for Indonesia during the colonial period. As expected, compared to low identifiers, high identifiers felt much less guilty and were less willing to compensate for behaviors, but only when the past treatment of the outgroup was open to some interpretation. The data also confirmed that the feeling of guilt mediated participants' compensatory behavior. As can be seen, the nice twist about these studies is that they minimized as much as possible participants' personal responsibility for the actions of their group. To sum up, several lines of research—like the work on fraternal relative deprivation, basking in reflected glory, and extended intergroup contact—are all congruent with the idea that people may experience emotions as a result of their taking into consideration their association with a particular group. Perhaps closest to our present argument, however, the work on collective guilt offers suggestive evidence that people may be sensitive to information regarding their fellow group members. As a result, they manifest specific emotional reactions and engage in significant behaviors. Despite this, there is a shortage of research that meets with the ambition of the appraisal model of social emotions. The next section deals with a research program specifically aimed at showing the importance of group membership, whether contextually salient or chronically accessible, in the emergence of specific social emotions and their associated action tendencies.

SELF-CATEGORIZATION, ANGER, AND OFFENSIVE TENDENCIES According to appraisal theories, anger at another person is likely to result when the individual believes that the other has harmed the self and the self has the proper resources to react. Applying this idea to the group situation, Mackie, Devos, and Smith (2000; Devos, Silver, Mackie, & Smith, this volume) predicted and, indeed, found that group members experience anger and report offensive tendencies if they had the impression that their ingroup benefited from greater collective support than the outgroup. Although we have also been interested in anger as a critical intergroup emotion, our research agenda took a somewhat different path and aimed at addressing the following related questions: Are people capable of experiencing anger when other individuals are harmed by unfair and intentional outgroup behavior, and is the role of categorization of those victims as ingroup or as outgroup at all relevant? A quick look at the literature on emotions indicates

75

76

FROM PREJUDICE TO INTERGROUP EMOTIONS

that people seem to experience some difficulty in feeling angry on behalf of somebody else. In fact, previous research has shown that observers prefer to believe that other people get what they deserve, especially when they do not have the possibility to compensate the victims (Lerner, 1980). Consequently, observers are not very likely to empathize with victims. One way of inducing more empathy with victims is by telling observers to imagine themselves in the negative circumstances of the victims (Aderman, Brehm, & Katz, 1974). All in all, however, previous research did not pay much attention to the experience of anger on behalf of other people. This is hardly surprising if one looks closely at the specific characteristics of anger. Indeed, emotion theories generally assume that anger is only experienced when one perceives a situation that concerns oneself (Frijda et al., 1989). Building upon Smith s (1993) model, we hypothesized that observers may also be likely to experience anger toward the perpetrator, especially if the victims can be seen as part of the ingroup and the perpetrator can be seen as part of an outgroup. One obvious approach to test our model would be to rely on individual differences in group identification. Our analysis suggests that, compared to low identifers, high identifiers would be more likely to feel angry when confronted with victims who are members of their ingroup. There is one difficulty with this strategy, as people s group identification may be related to their aptitude to empathize with others in general. Therefore, we initially wanted to take a more radical approach by trying to alter in a straightforward manner the way people construct their surrounding environment. By manipulating the social context, we hoped to show that the appraisal of the social situation significantly influences people s reactions to intergroup events over and above the impact of existing differences in people s personality. In a first study (Gordijn, Wigboldus, Hermsen, & Yzerbyt, 1999), we used the crosscutting categorization paradigm to categorize the victims as part of the same group as the participants or as part of an outgroup (for a meta-analysis, see Urban & Miller, 1998). In the crosscutting categorization paradigm, the target is part of the outgroup on one dimension and part of the ingroup on another dimension. In order to manipulate the categorization of the victims, psychology students from the University of Amsterdam were told that the researchers were interested in differences in impression formation between students of different majors (for example, math students versus psychology students), or that the researchers were interested in differences in impression formation between students of different universities (for example, University of Amsterdam versus Free University). Participants in both conditions were then asked to read an article reporting on a math student of the Free University who had used the Internet facilities of the psychology department to illegally enter the Pentagon computer. The student had been caught, and the Free University had decided to close the computer room at the psychology department for some period of time. In sum,

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

the psychology students of the Free University were the people harmed by the behavior of a math student of the Free University. A focus on differences between students of different majors should make the target appear to belong to the same group as the participants (psychology students). In contrast, a focus on differences between students of different universities should make the target appear to belong to a different group than the participants (students of the Free University). Participants then rated their feelings on a series of 7-point scales, three related to anger (outraged, aggressive, and angry) and three to happiness (happy, elated, and cheerful). Globally, participants reported feeling more happy than

Categorization of Participants and Victims FIGURE 5.1. Intensity of happiness and anger as a function of the categorization of the participants and the victims.

77

78

FROM PREJUDICE TO INTERGROUP EMOTIONS

anger upon reading the newspaper article (see Figure 5.1). More importantly, and in line with predictions, results showed that the same negative behavior of an outgroup member that harms others led to more anger among participants when their perception was focused on similarities rather than on differences between the harmed group and themselves. This pattern implies that an emotion such as anger can be influenced by the way one perceives the people being harmed but only when similarities rather than differences between oneself and the victims are salient, implying that they belong to one s ingroup. Although this experiment offers initial support for our analysis, it also has some limitations. First of all, the absence of a control group makes it difficult to see whether people feel more angry when similarities with the victims are made salient or whether they feel less angry when differences are made salient. Indeed, according to social identity theory (Tajfel, 1982), people are inclined to discriminate against groups they do not belong to in order to enhance their selfesteem. People could thus feel satisfied and not angry to see an outgroup being harmed. A second limitation is that we restricted ourselves to an evaluation of anger and happiness. It is not clear however whether negative affect in general is influenced or whether it is just anger-related feelings that are affected by our manipulation. A third limitation concerned the fact that it is not obvious whether the observers were angry as a result of the math student's behavior, or if they were angry with the university because it closed down the computer room. It is thus important to disambiguate the target of participants' emotional reaction. To deal with these issues, we conducted another study (Gordijn, Wigboldus, & Yzerbyt, 2001). A control group in which the focus of attention of participants was not manipulated was added to the design. Further, a situation was chosen which was likely to cause anger rather than anxiety. As a matter of fact, although anger and anxiety share the same negative valence, they are based on very different appraisals of the situation. If the situation is appraised as if the perpetrator is behaving in an intentional and unfair way with respect to the victims and these victims have a sense that they have the power to do something about it, anger rather than anxiety should be experienced. Finally, a story was selected in which it was clear who was causing anger in the participants. We expected that anger rather than anxiety would be influenced as a function of categorization of the victims. In addition, the direction of influence was explored: that is, it was tested whether the perception of the victims as outgroup causes happiness or whether the perception of the victims as ingroup causes anger in comparison to the control group. In order to manipulate the categorization of the people who were harmed (target), one third of the participants, the different group condition, were told that the researchers were interested in differences in impression formation between students of different universities (for example, University of Amsterdam versus Leiden University). A second third, the same group condition, learned that the researchers were interested in differences in impression formation

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

between students and professors. Finally, the rest of the participants, the control condition, were just told that the study was about impression formation with no indication of a particular focus. Next, participants were asked to carefully read an article that allegedly had appeared in the Leiden University newspaper. This time, a story was developed in which it was described that the board of Leiden University was thinking about various ways to help the University become a smaller and more exclusive university with only the best students of the Netherlands. Students at Leiden University were shocked by such possible decisions as introducing admission exams, increasing tuition for slow students, removing students who do not pass their exams, and the like. Because they had not been informed about the decision, and especially because they were not allowed to participate in the decision-making process, students were planning various forms of protest. Thus, the negatively behaving source was described as a committee of professors of Leiden University, and the Leiden students were the victims. Various pieces of information collected in a pilot study confirmed that the story was inducing anger as intended. After reading the information, participants were asked to rate their feelings on 7-point scales, three related to anger (angry, outraged, and aggressive), three to happiness (happy, elated, and cheerful), and three to anxiety (anxious, powerless, and helpless). In order to check our manipulation, we asked participants in the different and in the similar target conditions among which groups the study was carried out. Clearly, our manipulations were found to be highly successful. We also asked participants to what extent they felt similar to students of Leiden University, and to what extent they saw students of Leiden University and of the University of Amsterdam as similar. Interestingly, participants indicated that they felt more similar to the Leiden students in the similar target condition than in the different target and control conditions, these two conditions showing no differences in the perception of similarity. Results strongly supported our hypotheses: Participants felt more angry than happy. Anxiety was intermediate (see Figure 5.2). The critical interaction effect between feelings and categorization of the target also emerged. Additional analyses confirmed that anger was differentially influenced by the manipulation of similarity. Participants felt more angry when their attention was focused on the fact that they belonged to the same category as the target than when their attention was focused on differences or when their focus of attention on category was not manipulated, the latter conditions not being different from each other. Happiness was also differentially influenced by the manipulation of similarity. Indeed, participants felt less happy when their attention was focused on the fact that they themselves and the target belonged to the same category than when their attention was focused on differences or when their focus of attention on category was not manipulated. Again, the latter two conditions were not different from each other. Finally, the data revealed no significant differences for anxiety as a function of the manipulation of target category.

79

80

FROM PREJUDICE TO INTERGROUP EMOTIONS

Categorization of Participants and Victims FIGURE 5.2. Intensity of anxiety, happiness, and anger as a function of the categorization of the participants and the victims.

Clearly, these findings show that the unfair and intentional behavior of the perpetrator influenced anger rather than anxiety as a function of categorization. Interestingly, the data obtained for the control group suggest that a focus on differences is the default option and that observers are more likely to spontaneously categorize victims as different rather than as similar, a pattern reminiscent of the classic just world findings (Lerner, 1980). The first couple of studies provides unambiguous evidence of the dramatic consequences of a subtle change in the way people are led to categorize themselves with respect to victims of the harmful behavior of an outgroup. The merits of these experiments notwithstanding, a series of interesting questions remained unanswered. First, we included a limited set of negative emotions

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

and did not put anger in competition with a closely related emotion—namely fear. Second, we had no evidence regarding the impact of the categorization context on factors other than emotions. Still, in line with appraisal theories of emotion, it would be most important to see whether the emotional experience would also translate into a specific action tendency. A related objective is to examine whether the emotional experience indeed mediates the impact of selfcategorization factors on some specific action tendency. Finally, according to self-categorization theory, people are likely to self-stereotype as a function of their level of identification with the group. High identifiers are expected to adopt the prototypical behavior of the ingroup to a larger extent than low identifiers. Clearly, thus, it would be important to examine the moderating role of group identification in the emergence of the emotional experience and, ultimately, in the occurrence of the behavioral reactions toward the outgroup (Branscombe & Wann, 1992; Doosje et al., 1998). To address these various issues, we designed a new study (Yzerbyt, Dumont, Wigboldus, & Gordijn, 2000). Concretely, participants were contacted in university libraries and invited to take part in a study that allegedly aimed at surveying people s opinions about a series of events that had recently appeared in national newspapers. For half of the participants, the written instructions presented on the first page of the booklet made the aims of the study (i.e., comparing the opinions of students and those of professors) very explicit. On the bottom of the page, these same participants were asked to indicate whether they were a student, a professor, or something else. For the remaining participants, the instructions unambiguously indicated that the study aimed at comparing the opinions of the students at University of Louvain at Louvain-la-Neuve (UCL) and those at the University of Gent. On the bottom of the page, participants were asked to report whether they were a student at UCL, at the Free University of Brussels, at the University of Gent, or elsewhere. Next, participants filled in a group identification scale. Depending on the experimental condition, the items were written with respect to the group of students or to the group of students at the UCL. On the next page, participants then read a photocopy of a text allegedly taken from a national newspaper. The story was that the Board of the University of Gent had decided to start using English as the sole language in the third, fourth, and fifth year of University. The same general components as before had been used as means to ensure that anger was the most likely reaction to the text. After they had read the text, participants were asked to indicate their feelings on a series of 12 rating scales. Three items concerned anger, three were related to depression, three had to do with fear, and three were associated with happiness. Finally, participants used rating scales to indicate to what extent they endorsed each one of 12 action tendencies. As for the emotions, three of the action tendencies concerned anger, three were related to depression, three had to do with fear, and three were associated with happiness. Finally, a series of filler questions were included in order to ascertain the success of the manipulation.

81

82

FROM PREJUDICE TO INTERGROUP EMOTIONS

Our results fully corroborated our hypotheses. Looking at participants facing a context in which the distinct memberships of themselves and the victims were emphasized, we only found a significant main effect of emotion. Not surprisingly, participants reported feeling more anger than any other emotion. Also, they seemed somewhat less depressed and happy than fearful. Turning to those participants confronted with a context stressing their common group membership with the victims, the main effect of emotion was significant (see Figure 5.3). Once again, participants reported feeling more anger than any other emotion. Interestingly, they also reported feeling less happiness than any other

FIGURE 5.3. Intensity of emotion as a function of the participants' identification with the group (when the context stresses the common group membership with the victims).

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION emotion. Fear and depression fell in between these two extremes. More importantly, the data also reveal the presence of a significant emotion by identification interaction, thereby confirming the fact that the impact of the story was more pronounced among high identifiers than among low identifiers. In complete agreement with our predictions, significant comparisons for anger and happiness indicated that the story made high identifiers more angry and less happy than their less identified colleagues. A distinct goal of this study was to examine the joint impact of category salience and group identification on action tendencies. Paralleling the data for

Type of Emotion FIGURE 5.4. Action tendency as a function of the participants' identification with the group (when the context stresses the common group membership of the victims).

