Choice of Scapegoats

244 Glick 15 Chapter Fifteen Choice of Scapegoats Peter Glick A decade before Allport wrote The Nature of Prejudice, the Nazis had attempted to exterminate Europe’s Jews, whom they blamed for Germany’s woes. A decade before this chapter was written, hundreds of thousands of Tutsi were murdered by Hutu neighbors, who blamed the Tutsi for Rwanda’s economic and social problems. Both are examples of scapegoating, here defined as an extreme form of prejudice in which an outgroup is unfairly blamed for having intentionally caused an ingroup’s misfortunes. Allport considered scapegoating to be one of the most important theories of prejudice, devoting considerable space to it (chapter 15, “Choice of Scapegoats,” plus large portions of Part IV, “The Dynamics of Prejudice”), focusing (not surprisingly, given how recent the Holocaust was) on the Jews’ long history of having been scapegoated. Although he invoked “normal” cognitive (e.g., categorization) and social (e.g., conformity) processes to explain why people differentiate between groups (see Fiske, ch. 3 this volume), Allport believed that only scapegoating, which he attributed to irrational and maladaptive ego-defense mechanisms, could account for the most extreme forms of prejudice, such as genocide. Although subsequent prejudice researchers took up the cognitive-social strand of Allport’s thought with zeal, research on scapegoating languished as psychodynamic views fell out of favor. This chapter first reviews Allport’s synthetic view and incisive critique of scapegoat theory and then presents an alternative model of scapegoating that (a) is more in keeping with the dominant cognitive-social approach, (b) focuses on social conditions and group-level processes, rather than individual psychopathology, as root causes of scapegoating (and its most extreme manifestation, genocide), and (c) solves the problems Allport identified as major stumbling blocks for scapegoat theory. Choice of Scapegoats 245 Allport’s View of Scapegoating Allport noted the biblical origins of the term “scapegoat” – a sacrificial goat to which the ancient Israelites had symbolically transferred their sins. This anecdote illustrates the psychodynamic view of scapegoating: negative traits that people do not wish to acknowledge in themselves are psychologically projected onto others, thereby relieving feelings of guilt and frustration (see Newman & Caldwell, ch. 23 this volume). Allport embraced both the psychodynamic theory of scapegoating and a less complicated model in which external frustrations (not necessarily due to one’s own sins) are displaced onto an innocent target, regarding these two versions as complementary. Freudian Psychodynamics versus Frustration–Aggression The two approaches to scapegoating Allport described, though overlapping, stem from different psychological approaches, one rooted in Freudian theory (the psychodynamic view) and the other in later drive-reduction models (the frustration–aggression view). Freud believed that frustration and guilt are the inevitable consequences of society’s (necessary) restrictions of individuals’ instinctual drives, most prominently for sex and aggression, which are collectively labeled as the id. To avoid social chaos, people must be socialized (predominantly by their parents) to restrict their natural impulses and to redirect them in socially approved ways. These parental and societal strictures become internalized, creating the superego, the individual’s conscience, which strives to enforce good behavior. Thus an overly punitive upbringing guarantees continuing conflicts between a highly restrictive superego and the instinctual drives of the id. The superego (with limited success) attempts to repress even simple awareness of the id’s unacceptable impulses, resulting in frustration, guilt, and aggressive feelings that need to be vented, setting the stage for scapegoating. Later drive-reduction models agreed with Freud that various drives sought discharge in behavior. In contrast to the psychodynamic view, however, frustration–aggression theory characterized aggression as a response to circumstances that block goal-directed behavior, not as an innate drive (Dollard et al., 1939). External frustrations (rather than internal psychodynamic conflict) evoke aggression, typically against the source of frustration. Like Freud, however, Dollard et al. believed that aggression would be displaced 246 Glick onto an innocent target if the true source of frustration was powerful and therefore likely to retaliate. Both the psychodynamic and frustration–aggression views of scapegoating invoke the notions of displaced aggression (directing aggressive impulses toward an innocent target) and projection (attributing negative characteristics and blame to targets to justify aggressing against them). The Freudian approach, however, emphasized what Allport termed “direct projection,” in which the individual’s own unacceptable impulses are projected onto others, allowing people to deny their own faults. Direct projection requires a complicated set of ego-defense mechanisms, including repression (eliminating conscious awareness that the faults are one’s own) as well as displacement. The frustration–aggression approach concentrated instead on what Allport termed “complementary projection” in which perceivers explain feelings (of which they are consciously aware) by projecting complementary, causal traits onto others (e.g., I am fearful, therefore you must be hostile). Because the source of frustration is external (circumstances that block attainment of goals, rather than “character-conditioned” internal psychodynamic conflict) there is no need for repression. In both views, however, derogatory beliefs about scapegoats reveal more about the perceiver (either their faults or their fears) than they do about the actual traits of the scapegoated group, which merely serves as a projective screen. An example of the psychodynamic view is the idea that antihomosexual prejudice results from the perceiver’s own “latent homosexuality.” Individuals who have internalized the notion that homosexuality is immoral yet themselves experience homosexual impulses would try to deny these desires in themselves and displace their self-loathing onto others. Adams, Wright, and Lohr (1996) found that avowedly heterosexual male participants’ antihomosexual attitudes were correlated with greater penile tumescence when viewing videos of homosexual sex. This study is consistent with the idea that men who had repressed their own homosexual impulses had (at least in part) transformed self-loathing into hostility toward others. In contrast, Bettleheim and Janowitz’s (1950) findings (cited by Allport) that men experiencing downward social mobility (an external source of frustration) expressed both more anti-Semitism and racism, is a prime example of the frustration–aggression approach. Presuming that Jews and Blacks were not the direct cause of their downward mobility, these men’s greater anti-Semitism and racism was an example of displaced aggression. The targets of their aggression were assumed to be chosen because they were weak, vulnerable, and already deprecated minority groups. In sum, the major difference between the psychodynamic and frustration– aggression theories of scapegoating is whether frustration is due to internal Choice of Scapegoats 247 psychodynamic conflict or to external sources that block goal-directed behavior. Otherwise, both views suggest that: (a) aggression against the true source of frustration is inhibited, (b) aggression is displaced toward others who are weak and therefore unable to retaliate, and (c) negative stereotypes of scapegoated groups are mere projections or rationalizations that obscure the real reasons for hostility toward them. Overall, both accounts view scapegoating as irrational – an infantile, maladaptive, and emotionally driven response to (either internal or external) frustration. Despite his frequent use of scapegoat explanations and willingness to synthesize psychodynamic and frustration–aggression approaches, Allport provided an extensive critique of scapegoat theory (based partly on concerns expressed by Zawadzki, 1948), summarized below: (1) Frustration does not always lead to aggression . . . (2) Aggression is not always displaced [but rather is more likely to be directed at the true source of frustration] . . . (3) Displacement does not . . . actually relieve the feeling of frustration [because it does not address frustration’s underlying causes] . . . (4) The theory says nothing concerning the choice of scapegoats . . . (5) It is not true that a defenseless minority is always chosen for displacement purposes [powerful groups are sometimes blamed] . . . (6) Available evidence does not indicate that the displacement tendency is any more common among people high in prejudice than among those low in prejudice . . . (7) Finally, the theory itself overlooks the possibility of realistic social conflict. (1954/1979, pp. 350–1) In the decades following Allport’s book, progress addressing the problems listed above was slow in coming, but recent theoretical advances are beginning to suggest resolutions. Developments Since Allport Theoretical developments since the 1950s suggest that scapegoating ought to be viewed as a collective rather than individual process (though for more on the latter, see Newman & Caldwell, ch. 23 this volume). Subsequent work has accepted the fundamental role of frustration in scapegoating, but with a focus on the roles of shared (collective) frustrations (e.g., economic collapse, social disorder) that produce social movements with scapegoating ideologies that lead to organized persecution. Frustration’s role thus becomes indirect, with ideology being the proximal cause of aggression. This section reviews post-Allport developments related to collective 248 Glick scapegoating and explains how they begin to address some of Allport’s criticisms. Staub’s (1989) refinement of the frustration–aggression perspective suggests one answer to Allport’s first criticism that (contrary to frustration– aggression theory) frustration does not always turn into aggression against scapegoats. Staub theorized that genocides and mass killings require widespread, extremely “difficult life conditions” (due to social, political, and/or economic upheavals) that frustrate a host of basic needs, for physical wellbeing, safety, a sense of belonging, self- and group esteem, and hope for the future. Mundane frustrations are insufficient to spawn scapegoating movements, which gain adherents by promising to fulfill the heightened needs of a mass of people (e.g., safety and prosperity will be achieved by attacking the “enemy”). Terror management theory (Pyszczynski, Greenberg, & Solomon, 1999), which proposes that asserting their ingroup’s superiority is one means by which people defend against awareness of their own mortality, can be incorporated into Staub’s view. The difficult life conditions that precede scapegoating movements often include widespread disease or wars, making mortality highly salient. Deeply frustrated basic motives can attract masses of people to movements that blame an outgroup for current difficulties, affirm