With malice toward none and charity for some: UC Santa Cruz UC Santa Cruz Previously Published Works Title With malice toward none and charity for some: ingroup favoritism enables discrimination. Permalink https://escholarship.org/uc/item/14x6w6t4 Journal The American psychologist, 69(7) ISSN 0003-066X Authors Greenwald, Anthony G Pettigrew, Thomas F Publication Date 2014-10-01 DOI 10.1037/a0036056 Peer reviewed eScholarship.org Powered by the California Digital Library University of California https://escholarship.org/uc/item/14x6w6t4 https://escholarship.org http://www.cdlib.org/ “With malice toward none . . .” Draft of 9 November 2013 -1- With malice toward none and charity for some: Ingroup favoritism enables discrimination Anthony G. Greenwald & Thomas F. Pettigrew University of Washington University of California, Santa Cruz Contact information: Anthony G. Greenwald Department of Psychology University of Washington Box 351525 Seattle WA 98195-1525 Email: agg@u.washington.edu Phone: (206) 543–7227 Fax: (206) 329-0268 RUNNING HEAD: With malice toward none mailto:agg@u.washington.edu “With malice toward none . . .” Draft of 9 November 2013 -2- With malice toward none and charity for some: Ingroup favoritism enables discrimination Abstract Dramatic forms of discrimination, such as lynching, property destruction, and hate crimes, are widely understood to be consequences of prejudicial hostility. This article focuses on what has heretofore been only an infrequent counter-theme in scientific work on discrimination—that favoritism toward ingroups can be responsible for much discrimination. We extend this counter- thesis to the strong conclusion that ingroup favoritism is plausibly more significant as a basis for discrimination in contemporary American society than is outgroup-directed hostility. This conclusion has implications for theory, research methods, and practical remedies. “With malice toward none . . .” Draft of 9 November 2013 -3- Imagine: You are a well-positioned manager in a large business. You supervise several other managers who also have substantial responsibility. One of your subordinate managers, Sylvia, mentions that her daughter, Kate, who is a school classmate of your daughter, was just sent home from school with the flu. You encourage Sylvia to take time off until Kate can return to school. When it later becomes time for you to conduct Sylvia’s annual performance review you have a problem because her above-average performance falls just between levels that could justify your giving her an overall judgment of “meets expectations” or “exceeds expectations.” You opt for “exceeds expectations,” which ultimately helps Sylvia to qualify for a promotion and a salary raise. Another employee, Robert, is equally above average. Robert’s records show that he too missed several days of work, but you do not know him as well and do not know why he missed work. You give Robert a “meets expectations” evaluation, and he gets a smaller raise and no promotion. It is not difficult to understand why you, in your managerial role, would resolve doubt more favorably for a supervisee with whom your daughter provides a personal connection. Theories to explain effects of such personal connections on social judgment have existed at least since Heider’s (1958) analysis of interpersonal relations. Related explanations appear in more recent theoretical analyses of social identity (e.g., Brewer, 1999; Gaertner et al., 1997; Greenwald et al., 2002; Tajfel & Turner, 1979). If your favorable judgment of Sylvia’s work performance is understandable, then it is also understandable that many other types of connections between people—including ones due to shared race, ethnicity, age, religion, or perhaps even just a shared birthday (Finch & Cialdini, 1989)—can likewise result in tipping the balance toward a favorable judgment, giving “the benefit of the doubt.” This role of ingroup connections in shaping favorable feelings, judgments, and actions underlies this article’s thesis that ingroup-directed favoritism is, in the United States, a more potent engine for discriminatory impact than is outgroup-directed hostility. Quite often ingroup favoritism is hidden even from those who practice it. Consider the way much job recruitment occurs. Good workers are asked frequently to seek out others for job “With malice toward none . . .” Draft of 9 November 2013 -4- openings. Because of extensive (so-called “de facto”) racial segregation in residences, schools, and workplaces, this practice often leads to White workers, drawing on virtually all-White acquaintanceship networks, to seek out only other Whites for vacancies (Reskin, 1998; Rivera, 2012). By drawing heavily on ingroup ties, this unremarkable process can sustain or exacerbate racial or other imbalances, entirely without involvement of hostility toward minorities. In qualitative studies of White workers, DiTomaso (2012) shows how such ingroup-enabled networking affords increased access to job openings. Prejudice, Hostility, Discrimination, and Ingroup Favoritism The Oxford English Dictionary defines prejudice as “. . . dislike, hostility, or unjust behaviour deriving from preconceived and unfounded opinions.” 