83

84

FROM PREJUDICE TO INTERGROUP EMOTIONS

emotions, the answers given to the action tendency questions revealed the presence of a significant interaction between emotion and identification for those participants who were confronted with a common group membership (see Figure 5.4) but not for those who were led to think about a distinct group membership. Additional analyses confirmed that high identifiers who were reminded of a common group membership stated that they wanted to "move against" the perpetrator more than low identifiers. The next step in our analysis sought to test the highly specific prediction that anger, along with its associated action tendencies, would be more extreme when both the contextual forces and the personal characteristics combine to exacerbate the inclusion of the victims in the self. An a priori contrast opposing the common group membership/high identification condition to the three others confirmed the validity of this prediction both for the emotion and for the offensive action tendency. Finally, we wanted to test the hypothesis that emotions would mediate their respective action tendency. Indeed, a mediational analysis confirmed that the offensive action tendencies were entirely mediated by the emotional reactions of anger. In sum, this third study not only replicates our earlier studies, but it also extends our findings in several important ways. First, we were again able to establish the distinctive impact of the observed event on the emotional experience of our participants. In contrast to what was observed for anger, two other negative emotions, namely fear and depression, proved to be largely unaffected by our manipulation of category salience and by participants' level of identification with the salient category. Second, and more importantly, we found supportive evidence for the combined impact of category salience and identification on the emotional experience. Specifically, the salience of similarity was found to generate angry feelings among participants only to the extent that they strongly identified with the relevant category. Third, we were able to show that the impact of the independent variables was not limited to emotional consequences but had an influence at the conative level as well. Finally, we also showed that the tendency to oppose the perpetrator and react to the event was largely mediated by the degree to which angry feelings had been triggered in the first place. To sum up, we accumulated an impressive amount of evidence showing that the extent to which people perceive themselves to be similar to the victims of harmful behavior influences both their emotions and their action tendencies. In complete agreement with Smiths (1993,1999) model of social emotions, we found the emotional experiences to be extremely specific and to play a mediational role in the emergence of behaviors.

CONCLUSIONS One of the key messages of the present chapter is that people experience emotions on behalf of others as long as they see others as fellow ingroup members.

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

Because human beings are fundamentally inclined to merge the characteristics of their self with those of the members of their group, they appraise the situations confronted by those people, experience feelings on their behalf, and manifest behavioral tendencies just as if they themselves were going through the episode. Over the course of several studies, our findings deliver a highly consistent message. Our initial demonstration ascertained the impact of the contextual salience of social categories in the emergence of negative emotions (Gordijn et al., 1999). In a subsequent experiment, we also established the importance to discriminate between several negative emotions (Gordijn et al., 2001). Finally, we extended our analysis to show the joint impact of contextual salience and chronic social identification in the emergence of specific affective and conative reactions (Yzerbyt et al, 2000). As a set, these studies offer strong evidence that appraisal theories of emotion, when applied to the intergroup level, offer much promise not only to further our understanding of the phenomenology of group membership but also to improve the quality of our prediction with respect to intergroup behavior. As a matter of fact, people s reactions in an intergroup context can only rarely be traced back to events that they experienced for themselves. The extent to which people see others as belonging to the same category as themselves will be of utmost importance for such critical things as collective action, improvement of the intergroup relations, and the like. We started this chapter with anecdotal evidence from a TV show. Clearly, Belgian viewers were induced to see civilians undergoing dreadful events in their lives hundreds of miles away as people who were pretty much like them. Whether there was a conscious strategy on the part of the media people or on the part of some parties involved in the conflict remains of course an open question. As such, the factors promoting the specific demarcation of the social environment should be a most fascinating issue for future research. Once a particular social landscape was being adopted, however, there was a general momentum in the coverage of the events that tended to associate Western European citizens with the victims of the Serbian regime. Despite this, the research reviewed in the present chapter suggests that a systematic reference to the fact that these people may be in some profound and significant way different from the observers would have tempered the generosity that manifested itself during that historic TV show. Such is the power of social emotions!

REFERENCES Aderman, D., Brehm, S. S., & Katz, L. B. (1974). Empathie observation of an innocent victim: The just world revisited. Journal of Personality and Social Psychology, 29, 342347. Aron, A., Aron, E. N., & Norman, C. (2000). Self-expansion model of motivation and cog-

nition in close relationships and beyond. In G. J. O. Fletcher & M. S. Clark (Eds.), Blackwell handbook of social psychology: Interpersonal processes (pp. 478-501). Oxford, England: Blackwell. Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships as including

85

86

FROM PREJUDICE TO INTERGROUP EMOTIONS

other in the self. Journal of Personality and Social Psychology, 60, 241-253. Aron, A., Norman, C. C , Aron, E. N., McKenna, C , & Heyman, R. (2000). Couples' shared participation in novel and arousing activities and experienced relationship quality. Journal of Personality and Social Psychology, 78, 273-283. Batson, C. D. (1997). Self-other merging and the empathy-altruism hypothesis: Reply to Neuberg et al. (1997). Journal of Personality and Social Psychology, 73, 517-522. Batson, C D . (1998). Altruism and prosocial behavior. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (Vol. 2, pp. 282-316). Boston: McGraw-Hill. Batson, C D . , Sager, K., Garst, E., Kang, M., Rubchinsky, K., & Dawson, K. (1997). Is empathy-induced helping due to self-other merging? Journal of Personality and Social Psychology, 73, 495-509. Boen, E, Vanbeselaere, N., & Feys, J. (2000). Behavioral consequences of fluctuating group success: An internet study on soccer teams. Unpublished manuscript, Catholic University of Louvain at Leuven, Belgium. Branscombe, N. R., & Wann, D. L. (1992). Physiological arousal and reactions to outgroup members during competitions that implicate an important social identity. Aggressive Behavior, 18, 85-93. Brewer, M. B. (1999). The psychology of prejudice: Ingroup love or outgroup hate? Journal of Social Issues, 55, 429-444. Cadinu, M. R., & Rothbart, M. (1996). Selfanchoring and differentiation process in the minimal group setting. Journal of Personality and Social Psychology, 70, 661-677. Cialdini, R. B., Borden, R. J., Thorne, A., Walker, M. R., Freeman, S., & Sloan, R. L. (1976). Basking in reflected glory: Three (football) field studies. Journal of Personality and Social Psychology, 34, 366-375. Cialdini, R. B., Brown, S. L., Lewis, B. P., Luce, C , & Neuberg, S. L. (1997). Reinterpreting the empathy-altruism relationship: When one into one equals oneness. Journal of Personality and Social Psychology, 73, 481-494. Davis, M. H., Conklin, L., Smith, A., Luce, C. (1996). Effect of perspective taking on the

cognitive representation of persons: A merging of self and other. Journal of Personality and Social Psychology, 70, 713-726. Dijker, A. J. M. (1987). Emotional reactions to ethnic minorities. European Journal of Social Psychology, 17, 305-325. Doosje, B., Branscombe, N. R., Spears, R., & Manstead, A. S. R. (1998). Guilty by association: When one's group has a negative history. Journal of Personality and Social Psychology, 75, 872-886. Dovidio, J. E, & Penner, L. A. (2000). Helping and altruism. In G. J. O. Fletcher & M. S. Clark (Eds.), Blackwell handbook of social psychology: Interpersonal processes (pp. 162-195). Oxford, England: Blackwell. Ellemers, N., Spears, R., & Doosje, B. (1999). Social identity: Context, commitment, content. Oxford, England: Blackwell. Frijda, N. H., Kuipers, P., & ter Schure, E. (1989). Relations among emotion, appraisal, and emotional action readiness. Journal of Personality and Social Psychology, 57,212228. Galinsky, A. D. (in press). Creating and reducing intergroup conflict: The role of perspective-taking in affecting out-group evaluations. In M. A. Neale, E. A. Mannix, & H. Sonkak (Eds.), Research on managing in teams and groups (Vol. 4). Greenwich, CT: JAI Press. Galinsky, A. D., & Moskowitz, G. B. (2000). Perspective-taking: Decreasing stereotype expression, stereotype accessibility, and ingroup favoritism. Journal of Personality and Social Psychology, 78, 708-724. Gordijn, E. H., Wigboldus, D., Hermsen, S., & Yzerbyt, V. (1999). Categorisatie en boosheid: De invloed van negatief outgroup gedrag. In D. Van Knippenberg, C K. W De Dreu, C Martijn, & C Rutte (Eds.), Fundaméntele Sociale Psychologie, 13. Tilburg, NL: Tilburg University Press. Gordijn, E. H., Wigboldus, D., & Yzerbyt, V (2001). Emotional consequences of categorizing victims of negative outgroup behavior as ingroup or outgroup. Group Processes and Intergroup Relations, 4, 317-326. Haslam, S. A., Turner, J. C , Oakes, P. J., McGarty, C , & Reynolds, K. J. (1998). The group as a basis for emergent stereotype consensus. European Review of Social Psy-

INTERGROUP EMOTIONS AND SELF-CATEGORIZATION

chology, 9, 203-239. Heider, F. (1958). The psychology of interpersonal relationships. New York: Wiley. Lerner, M. J. (1980). The belief in a just world. New York: Plenum. Mackie, D. M., Devos, T., & Smith, E. R. (2000). Intergroup emotions: Explaining offensive action tendencies in an intergroup context. Journal of Personality and Social Psychology, 79, 602-616. Macrae, C. N., Bodenhausen, G. V, Milne, A. B., & Jetten, J. (1994). Out of mind but back in sight: Stereotypes on the rebound. Journal of Personality and Social Psychology, 67, 808-817. Neuberg, S. L., Cialdini, R. B., Brown, S. L., Luce, C , Sagarin, B. J., & Lewis, B. P. (1997). Does empathy lead to anything more than superficial helping? Comments on Batson et al. (1997). Journal of Personality and Social Psychology, 73, 510-516. Oakes, P. J., Haslam, S. A., & Turner, J. C. (1994). Stereotyping and social reality. Oxford, England: Basil Blackwell. Otten, S., & Moskowitz, G. B. (2000). Evidence for implicit evaluative ingroup bias: Affect-biased spontaneous trait inference in a minimal group paradigm. Journal of Experimental Social Psychology, 36, 77-89. Otten, S., & Wentura, D. (1999). About the impact of automaticity in the minimal group paradigm: Evidence from affective priming tasks. European Journal of Social Psychology, 29,1049-1071. Roseman, I. J. (1984). Cognitive determinants of emotion: A structural theory. In P. Shaver (Ed.), Review of personality and social psychology: Emotions, relationships, and health (pp. 11-36). Beverly Hills, CA: Sage. Sande, G. N., Goethals, G. R., & Radloff, C. E. (1988). Perceiving one's own traits and others': The multifaceted self. Journal of Personality and Social Psychology, 54, 1320. Scherer, K. R. (1988). Criteria for emotionantecedent appraisal: A review. In V. Hamilton, G. H. Bower, & N. H. Frijda (Eds.), Cognitive perspectives on emotion and motivation (pp. 89-126). Norwell, MA: Kluwer Academic Publishers. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Jour-

nal of Personality and Social Psychology, 48, 813-838. Smith, E. R. (1993). Social identity and social emotions: Toward new conceptualizations of prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 297-315). San Diego, CA: Academic Press. Smith, E. R. (1999). Affective and cognitive implications of a group becoming part of the self: New models of prejudice and of the selfconcept. In D. Abrams & M. A. Hogg (Eds.), Social identity and social cognition (pp. 183196). Oxford, England: Basil Blackwell. Smith, E. R., Coats, S., & Walling, D. (1999). Overlapping mental representations of self, in-group, and partner: Further response time evidence and a connectionist model. Personality and Social Psychology Bulletin, 25, 873-882. Smith, E. R., & Henry, S. (1996). An in-group becomes part of the self: Response time evidence. Personality and Social Psychology Bulletin, 22, 635-642. Smith, H. J., Spears, R., & Hamstra, I. J. (1999). Social identity and the context of relative deprivation. In N. Ellemers, R. Spears, & B. Doosje (Eds.), Social identity : Context, commitment, content (pp. 205229). Oxford, England: Blackwell. Tajfel, H. (1982). Social identity and intergroup relations. Cambridge, England: Cambridge University Press. Tropp, L. R., & Wright, S. C. (1999). Ingroup identification and relative deprivation: An examination across multiple social comparisons. European Journal of Social Psychology, 29,707-724. Turner, J. C , Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory, Oxford, England, & New York: Basil Blackwell. Turner, J. C , &Onorato, R. (1999). Socialidentity, personality, and the self-concept: A selfcategorization perspective. In T. R. Tyler, R. Kramer, & O. John (Eds.), The psychology of the social self. Hillsdale, NJ: Lawrence Erlbaum. Turner, J. C , & Reynolds, K. J. (2000). The social identity perspective in intergroup re-

87

88

FROM PREJUDICE TO INTERGROUP EMOTIONS

lations: Theories, themes, and controversies. In R. J Brown & S. Gaertner (Eds.), Blackwell handbook of social psychology: Intergroup processes (pp. 133-152). Oxford, England: Blackwell. Urban, L. M. & Miller, N. (1998). A theoretical analysis of crossed categorization effects: A meta-analysis. Journal of Personality and Social Psychology, 74, 894-908. Wenzlaff, E. M., & Wegner, D. M. (2000). Thought suppression. Annual Review oj Psychology, 51,59-91. Wright, S. C , Aron, A., McLaughlin-Volpe, T., & Ropp, S. A. (1997). The extended contact

effect: Knowledge of cross-group friendships and prejudice. Journal of Personality and Social Psychology, 73, 73-90. Yzerbyt, V. Y, Castaño, E., Leyens, J.-Ph., & Paladino, P. (2000). The primacy of the ingroup: The interplay of entitativity and identification. European Review of Social Psychology, 11, 257-295. Yzerbyt, V. Y, Dumont, M., Wigboldus, D., & Gordijn, E. (2000). I feel for us: A self-categorization account of intergroup emotions and action tendencies. Catholic University of Louvain at Louvain-la-Neuve. Manuscript submitted for publication.