the ingroup’s superiority, and offer the promise of a better future. Consistent with Staub’s (1989) viewpoint, social-identity theorists (Billig, 1976; Tajfel, 1981) have posited that only collective frustrations (those shared by many) lead to organized scapegoating. Based upon socialidentity theorists’ general critique of Allport’s (and others’) individualistic approach to prejudice (as rooted in personality), Tajfel and Billig argued that scapegoating is a group-based process. These theorists highlighted a problem Allport had not articulated – why does the scapegoating of a particular outgroup become socially shared or consensual within a society? Although Freudian psychodynamics or the frustration–aggression hypothesis might explain why a particular individual lashes out against a convenient target, how can this account for mass movements in which a significant portion of a society’s members join together in attacking a specific group? As Billig observed about the Holocaust, “It is too fanciful to imagine that the Germans were kept in an increasing state of emotional arousal for fifteen years, and at the end of this time simultaneously millions happened to rid themselves of these tensions in an identical manner” (1976, p. 150). The most horrendous examples of scapegoating (from medieval witch-hunts to the attempted genocides of the twentieth century) involve coordinated mass movements. Allport imagined that all levels of frustration could promote scapegoating, including intensely personal causes. For instance, he noted that a “short Choice of Scapegoats 249 stature” could be “a life-long cause for irritation” (1954/1979, p. 344). In contrast, Tajfel (1981) hypothesized that individual frustrations lead people to blame other individuals for their problems, whereas shared (grouplevel) frustrations predispose people to blame other groups. For example, a woman might blame her boss for firing her if other members of her ethnic group remain gainfully employed, but might blame an outgroup if she loses her job during an economic downturn that affects many members of her (but not of another) ethnic group. Events that people interpret as applying to their group as a whole (rather than to themselves as individuals) arouse group-based emotions (see Smith & Mackie, ch. 22 this volume). Tajfel termed the process by which frustration becomes group blame “social attribution,” situating it firmly within the social-cognitive approach favored by social psychologists. Attribution theory (e.g., Kelley, 1967) posits that people seek to understand the causes of events, especially negative ones. This makes adaptive sense: when bad events happen, people try to diagnose why they occurred, which may indicate how to correct the problem. When goals are blocked, frustration energizes the individual to diagnose and overcome obstacles. Like physical pain, frustration is an aversive state that indicates something is amiss and motivates the individual to attend to the problem. Although Tajfel did not get the chance to further refine or empirically test the concept of social attribution, he pointed out how “normal” socialidentity and cognitive-social processes, rather than infantile rage or the random venting of frustrations, could precipitate scapegoating. For example, when people are classified as members of a salient outgroup, they are perceived in a less individualized fashion so that their similarities to one another are exaggerated (see Gaertner & Dovidio, ch. 5 this volume), and are often ascribed less “human” qualities than are ingroup members (Leyens et al., 2000). These consequences of social categorization facilitate negative stereotyping and dehumanization and often provide system-justifying ideologies (see Jost & Hamilton, ch. 13 this volume) that act as catalysts for scapegoating. Staub (1996) identified the systematic devaluation and dehumanization of members of another group as key elements in the “psychology of evil” leading to open, and often mass, violence against the group (see also Sternberg’s, 2003, duplex model of intergroup hate). Further, social-identity theory links these cognitive processes to powerful motivations (e.g., the need for a positive group identity) of the sort Staub (1989) also hypothesized shared frustrations to arouse. The link between attributions and emotions can also be understood using appraisal theories of emotion, which Smith and Mackie (ch. 22 this volume; see also Cottrell & Neuberg, 2004) have incorporated in their intergroup 250 Glick emotions theory. For example, if another group’s actions are appraised as a deliberate attempt to harm the ingroup, ingroup members will experience fear (if the other group is powerful) or anger (if the other group is weak). Social attribution is an initial link in a chain that produces intergroup emotions. The theories outlined above suggest that group-level frustrations create shared attributions and appraisals that, in turn, lead to scapegoating. They effectively address Allport’s first criticism (that not all frustrations lead to scapegoating), but not the other questions Allport asked: Why are innocent groups targeted? Why is aggression ever displaced? How are scapegoats chosen? Are scapegoats always defenseless and vulnerable? Heider’s (1958) foundational insights into the attribution process and recent work on the stereotype content model of prejudice (Fiske, Cuddy, Glick, & Xu, 2002; see Fiske, ch. 3 this volume) help to address many of Allport’s concerns. Heider reasoned that to explain the cause of an event, people look for a causal agent (e.g., another person) who possesses the power (or ability) and intent to bring about that event. Thus, for example, an infant is unlikely to be blamed for knocking a plate off of a 6-foot-high, secure shelf because of her inability to reach it, even if she is known generally to have the intent to knock things over. Extending Heider’s analysis to social (or intergroup) attribution suggests that only groups perceived as having the power and intent to cause widespread frustrations will be scapegoated. The stereotype content model (see Fiske, ch. 3 this volume) suggests which groups are likely to be scapegoated. This model states that the structural relations between groups – their relative status and type of interdependence (competitive or cooperative) – determine the content of group stereotypes. High status or powerful (e.g., socioeconomically successful) minorities that are viewed as competing with the dominant group are subjected to envious prejudice: they are admired for their success, but also resented for it; stereotyped as highly competent, but as having hostile motives. Because envied minorities are viewed as having the power and intent to harm, they are at risk of being blamed for causing group-level frustrations. Additionally, a sense of relative deprivation (Runciman, 1966) can play into the scapegoating of successful minorities, who may be perceived as being less affected by or even as benefiting from the suffering endured by the ingroup, inflaming resentment and lending plausibility to the belief that the scapegoated group intentionally brought about the ingroup’s misfortunes. In contrast, low-status groups – whether paternalized (see Jackman, ch. 6 this volume; Rudman, ch. 7 this volume) or treated with contempt – are stereotyped as incompetent, suggesting less ability and power to Choice of Scapegoats 251 cause widespread misfortunes, making them less likely to be chosen as scapegoats. Allport identified the Jews as history’s most frequently scapegoated group. In retrospect, his historical analysis of European anti-Semitism (from the Jews’ special theological position as the Bible’s “chosen people” to their professional success in pre-Nazi Germany) is consistent with the notion of envious prejudice. The nineteenth-century anti-Semitic forgery, The Protocols of the Learned Elders of Zion (which purports to describe a meeting of a Jewish group bent on world domination), perfectly illustrates the character of envious prejudice. Far from being disparaged as inept, the Jews’ power and abilities are exaggerated to a hysterical degree. Currently, difficult life circumstances in the Muslim world combined with Israel’s military might, close alliance with the United States, and occupation of the West Bank and Gaza have reinvigorated these anti-Semitic canards. For example, at a summit of 57 Islamic nations the Malaysian Prime Minister contended that the Jews control the world (Rhode, 2003). Conspiracy theories and related rumors that blame disliked events on the machinations of a group are the hallmark of envious prejudice and scapegoating. Intergroup emotion theory (Smith & Mackie, ch. 22 this volume) suggests that an outgroup that is perceived as intentionally having brought about calamities for the ingroup will elicit strong (group-based) anger. If a scapegoated group is viewed as both powerful and malevolent, even the most extreme actions against them (e.g., murder) can be rationalized as self-defense. A New Framework for Scapegoating Theory The theoretical developments described above can be synthesized into a new model of scapegoating that addresses the problems Allport identified with older versions of the theory, an ideological model of scapegoating (Glick, 2002). This model views scapegoating, even in its most extreme and unwarranted versions, as stemming from ordinarily adaptive cognitive tendencies and motives, rather than inherently irrational, maladaptive processes. It downplays Freudian defense mechanisms, emphasizing instead the role of culturally shared beliefs, stereotypes, and ideologies. Finally, it focuses on scapegoating as a collective process by which commitment to hateful ideologies becomes widely shared within a community, creating a consensus that spawns political movements (e.g., Nazism) and coordinated hostile actions (e.g., discriminatory laws, organized mass killings, etc.). 252 Prior envious stereotypes that target group has the power and intent to have caused collective frustrations (D) Heightened frustration, needs, and anxieties (B) Collective search for plausible causes of and solutions to shared difficulties (C) Perceived ability of movement to fulfill frustrated needs (e.g., security, esteem, hope) (E) Commitment to social movement/ scapegoat ideology (G) Individual/cultural propensity toward prejudice (e.g., Social Dominance Orientation and Right-Wing Authoritarianism) (F) Figure 15.1 An ideological model of scapegoating (adapted from Glick, 2002) Actions against scapegoat, both spontaneous (willing) and organized (conformity, obedience to movement) (I) Perceived success of movement at lessening frustrations (e.g., improved economic conditions for the ingroup) (J) Glick Collective difficulties (A) Hostility toward and exaggerated conspiracy theories about the scapegoated group (H) Choice of Scapegoats 253 Causes of Commitment to Scapegoating Ideology Figure 15.1 illustrates the proposed ideological model of scapegoating (the letter labels in the figure correspond to the text below). The model begins with (A) widespread, shared frustrations, the kind of group-level events that Staub (1989) and Tajfel (1981) viewed as a prerequisite for scapegoating. These