1 This definition, which links prejudice to both hostile intergroup attitude and discriminatory behavior, fits well with scholarly scientific analyses. To confirm that this understanding of “prejudice” was not merely our own caricature of scientific treatments, we searched for authoritative definitions in published research on prejudice and also in well-used past and current social psychology texts that we found on our bookshelves. We retrieved 24 definitions, 18 of which explicitly connected prejudice to negative attitude or negative evaluation; four others identified prejudice with either positive or negative attitudes; the remaining two identified prejudice with emotional reactions without specifying the emotions’ affective character. Only two of the 24 definitions we retrieved included ‘discrimination’ as part of the definition of prejudice. Following Gordon Allport’s (1954) lead, social psychologists have long held that the connection of prejudicial attitude to discriminatory behavior is not something to be assumed but, rather, something that requires empirical demonstration. Complementing this scientific understanding that the link between prejudicial attitude and discrimination is not obligatory, politicians and lawyers have likewise treated them as separable. With the exception of a few references to “hostile work environments,” America’s civil rights laws interpret discrimination in non-emotional terms, making it illegal to treat people unequally “because of” race, skin color, sex, religion, national origin, age, or disability status. Civil rights laws do not take either hostility or negative prejudicial attitude to be a necessary feature of discrimination. Likewise, in this article, we hold that discrimination does not require hostility. Unequal treatment can be 1Retrieved at http://www.oxfordreference.com, July 1, 2012. http://www.oxfordreference.com/ “With malice toward none . . .” Draft of 9 November 2013 -5- produced as readily (or, as we will conclude, more readily) by helping members of an advantaged group as by harming members of a disadvantaged group. The scientific study of prejudice has been pursued uninterruptedly since the introduction of the first measures of intergroup attitudes by Bogardus (1925) and Thurstone (1928). In this (now) massive body of scientific work, one is unlikely to encounter completely new ideas. True to that expectation, this article’s central thesis—that ingroup favoritism is a prime cause of discrimination—is not new. The importance of ingroup favoritism in discrimination was described especially clearly by both Gaertner et al. (1997) and Brewer (1999). Gaertner et al. wrote that “. . . racial bias, particularly in its contemporary manifestations, may reflect a prowhite, not simply [the] antiblack sentiment that many traditional theories and measures have implied” (p. 175). Brewer wrote: “Ultimately, many forms of discrimination and bias may develop not because outgroups are hated, but because positive emotions such as admiration, sympathy, and trust are reserved for the ingroup and withheld from outgroups” (p. 438). In a review article, Hewstone, Rubin, and Willis (2002) carefully considered the interplay of ingroup favoritism and outgroup hostility as components of intergroup bias, in the process presaging several topics central to this article. More recently, Dixon, Levine, Reicher, and Durrheim (2012) asked, “Has the time come to challenge the assumption that negative evaluations are inevitably the cognitive and affective hallmarks of discrimination?” (p. 411). We proceed further in this direction, concluding that, at least in North America, ingroup favoritism is the prime mechanism of discrimination.2 Perhaps because the most dramatic forms of discrimination contain no hint of ingroup favoritism, statements such as those by Gaertner et al. (1997) and Brewer (1999) have not displaced the view that outgroup hostility is discrimination’s primary antecedent. Studies collected for two meta-analyses confirm this observation. Among 1,351 individual tests of effects of intergroup contact on intergroup relations (Pettigrew & Tropp, 2006; 2011) not even one of these employed a measure of ingroup favorability as a dependent variable. Likewise, among 370 tests analyzed by Smith, Pettigrew, Pippen & Bialosiewicz (2012) to determine the effects of group relative deprivation on prejudice, only two used ingroup favorability as a 2 At about the same time that it began to appear in the psychological literature, the proposition that workplace discrimination could occur importantly as ingroup favoritism was also beginning to appear in legal scholarship (see Krieger, 1998). “With malice toward none . . .” Draft of 9 November 2013 -6- dependent variable (Amiot, Terry, & Callan, 2007; Terry & O’Brien, 2001). The only ingroup- related measures—used in a small minority of tests in these two meta-analyses—were measures of collective self-esteem and ingroup identification. Consider, too, the Journal of Social Issues special issue in 2012 that focused on discrimination (Nier & Gaertner, 2012). Several authors in that issue acknowledged that hostility is not a necessary precondition for discrimination. For example, Nier and Gaertner (2012, p. 218) wrote: “In many cases, . . . discrimination is likely to be subtle and difficult to detect, and in some instances, may be unintentional.” Nevertheless, throughout the special issue the dominant assumption was that, in most instances, hostile