6

Intergroup Encounters and Threat A Multi-Method Approach

JIM BLASCOVICH WENDY BERRY MENDES MARK D. SEERY

V

r

eridically assessing emotional responses that occur during intergroup interactions often proves difficult for researchers. Educational, institutional, and cultural socialization that promotes the value of, and sensitivity toward, ethnic and racial diversity may exaggerate differences between selfreported and actual emotions and attitudes toward members of minority groups. Not surprisingly, social psychologists have become increasingly, but not exclusively, attracted to less consciously controlled or implicit measures of affect within the context of intergroup interactions. Here, we focus on the importance of using multiple measures, including both less consciously controllable, or covert, and more consciously controllable, or overt, ones for the study of intergroup interactions. This chapter has four major sections. First, we provide a brief account of intergroup encounters within the context of our biopsychosocial model and cardiovascular measures of challenge and threat. Second, we describe four studies that examined perceivers' emotional and motivational reactions during social interactions with White and Black partners, which included both covert and overt measures. Third, we describe the results from a meta-analysis of these This research was supported in part by National Science Foundation Award SBR9596222, National Research Service Award MH12013, and a National Science Foundation Graduate Fellowship.

89

90

FROM PREJUDICE TO INTERGROUP EMOTIONS

studies organized within a multi-method matrix that demonstrates convergences and divergences among covert and overt measures as a function of the interaction partners race. Finally, based on the results of the meta-analysis, we review the utility of different types of measures and describe some apparent advantages of emotion-oriented covert measures over overt measures for the study of intergroup interactions.

A BIOPSYCHOSOCIAL PERSPECTIVE OF INTERGROUP INTERACTIONS In our approach, we capitalize on the value of covert psychophysiological indexes for assessing emotional and motivational responses during intergroup interactions (Blascovich, 2000). As Cacioppo, Tassinary, and Bernston (2000) noted, the most meaningful physiological indexes bear a one-to-one relationship to the psychological construct they purportedly index. A one-to-one relationship provides the strongest basis for inferring psychological meaning from physiological responses. This type of relationship allows for the relatively unambiguous interpretation of changes in physiology in terms of changes in psychological processes. Physiological indexes have several advantages over other types of measures. Specifically, they are on-line (i.e., parallel to in vivo behavior), covert, and can be continuous (Blascovich, 2000). These qualities allow researchers to track emotional and motivational changes continuously even during actual interaction episodes. Furthermore, the use of covert measures, such as physiological ones, reduces concerns about demand characteristics and self-presentational factors that can be evoked during intergroup interactions. Hence, physiological measures can provide veridical assessments of emotions during intergroup encounters. Two recent research efforts demonstrate the utility of psychophysiological indexes for examining perceivers' reactions to minority group members. The first, by Vanman and colleagues (Vanman, Paul, Ito, & Miller, 1997), employed facial electromyography (EMG) to assess affective reactions to White and Black targets. These researchers found evidence for greater negative affect (increased corrugator supercilli and decreased zygomaticus major activity—physiological responses validated as indexing affect; for a review see Blascovich, 2000) displayed by White participants when exposed to photographs of Black faces compared to White faces. The second, by Phelps and colleagues (2000), capitalized on recent advances in functional magnetic resonance imaging (fMRI) techniques to contrast activation occurring in the amygdala (a physiological response validated as an index of fear; LeDoux, 1998) of perceivers exposed to Black and White faces. Results of this study demonstrated that upon presentation of unfamiliar Black faces, more amygdalar activity was associated with greater implicit racial bias (e.g., implicit associates test and startle eyeblink).

MULTI-METHOD APPROACH

Challenge and Threat We also employ psychophysiological indexes to study intergroup interactions. We have and continue to do so within the context of our biopsychosocial model of challenge and threat (see Blascovich & Mendes, 2000, for an early review). Although the purpose of our work in this area initially was to test theoretical aspects of our model, it soon became clear that our model and research had import for the study of intergroup encounters (e.g., Blascovich, Mendes, Hunter, & Lickel, 2000; Blascovich, Mendes, Hunter, Lickel, & Kowai-Bell, 2001). Our theoretical model of challenge and threat (Blascovich & Mendes, 2000; Blascovich & Tomaka, 1996; Tomaka, Blascovich, Kelsey, & Leitten, 1993) draws heavily from the coping research (e.g., Lazarus & Folkman, 1991) and cardiovascular reactivity to stress literatures (for a review, see Blascovich & Katkin, 1993). Empirically, we have generally adopted a multi-method approach that includes the examination of cardiovascular (CV) responses, task performance, and self-reported evaluations of demands and resources, performance, affect, and subjective well-being to test our model of challenge and threat. Our challenge and threat model to date has been context-bound, studied only during motivated performance situations, which we define as episodes that are self- or goal-relevant, require instrumental cognitive responses, and are active rather than passive. Because many, if not most, motivated performance situations outside the laboratory occur in social contexts, we believe that actual social interactions provide a rich and ecologically valid context in which to study intergroup interactions. That is, work and school settings abound with examples of intergroup interactions that constitute motivated performance situations, such as negotiations, cooperative and competitive tasks, and, in some cases, conversations. Perceived challenge and threat can be identified by responses gathered from pre-task evaluations of demands and resources related to the task situation. In validational studies, individuals who perceived the demands of a situation as outweighing their personal resources were characterized as "threatened"; individuals who evaluated resources as exceeding demands were characterized as "challenged" (Tomaka et al., 1993). In subsequent iterations of the theory (Blascovich & Mendes, 2000), demand evaluations were broadened to include danger, uncertainty, and required effort; resource evaluations included perceptions of knowledge and abilities relevant to situational performance, as well as dispositional characteristics and external support. It is possible that in a given situation one of these elements can trigger high overall demand or resource evaluations, similar to the thinking of Lazarus and colleagues (Lazarus, DeLongis, Folkman, & Gruen, 1985) who argued, "no single variable . . . can stand for stress" (p. 777); or in our terminology, "threat" however, we simultaneously consider all demand and resource elements and their potentially additive or synergistic effects. Challenge and threat can also be indicated by specific CV responses. Fol-

91

92

FROM PREJUDICE TO INTERGROUP EMOTIONS

lowing Dienstbier's (1989) work on physiological toughness, activation of the sympathetic adrenomedullary (SAM) axis is implicated in positive coping, whereas SAM activation and activation of the pituitary-adrenal-cortical (PAC) axis are associated with "stress" or negative responses (we label the former reaction challenge and the latter reaction threat). In our terms, challenge is marked by SAM activation, which enhances cardiac performance, particularly left ventricular contractility and cardiac output, and decreases systemic vascular resistance. In contrast, threat is marked by activation not only of the SAM axis, again increasing contractility, but also activation of the pituitary-adrenal-cortical (PAC) axis, which inhibits decreases in systemic vascular resistance (Blascovich & Tomaka, 1996). Thus, different patterns of CV responses differentiate challenge and threat states. We use three cardiovascular responses, based on Dienstbier's work, to index challenge and threat: left-ventricular contractility (VC), measured as preejection period (PEP), or the time from the initiation of left-ventricular contraction until the opening of the aortic valve (VC = PEP x -1); cardiac output (CO), which is the amount of blood being pumped by the heart expressed in liters per minute; and total peripheral resistance (TPR), which is the amount of overall vasoconstriction or vasodilation occurring in peripheral blood vessels.1 Challenge responses are marked by significant increases in VC and CO from baseline coupled with a significant decrease in TPR, whereas threat responses are marked by increases in VC (although typically not as large as increases during challenge), no change or a decrease in CO, and no change or an increase in TPR (see Figure 6.1). These markers have been used successfully to investigate challenge and threat processes in many areas, including stigma (Blascovich et al., 2001), social facilitation (Blascovich, Mendes, Hunter, & Salomon, 1999), social comparisons (Mendes, Blascovich, Major, & Seery, 2001), and dispositions (Tomaka, Palacios, Schneider, Colotta, Concha, & Herrald, 1999).

APPLICATION OF CHALLENGE AND THREAT THEORY TO THE STUDY OF INTERGROUP INTERACTIONS In the past, we argued that our theoretical challenge and threat framework is well suited for the examination of emotionally charged responses that occur during dyadic interactions because such interactions often occur within the con1. We do not include heart rate (HR) as a specific component because HR contributes little to the differentiation of challenge and threat, though HR increases significantly during both. This is not surprising given the complexity of neural sympathetic and parasympathetic as well endocrine controls affecting HR. Nevertheless, HR itself is informative within our motivated performance situation paradigm; we use it as an indication of goal-relevance (Obrist, 1981).

MULTI-METHOD APPROACH

FIGURE 6.1. Cardiovascular responses indicating challenge and threat. .

text of motivated performance situations; that is, they are goal-relevant and involve instrumental cognitive responses on the part of interactants (Blascovich et al., 2000). Consequently, we brought our theory and cardiovascular indexes to bear on the issue of intergroup and intragroup dyadic interactions. To date, we have completed more than 20 experiments that include the examination of emotional and motivational responses that occur during dyadic interactions focusing on the differences between intergroup and intragroup pairings. In the following section, we integrate intergroup theories with our challenge and threat appraisal theory to generate predictions regarding emotional responses of interactants during an intergroup encounter. We then briefly describe the basic paradigm we used for these experiments followed by a more detailed consideration and analyses of several intergroup studies germane to the issue of intergroup emotion.

Challenge and Threat Responses During Intergroup Interactions Intergroup theories provide considerable grist for formulating predictions regarding challenge and threat responses during intergroup interactions. These theories suggest ways in which demands (i.e., danger, uncertainty, and required effort) and resources (knowledge and abilities, dispositions, external support) may be affected during intergroup encounters. Some theoretical perspectives address how perceived demands and resources are affected at the individual or

93

94

FROM PREJUDICE TO INTERGROUP EMOTIONS

interpersonal level, whereas other perspectives address how evaluations are affected at the group level. In our model, perceptions of either or both individual and group harm or loss might trigger high demand evaluations. Many theories suggest that intergroup interactions result in an increase in perceived demands. At an individual or interpersonal level, some theories suggest that intergroup interactions create anxiety or tension (Devine, Plant, & Buswell, 2000; Stephan & Stephan, 1985; Wilder, 1993). To the extent that such anxiety represents aversive psychological states, intergroup interactions can be regarded as dangerous. At a group level, social dominance (Sidanius & Pratto, 1993) and system-justification (Jost & Banaji, 1994) theories maintain that to the extent that individuals belong to groups perceived as lower status, they represent a danger to members of the dominant or powerful group in a culture. The second element of demand evaluations is uncertainty, which would typically function at the individual level. Uncertainty, referring to the novelty and unfamiliarity of outgroup partners, may be a critical factor during intergroup interactions. Due to the relative infrequency of outgroup compared to ingroup interactions (Charles, 2000), we can expect that the more novel or unfamiliar an interactant is the more uncertainty surrounding that interaction. This uncertainty then may trigger a variety of compensatory actions that direct attention from the task at hand, thus leaving fewer task-related resources. Thus, in this case, uncertainty may be inextricably linked to required effort. Vigilance efforts may increase during an intergroup interaction because the subtle nonverbal cues that govern two-way communication may be unfamiliar to interactants in a cross-cultural context (Gundykunst, 1984). Intergroup interactions may require increased cognitive effort in terms of self-monitoring because of additional or hidden agendas. At one extreme, perceivers may strive to present themselves as unbiased or nonprejudiced toward outgroup members (Devine, Evett, & Vasquez-Suson, 1996; Stephan & Stephan, 1985). At the other extreme, members of higher status groups than those of their interaction partners may seek to justify or preserve this imbalance (Jost & Banaji, 1994; Sidanius & Veniegas, 2000). Such an agenda requires perceivers to strive to perform in a clearly superior fashion to their partner. Finally, because intergroup interactions may evoke relevant negative stereotypes even in nonprejudiced individuals, increased effort may be expended to suppress stereotypes (Devine, 1989; Wyer, Sherman, & Stroessner, 2000). Perceived resources during an intergroup interaction may not offset the increased demands and in some cases may be diminished, especially resources associated with knowledge and abilities. At the interpersonal level, individuals may perceive that they do not know the most appropriate way to communicate during intercultural interactions (Wiseman, 1995). Insofar as individuals perceive outgroup partners to possess different conversational and interpersonal norms than their own, they may perceive less knowledge and abilities in terms of interaction skills with outgroup members, resulting in awkward and strained

MULTI-METHOD APPROACH

responses when confronted with an intergroup partner. At the group level, stereotype threat may adversely affect knowledge and abilities during an intergroup interaction if group membership is primed and negative stereotypes are relevant (Steele & Aronson, 1995). These arguments lead us to predict that interactions with outgroup members will result in greater perceived demands and potentially lower perceived resources for perceivers than the same interactions with ingroup members.

Basic Dyadic Interaction Paradigm Each experiment reported here begins with two same-sex undergraduates (a naïve participant and a confederate) waiting about 10 meters apart in a hallway outside the entrances to our social psychophysiology laboratory. The experimenter introduces the participants to each other and confirms that they do not know each other. The experimenter provides them information about each other as a means of ensuring their perceived similarity or dissimilarity as a particular experiment requires. Immediately following this initial meeting, the experimenter takes participants to separate rooms to complete a variety of study-related forms. In some studies, we bring the participants back together to exchange some additional limited, but informative background information about themselves. The target confederate provides a prescripted background designed to manipulate a key variable such as socioeconomic status, similarity, typicality, and so forth within a particular study protocol. In other studies, we do not implement the exchange of such background information, leaving participants only the visual information they received upon arrival such as their own and their partners' race, or whatever similarity or dissimilarity information was initially provided by the experimenter. Next, if they have been brought back together, participants are returned to their own recording rooms where appropriate sensors are applied for recording cardiovascular responses during the experiment. This is followed by an adaptation/rest period during which we instruct participants to sit at ease while we record baseline cardiovascular responses. Finally, we connect the participant's and confederate's recording rooms so that each can see and hear the other via audio speakers and video monitors. The participants then engage in several motivated performance tasks (e.g., speech delivery, word games) while cardiovascular responses are recorded. If multiple tasks are required, we disconnect the recording rooms to allow for a rest/recovery period prior to connecting the rooms again.

Relevant Intergroup Interaction Studies Our research program has included the study of how a variety of different partner characteristics affect perceivers' responses during dyadic interactions. For

95

96

FROM PREJUDICE TO INTERGROUP EMOTIONS

example, we have manipulated whether or not confederates were physically stigmatized by facial birthmarks, by their ostensible socioeconomic backgrounds, and even by their speech accents. We have employed confederates from a variety of racial and ethnic backgrounds, including Black, White, Asian, and Latino targets. In this review, we focus exclusively on perceive rs' responses during dyadic interactions comparing those involving Black to White partners.