difficult life conditions (B) frustrate basic needs and arouse shared anxieties (Staub, 1989) that (C) lead to a collective search for explanations and solutions (Tajfel, 1981). How and why the third step leads to the heart of the model, (G) commitment to scapegoat ideologies, is the unique contribution of this framework. Allport suggested that “the commonest reaction to frustration is . . . a simple and direct attempt to surmount the obstacle in our path” (1954/ 1979, p. 348), but this first requires identifying the obstacle. In cases of widespread misfortune, the nature of the “obstacle” and how to overcome it is often obscure. In their collective search for plausible explanations and courses of action, people may come to inaccurate, even (to those with better or different information) objectively ridiculous, conclusions that scapegoat an innocent group. Incorrect attributions may occur because information and people’s cognitive abilities to process it are limited, especially when coping with large-scale problems in complex, modern societies. For example, even professional economists may be unable adequately to explain an economic crisis. Complex explanations involving largely uncontrollable, impersonal forces are difficult to fathom and may fail to suggest a clear solution, rendering them unattractive (because they do not fulfill heightened psychological needs for control and optimism). Scapegoat movements attract followers by offering simpler, culturally plausible explanations and solutions for shared negative events. The ideological model suggests that the choice of scapegoat is primarily determined by (D) prior envious stereotyping (Fiske et al., 2002) of groups that are viewed as having the power and intent to cause widespread harm. Thus the perceived plausibility of a scapegoat ideology is influenced by cultural context. When Medieval Europe was decimated by plague, people blamed witches because witches were already believed to have the capability (via supernatural powers) and desire to cause such harm. Within the cultural context, blaming witches made “sense.” Even though blame was wholly (and horribly) misplaced, rational faculties were not abandoned. For example, the well-documented witchcraft claims made in Salem, Massachusetts, in the seventeenth century were subjected to rigorous 254 Glick and, if basic premises about witches are granted, arguably rational tests, as well as trials that adhered to the legal standards of the day (Karlsen, 1987). Similarly, a history of envious stereotyping is evident in most of the large-scale genocide attempts of the twentieth century. The Armenians in Turkey and the Jews in Germany filled strikingly similar social niches; both were economically successful minorities, often in “middleman” roles as merchants and traders. Perceived as a powerful minority, the Armenians were blamed for weakening the Ottoman Empire during the years of its decline (Adanir, 2001) much as Nazis blamed the loss of the First World War on a Jewish conspiracy (the “stab in the back” theory). In Rwanda (see Prunier, 2001), the Tutsi are a high-status minority, who (having been favored during Belgian colonization) are typically better educated, wealthier, and more likely to hold high-status jobs in comparison to the Hutu majority. The genocidaires among Rwanda’s Hutus characterized the Tutsi as a powerful minority profiting off of the majority’s misery. Cambodia’s Khmer Rouge, who had their roots in the peasantry, resented and specifically targeted those with wealth and social influence, such as intellectuals (Kiernan, 2001). Thus, Allport’s critique that it “is not true that a defenseless minority is always chosen” (p. 351) is turned into a much stronger claim in the model presented here – it is precisely the perceived power of a group (not its perceived weakness) that makes it likely to be scapegoated. Earlier views dismissed perceptions of the scapegoat’s power and malevolence as mere projections (of perpetrators’ own megalomania and murderous intentions), but for the genocides listed above, envious stereotypes can be traced to prior historical circumstances (e.g., a middleman role) that conditioned the structural relations between groups. Envious stereotypes are here viewed more as a cause (increasing the likelihood that a group will be blamed for misfortunes) than a consequence of scapegoating. When a history of envious prejudice collides with a precipitous decline in a society’s life conditions, successful minorities are at grave risk. In the ideological model, the tendency to seek a scapegoat is also conditioned by (E) a scapegoat movement’s ability to address a host of psychological and social needs. Although past theorists focused on the destructiveness of scapegoating, a utopianism that promises the fulfillment of people’s basic needs is an equally important feature of scapegoating movements (see Staub, 1989). For example, Friedländer (1997) argued that Nazi anti-Semitism was “redemptive” – the Nazis promised Germans a “Thousand Year Reich” in which the problems of the present would be Choice of Scapegoats 255 solved by eliminating their source, the Jews. Similarly, Fritzsche (1999) contends that most Germans were more directly attracted by the Nazis’ promise to satisfy their current needs and provide a better tomorrow than by their hostility toward the Jews per se (though the former was inextricably intertwined with the latter in Nazi ideology). As Staub (1989) has noted, difficult times heighten needs for security and physical well-being, for self- and group esteem, to believe that life is controllable and predictable, and to