prejudice is the wellspring of discrimination. And common throughout the research literature are studies that show how hostile prejudice is linked with discriminatory intentions without regard for ingroup favorability (e.g., Wagner, Christ & Pettigrew, 2008). In three steps, the remainder of this article builds a case for understanding ingroup favoritism as not just a cause but as the prime cause of American discrimination. First, we review findings supporting the two phenomena that are merged in this article’s main thesis: existence of strong positive dispositions toward ingroup members (i.e., ingroup favoritism), and evidence that discrimination occurs more often as differential favoring than as differential harming. Although both of these phenomena are thoroughly established empirically, the connection between them has had very little recognition. Second, we consider theories that explain the psychological antecedents of favoritism. Third, we review research methods that have been used in studies of discrimination to understand why there are so few available tests of the link between ingroup favoritism and discrimination. In the concluding discussion, we reflect on implications of this article’s ingroup favoritism thesis. Findings: Favoring Ingroup Members Similarity and Attraction Byrne (1961) introduced a method for investigating attraction as a function of attitude similarity. In an initial session, experimenters obtained subjects’ responses to 26 attitude questions. Two weeks later, the same subjects were asked to evaluate an otherwise unknown person for whom the only available information consisted of that person’s responses to the same “With malice toward none . . .” Draft of 9 November 2013 -7- 26 questions. Unknown to subjects, the attitude responses of these “strangers” had been filled out by researchers so as to vary systematically, in four levels, ranging from exactly agreeing with all of the subjects’ own responses to exactly disagreeing with all of them. Byrne’s finding, which proved to be robustly replicable, was that liking and attraction toward the strangers were strongly a function of attitude similarity. In Byrne’s (1961) report, across six dependent measures, effect sizes for the greater positivity of evaluations for most versus least similar strangers averaged a Cohen’s d of 3.40, constituting a very large effect. As a reference point, Cohen (1977) described d = 0.80 as a “large” effect. Subsequent studies showed that Byrne’s similarity–attraction principle was not limited to effects of attitude similarities; it occurred equally for similarities in personality traits and for similarities in behavior (e.g., Byrne, Ervin, & Lamberth, 1970). Furthering Byrne’s contentions, Rokeach proposed that the perception of conflicting beliefs and values triggered race prejudice more than does race itself (Rokeach, 1960; Rokeach & Mezei, 1966). This is plausible if, as Rokeach supposed, outgroup members were typically assumed to harbor beliefs and values conflicting with those of the ingroup. Rokeach’s position was initially controversial (see Stein, Hardyck, & Smith, 1965; Triandis, 1961; Triandis & Davis, 1965), but the controversy gradually disappeared. And, as already mentioned, by 1970 the similarity–attraction principle had been extended beyond attitudes to other characteristics (Byrne et al, 1970). Nevertheless, the effect of (especially) race similarity on interpersonal attitudes has continued to be interest. In work settings, evaluations have often been shown to be more favorable when the evaluator and evaluatee (e.g., a hiring manager and a job applicant) are similar, rather than different, in race or gender (e.g., Riordan, 2000). Interpretation of this demographic similarity effect as a form of discriminatory bias has been made plausible by reports that the effect can be minimized or eliminated when highly structured interview methods are used (e.g., McCarthy, Van Iddekinge, & Campion, 2010; Sacco, Scheu, Ryan, & Schmitt, 2003). Highly structured interviews are understood to minimize discriminatory effects because they leave little to the interviewer’s subjectivity or discretion (e.g., Heilman & Haynes, 2008). The interesting question as to whether effects of demographic similarity in the workplace are due to ingroup favoritism or outgroup hostility has not been directly addressed in most of the available research. However, the (earlier mentioned) studies by DiTomaso (2012), Reskin (2008), and Rivera (2012) are “With malice toward none . . .” Draft of 9 November 2013 -8- supportive of a favoritism interpretation. In sum, the similarity–attraction principle is consistent with an expectation that attitudes toward members of one’s own group (ingroup) will typically be more positive than attitudes toward members of other groups (outgroups). The Minimal Group Paradigm (MGP) “With malice toward none . . .” Draft of 9 November 2013 -9- More than 40 years after the discovery of minimal group paradigm (MGP) effects, the original report of that finding by Tajfel, Billig, Bundy and Flament (1971; see also Tajfel, 1970) continues to shape research and theory on intergroup relations. Tajfel et al. found that, even when subjects were assigned arbitrarily to groups in a laboratory study, they preferentially allocated resources to members of their own group rather than