Social interactions with Black versus White Male Partners. In one of our first intergroup experiments, we examined cardiovascular responses of non-Black male participants during social interactions with White or Black confederates (Mendes, Blascovich, Lickel, & Hunter, 2002). We predicted that non-Black participants interacting with Black partners would exhibit cardiovascular responses consistent with threat responses, whereas participants interacting with White partners would exhibit cardiovascular responses consistent with challenge responses. In addition to the race of the target, we also manipulated the target s socioeconomic status (SES) by having confederates describe their backgrounds as either advantaged or disadvantaged. We observed differences in cardiovascular responses based on the race and SES of the confederates. Specifically, participants paired with Black or disadvantaged partners exhibited cardiovascular responses consistent with threat during the speech and word-finding tasks; participants paired with White or advantaged partners exhibited cardiovascular responses consistent with challenge. Participants who were paired with Black-disadvantaged confederates exhibited larger threat reactivity than any of the other conditions (an additive effect of the two main effects). In addition, participants paired with Black partners performed worse during a cooperative word-finding task than participants paired with White partners. That is, participants cooperating with Black confederates generated fewer words than did participants paired with White confederates even though the confederates always performed at the same level. In contrast to their physiological and behavioral data, participants' selfreported attitudes painted a very different picture of the interaction. Participants interacting with Black confederates rated their partner more positively than did participants interacting with White confederates. That is, Black confederates were rated as more likable, independent, trustworthy, and hardworking than were White confederates. Participants also rated their partners on a variety of negatively valenced traits, and again Black confederates were rated as less unintelligent and unfriendly than were White confederates. Social Interactions with Black versus White Female Partners: Moderating Factors. In this second study (Experiment 3 in Blascovich et al., 2001), we employed the same paradigm and experimental factors as in the previous study with two changes. First, we used female perceivers and targets (i.e., confederates). Second, we recruited participants who had completed a variety of

MULTI-METHOD APPROACH

questionnaires in an earlier mass-testing session. These questionnaires focused on attitudes, beliefs, and contact regarding African Americans. Participants completed the Modern Racism Scale (MRS), the Motivation to Control Prejudiced Reactions Scale (MCPRS; Dunton & Fazio, 1997), and the Intergroup Contact Scale (adapted from Islam & Hewstone, 1993). The latter queried participants regarding the quality and quantity of contact with African Americans. The results of this experiment generally replicated the first study. Participants interacting with Black confederates during a cooperative word-finding task exhibited greater threat responses and performed worse than participants interacting with White targets. In addition, we found that responses from the MRS and the MCPRS were not related to cardiovascular responses during the actual interaction with a Black partner. That is, participants' self-reported attitudes regarding African Americans in general were not related to their cardiovascular responses. However, among participants interacting with Black partners, intergroup contact was significantly related to participants' cardiovascular responses during the word-finding task (see Table 6.1). More historical contact with African Americans was associated with greater VC, greater CO, and lower TPR—indicating less threat—during an actual interaction with a Black partner. The correlations among intergroup contact and cardiovascular responses during interactions with White partners were not significant. Hence, we would speculate that as familiarity increased then uncertainty decreased, leading to overall decreases in demand evaluations, and hence less threat. These findings point to the important role that quality intergroup contact may have in ameliorating otherwise threatening intergroup interactions. Social Comparisons with Black and White Partners. Social comparisons are often the inevitable consequences of cooperation. In our social interaction studies, the word-finding task allowed for the possibility for participants to compare their own performance with their partners' performance. Because of this inevitability, we carefully orchestrated the confederates' responses during the

TABLE 6.1. Correlations between Intergroup Contact and Cardiovascular Reactivity as a Function of Partner's Race Cardiovascular Reactivity

Historical Contact with African Americans Partner was White

VC CO TPR t

p stereotypes (which emerge as rationalizations or summaries of emotional reactions). The differentiation and specificity of responses to outgroups, and particularly their differentiation by context, raises a considerably broader point with major implications for the traditional views of prejudice as an attitude. We have argued that responses depend on the outgroup s appraised relationship to an ingroup, such as on the type of threat it poses (Fiske et al., this volume; Neuberg & Cottrell, this volume; Stephan & Renfro, this volume). The responses triggered by this relationship or degree of threat may have little to do with the outgroup s perceived intrinsic qualities (the outgroup s stereotype as commonly conceptualized). Indeed—and this is a major thrust of the chapter by Fiske et al.—there may be an inverse relation, with outgroups characterized by positive qualities such as strong, ambitious, smart, and so on being viewed as particularly potent competitors and exciting particularly negative reactions. If responses to social groups are based on those groups' perceived relations to the perceiver's own group, rather than on the evaluation of the outgroups' own inherent characteristics, a radical implication follows. At least for reactions to social groups, evaluation is not fundamental, but instead is secondary to the self/nonself distinction. Social psychologists at least since the 1920s have assumed that attitudes (i.e., positive or negative evaluations) are the primary dimension of meaning that people attach to objects or situations. This assumption naturally led to the view that evaluations of groups were driven by the positive or negative nature of the group's perceived characteristics (stereotypes; Eagly & Mladinic, 1989). However, we now see that the positivity of a trait or other group characteristic depends on whether that trait is attached to an ingroup, a nonthreatening outgroup, or a threatening, competitive outgroup (Brewer, 1979; Brewer & Alexander, this volume; Fiske et al., this volume). Supporting this point, studies by Perdue, Dovidio, Gurtman, and Tyler (1990) and Otten and Wentura (1999) have demonstrated that simply linking an object or group membership to the self automatically gives rise to a positive evaluation, at the im-

295

2 9 6 FROM PREJUDICE TO INTERGROUP EMOTIONS plicit as well as explicit level. It follows that evaluation is less fundamental as a dimension of meaning than is an object's relationship to the self or ingroup. We can go even further. We have said that traits and other group characteristics cannot be viewed as positive or negative in themselves but acquire evaluation only in terms of their appraised relationship to the ingroup. The same can be argued even regarding emotional reactions. For example, Fiske et al. (this volume) point out that in certain intergroup situations, subjectively positive emotions can be part of, and support, objectively negative and exploitative intergroup relations. Thus, paternalistic benevolence toward a nonthreatening, low status group can be part of a negative or exploitative intergroup scenario. This positive emotion can even function to stabilize that situation, through the implicit threat of withdrawal of the paternalistic treatment if the lower status group should begin to threaten the status quo. It seems that just as a group s quality (such as ambitiousness) can be either positive or negative depending on the context, so too an emotional response can be interpreted as positive or negative in its implications for an intergroup situation, independent of its subjective valence. In summary, chapters in this volume are clear on the many ways that reactions to outgroups are shaped by concrete intergroup relations and social structural factors. Outgroups elicit reactions that generally are not based on their perceived characteristics, but more fundamentally on their appraised relation to the ingroup. Emotions follow from those relational appraisals, and stereotypes are attributed to outgroups as rationale or justification for those emotions. From these ideas come a more radical suggestion, that evaluation is less fundamental as a dimension of reactions to groups than is the group s relation to the self: is it an ingroup, a nonthreatening outgroup, or a threatening outgroup?

IMPLICATIONS OF THIS PERSPECTIVE The general perspective outlined here has important implications both for researchers and for those who seek concrete ways to reduce prejudice in society. We recommend that researchers studying stereotyping, prejudice, and intergroup relations consider broadening their focus to assess emotional responses to ingroups and outgroups, as well as the traditional measures (e.g., ascription of stereotypic traits and evaluations). When possible, measures of emotional reactions should be supplemented with additional measures of intergroup appraisals and action tendencies, to give a fuller picture of the ways people assess and respond to intergroup situations. Finally, researchers should heed the message of the chapter by Blascovich et al. (this volume), whose studies show several dissociations between emotional responses (indexed by physiological measures and also by actual task performance) and verbal reports. In some cases, the people who experience the strongest threat reactions in intergroup encoun-

COMMENTARY 2 9 7 ters actually rate the interaction partner the most positively! In the clearest possible way, this finding suggests the importance of assessing emotional responses and not solely verbal reports in the study of prejudice and intergroup relations (cf. Guglielmi, 1999). The perspective advanced here also has implications for reducing prejudice in society. Based on the traditional conceptualization of prejudice as a negative evaluation, many programs seek to emphasize the positive aspects of outgroups (e.g., by focusing on positive stereotypes) or to provide concrete information intended to challenge negative stereotypes. Our perspective questions the basis of that seemingly commonsense approach (which in any case does not enjoy much success empirically; Stephan & Stephan, 1996). If prejudice against outgroups is essentially an emotional reaction, it is not directly based on group stereotypes (and so changing stereotypes may have little effect on it). Instead, programs should seek to change relevant appraisals of the outgroup s relationship to the ingroup (e.g., its potential as a competitor). Alternatively, they could intervene directly at the level of emotions, with misattribution or reattribution manipulations, or by encouraging the type of positive personal contact that reduces negative emotions such as anxiety (Stephan & Renfro, this volume). Finally, this perspective encourages the use of recategorization or friendship manipulations that have been shown to bring an individual outgroup member or the outgroup as a whole within the compass of the psychological self (Pettigrew & Tropp, 2000; Wright, Aron, McLaughlin-Volpe, & Ropp, 1997). These potential approaches to prejudice reduction have not been well explored to date, but their investigation and development should be encouraged by the underlying intergroup appraisal/emotion perspective shared by most of the chapters in this volume.

CONCLUDING COMMENTS We and the other authors represented in this volume have discussed diverse implications of the notion that group membership allows emotional and selfregulatory processes to operate at the level of the collective self, with important implications for intergroup relations. In closing, we would like to briefly discuss one tantalizing implication of this perspective that in a sense reverses its causal flow. We propose that self-regulatory processes operating at multiple levels may actually contribute to defining the boundaries of the psychological self. The affective and motivational consequences of self-regulatory systems are often subjectively accessible and noticeable, since one function of these systems is to grab our attention when important events occur. These subjective emotional experiences may at times be cues that tell us what is (and what is not) incorporated within the self. Since self-regulatory systems, by the hypothesis of this volume, operate at both individual and group levels, this process should oper-

2 9 8 FROM PREJUDICE TO INTERGROUP EMOTIONS ate in conceptually the same way at each level—and we could add a third, relational level as well (Sedikides & Brewer, 2001). Here are three examples. (a) I know it's my foot because when someone steps on it, it hurts. (b) I know she is my partner because when someone pays her a compliment, I feel warm inside. (c) I know it's my group because when someone disrespects us, I feel angryFurther research needs to be conducted to examine the possibility that self-regulation in this extended sense actually constitutes the subjective sense of self. It is but one example of the new areas opened up for empirical and conceptual investigation by the theoretically integrative endeavor represented in this book.

REFERENCES Arnold, M. B. (1960). Emotion and personality, Vol. 1: Psychological aspects. New York: Columbia University Press. Banaji, M. R., & Prentice, D. A. (1994). The self in social contexts. Annual Review of Psychology, 45, 297-332. Bodenhausen, G. V. (1993). Emotions, arousal, and stereotypic judgments: A heuristic model of affect and stereotyping. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 13-37). San Diego, CA: Academic Press. Bodenhausen, G. V, Mussweiler, T., Gabriel, S., & Moreno, K. N. (2001). Affective influences on stereotyping and intergroup relations. In J. P. Forgas (Ed.), Handbook of affect and social cognition (pp. 319-343). Mahwah, NJ: Lawrence Erlbaum. Brewer, M. B. (1979). Ingroup bias in the minimal intergroup situation: A cognitive-motivational analysis. Psychological Bulletin, 86, 307-324. Buck, R. (1994). Social and emotional functions in facial expression and communication: The readout hypothesis. Biological Psychology, 38, 95-115. Carver, C. S., & Scheier, M. F. (1998). On the s elf-regulation of behavior New York: Cambridge University Press.