have hope for the future. Pyszczynski et al. (1999) add psychological defense against mortality salience to this list. Scapegoat movements address frustrated needs for esteem, belonging, and transcendence by reasserting the ingroup’s superiority and offering membership in a wider social movement that promises symbolic immortality. Security, physical well-being, and hope are addressed by identifying and combating the “enemy” that has putatively caused the ingroup’s misfortunes. The Nazis, for example, were masterful at creating a sense of belonging (everyone, children included, could belong to a subgroup, e.g., the Hitler Youth), group esteem and transcendence (through pageantry and an ideology of racial superiority), and optimism (through economic progress spurred by re-armament). Thus parts A–E of the current model answer many of Allport’s critiques of earlier scapegoat theory (corresponding to Allport’s numbered critiques): (1) Why does frustration not always lead to intergroup aggression? Because only widely shared (not individual) frustrations are likely to lead to scapegoating. (2) Why is aggression ever “displaced”? Aggression may be better characterized as misplaced as a result of prior cultural stereotypes that characterize an outgroup as having the ability and intent to cause harm to the ingroup. (3) Why would human nature include an inherently maladaptive displacement tendency? It does not; a generally adaptive attribution process, due to limited cognitive capacity, imperfect knowledge, and motivational and cultural biases, results in misplaced blame. (4) Which groups are likely to be scapegoated? Successful but resented minorities. (5) Are groups always chosen for their perceived vulnerability? No, precisely the opposite – groups are scapegoated because they are (often falsely) perceived to be powerful and malevolent. Allport’s criticism (6), that “[a]vailable evidence does not indicate that the displacement tendency is any more common among people high in prejudice than among those low in prejudice” (p. 351), is also solvable if ideological commitment, not projection and displacement, is the central mediator of scapegoating. For Allport, the prejudiced personality was high in authoritarianism, a personality syndrome then thought to be dominated 256 Glick by low frustration tolerance, repression of one’s own desires, and psychodynamic conflict. Authoritarians were assumed to be especially prone to displace and to project their own failings onto others. As Allport noted, however, there has been little support for this view. In contrast, the current model (F) is consistent with contemporary theories of the prejudiced personality (see Duckitt, ch. 24 this volume). Altemeyer (1981) has reconceptualized the authoritarian personality as being rooted in ideology, rather than psychodynamics. Right-Wing Authoritarianism (Altemeyer, 1981) assesses a tendency toward group conformity and the acceptance of conservative political ideologies that demean cultural outgroups. People high in Right-Wing Authoritarianism are hypervigilant toward perceived threats to the ingroup, and therefore (especially during difficult life conditions that threaten the ingroup) are likely to scapegoat outgroups. The other most prominent individual difference measure of prejudice, Social Dominance Orientation (Sidanius & Pratto, 1999), assesses a preference for group-based hierarchy. People high in Social Dominance Orientation seek to preserve established dominance hierarchies (which are threatened during times of social upheaval) and may scapegoat to restore ingroup dominance. Authoritarianism and Social Dominance Orientation are strongly related to the mediator of scapegoating proposed here, endorsement of nationalistic and xenophobic political ideologies (Duckitt, ch. 24 this volume). Further, rather than indicating individual psychopathology, current research reveals these individual differences to be socialized belief systems (Duckitt, this volume), suggesting that cultures (as well as individuals) differ on these dimensions and therefore in their proneness to scapegoating (Staub, 1989). As Duckitt (this volume) notes, people’s endorsement of authoritarianism and social dominance may increase when the ingroup is threatened (authoritarianism) or the status of the ingroup is diminished (social dominance), increasing attraction to and social consensus in scapegoating. Thus, people who endorse these belief systems are likely to be prominent among the willing adherents of scapegoat movements (Staub, 1989). Allport’s last criticism, (7), was that scapegoat theory ignores the possibility of realistic social conflict. In the model proposed here, realistic conflict is part of a continuum that can result in scapegoating. For instance, during war the enemy is reasonably defined as a source of problems. At the same time, exaggerated stereotypes and inaccurate conspiracy theories are also likely to arise. Realistic conflict increases the perceived plausibility that an outgroup caused the ingroup’s woes and exacerbates envious stereotyping. Just when realistic conflict becomes unreasonable scapegoating is a Choice of Scapegoats 257 difficult judgment to make, but similar psychological processes are at work and the former can easily develop into the latter. Consequences (and Reinforcers) of Commitment to Scapegoat Ideologies Commitment to a scapegoat ideology (G) mediates the negative consequences of scapegoating. Although already resented due to prior envious stereotypes, hostility toward scapegoated groups (H) increases in direct proportion to prejudiced perceivers’ endorsement of an