to those in another group. Later studies found that this occurred even when subjects knew that the basis for assignment was random. It also occurred when subjects did not know which, of the other subjects who were present, were members of their own group or members of the other group (Billig & Tajfel, 1973). Platow and Van Knippenberg (2001), using the MGP, showed that subjects expect and believe it is fair for an ingroup member to treat all ingroup members fairly. But they also tend to expect and believe it fair for an ingroup member to favor another ingroup member over an outgroup member. Not surprisingly, ingroup members cannot be expected to identify their ingroup favoritism as discrimination when they see their behavior as legitimate, normative and even procedurally fair. A further study by L. Gaertner and Insko (2001) showed that distributions of monetary payments in the MGP are partly constrained by equity norms—which prescribe giving equal rewards to all. Nevertheless, these investigators also observed ingroup favoritism when the monetary distributions were described as bonus payments (making equity irrelevant) and when subjects were asked to describe their feelings toward unspecified members of each group. Their article also introduced a new dependent measure format (“multiple alternative matrices”) that avoided a strict inverse relationship between outcomes to ingroup and outgroup members. This measure allowed the conclusion that, on average, subjects were more motivated by ingroup favoritism (higher payments to ingroup members) than by outgroup hostility (lower payments to outgroup members). In the context of similarity–attraction research, MGP research indicates that similarity that is solely due to membership in the same group suffices to provide a basis for both attraction and favoritism, even when no specific attributes are known to be shared with members of that (minimal) ingroup. Greater Differences between Ingroup and Outgroup in Positive than Negative Feelings If negative outgroup attitudes are expressed primarily in hostile form, we can expect that negative emotions should be more readily expressed toward outgroups than ingroups—a “With malice toward none . . .” Draft of 9 November 2013 -10- phenomenon that is indeed observed (cf. Smith, 1993; Smith & Mackie, 2005). A rarely addressed empirical question (but cf. Dovidio, Mann & Gaertner, 1989; Gaertner & McLaughlin, 1983) is how ingroup–outgroup differences in negative feelings compare to ingroup–outgroup differences in positive feelings. A study of racial attitudes in the 2012 American Presidential election (Ziegler, Kirby, Xu, & Greenwald, 2013) included data from more than 45,000 volunteers who responded to two measures of emotional responses to Black and White persons. One of these was a standard feeling thermometer, which asked “Please rate how warm or cold you feel toward White [or Black] people.” We limited analysis to Whites who expressed strong White preference and Blacks who expressed strong Black preference to be sure that we were appropriately focusing on Whites and Blacks for whom their racial group was justifiably considered an “ingroup”. For these participants, thermometer responses of strong warmth toward the ingroup exceeded expressions of strong cold feelings toward the outgroup by a ratio of approximately 4:1 for Whites and more than 50:1 for Blacks. That is, expressions of warmth toward the ingroup greatly exceeded expressions of coldness toward the outgroup. The second emotion measure used by Ziegler et al. (2013) was an adaptation of Pettigrew and Meertens’s (1995) measure of subtle racism. It consisted of two items asking “How often have you felt _____ for African Americans who grew up in slums and poverty?” The blank was replaced by “sympathy” or “admiration” in the two items. The same items were also asked about “Americans,” in place of “African Americans.” Subjects of both races whose Likert-item responses stated “very strong” racial ingroup preference were more disposed to feel positively toward impoverished members of their own racial group than toward comparable members of the outgroup.3 Similar findings with the same two items have emerged in European research. Pettigrew and Meertens (1995; Meertens and Pettigrew, 1997) found in seven independent samples from France, Germany, Great Britain and the Netherlands that these items correlated highly with other measures of prejudice. Although the sympathy and admiration items are typically not seen by native Europeans as reflecting prejudice, they nevertheless predicted pro-discrimination beliefs. 