Cosmides, L., & Tooby, J. (2000). Evolutionary psychology and the emotions. In M. Lewis & J. M. Haviland-Jones (Eds.), Handbook of emotions (2nd éd., pp. 91-115). New York: Guilford Press. Devine, P. G., Evett, S. R., & Vasquez-Suson, K. A. (1996). Exploring the interpersonal dynamics of intergroup contact. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motivation and cognition, Vol. 3: The interpersonal context (pp. 423-464). New York: Guilford Press. Eagly, A. H. (1987). Sex differences in social behavior: A social-role interpretation. Hillsdale, NJ: Lawrence Erlbaum. Eagly, A. H., & Mladinic, A. (1989). Gender stereotypes and attitudes toward women and men. Personality and Social Psychology Bulletin, 15, 543-558. Frijda, N. H., Kuipers, P., & ter Schure, E. (1989). Relations among emotion, appraisal, and emotional action readiness. Journal of Personality and Social Psychology, 57, 212228. Guglielmi, R. S. (1999). Psychophysiological assessment of prejudice: Past research, current status, and future directions. Personality and Social Psychology Review, 3, 123157. Higgins, E. T. (1989). Self-discrepancy theory:

COMMENTARY

What patterns of self-beliefs cause people to suffer? In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 22, pp. 93-136). San Diego, CA: Academic Press. Higgins, E. T (1998). Promotion and prevention: Regulatory focus as a motivational principle. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp. 1-46). New York: Academic Press. Higgins, E. T, Rholes, W S., & Jones, C. R. (1977). Category accessibility and impression formation. Journal of Experimental Social Psychology, 13, 141-154. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: SpringerVerlag. Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an interpersonal monitor: The sociometer hypothesis. Journal of Personality and Social Psychology, 68, 518-530. LeVine, R. A., & Campbell, D. T. (1972). Ethnocentrism: Theories of conflict, ethnic attitudes, and group behavior New York: Wiley. Linville, P. W (1987). Self-complexity as a cognitive buffer against stress-related illness and depression. Journal of Personality and Social Psychology, 52, 663-676. Mackie, D. M., & Hamilton, D. L. (Eds.). (1993). Affect, cognition, and stereotyping: Interactive processes in group perception. San Diego, CA: Academic Press. Mackie, D. M., & Smith, E. R. (1998). Intergroup relations: Insights from a theoretically integrative approach. Psychological Review, 105, 499-529. Otten, S., & Wentura, D. (1999). About the impact of automaticity in the Minimal Group Paradigm: Evidence from affective priming tasks. European Journal of Social Psychology, 29, 1049-1071 Perdue, C. W, Dovidio, J. F., Gurtman, M. B., & Tyler, R. B. (1990). Us and them: Social categorization and the process of intergroup bias. Journal of Personality and Social Psychology, 59, 475-486. Pettigrew, T. R, & Tropp, L. R. (2000). Does

intergroup contact reduce prejudice: Recent meta-analytic findings. In S. Oskamp (Ed.), Reducing prejudice and discrimination (pp. 93-114). Mahwah, NJ: Lawrence Erlbaum. Roseman, I. J. (1984). Cognitive determinants of emotion: A structural theory. Review of Personality and Social Psychology, 5,11-36. Sedikides, C , & Brewer, M. B. (Eds.). (2001). Individual self relational self collective self. Philadelphia: Psychology Press. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality and Social Psychology, 48, 813-838. Smith, E. R. (1993). Social identity and social emotions: Toward new conceptualizations of prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping: Interactive processes in group perception (pp. 297-315). San Diego, CA: Academic Press. Smith, E. R., Coats, S., & Walling, D. (1999). Overlapping mental representations of self, ingroup, and partner: Further response time evidence and a connectionist model. Personality and Social Psychology Bulletin, 25, 873-882. Stephan, W C , & Stephan, C. W. (1985). Intergroup anxiety. Journal of Social Issues, 41, 157-175. Stephan, W G., & Stephan, C. W (1996). Intergroup relations. Madison, WI: Brown & Benchmark Publishers. Tajfel, H. (1969). Cognitive aspects of prejudice. Journal of Social Issues, 25, 79-97. Turner, J. C , Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. S. (1987). Rediscovering the social group: A self-categorization theory. Oxford, England: Blackwell. Wiggins, J. S. (1996). An informal history of the interpersonal circumplex tradition. Journal of Personality Assessment, 66, 217-233. Wright, S. C, Aron, A., McLaughlin-Volpe, T, & Ropp, S. A. (1997). The extended contact effect: Knowledge of cross-group friendships and prejudice. Journal of Personality and Social Psychology, 73(1), 73-90.

299

Page Intentionally Left Blank

Index Aaker, J. L., 45 Abalakina-Paap, M., 201 Aberson, C. L., 34 Abrams, D., 1, 4, 34, 227-228, 241 Ackerman, J., 265n Action tendencies, 117-119 image theory, 214-215 role of emotions, 262-269 Actual group collective self-esteem and, 23-25 controlling for variables, 21-23 fear of negative evaluation, 25-26 regulatory focus, 36-38 vs. ideal group, 17-20 vs. ought group, 17-20 Aderman, D., 76 Adetoun, B., 227-228, 241 Admiration, 257-258 behavioral implications, 258 Adorno, T. W., 191, 193 Affect as basis of prejudice, 2, 4 as predictor of contact, 157-163 attitude structure, 153-154 biocultural approach, 265-284 image theory and, 209-225 incidental vs. integral, 286-287 increasing contact, 163-166 interracial attitudes, 155-157 racial attitudes, 154-155 role in intergroup behavior, 153-172 Affection, 259 Affective preferences higher-status groups, 239-241 Affective structure, 153-154 Affirmative action, 196

African Americans. See also Black-White interactions prejudice toward, 273-276 stereotypes of, 40-42, 247 "Afro-Caribbean" people stereotypes of, 40-42 Aggression, 270, 272 Agitation-related emotions, 16, 38-39, 44 group-based, 21-23 Akande, A., 227-228, 241 Alao, A., 227-228, 241 Alderfer, C., 176 Alexander, M. G., 9, 21, 213, 214, 216, 219, 223, 292-295 Allen, H. M., 160 Allen, V. L., 31 AUport, G. W, 1,4,32,45,156,191,193,227228, 232, 235-237, 241-242, 247-248, 250, 261 Ally image, 211, 214-215, 217, 219-223 Altemeyer, B., 201 Ambivalence, 3, 9-10, 291-293 about stereotypes, 248-252 implications of emotions, 260-261 reactions to ingroup, 234-235 reality of, 9, 247-264 system justification theory, 9, 227-245 Ambivalent Sexism Inventory, 249-250 Amodio, D., 41, 43 Anastasio, P. A., 166 Ancona, D. G., 176 Andersen, S., 175, 180, 182 Andreoletti, C., 250 Anger, 5, 10, 111, 116-117, 129, 270-272 behavioral implications, 258 contempt, 254-255

301

3 0 2 INDEX Anger (continued) image theory, 214-215 offensive tendencies, 116-122 self-categorization and, 75-84 social identity and, 7-8, 135-151 triggering, 119-122 Anti-Negro scale, 161 Anti-Semitism, 1, 57, 61-62, 186, 259 Pope apologizes for, 60 Antoniou, A. A., 121 Anutei, A.-M., 52, 57, 60-61 Anxiety, 39, 79-80, 115, 179, 258, 271, 287, 297 intergroup, 194 Apartheid, 61 Appraisal processes, 7, 89-109 Appraisal theory, 72-75, 112-113, 286-289 image theory, 210-215 Apprehension, 21-23 Argentina collective guilt, 61 Armstrong, T. L., 196, 201 Arnold, M.B., 112, 270,289 Aron, A., 69-70, 74, 127, 297 Aron, E. N., 69-70 Aronson, E., 232 Aronson, J., 44, 95, 199 Asch, S. E., 34, 249 Asher, T., 265n. , 267, 269, 273 Ashkenazi Jews stereotypes of, 238 Asian Americans, 195-196 (See also Japanese Americans) as "model" minority, 2-3 prejudice toward, 273-276 stereotypes of, 240, 247, 250, 252, 257-259 Athenstaedt, U., 255 Attitudes, 295-296 affect-based, 292 as predictors of contact, 157-163 belief-based, 292 increasing contact, 163-166 interracial, 155-157 racial, 154-155 role in intergroup behavior, 153-172 structure, 153-154 Attitudinal theory emotional theory parallels, 4-5 limits of, 2-3 Attributes group, 20-21, 43 individual vs. social, 17-20

Attribution theory, 229 Aufderheide, B., 112 Australia collective guilt, 52-53, 60-61 Avoidance, 270 Avolio, B. J., 250 Bachman, G., 195 Bachman, R., 166, 173 Bakan, D., 249 Banaji, M. R., 90, 94, 100, 106, 137, 227-229, 236, 239, 241, 285 Banner, M. J., 112 Barbarian image, 211-212, 214-215, 219-223 Barchas, R, 268 Bargh, J. A., 41 Barkan, E., 51, 53, 59, 61-62 Barnett, M. A., 255 Baron, R. M., 271 Barrett, G. J., 250 Bar-Tal, D., 57 Basking in Reflected Glory, 14-15, 73 Batson, C. D., 59, 69, 155, 164, 167 Baumeister, R. F., 57 Beach, K. R., 8, 292 Bednar, L. L., 155, 164, 167 Behavioral relations, 1-2, 8-9, 102, 115-116 affect in intergroup behavior, 153-171 biocultural approach, 265-286 higher-status groups, 239-241 image theory, 209-225 implications of emotions, 257-258 intensity of, 277-278 manifestation of bias, 33 outgroup paranoia, 173-189 perceived threat, 247-264 regulatory focus, 36-38 role of threat, 191-207 system justification motive, 227-245 Behavioral structure, 153-154 interracial attitudes and, 155-157 Bell, D. W, 196 Bernston, G. G., 3, 32, 90 Berrenberg, J. L., 195 Bettencourt, A., 196 Biernat, M., 114-115, 196 Billig, M. G., 31, 33 Bilsky, W, 114 Biocultural approach, 265-284, 291 acute environmental events, 279-280 empirical findings, 273-276 group-level resources, 277-278

INDEX 3 0 3 intensity of emotions, 277-278 investment in group, 278-279 moderator variables, 276-280 nested theories, 280-281 overview, 267-269 quality of emotions, 278 role of emotions, 269-272 threat hierarchy, 276-277 Biopsychosocial framework, 7, 292 assessing threat, 90-92 intergroup interactions, 92-101 meta-analysis, 101-106 Bizman, A., 5, 17, 195, 280, 289 Black-White interactions, 95-106 evaluator feedback, 99-100 expressing emotions, 137-138 female partners, 96-97 image theory, 216-223 increased willingness, 163-166 intergroup emotion theory, 127-128 interracial attitudes, 155-163 male partners, 96 opposition to equality, 235-236 perceived relationships, 218 social comparisons, 97-99 Blaine, B., 14, 173, 181, 229, 232-233, 242 Blake, R. R., 173 Blanton, H., 228 Blanz, M., 37 Blascovich, J., x, 7,90-93,96,98-99,137,149, 292 Bobo, L., 114, 156-157, 160-161 Bodenhausen, G. V, 2, 23, 37, 71, 135, 137, 286-287 Boen, E, 73-74 Bogardus Social Distance Scale, 162 Bonacich, P., 174-175 Bond, R. N., 19, 41 Boniecki, K. A., 115, 196 Bootzin, R., 158 Borden, R. }., 14, 50, 73, 126 Boulding, K., 209-210 Bourhis, R. Y., 34 Boyle, R., 174-175 Bradac, J. J., 250 Branscombe, N. R., 6, 15, 49-53, 55-58, 60, 63, 74-75, 81, 115, 125, 196, 201, 230231, 289 Brauer, M , 216 Brazy, P. C., 5-6, 36, 38-39, 41, 280, 289 Brehm, J. W, 58-60 Brehm, S. S., 76

Breinlinger, S., 113 Brewer, M. B., 2, 9,14, 31-32, 36, 70, 98, 111, 173n, 174, 213, 214, 216, 219, 223, 227228, 265, 267-268, 286, 288, 291-295 Brigham, J. C , 155-156, 158, 166 Britt, T. W., 115, 196 Broadnax, S., 14, 173, 181, 229, 232-233, 242 Brockner, }., 177, 229 Bromley, R, 248 Bromley, S., 1 Bronfenbrenner, U., 211 Brown, I. D., 50 Brown, J. D., 34 Brown, L. M., 115, 196 Brown, R. J., 2, 31, 36, 40-41, 111, 265 Brown, R., 113, 115, 128, 130, 236, 241, 279 Brown, S. L., 69-70 Brunner, A., 227-228, 241 Buck, R., 290 Buidin, G., 200 Bundy, R. R, 31, 33 Burgess, D., 184, 228-235, 236-239, 241 Burt, R. S., 185 Buruma, I., 64, 56 Buswell, B. N., 94 Cacioppo, J. T., 3, 32, 90 Cadinu, M. R., 70 Calmness, 21-23 Cameron, J. E., 115 Cameron, N., 175 Campbell, D. T., 32, 191-192, 239, 267, 278, 294 Canadians collective guilt, 53 prejudice toward, 273-276 Canarians, 139 Caporael, L. R., 268 Career women stereotypes of, 247, 249-250 Carnevale, P., 176 Carter, J. E., 259 Carvallo, M. R., 184, 228, 236, 239-240 Carver, C. S., 16, 23, 28, 270, 289 Castaño, E., 70 Categorization processes. See also Self-categorization appraisal, 7, 89-109 intergroup interactions, 6, 67-88 role in collective guilt, 6, 49-66 Cejka, M. A., 3

3 0 4 INDEX Chaiken, S., 2, 154, 159, 162, 166 Challenge and threat, 91-92 intergroup interaction responses, 93-95 vs. threat, 7, 89-109 Chandler, C , 253, 255 Charles, C. Z., 94 Cheerfulness, 35-36, 38-39, 77-78 Chen, E. S., 230, 241 Chile collective guilt, 61 China, 279 prejudice toward, 273-276 Chipeta, K., 227-228, 241 Christians prejudice toward, 273-276 stereotypes of, 252, 257-258 Cialdini, R. B., 14, 50, 69, 70, 73, 126 Clark, L. A., 99 Clark, M. S., 44 Clason, D. L., 196, 199 Claypool, H. M., x, 112 Clinton, W. J., 60 Clore, G. L., 27 Coats, S., 43, 70, 112, 286 Cognitive dissonance, 229-231 reduction of, 231-233 Cognitive preferences higher-status groups, 239-241 Cognitive structure, 153-154 Colby, K. M., 175-176 Collective guilt, 6, 49-66, 125-126, 289-290, 294 antecedents to accepting, 53-56 attenuating factors, 56-59 augmenting factors, 59-60 distinguishing from personal, 51-53 political orientation, 52 political ramifications, 60-63 system justification theory, 230-231 vs. moral outrage, 54-55 Collective regulatory focus, 45 Collective Self-Esteem Scale, 14, 24 Collective self-esteem, 23-25, 201 ingroup bias, 33-35 scale, 14, 24 Collectivist societies, 279 Collins, N., x Collins, R. L., 34 Colotta, M., 92 Compassion, 270 Competence, 249-254, 257-258