ideology that blames them for current shared misfortunes. Such ideologies (and increased hostility) cohere around political movements that act against the scapegoated group (I). Genocides, for example, are not spontaneous events, but top-down affairs, organized and ordered by a state. The perpetrators may be a mixture of ideologically-motivated true-believers (Goldhagen, 1997) and those who merely follow orders (Milgram, 1974). These motivations, however, may typically act in concert to produce a willing obedience to authority in which perpetrators’ acceptance of a scapegoat ideology and commitment to a political movement motivates zealous compliance with orders (and, for some, spontaneous sadistic acts that exceed those orders). The persecution of scapegoats launches perpetrators along a “continuum of destruction” in which initial, less severe actions spiral into increasingly violent behavior (Staub, 1989). Harming another creates psychological pressures to rationalize one’s behavior, resulting in further derogation of victims (Lerner & Miller, 1978) and increased commitment to scapegoat ideologies that portray targets as powerful, implacable enemies who deserve what they get (hence the double-tipped arrows in figure 15.1). A vicious loop between ideological commitment, hostility, and violence, can devolve into mass murder, which is justified by the movement as moral and necessary, a matter of self-defense against an inherently malevolent “enemy.” Finally, if ideological commitment is the proximal cause of scapegoating, an important implication follows. Older scapegoat theories assumed that the lessening of frustrations would decrease subsequent scapegoating. The current model suggests, however, that if the scapegoating movement itself is viewed as having improved general life conditions ( J), thereby “proving” its effectiveness, ideological commitment is likely to increase. For example, the Nazis’ economic successes in the 1930s increased Germans’ faith in Hitler’s policies, encouraging the Nazis to take increasingly harsh and, finally, murderous actions against the Jews. 258 Glick Has Allport Been Supported? The model proposed here is quite different from the scapegoating theory Allport described. Importantly, it yields several implications that directly contradict earlier versions. First, only collective, not individual-level frustrations are likely to foster scapegoating. Second, groups that are perceived to be powerful (not weak and vulnerable) are likely to be victimized. Third, because frustration is an underlying, but not the proximal, cause of aggression, the lessening of frustrations may actually increase (not decrease) scapegoating (if improved conditions are attributed to the scapegoat movement). Although the model proposed here differs significantly from Allport’s view, frustration is still conceived to be a critical precondition for scapegoating. Further, the current model is strongly indebted to Allport’s incisive critique of the original theory, which pointed the way forward. Although this model eschews projection as a central mechanism in scapegoating, it is important to note that researchers have recently constructed careful empirical tests of projection, yielding evidence that it may play a role in prejudice (see Newman & Caldwell, ch. 23 this volume). This revival of psychodynamic views is exciting, but although projection may account for individual differences in prejudice, it may not be a central mediator of consensual, organized scapegoating. Future Directions Although this chapter has focused on scapegoating as a collective process, individual-level scapegoating ought not to be abandoned as a topic of study. At the individual level, momentary frustrations, such as threats to self-esteem, can exacerbate derogation of outgroups (Fein & Spencer, 1997) and daily frustrations can lead individuals to respond disproportionately to minor provocations, especially by outgroup members (Miller, Pederson, Earleywine, & Pollock, 2003). However, such individual-level processes must be distinguished from collective processes; whereas the former explain when individuals are likely spontaneously to act on their prejudices, the latter explain when frustrations lead to organized persecution of outgroups. Studying collective scapegoating presents significant methodological challenges, especially to laboratory-based experimental approaches. Specific Choice of Scapegoats 259 components of the theory, such as the attribution of blame, may be amenable to experiments, but laboratory simulations of the entire model may be difficult or impossible to construct. To study the most extreme manifestations of scapegoating, such as genocide, requires a social-historical analysis of stereotypes and mass political movements. Allport himself suggested that “[i]t is chiefly the historical method that helps us to understand [scapegoating]” (1954/1979, p. 246). A rigorous social-scientific approach could, however, be applied to archival data. For example, content analysis of the propaganda directed at scapegoated groups, either historically (e.g., Mein Kampf and Der Stürmer) or currently (e.g., conspiracy theories on the internet) could test whether scapegoats are stereotyped in the manner the model predicts. Comparisons of scapegoat movements from different cultures and time periods might reveal (or disconfirm) the underlying socialpsychological similarities the ideological model proposes. The twentieth century was marred by increasingly frequent genocide attempts inspired by nationalism and fascism. In the current century, terrorism rooted in religious fundamentalism