3 Note that the researchers assumed that “Americans” would be assumed to be White Americans. This is presumably a valid assumption, but we are not aware of direct empirical tests. “With malice toward none . . .” Draft of 9 November 2013 -11- For example, those who reported that they rarely or never felt sympathy or admiration for immigrants were significantly more likely to support expelling immigrants who have committed crimes or who have no immigration papers. These survey respondents are more appropriately described as withholding positive emotions (sympathy and admiration) from immigrants than as expressing negative feelings toward immigrants. Findings: Discrimination Often Occurs as Differential Favoring Studies of intergroup behavior in field settings have examined the extent to which significant discriminatory effects result from differential helping or favoring. Although there are limitations on conclusions from these studies that we will mention, nevertheless they are quite consistent in showing the potential for discrimination to result from differential favoring. Helping Behavior Experiments on unobtrusively observed helping of ingroup and outgroup members began with an ingenious study using a “wrong number” method devised by Gaertner and Bickman (1971). Researchers, speaking with accents that were racially identifiable as Black or White, placed telephone calls in which they claimed to be stranded drivers calling an automobile mechanic and urgently needing help for their disabled cars. Because of strong residential segregation in the Brooklyn, New York, neighborhoods to which those calls were directed, researchers could know whether the call recipients were racially White or Black. Claiming to have used his last coin in a pay phone, the caller asked the recipient to help by calling the mechanic to relay the emergency request for road service. The key finding: White call recipients discriminated by race—they were less likely to help Black callers (53 percent) than White callers (65 percent). Three later repetitions of the experiment replicated Gaertner and Bickman’s finding (see Crosby, Bromley, & Saxe, 1980; Saucier, Miller, & Doucet, 2005). A crucial element of Gaertner and Bickman’s (1971) wrong-number method was that potential helpers could not know that their helping or non-helping was being monitored. Dozens of later experiments in the 1970s similarly used unobtrusive measures to compare the amount of help that Black and White help seekers would receive from White potential helpers. In reviewing the accumulated collection of more than 30 such studies, Crosby et al. (1980) “With malice toward none . . .” Draft of 9 November 2013 -12- concluded, “Discriminatory behavior is more prevalent in the . . . unobtrusive studies than we might expect on the basis of survey data” (p. 557). Similar results have more recently been found in studies of tipping behavior. Obtaining the cooperation of taxi drivers in New Haven, Connecticut, Ayres, Vars, and Zakariya (2005) asked the drivers to keep records of fares and tips. White drivers received tips that were 51% larger than those received by Black drivers, a ratio that was even greater when computed as a percentage of fare. A study of restaurant tipping behavior by Lynn et al. (2006; 2008) likewise revealed a race difference—White waiters received tips that, on average, were 22 percent larger than those received by Black waiters. Hiring and Housing Audits Many field experiments using audit methods have assessed discrimination in employment and housing. The standard audit method uses paired testers who differ in race or ethnicity while being matched in relevant qualifications such as (for housing audits) income, assets, debt levels, family circumstances, employment history, credit record, and neatness of appearance. Randomizing which of the two members of each pair arrives first, the two testers apply for work or housing to each of a large sample of hiring managers or real estate agents. This method almost invariably reveals discrimination against Blacks and Hispanics in access to jobs (reviewed by Bendick, 2004) and housing (e.g., Turner, Ross, Galster & Yinger, 2002). Results obtained from field audit experiments strongly suggest that significant acts of discrimination in housing and employment can often occur without expression of hostility toward the people who are disadvantaged by those acts. Some of the individual hiring managers or real estate agents might have been hostile in their denials of consideration for job interviews or housing. However, the great majority of declinations of hiring and housing applications involve simply the non-occurrence of a helpful act—either the act of inviting the job seeker for an interview or the act of escorting the housing seeker to view an apartment or home. The greater non-occurrence of those helpful actions when applicants are Black or Hispanic than when they are White can effectively cause substantial discrimination in housing or hiring.4 Recently, 4 A necessary qualification in regard to ingroup favoritism follows from audit studies’ lack of systematic report of effects due to variations in race or ethnicity of real estate agents and hiring managers. Although it is a near certainty that the majority of agents and managers in all of the published audit studies were White, it remains unknown whether ingroup favoritism occurred “With malice toward none . . .” Draft of 9 November 2013 -13- Bendick (2007) and colleagues (Bendick, Rodriguez, & Jayaraman, 2010), using a method of situation testing in hiring studies, have shown that discrimination in hiring interactions is linked more to occurrences of favorable than of hostile actions. Policing Large bodies of data on discrimination in policing have been accumulated in studies of profiling by police in their interactions with pedestrians or with drivers who have been stopped for driving violations or vehicle maintenance infractions. Discrimination is evident when there is a greater probability of searching or issuing a citation when the driver is Black or Hispanic