Competitive threat, 293-296 envy, 254 Component models, 2 Concha, J. B., 92 Conditioning mechanism, 2, 287 Conklin, L., 69 Consensual stereotyping, 236-239 system justification function, 238-239 Conspiracy theories, 181-182, 185, 201 Contempt, 247, 254-255, 257-258, 272, 292 behavioral implications, 258 image theory, 214-215 Cook, S. W., 155 Coon, C. S., 267 Cooper, J., 137 Corenblum, B., 196 Cornielle, O., 200 Cosmides, L., 270, 291 Cottam, R., 210 Cottrell, C. A., 10,198,273,286,291-293,295 Craig, T., 201 Crocker, J., 14, 23-25, 34, 45, 98, 135, 173, 181, 201, 228-229, 232-233, 241-242 Crosby, F. J., 1, 248 Crowe, E., 35 Cuddy, A. J. C , 9-10, 50, 136, 148-149, 234235, 241, 248, 250-251, 261, 291-296 Cunningham, W. A., 90, 100, 137 Daly, J. P., 229 Dardenne, B., 227-228, 241 Darley, J. M., 141 Darwin, C , 269 Davidson, R. J., 266, 269 Davies, J. C , 233 Davis, C , 179 Davis, M. H., 69 Dawes, R. M., 182 Dawson, K., 69 Defensive tendencies triggering, 122-124 vs. offensive, 116-117 Dejection-related emotions, 16, 35-36, 38-39 group-based, 21-23 DeLongis, A., 91 Demand evaluations, 94-95 Demitrakis, K., 196 Demoulin, S., 7-8, 139, 145, 290 Dependent image, 211-212, 214-215, 217, 219-223 Depersonalization, 57, 70-71 Depression, 82-83

INDEX 3 0 5 Désert, M., 7-8, 290 Deutsch, M., 175 Devereux, E., 201 Devine, P. G., 41, 43, 50, 94, 105, 135-137, 155, 164, 167, 287 Devos, T., x, 7, 15, 31, 33, 63, 75, 112, 117, 128, 136, 209, 214-215, 221, 265-166, 277-278, 280, 290, 292, 295 Diaz-Loving, R., 114, 196 Dienstbier, R. A., 92 Dietze, C , 33 Differentiated reactions, 1-12, 291-293 Dijker, A. J. M., 73, 115, 136-137, 144, 165, 253-255, 257-258, 265n Dijksterhuis, A., 227-228, 241 Dion, K. L., 113, 115, 162 Disabled persons pity for, 253-254 stereotypes, 247, 250, 252, 257-258 Disappointment, 21-23 Discrepancies implications of, 31-48 social, 13-30 theory, 5 Discrimination as evaluation-consistent behavior, 2 prejudice cause of, 1-2 system justification theory, 227-245 Disgust, 5, 10, 111, 194, 265-255, 270-272, 295 image theory, 214-215 Dissatisfaction, 21-23 Distrust, 173-189, 201, 229-230 Dodge, K. A., 182 Dominant ideology theory, 236-239 Doosje, B., 6,15,50-53,55-56,58-60,63, 7475, 81, 121, 125, 228, 230-231, 289 Dovidio, J. R, 4, 8, 69,106,135,137-138,142, 145, 154-156, 159, 164, 166-167, 194, 196, 204, 292, 295 Downs, D. L., 285 Dubé-Simard, L., 113 Duckett, J., 32 Dumont, M , 6-7, 63, 81, 85, 290, 292-293 Dunton, B. C., 97, 101 Duran, A., 114, 196, 199 Dyadic interaction paradigm, 95 Eagly, A. H., 154, 162, 294-295 Eastern Europe stereotypes, 250 Eayrs, C. B., 258

Eberhardt, J. L., 135 Eckes, T., 227-228, 241 Eckman, P., 266, 269-270 Edwards, K., 2, 163-164 Egan, T. D., 240 Ego justification, 230, 233-234 Ekman, P., 146 Elderly people pity for, 253-254, 257-258 stereotypes, 247, 250 Electromyography, 90, 100-101 Ellemers, N., 50, 55, 115, 121, 235, 237, 241 Elliot, A. J., 50, 105 Ellis, N., 258 Ellsworth, P. C., 4, 15, 72, 112, 147, 209, 214, 286 Emotion, 135-151, 285-291 agitation-related, 16 appraisal theories, 112-113 as basis of prejudice, 4 as predictor of contact, 157-163 aspect of prejudice, 291-293 attitude structure, 153-154 biocultural approach, 265-284 collective guilt, 50-66 competition and, 247-264 decoding, 143-148 dejection-related, 16 determined by intergroup interactions, 8-9 (chaps 11-12) differentiated reactions, 5 expressing, 137-143 , 290 group-based, 14-15 image theory and, 209-225 increasing contact, 163-166 intensity of, 277-278 intergroup, 67-88, 72-75, 111-134 interracial attitudes, 155-157 manifestation of bias, 33 perceived status effect, 247-264 perception of, 290 racial attitudes, 154-155 regulatory focus, 35-36 role in intergroup behavior, 153-172 self-regulatory, 5, 13-154 theory, ix-x, 4-10 Emotional manifestation of bias regulatory focus, 38-39 Empathy anger and, 75-76 impact of perspective-taking, 67-88 Enemy images, 210, 211, 214-215, 219-223

306

INDEX Envy, 136, 247, 254, 256-259, 272, 292 behavioral implications, 258 image theory, 214-215 Erickson, G. A., 14-15 Ervin, K. S., 158, 196 Esses, V. M., 4,8,114-115,135,153-156,161, 166-167, 193, 196, 201, 265, 292 Ethiopians prejudice toward, 273-276 Evaluation. See also Appraisals ambivalence in, 3 fear of negative, 25-26, 43-44 manifestations of bias, 32-33 multiple, 3 Evett, S. R., 94, 137, 287 Evolutionary perspective, 10, 265-284, 291 acute environmental events, 279-280 empirical findings, 273-276 group-level resources, 277-278 intensity of emotions, 277-278 investment in group, 278-279 moderator variables, 276-280 nested theories, 280-281 overview, 267-269 quality of emotions, 278 role of emotions, 269-272 threat hierarchy, 276-277 vs. social identity perspective, 10 Expansion of the self, 69 Expósito, E, 227-228, 241 Extended self, 289-291 Fabrigar, L. R., 3 Fazio, R. H., 3, 50, 97, 101 Fear of Negative Evaluation Scale, 16, 25-26 Fear, 5, 10, 82-83, 111, 116-117, 129, 194, 270-272 defensive tendencies, 116-117, 122-124 image theory, 214-215 of negative evaluation, 25-26, 43-44 outgroup paranoia, 173-189 triggering, 122-124 Fein, S., 1, 34, 44, 174-175, 199, 227-229 Feldman, R. S., 138, 149 Feminists ideology, 229 prejudice toward, 273-276 Fenigstein, A., 175, 178-180 Ferrari, L., 210 Ferreira, M. C., 227-228, 241 Festinger, L., 34 Feys, J., 73-74

Fincham, F , 33 Finchilescu, G., 34 Finlay, K., 155, 164, 167, 195 Fischer, K. W., 50 Fischerkeller, M., 210 Fiske, S. T., 2, 9-10, 31, 111, 135-136, 148149, 176, 184, 204, 227-228, 231, 234235, 239, 241, 248-251, 256-259, 261, 291-296 Flament, C., 31, 33 Foddy, M., 227-228, 241 Folkman, S.,91, 112, 178,289 Ford, T. E., 115, 135, 154, 158 Form, W. H., 248 Freeman, S., 14, 50, 73, 126 Frenkel-Brunswik, E., 191, 193 Fridlund, A. }., 139 Friedman, R., 21, 27, 35-36 Friend, R., 16, 25 Frijda, N. H., 4,50, 72, 76,112-113,115-116, 214, 254, 257-258, 265, 270, 286 Funayama, E. S., 90, 100, 137 Fundamentalist Christians prejudice toward, 273-276 Gabriel, S., 2, 23, 287 Gaertner, S. L., 2, 8, 106, 135, 154-156, 159, 164, 166-167, 194, 204, 292 Galinsky, A. D., 71-72, 167 Gamble, I., 115 Garcia, A., x Gardner, W. L., 14, 32, 45 Garst, E., 69 Garstka, T. A., 57 Gatenby, J. C., 90, 100, 137 Gaunt, R., 139 Gay men prejudice toward, 273-276 Gender differences Black-White interactions, 96-97 collective guilt, 57-58 George, G., 228 Gerard, D. A., 158 Germany collective guilt, 56, 61 Nazis, 1, 56, 62, 186, 259 reparations, 62 Gibson, J. J., 271 Gilbert, D. T., 182 Gilens, M., 160 Gilovich, T., 140, 180, 182 Giner-Sorolla, R., 2, 159, 162, 166

INDEX 3 0 7 Glass, D. C , 3 Glick, P. S., 9-10,136,148-149,184,227-228, 231, 234-235, 239, 241, 248-251, 258261, 291-296 Goal attainment, 27-28 Goethals, G. R., 69, 210 Goldhagen, D. J., 51 Gordijn, E. H., 6-7, 63, 76, 78, 81, 85, 290, 292-298 Gore, J. C , 90, 100, 137 Graham, S., 253, 255 Grant, P. R., 113 Green, J. A., 158 Greenberg, J., 179, 199 Greenland, K., 115 Greenwald, A. G., 106, 240 Gregory, W. L., 201 Groenewoud, J. T., 128 Gross, J. T., 1 Group justification, 230, 233-234 Group membership appraisal processes, 89-109 collective guilt, (chap 4) intergroup interactions, 67-88 investment in, 278-279 role in emotional reactions, 6 (chaps 4-6) social discrepancies, 25-26 Group threats, 197-199, 272 integrity, 272 resource, 272 Grove, J. B., 239 Gruen, R., 91 Gruenert, M., 33 Gudykunst, W. V, 194, 201 Guerra, P., 167 Guglielmi, R. S., 106, 137, 148-149, 297 Guilt, 270, 272 {See also Collective guilt) defining, 50-51 personal, 51-53 Guimond, S., 113 Gundykunst, W. B., 94, 106 Gurtman, M. B., 295 Haddock, G., 114-115, 135, 153-155, 161, 166, 193, 265 Haeger, G., 33 Haidt, J., 255, 258 Haines, E. L., 184, 228-229, 241 Hamilton, D. L., x, 114,286 Hamilton, W. D., 268 Hamstra, I. }., 3 Happiness, 21-23, 73, 77-80, 82-83, 270

Hardin, C. D., 34 Hardin, R., 173n Harel, I., 13 Harmon-Jones, E., 41, 43, 155, 164, 167 Harvey, O. J., 174 Haslam, S. A., 50, 55, 70 Hass, R. G., 3 Hastedt, C., 112 Havel, V, 229 Healy, M., 34 Heider, E, 68, 73-74, 210 Helmreich, R. L., 249 Hendres, D., 52, 57, 60-61 Hendricks, M., 158 Henry, S., 50, 70-72, 112 Hermsen, S., 76 Herrald, M. M., 92 Herrmann, R. K., 210, 213-214, 216, 219, 223 Hetts, J. J., 157, 228 Hewstone, M., 33, 97, 115, 128, 196, 204, 236 Heyman, R., 69 Hierarchical trust dilemmas, 173-189, 294 as context for outgroup paranoia, 176-178 outgroup paranoia in, 178-183 Higgins, E. T., 5-6, 13, 15-17, 19, 21-28, 3436, 38-39, 50, 270, 280, 285-287, 289 Higgins' Selves Questionnaire, 17-18 Highberger, L., 155, 164, 167 Higher-status groups preferences for, 239-241 Hilden, L. E., 50 Hilton, }. L., 174, 229 Hinkle, S., 236, 241 Hirt, E. R., 14-15 Hispanic Americans, 240 Ho, C., 113 Hodge, C. N., 158 Hodges, S. D., 3 Hofstede, G., 201 Hogg, M. A., 1, 4, 14, 34, 50, 55, 70, 72, 112, 227, 230, 277n, 280, 286 Holocaust, 56-57, 61 reparations, 62 Holsti, O., 210 Homeless persons stereotypes, 252, 257-258 Homophobia, 255 Hood, W. R., 174 Hopelessness, 21-23 Hough, J. C., 217

3 0 8 INDEX Housewives pity for, 253-254 Huber, J., 248 Hughes, M., 161 Hummert, M. L., 57 Hunter, S. B., 91-93, 96 Hunyady, O., 231, 241 Hur, T., 21 Hurh, W. M., 250 Hutchings,V L., 114 Hymes, C , 35 Hypervigilance, 178-183 Ibarra, H., 240 Ideal group collective self-esteem and, 23-25 controlling for, 21-22 fear of negative evaluation, 25-26 regulatory focus, 36-38 vs. actual, 17-20 Idealization, 229-230 Identity threats, 10, 265-284 Imada, S., 255, 258 Image theory, 9, 209-225, 292, 294-295 appraisals, 210-212 emotions and, 214-215 experiment, 213-214 race relations, 216-223 Imhoff, H. J., 155, 164, 167 Imperialist image, 211-212, 214-215, 217, 219-223 Implicit Association Task, 239-240 Implicit associations, 40^42 Implicit preferences, 239-241 Incidental affect, 286-287 Ingerman, C , 34 Ingram, J. M., 158 Ingroup bias, 31-38 as mood regulation, 44 behavioral manifestations, 33, 36-38 collective, 45 decoding emotions, 143-148 discrepancies, 25-26 evaluative manifestations, 8-39, 333 expressing emotions, 137-143 fear of negative evaluation, 25-26 focus of, 35-36 implicit associations and, 40-42 implicit preferences for, 239-241 motivational determinants, 33-35 regulatory focus, 35-39 situational differences, 39-42

Insko, C. A., 2, 174 Integral affect, 286-287 Integrated threat theory, 9, 191-207, 280 antecedents, 195 intergroup anxiety, 194 negative stereotypes, 194 prejudice, 194-195 realistic threats, 192-193 revised, 196-203 support for, 195-196 symbolic threats, 193-194 Interdependency, 278-279 Intergroup Contact Scale, 97 Intergroup emotions theory, ix-x, 111-134, 277-278, 280, 295 and prejudice, 127-128 anxiety, 194 biocultural approach, 265-284 competition and, 247-264 defensive action tendencies, 122-124 defined, 112-113 direct evidence for, 116-117 image theory and, 209-225 indirect evidence, 113-116 interpersonal contexts, 125-127 offensive action tendencies, 119-112 perceived status effect, 247-264 triggering anger, 119-122 triggering fear, 122-124 vs. action tendencies, 117-119 Intergroup interactions ambivalence and, 9-10, 247-264 attitude as predictor of, 157-163 attitude structure, 153-154 biopsychosocial perspective, 90-92 categorization processes, 6, 67-88 challenge and threat theory, 92-101 collective self-esteem, 23-25 construing, 200 cultural dimensions, 201 decoding emotions, 143-148 determining emotions, 8-9, 191-225 discrepancies theory, 31-48 emotions affecting, 8, 67-88, 153-171 evolutionary perspective, 10, 265-284 experiencing emotions, 111-134 expressing emotions, 137-142 fear of negative evaluation, 25-26 group attributes, 27-21 group-based agitation, 21-23 group-based emotions, 14-17 image theory, 9, 209-225