poses an equally grave threat. The ideologies differ, but their underlying dynamics have something in common – the scapegoating of groups who are viewed as powerful and malevolent, and who therefore must be annihilated. The persistence and destructiveness of such movements makes it tempting to conclude that people are inherently irrational, aggressive, and malicious. Despite his appeal to psychodynamics to explain scapegoating, however, Allport concluded that “human nature seems, on the whole, to prefer the sight of kindliness and friendliness to the sight of cruelty” (1954/1979, pp. xv– xvi). The perspective presented here – which explains fanatical ideological commitment to destructive ideologies as a result of “normal” social psychological processes in response to extreme circumstances – is consistent with Allport’s cautious optimism that intergroup hostility is not an inevitable outcome of the human condition. REFERENCES Adams, H. E., Wright, L. W., & Lohr, B. A. (1996). Is homophobia associated with homosexual arousal? Journal of Abnormal Psychology, 105, 440–5. Adanir, F. (2001). Armenian deportations and massacres in 1915. In D. Chirot & M. E. P. Seligman (eds.), Ethnopolitical warfare: Causes, consequences, and possible solutions (pp. 71–81). Washington, DC: American Psychological Association. Allport, G. W. (1954/1979). The nature of prejudice. Cambridge, MA: Perseus Books. 260 Glick Altemeyer, B. (1981). Right-wing authoritarianism. Winnipeg: University of Manitoba Press. Bettelheim, B. & Janowitz, M. (1950). Dynamics of prejudice: A sociological study of veterans. New York: Harper and Brothers. Billig, M. (1976). Social psychology and intergroup relations. London: Academic Press. Cottrell, C. A. & Neuberg, S. L. (2004). Different emotional reactions to different groups: A sociofunctional threat-based approach to “prejudice.” Unpublished MS, Arizona State University. Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H., & Sears, R. R. (1939). Frustration and aggression. New Haven, CT: Yale University Press. Fein, S. & Spencer, S. (1997). Prejudice as self-image maintenance: Affirming the self through derogating others. Journal of Personality & Social Psychology, 73, 31–44 Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878–902. Friedländer, S. (1997). Nazi Germany and the Jews: The years of persecution, 1933–39 (vol. 1). New York: Harper Collins. Fritzsche, P. (1999). Germans into Nazis. Boston: Harvard University Press. Glick, P. (2002). Sacrificial lambs dressed in wolves’ clothing: Envious prejudice, ideology, and the scapegoating of Jews. In L. S. Newman & R. Erber (eds.), Understanding genocide: The social psychology of the Holocaust (pp. 113– 42). New York: Oxford University Press. Goldhagen, D. J. (1997). Hitler’s willing executioners: Ordinary Germans and the Holocaust. New York: Random House. Heider, F. (1958). The psychology of interpersonal relations. New York: John Wiley and Sons. Karlsen, C. F. (1987). The devil in the shape of a woman: Witchcraft in colonial New England. New York: W. W. Norton & Co. Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine (ed.), Nebraska symposium on motivation (vol 15., pp. 192–238). Lincoln: University of Nebraska Press. Kiernan, B. (2001). The ethnic element in the Cambodian genocide. In D. Chirot & M. E. P. Seligman (eds.), Ethnopolitical warfare: Causes, consequences, and possible solutions (pp. 83–91). Washington, DC: American Psychological Association. Lerner, M. J. & Miller, D. T. (1978). Just world research and the attribution process: Looking back and ahead. Psychological Bulletin, 85, 1030–51. Leyens, J. P., Paladino, P. M., Rodriguez, R. T., Vaes, J., Demoulin, S., Rodriguez, A. P., & Gaunt, R. (2000). Personality and Social Psychology Review, 4, 186–97. Milgram, S. (1974). Obedience to authority. New York: Harper and Row. Miller, N., Pedersen, W. C., Earleywine, M., & Pollock, V. E. (2003). A theoretical model of triggered displaced aggression. Personality and Social Psychology Review, 7, 75–97. Prunier, G. (2001). Genocide in Rwanda. In D. Chirot & M. E. P. Seligman (eds.), Ethnopolitical warfare: Choice of Scapegoats Causes, consequences, and possible solutions (pp. 109–16). Washington, DC: American Psychological Association. Pyszczynski, T., Greenberg, J., & Solomon, S. (1999). A dual process model of defense against conscious and unconscious death-related thoughts: An extension of terror management theory. Psychological Review, 106, 835– 45. Rhode, D. (2003). Radical Islam gains a seductive new voice. The New York Times, Oct. 26, D4–5. Runciman, W. G. (1966). Relative deprivation and social justice: A study of attitudes and social inequality in twentieth century England. Berkeley: University of California Press. Sidanius, J. & Pratto, F. (1999). Social dominance: An intergroup theory of social hierarchy and oppression. Cambridge: Cambridge University Press. 261 Staub, E. (1989). The roots of evil: The psychological and cultural origins of genocide. Cambridge: Cambridge University Press. Staub, E. (1996). Cultural-societal roots of violence: The examples of genocidal violence and of contemporary youth violence in the United States. American Psychologist, 51, 117–32. Sternberg, R. J. (2003). A duplex theory of hate: Development and application to terrorism, massacres, and genocides. Review of General Psychology, 7, 299– 328. Tajfel, H. (1981). Human groups and social categories: Studies in social psychology. Cambridge: Cambridge University Press. Zawadzki, B. (1948). Limitations of the scapegoat theory of prejudice. Journal of Abnormal and Social Psychology, 43, 127–41.