rather than White, or greater probability of subjecting Black or Hispanic pedestrians to search (Lamberth, 1994; Office of the Attorney General, 1999; Verniero & Zoubek, 1999; Weiss & Rosenbaum, 2009). In summarizing available profiling data, Chanin, Rintels et al. (2011) concluded that: (a) Blacks and Hispanics were stopped more frequently than Whites; (b) among those stopped, higher proportions of Blacks and Hispanics than Whites received citations; (c) among those stopped, higher proportions of Blacks and Hispanics than Whites were subjected to searches; and (d) among those searched, a smaller proportion of the searches of Blacks and Hispanics than Whites yield contraband (e.g., drugs or weapons). The lower yields of contraband from searches of Blacks and Hispanics establishes that the greater searching of Blacks and Hispanics is not justified by greater criminal activity of Blacks or Hispanics than of Whites among those who are stopped. The greater rate of discovering contraband from searches of vehicles driven by Whites suggests that White drivers are being stopped and searched at inappropriately low rates. This is consistent with the proposition that discrimination reflected in profiling data is in part—perhaps large part—due to favorable acts of either not stopping White drivers or (as the data show) not searching their vehicles after they are stopped. Public Opinion Surveys Surveys of White Americans’ racial attitudes over the past 50 years show that Whites have steadily increased their support for policies that provide educational or housing opportunities for African Americans (this and the other findings mentioned in this paragraph are documented in equally across variations in agent or manager race and ethnicity. This same qualification applies to the following section on policing. “With malice toward none . . .” Draft of 9 November 2013 -14- Schuman, Steeh, Bobo & Krysan, 1997). Surveys also have revealed White Americans’ increasing support for racial intermarriage, for equal opportunity in employment, and for the acceptability of an African-American candidate for U.S. President. At the same time, White Americans’ levels of support for policies that provide governmental help to minorities have been largely unchanged during the past half century. In particular, majorities of White Americans have steadily opposed financial assistance to minorities, social services to minorities, and affirmative action to benefit minorities in hiring or college admissions. Put differently, national American surveys across the past five decades have found that most White Americans accept basic principles of equal opportunity while, at the same time, resisting the implementation of policies that would increase equality directly by helping outgroups (see also DiTomaso, 2012; Pettigrew, 1979). In combination, these evidences of White Americans’ current high levels of opposition to both anti-minority segregation policies and pro-minority assistance policies suggest that the preferred policies of White Americans amount to a generalized anti-discrimination stance. That is, they favor neither segregation policies that could potentially harm Black Americans nor assistance policies that could potentially favor impoverished Black Americans over other impoverished Americans. At the same time, there are some government assistance programs that many White Americans do support. As was documented by political opinion polling during the American presidential campaign of 2012, White Americans (more than other demographic categories) supported tax laws that assist relatively wealthy Americans. Because America’s Black and Hispanic minorities are underrepresented at high income levels, benefits received via tax laws necessarily help Whites more than racial minorities. It appears that many White Americans’ are more likely to oppose laws that would disproportionately benefit relatively impoverished minorities than ones that disproportionately benefit relatively wealthy Whites. Although some of Whites’ support for benefits via tax laws can be attributed to economic self- interest, the supported laws often include benefits that affect only Americans much wealthier than those laws’ supporters. “With malice toward none . . .” Draft of 9 November 2013 -15- Theories: Roots of Favoritism A variety of theories explain conditions that promote and sustain favoritism—not limited to ingroup favoritism. This section describes theoretical accounts of favoritism stated at four levels of psycho-social analysis: intrapersonal, interpersonal, intergroup, and societal. Balance Theory and Balanced Identity Theory In explaining the powerful similarity–attraction phenomenon revealed in his research, Byrne (1961) proposed that “. . . any time that another person offers us validation by indicating that his percepts and concepts are congruent with ours, it constitutes a rewarding interaction and, hence, one element in forming a positive relationship” (p. 713). Byrne advanced this reward theory toward the end of an era in which learning–reinforcement theories were psychology’s dominant theories. By the 1960s, however, those reinforcement theories were in decline. At that same time, affective–cognitive consistency theories—especially Heider’s (1958) balance theory, Osgood and Tannenbaum’s (1955) congruity theory, and Festinger’s (1957) dissonance theory— were on the ascent in social psychology (see Abelson et al., 1968). Although consistency theories themselves declined in the 1970s, they have recently experienced a resurgence (cf. Gawronski & Strack, 2012). The recent “balanced identity” theory (BIT), which was developed as an extension of Heider’s balance theory (Greenwald et al., 2002) offers a cognitive consistency interpretation of similarity–attraction. BIT’s balance–congruity principle holds that two concepts that are both associated with the same third concept will become associated with each other. When a newly encountered person (P) is an ingroup member, both self and P are associated with the ingroup (a third concept) that they share. BIT’s balance–congruity principle therefore predicts that the association between self and P will strengthen. The same principle then extends to the combination of the new (self–P) association and the pervasive association of self with positive valence (i.e., self-esteem). When P and positive valence are thus both associated with self (third concept), the association between P and positive valence should itself strengthen, theoretically explaining attraction to the ingroup member (P). “With malice toward none . . .” Draft of 9 November 2013 -16- Social Identity Theory Social identity theory (SIT) was developed by Tajfel and Turner (1979) in part to account for Tajfel et al.’s (1971) findings of intergroup discrimination in the minimal group paradigm. In contrast to BIT’s association-formation interpretation of the relation between self-esteem and ingroup favoritism, SIT offers a motivational interpretation rooted in understanding self-esteem as a motive with the goals of achieving and sustaining positive self-regard. SIT links intergroup discrimination in the MGP to increased self-esteem in two ways (Abrams & Hogg, 1988): Either (a) a self-esteem increase is achieved as a consequence of perceiving one’s own group as superior to the other in the MGP, or (b) approximately the reverse—the motive to elevate self- esteem is the cause of perceiving one’s group as superior to the other (Abrams & Hogg, pp. 320– 321). Hewstone, Rubin, and Willis (2002) summarized SIT’s self-esteem hypothesis this way: “(1) successful intergroup bias enhances self-esteem and (2) depressed or threatened self-esteem motivates intergroup bias” (p. 41). However, Hewstone et al. also concluded that the evidence for this SIT theorization is at best mixed. System Justification Theory If ingroup favoritism is practiced equally by all, then greatest benefits will necessarily flow to members of a society’s more powerful groups; their greater power, along with their (typically) greater numbers translate to their being better positioned to benefit from ingroup helpers. System justification theory (SJT; Jost & Banaji, 1994) explains how a complementary form of favoritism, rooted in existing status, adds to the benefits accruing to high status groups. Jost and Banaji defined system justification as a “process by which existing social arrangements are legitimized, even at the expense of personal and group interest” (p. 2). There has been substantial empirical support for this theorized reversal of ingroup favoritism for low- status groups, including the prediction that “[a]s system justification tendencies increase . . . members of low-status groups will exhibit increased outgroup favoritism” (Jost, Banaji & Nosek, 2004, p. 901, emphasis added). In their review of 10 years of SJT research, Jost et al. (2004) cited substantial evidence for SJT’s outgroup favoritism hypothesis. This evidence took the form of finding outgroup-favoring attitudes among members of low-status minorities, sometimes assessed with unobtrusive “With malice toward none . . .” Draft of 9 November 2013 -17- behavioral indicators (Jost, Pelham, & Carvallo, 2002). Behavioral evidence of outgroup favoritism evidence was also obtained in some of the unobtrusive-measure studies of helping of the 1970s and in the more recent studies of tipping behavior. Although the first of the unobtrusive helping studies (Gaertner & Bickman, 1971) found (nonsignificantly) that Blacks helped Whites more than they helped fellow Blacks, subsequent replications did not show that pattern. On the other hand, the two tipping studies—by Ayres et al. (2005) with taxi passengers and by Lynn et al. (2006, 2008) with restaurant patrons—both found, consistent with SJT’s expectations, that Black customers gave larger tip percentages to White than to Black service providers. When, as theorized in SJT, favoritism thus extends to an advantaged outgroup, the consequence is to exacerbate the relative disadvantage of lower status groups. SJT thus explains an additional source of favoritism directed toward a society’s highest status groups. For a high status group that constitutes a societal minority, this second source explained by SJT might exceed ingroup favoritism as a basis for sustaining the group’s advantage. Unrecognized Discrimination and Illusory Individuation Learning about a person’s distinctive characteristics—“individuating” that person—is widely understood as a means of overcoming the disadvantaging effects of stereotypes (Fiske & Neuberg, 1990; Kunda & Thagard, 1996; Locksley, Borgida, Brekke, & Hepburn, 1980). An unfamiliar person whose distinctive