INDEX 3 0 9 increasing, 163-166 individual difference variables, 201 interracial attitudes, 155-157 meta-analysis, 101-106 new research, 3-10 outgroup paranoia, 8, 173-191 overly personalistic construal, 180, 185 racial attitudes, 154-155 role of affect, 153-172 self-categorization and, 67-88 self-other overlap, 68-72 significant "others," 26-27 situational factors, 202 social structures and, 293-296 studies, 95-101 system justification motive, 9, 227-245 threat and, 89-109 Internal motivation prejudice and, 41-42 Internalized inferiority, 236, 239-241 Interracial attitudes, 155-157 Intragroup relations, 265-284 Iraqis prejudice toward, 273-276 Isen,A. M., 167 Islam, M. R., 97, 115, 196, 204 Israel actual vs. ought group, 17-20 collective guilt, 56 German reparations, 62 group attributes, 20-21 group discrepancies, 21-23 Internal Security Agency, 20 Palestinian conflict, 13 significant "others," 26 stereotypes, 238 Italy stereotypes, 238 Ito, T. A., 90, 100 Iyer, A., 58 Izard, C. E., 270 Jackman, M. R., 239, 259, 261 Jackson, L. A., 158, 196 Jackson, L. M., 196, 201 Jacobson, C. K., 160 Janis, I. L., 177-178 Janoff-Bulman, R., 178-179

Japan

collective guilt, 56, 61 prejudice toward, 273-276 Jaspars, J., 33

Jerdee, T. H., 250 Jervis, R., 173 Jetten, J., 37, 71, 228 Jews prejudice toward, 273-276 stereotypes of, 247, 250, 252, 257-258 Johnson, B. T., 155-156, 166 Johnson, D. W., 204 Johnson, R. T., 204 Jones, C. R., 286-287 Jones, E. E., 1, 182 Jones, W. H., 52 Jose, R E., 112, 121, 209 Jost, J. T., 8-9, 94, 183-184, 227-241, 290292, 294 Jost, T., 184 Joy, 21, 270 Judd, C. M., 167, 204, 216 Just-world theory, 229 Kabat, L. G., 175, 180, 182 Kahn, R. L., 173, 180 Kang, M., 69 Kappen, D. M., 52-53 Katkin, E. S., 91 Katz, I., 3, 260 Katz, J., 3 Katz, L. B., 76 Kawakami, K., 113, 204 Kelley, H. H., 34, 175, 181 Kelley, J., 161 Kelly, C., 113 Kelman, H. C., 26 Kelsey, R. M., 91 Kennedy, C., 14-15 Kessler, S., 33 Kim, H.-J., 227-228, 241 Kim, K. C., 250 Kinder, D. R., 3, 114, 156-157, 160 Kipnis, D., 177 Kirkland, S., 199 Kitano, H. H. L., 250 Kitayama, S., 45 Kivetz, Y., 238 Kleck, R. E., 145 Klein, R., 16, 19, 21, 26 Klein, S., x Klein, T. R., 155, 164, 167 Kluegel, J. R., 248 Kluegel, Smith, 230-231, 236, 241 Kok, G., 253, 255, 258 Kooman, W., 115, 253, 255, 258, 265

3 1 0 INDEX Kovel, J., 154 Kowai-Bell, N., 91, 96 Kraft, D., 180 Kramer, R. M., 8-9,173^177,179-182,185,201, 229, 232-233, 235, 242, 290-292, 294 Kraus, S., 216 Kravitz, D. A., 160 Krotman, S., 17 Kruglanski, A. W., 34 Kugler, K. E., 52 Kuipers, P., 50, 72, 76, 113, 116, 214, 286 Kurzban, R., 269 Laing, R. D., 184 Lalonde, R. N., 115 Lameiras, M., 227-228, 241 Lane, C , 173 Lane, R. E., 229, 242 Larrick, R. P., 181 Larsen, K. S., 161 Larson, D. W., 173 Latané, B., 141 Lau, R. R., 160 Lawrence, J. W., 16, 23 Lazarus, R. S., 4, 91, 112, 178, 270, 289 Leach, C. W, 58 Leakey, R. E., 267-268 Leary, M. R., 269, 285 LeDoux, J., 90 Lee, A. Y., 45 Leitten, C. L., 91 Lemyre, L., 228 Lepore, L., 40-41 Lerner, M. J., 76, 80, 230, 241 Lettgen, S.,33 Leung, K., 202 Levenson, R. W., 270 LeVine, R. A., 32, 191-192, 278, 294 Levinson, D. J., 191, 193 Lewin, K., 34, 239 Lewin, R., 267 Lewis, B. P., 69-70 Lewis, M., 50, 56 Leyens, J.-P., 290 Li, K., 253 Lickel, B., x, 91, 93, 96 Lin, M., 250, 259 Lind, E. A., 177, 241 Lindskold, S., 174 Linville, P. S., 286 Lombardi, W. J., 41 Lopez, W. L., 227-228, 241

Lowery, L., 255, 258 Luce, C , 69-70 Luhtanen, R., 14, 23-25, 45, 173, 181, 201, 229, 232-233, 242 Lyon, D., 199 Lyubomirsky, S., 179 Mackie, D. M., ix-x, 1-2, 7, 10, 15, 31, 33, 36, 49, 63, 75, 112, 114, 117, 122, 125, 127129, 136, 167, 173n, 209, 214-215, 221, 227, 265n, 265-266, 277-278, 280, 286, 290-292, 295 Macrae, C. N., 37, 71 Maghrebis, 144 Magnusson, J., 253, 255 Maio, G. R., 161, 196 Major, B., x, 92,98-99,135,183-184,228,233, 241 Maner, J., 265n Mann, L., 177-178, 267 Manstead, A. S. R., 15, 51-53, 55-56, 58, 60, 74-75, 81, 125 Markus, H., 26, 45 Marquez, J. M., 55 Martin, J., 115 Martin, T., 196 Martinez, C. M., 114, 195 Masser, B., 227-228, 241 Maurice, J. K., 250 McArthur, L. Z., 271 McCarthy, J. D., 233 McConahay, J. B., 3, 114, 155, 157, 160-161, 193, 217 McCoy, S., 99 McFarland, C , 138, 140 McGarty, C , 6,50,52,57,60-61, 70,230-231, 289 McGhee, D. E., 240 McGraw, K. M., 34, 229-230, 236 McGroarty, C , 250 McGuire, W. J., 230 McKenna, C , 69 McLaughlin-Volpe, T., 74, 127, 297 McNatt, P. S., 196 Medvec, V. H., 140, 180 Meertens, R. W., 114, 154 Mendes, W. B., 7, 91-93, 96, 98-99, 137, 292 Meritocracy, 230-231 Mesquita, B., 147 Messick, D. M., 173n, 181 Meta-analyses, 101-106, 157-159 Mexican Americans, 195-196

INDEX prejudice toward, 273-276 Mexicans prejudice toward, 273-276 Meyerson, D., 179 Middle-class Americans stereotypes, 247, 252, 255, 257-258 Mikula, G., 255 Millar, K. U., 163 Millar, M. G., 4, 155-157, 163 Miller, D. A., 122, 125, 127 Miller, D. L., 52-53, 60, 160-161, 167 Miller, D. T., 138, 140, 174, 229 Miller, D., x Miller, N., 76, 90, 100 Mills, J., 44, 232 Milne, A. B., 37, 71 Mitchener, E. C., 155, 164, 167 Mladinic, A., 154, 227-228, 241, 295 "Model" minority groups, 2-3 Modern Racism Scale, 97, 155, 158 Moghaddam, F. M., 115,237 Moniz, A., 254 Montada, L., 53-54, 58 Monteith, M. J., 50, 105, 155, 164, 167 Montenegro, 67 Mood regulation, 44 Moral outrage, 54-55 Moreno, K. N., 2, 135, 137, 287 Moretti, M. M., 16, 23-24 Morris, M. W, 181 Moscovici, S., 59, 173, 181 Moskowitz, G. B., 71-72, 167 Mosso, C., 228, 230, 233, 236, 238 Motivation to Control Prejudiced Reactions Scale, 97 Motivation, 289-291 Mouton, J. J., 173 Moya, M., 227-228, 241 Mphuthing, T., 32 Mucchi-Faina, A., 227-228, 241 Mudditt, A., x Mueller, A., 279 Mullen, B., 279 Mullin, B.-A., 230 Multifactor Racial Attitudes Inventory, 155, 158, 161 Mummendey, A., 33, 37, 112, 116 Murphy, J., 43 Mussweiler, T., 2, 23, 287 Native Americans prejudice toward, 273-276

Nazis, 1, 56, 62, 186, 259 Nelson, C , 249 Nervousness, 21-23 Nested theories, 280-281 Netherlands collective guilt, 53, 55-56, 62 soccer fans, 73-74 stereotypes of immigrants, 254-255 Neuberg, S. L., 10, 69-70, 198, 267, 269, 273, 286, 291-293, 295 Nolan, J. M., 196 Nolen-Hoeksema, S., 179 Norman, C , 69-70 Nurius, P., 26 O'Connor, K. J., 90, 100, 137 O'Reilly, C. A., 240 Oakes, P. J., 4, 14, 50, 55, 70, 72, 112, 228, 277n, 280, 286 Obligatory interdependence, 167-269 Obrist, I. A., 92n Offensive tendencies, 75-84 outgroup paranoia, 182-183 triggering, 119-122 vs. defensive, 116-117 Onorato, R. S., 50, 63, 70 Osagie, J., 227-228, 241 Otten, S., 37, 70-72, 116, 295 Otto, S., 154 Ought group collective self-esteem and, 23-25 controlling for, 21-23 fear of negative evaluation, 25-26 regulatory focus, 36-38 vs. actual group, 17-20 Outgroup contempt for, 255 decoding emotions, 143-148 devalued, 1 eliciting reactions, 116-117 expressing emotions, 137-143 favoritism, 236-239 negative reactions to, 291-292, 295 pity for, 253-254 significant "others," 26-27 system justification theory, 9, 227-245 Outgroup paranoia, 8, 173-189, 290 conceptualizing, 174-176 context for, 176-178 in hierarchical trust dilemmas, 178-183 Oyen, M., 113 Ozer, D., 254

311

3 1 2 INDEX Paez, D., 55 Page, R., 1, 248 Palacios, R., 92 Paladino, M. P, 70, 139, 145 Pantaleo, G., 112 Paranoia, 229-230 outgroup, 8, 173-189, 290 Park, B., 167, 216 Park, J. H., 279 Parrott, W. G., 254 Paternalism, 253-254 Paul, B. Y., 90, 100 Peeters, G., 249, 259 Pelham, B. W, 184, 228-229, 231-233, 236, 239-241 Penner, L. A., 69 Pennington, G. L., 21 Perdue, C. W., 295 Perry, R., 253, 255 Perspective-taking, 67-88 self-categorization and, 68-72 Pettigrew, T. F., 113-114, 127, 154, 156, 166, 202, 297 Petty, R. E., 3 Petty, T. R, 163, 166 Pfeffer, J., 183 Phalet, K., 249-250 Phelps, E. A., 90, 100, 137 Philippot, P, 7-8, 138, 149, 290 Pierce, L. K., 255 Pity, 136, 247, 253-254, 256-258, 270, 272, 292 behavioral implications, 257-258 Plant, E. A., 41, 43, 94 Plutchik, R., 270 Polycarpou, M. P, 155, 164, 167 Pomerantz, E. M., 2, 159, 162, 166 Poor people stereotypes, 252, 257-258 Poppe, E., 249-250 Positive group interactions, 8, 153-173 self-other overlap, 68-72 Post, J. M., 185 Pozo, B. C., 148 Pratto, R, 94,141,157,160-161,194,201,227, 236, 241-242 Prejudice ambivalence, 3 emotional aspects of, 291-293 emotion-based, 4-10 historical focus on, 1-2 integrated threat theory, 194-195 intergroup emotion theory, 111-134

limits of attitudinal theory, 2-3 nature of, 261-262 new perspective, 285-288 reducing, 258-260, 296-297 regulatory bias and, 40-42 stereotypes and, 252-256 system justification theory, 227-245 universality of, 1 Prentice, D., 285 Prevention focus, 35-36, 39-45 Pride, 247, 255, 257, 271, 292 behavioral implications, 258 Promotion focus, 35-36, 359-45 Pruitt, D. G., 173, 181 Pyszczynski, T., 199 Quackenbush, S. W., 255 Quillan, L., 200 Quinn, D. M., 241 Racial attitudes, 154-155 Radloff, C. E., 69 Raeijmaekers, R, 258 Realistic Conflict Theory, 32 Realistic threats, 192-193 group, 198 individual, 199 Regulatory bias, 31-18, 289 as mood regulation, 44 behavioral manifestations, 33, 36-38 collective, 45 emotional manifestations, 333, 8-39 focus of, 35-36 implicit associations and, 40-42 situational differences, 39-42 Regulatory Focus Theory, 13-30, 35 attractiveness of group qualities, 43 collective, 45 experiencing members individually, 43-44 Reicher, S. D., 4,14, 50, 55, 70, 72,112, 277n, 280, 286 Relative deprivation, 113-114 Relaxation, 21-23, 38-39 Rempel, J. K., 2, 153 Renfro, C. L., 9, 149, 196, 199, 271, 280, 290, 292, 294-295, 297 Renfro-Rernandez, C. L., 201 Resentment, 10, 111, 259, 272 image theory, 214-215 Revised threat theory, 196-203 Reynolds, K. J., 70 Rholes, W. S., 286-287

INDEX 3 1 3 Robins, R. S., 185 Rodriguez, A. R, 139, 145, 148 Rodriguez, R. T., 139, 145, 148 Roese, N. J., 21 Rogier, A., 200 Romani, M., 227-228, 241 Romania collective guilt, 57, 61 Romero, V, 34 Roney, C , 35 Ropp, S. A., 74, 127, 297 Roseman, I. J., 4, 72, 112-113, 116, 121, 129, 209, 214, 270, 286 Rosen, B., 250 Rosenberg, M., 14, 23 Rosenberg, S., 249 Rosenblatt, A., 199 Ross, L., 185 Rothbart, M., 70 Rotter, J. B., 174, 201 Rozin, R, 255, 258 Rubchinsky, K., 69 Rubin, K. H., 250 Runciman, W. G., 200 Ruscher, J. B., 176 Rush, L. L., 253, 255 Rust, M. C , 166-167 Rutchick, A., x Rwanda, 55 prejudice in, 259 Ryan, C. S., 216 Sachdev, I., 34 Sack, J., x Sadness, 5, 21-23, 73, 270, 272 Sagarin, B. J., 70 Sager, K., 69 Saiz, J., 227-228, 241 Sakalli, N., 227-228, 241 Salancik, G. R., 183 Saloman, K., 92 Sande, G. N., 69, 210 Sanders, L. M., 156-157, 160 Sanford, R. N., 191, 193 Satisfaction, 10 Savitsky, K., 140, 180 Saxe, L., 1, 248 Schadron, G., 143 Schaferhoff, S., 33 Schaller, M., 279 Scheier, M. R, 16, 23, 28, 270, 289 Scherer, K. R., 4, 15, 72, 112, 116, 214,

Schmidt, G. W., 34 Schmitt, M. T., 49-50, 60, 63 Schneider, A., 53-54, 58 Schneider, K. T., 92 Schopler, J., 2, 174 Schudel, D., x Schwartz, I., 34, 201 Schwartz, J. L. K., 240 Schwartz, R. D., 239 Schwartz, S. H., 114 Schwarzwald, J., 114, 195 Sciaccitano, A. M., 98 Scott, J., 230, 239 Sears, D. O., 3, 114, 155-157, 160, 193 Sechrest, R. D., 239 Sedikides, C., 2, 286, 298 Seery, M. D., 7, 92, 98, 292 Self-categorization, 50, 75-84, 286, 290, 293 group-based emotions, 14-15, 67-88 offensive tendencies, 75-85 perspective-taking, 68-72 shifting, 54 theory, 4-5 Self-discrepancy theory, 5, 13-30 collective self-esteem, 23-25 extended, 15-17 fear of negative evaluation, 25-26 group attributes, 20-21 group-based agitation, 21-23 implications of, 31-48 individual vs. social attributes, 17-20 self-categorization theory, 14-15 significant "others," 26-27 Self-esteem, 285 actual-ideal-ought variables, 23 collective, 23-25 ingroup bias, 33-35 opposition to equality, 235-236 Self-other overlap, 68-72 ingroup emotions, 72-75 Self-regulation, 289, 289, 291 bias motivation, 33-35 discrepancy and, 13-30 emotional influence, 5 social, 285-288 Self-report attitude measures, 102-106, 296297 Selves Questionnaire, 17 Senter, M. S., 239 Sephardic Jews stereotypes, 238

3 1 4 INDEX Shah, J. Y., 5-6, 21, 27, 34-36, 38-39, 41, 280, 289 Sheldon, O., 229, 231-233, 241 Sheppard, L. A., 158 Shérif, C. W., 32, 174 Sherif, M., 32, 34, 174, 191-192 Sherman, J. W., 94 Shimko, K., 210 Sidanius, J., 94, 141, 156-157, 160-161, 194, 201, 227, 236, 241-242 Siegel, P. A., 177, 229 Sigall, H., 1, 248 Silver, L. A., x, 7, 63, 75, 122, 125, 277, 280, 290, 292, 295 Silver, M., x Silverman, R. A., 115 Silverstein, B., 210 Silvia, P., 50, 63 Simon, A. R, 180 Simon, B., 33, 112 Simon, H. A., 270 Singer, M. S., 250, 253-255 Six-Materna, I., 227-228, 241 Sloan, L. R., 14-15, 50, 73, 126 Slugoki, B., 52-53 Smith, A., 69 Smith, C. A., 4,15, 72,112, 209, 214, 279, 286 Smith, D. M., 267, 269 Smith, E. R., ix-x, 1-3, 5, 7, 10, 15, 31, 33, 33, 36, 43, 49-50, 63, 68, 70-73, 75-76,112113, 117, 122, 125, 127-129, 135-136, 143, 147, 149, 173n, 176, 191, 209, 214215, 221, 227, 248, 265n, 265-266, 277278, 280, 286, 289-292, 295 Smith, H. J., 73, 113 Smith, P. M., 228 Smith, R. H., 254 Snider, N., 58 Snyder, C. R., 178 Sober, E., 268n Social Avoidance and Distress Scale, 16 Social Distance Scale, 158, 162 Social dominance orientation, 201 Social Identity Theory, 4-^5,32,50,229,286,291 Social identity anger responses, 7, 75-84, 135-151 as antecedent to threat, 195 assessing threats, 7, 111-134 collective guilt, 55-56 group-based emotions, 14-15 ingroup bias, 31-38 intergroup emotion theory, 111-134

self-other overlap, 68-72 subordinate, 54 vs. evolutionary perspective, 10, 265-284 Social self-discrepancies, 13-30 collective self-esteem, 23-25 extended theory, 15-17 fear of negative evaluation, 25-26 group attributes, 20-21 group-based agitation, 21-23 group-based emotions, 14-15 individual vs. social attributes, 17-20 self-categorization theory, 14-15 significant "others," 26-27 Social self-regulation, 285-288, 289 Social structures, 293-296 Socioeconomic status, 96-97 system justification theory, 227-245 Solomon, S., 199 Sotelo, M. J., 227-228, 241 Spears, R., 15, 50-53, 55-56, 58-60, 63, 7 3 74,81, 113, 121, 125,228 Spence, J. T., 249 Spencer, S. J., 1, 34, 44, 199, 227-228 Spindel, M. S., 112, 209 Stangor, C , 115, 135, 154, 158, 230, 233, 239 Staub, E., 57, 260 Steele, C. M., 25, 44, 95, 135, 199, 228 Steele, S., 58 Stephan, C. W., 94, 106, 114-115, 154, 156, 191, 194-196, 202, 204, 257, 287, 297 Stephan, W. G., 9,94,106,114-115,149,154156, 164, 167, 191, 194-196, 199, 2 0 1 202, 204, 257, 271, 280, 287, 290, 292, 294-295, 297 Stereotype Content Model, 250-256 Stereotypes, 194, 247-248, 292 action tendencies, 144-148 ambivalence about, 248-252 as consequences, 9, 209-225 challenging, 297 consensual, 236-239 explained, 135-136 intergroup emotion theory, 111-134 internalizing, 81, 228-229 origins of, 294-295 prejudice, 252-256 regulatory bias, 40-42 social structure predicts, 250 suppression of, 71 system justification function, 238-239 systematic dimensions, 248-252 Stipek, D., 253

INDEX Strauman, T., 16, 19, 21, 25-26 Strenta, A., 145 Stroessner, S. J., 94 Struch, N., 114 Structural relations, 8-9 affect in intergroup behavior, 153-171 image theory, 209-225 outgroup paranoia, 173-189 perceived threat, 247-264 role of threat, 191-207 system justification motive, 227-245 Sue, S., 250 Sullivan, B., 229, 231-233, 241 Sullivan, L. A., 115, 135, 154, 158 Sumner, W. G., 31-32 Survival threats, 10, 265-284 Suspicion, 229-230 Swartz, T. S., 113, 116, 129, 214, 270 Swim, J. K., 52-53, 60, 160-161, 167 Symbolic Racism Scale, 155 Symbolic threats, 114, 193-194 group, 198-199 individual, 199 System justification theory, 9, 184, 227-245, 290-291, 294 ambivalence, 9-10, 247-264 consensual stereotyping, 236-239 implications of emotions, 258-260 implicit preferences, 239-241 outgroup favoritism, 236-239 overview, 228-236 palliative function, 230-231 System threat, 237-238 Tajfel, H., 1, 4, 14, 31-33, 50, 78, 115, 129, 227-228, 230, 280, 286 Tambor, E. S., 285 Tangney, J. P.,50 Tassinary, L. G., 90 Taylor, D. M., 115, 237 Tellegen, A., 99 Tension, 21-23 ter Schure, E., 50, 72, 76, 113, 116, 214, 286 Terdal, S. K., 285 Tesser, A., 4, 155-157, 255 Thakkar, V, x Theno, S. A., 114 Thompson, E. P., 34, 228-231, 233, 235-236, 239, 241 Thompson, L. L., 34 Thome, A., 14, 50, 73, 126 Threat, 21-23, 44, 288, 293, 295

antecedents, 195 biopsychosocial perspective, 90-92 competitive, 252-253 concept, 197-200 differentiated reactions, 9, 191-207 group integrity, 269, 272 hierarchy, 276-277 intergroup anxiety, 194 intergroup emotion theory, 111-134 intergroup interactions, 92-101 meta-analysis, 101-106 negative stereotypes, 194 outgroup paranoia, 173-189 prejudice, 194-195 realistic, 192-193 reconsidering the consequences, 202-203 resource, 272 revised theory, 196-203 role of emotions, 269-272 support for, 195-196 survival vs. identity, 10, 265-284 symbolic, 114, 193-194 system, 237-238 types, 197-200 vs. challenge, 7, 89-109 Tiedens, L. Z., 147 Tokusato, 127 Tomaka, J., 91-92 Tomkins, S. S., 270 Tooby, J., 270, 291 Tota, M. E., 41 Transparency, illusion of, 140-141 Triandis, H. C., 201 Tropp, L. R., 50, 73, 113, 127, 156, 297 Trotsky, L., 173 Trust, 229-230 dilemmas, 8, 173-189, 294 image theory, 214-215 Tsui, A. S., 240 Tuch, S. A., 161 Tudor, M., 69-70 Tunisia, 144 Tur-Kaspa, M., 114, 195 Turner, J. C., 4, 14, 32-33, 50, 55, 63, 70, 72, 112, 115, 227-228, 277n, 280, 286 Tyler, R. B., 295 Tyler, T. R., 160,177, 184, 229-230, 236, 241242 Udegbe, B., 227-228, 241 Ui, M., 227-228, 241 Uneasiness, 21-23

315

3 1 6 INDEX United States collective guilt, 52-53 individualistic society, 279 Japanese-American internment, 182-183 prejudice toward immigrants, 195-196 response to Kosovo war, 67-68 stereotypes, 238-240 Urban, L. M., 76 Vaes, J., 7-8, 139, 290 van den Heuvel, H., 115, 254, 257-258, 265 van Knippenberg, A., 237 Van Oudenhouven, J. P., 128 Vanable, P. A., 175, 179-180 Vanbeselaere, N., 73-74 Vance, S., 41, 43 Vanman, E. J., 90, 100 Vanneman, R. D., 113 Vasquez, K. A., 135 Vasquez-Suson, K. A., 94, 137, 287 Veeder, M., 199 Veniegas, R. C , 94 Vescio, T. K., 114-115, 196 Vivekanathan, P S . , 249 Vivian, J., 128 von Hippel, W., 2, 163-164 Wackenhut, J., 3 Walker, M. R., 14, 50, 73, 126 Walling, D., 70, 112,286 Wann, D. L., 50, 81, 196, 201 Ward, C , 236 Warmth, 247-248, 251-254, 257-258 Warren, E., 182-183 Watson, D., 16, 25, 99 Wealthy persons stereotypes, 252, 257-258 Webb, E. J., 239 Weber, J. G., 98 Wegener, D. T., 163, 166 Wegner, D. M., 71 Weiner, B., 50, 253-255, 257-258 Weiss, R., 242 Weitz, S., 158 Welfare recipients stereotypes, 247, 252, 257-258 Wentura, D., 70, 295 Wenzlaff, E. M., 71 Wetherell, M. S., 4, 14, 50, 55, 70, 72, 112, 277n, 280, 286 Whalen, H., 250 White Americans prejudice toward, 273-276

White, B. J., 174 White, R. K., 210 Wiest, C , 113, 116, 129, 214, 270 Wigboldus, D., 6-7, 63, 76, 78, 81, 85, 227228, 241, 290, 292-293 Wiggins, J. S., 294 Wilder, D. A., 31, 94, 180 Wilke, H., 237 Willemsen, T. M., 227-228, 241 Williams, J. E., 158 Wills, T. A., 98 Wilson, D. S., 268n Wilson, T. D., 3, 180 Wiseman, R. L., 94 Wittenbrink, B., 167 Wohl, M., 49 Wolsko, C , 167 Women stereotypes, 252, 257-258 Woodmansee, J. J., 155 Woodmansee-Cook Scale, 158 Woolf, V, 260 Woolf, V, 260 Word, C. O., 137 Worth, L. T., 167, 210 Wright, C , x Wright, S. C , 50, 73-74, 113, 115, 127, 237, 297 Wyer, N. A., 94 Wyer, R., 179 Xu, J., 136, 148-149, 248, 250-251, 261 Yabar, Y. C , 136, 144, 147 Yamada, A. M., 196 Yamamoto, M., 227-228, 241 Ybarra, O., 114, 195-196 Yinon, Y, 5, 17, 195, 280, 289 Yzerbyt, V, 6-7, 63, 70, 76, 78, 81, 85, 143, 200, 290, 292-293 Zald, M. N., 233 Zanna, M. P, 1-2,114-115,135,137,153-155, 161, 166-167, 193, 265 Zebel, S., 53, 59, 63 Zeller, E., 115 Zemore, S., 250 Zillmann, D., 14-15 Zimbardo, P., 173n, 175, 180, 182 Zubrinski, C. L., 160 Zucker, G. S., 253, 255, 257-259 Zucker, L. G., 176 Zuwerink, J. R., 50, 105