characteristics are unknown may be judged, unthinkingly, by applying stereotypes. However, knowing specific (individuating) characteristics of a person logically should pre-empt this use of stereotypes. This plausible theory notwithstanding, a series of studies conducted over the last thirty years has demonstrated that the expected reduction of stereotyping by individuation is rather easily thrown off track. Instead, stereotypes can insinuate themselves subtly into apparently individuated judgments that can prove disadvantageous to outgroup members. Darley and Gross (1983) found that their college-student subjects resisted applying stereotypes to judge the academic skills of a 9-year-old child (Hannah) for whom the only available information made clear that Hannah’s upbringing had been in either an impoverished or a well-to-do family environment. Perhaps the research setting put these subjects on alert not to “With malice toward none . . .” Draft of 9 November 2013 -18- let their knowledge of Hannah’s socioeconomic status influence their judgment of her academic skills. In two further conditions, subjects additionally observed a 12-minute videotape of Hannah’s responses to 25 ‘achievement-test’ questions. Her performance on these—showing Hannah giving a mixture of correct and incorrect responses—gave no clear impression of Hannah’s ability. Nevertheless, findings showed that, in these two individuated conditions, subjects interpreted the added information by applying social class stereotypes. They credited the well-to-do Hannah with having abilities at a higher grade level than the working-class Hannah. Exposure to the extra (presumably individuating) information apparently licensed subjects to apply stereotypes that they resisted applying when they had no opportunity to observe Hannah’s test performances. The process afforded by the videotape plausibly left subjects unaware that their knowledge of Hannah’s socioeconomic status had in any way affected their judgment. Yzerbyt, Schadron, Leyens, and Rocher (1994) conducted two experiments inspired by Darley and Gross’s (1983) finding. Remarkably, they found similarly that stereotype-confirming effects occurred even when subjects received no actual individuating information. Instead, they had merely been told that relevant information had been presented to one of their ears, outside of conscious awareness, in a selective listening (“shadowing”) task that required repeating an audible message presented to the other ear. Hodson, Dovidio, and Gaertner (2002) constructed a further variation on the use of illusory individuating information. They observed White subjects’ evaluations of two presumed college applicants, one White and one Black, whose qualifications differed. Although the two applicants were otherwise matched, one applicant was higher in high school grades and the other was higher on a standardized aptitude test. The two applicants therefore deserved, objectively, to be treated as approximately equally qualified. Hodson et al.’s noteworthy finding was that, in comparing the White and Black applicants, subjects who scored relatively high on a measure of prejudice attributed greater predictive weight to the measure on which the White applicant was superior. Again, this result reveals discrimination in the presence of actually uninformative, but presumably individuating, information. This discriminatory use of the information was apparent to experimenters who could compare the data from different conditions, but the subjects themselves had no basis for suspecting that stereotypes had influenced their judgments. “With malice toward none . . .” Draft of 9 November 2013 -19- Stronger biased-processing findings of the type obtained by Hodson et al. (2002) were obtained in subsequent studies by Norton, Vandello, and Darley (2004) and by Uhlmann and Cohen (2005). The biased processing observed in these studies was labeled variously hypothesis-confirming bias (Darley & Gross, 1983), social judgeability bias (Yzerbyt et al., 1994), differential weighting (Hodson et al., 2002), casuistry (Norton et al., 2004), and constructed criteria (Uhlmann & Cohen, 2005). These variations in labels notwithstanding, the several studies support each other in demonstrating the readiness with which people “engage in biased behavior while retaining a view of the self as objective” (Norton et al., 2004, p. 828). In each case the bias started with exposure to information that presumably afforded a basis for objective, individuated judgment, but was nevertheless used in biased fashion. These illusory individuation phenomena relate to this article’s main point in showing a subtle form of favoritism that can give the benefit of doubt to an ingroup member.5 Conformity to Social Norms Sociologists stress the importance of societal structures in producing intergroup discrimination (DiTomaso, 2012; Pettigrew, 1975; Pettigrew & Taylor, 2002). An important form of this theory is that historical realities such as past slavery and immigrant poverty are inevitably associated with differences in employment and wealth, and consequently with residential segregation. The resulting limitations of intergroup contact in turn provide a breeding ground for perceived differences that can take the form of stereotypes, intergroup threat, and wariness of the outgroup if not outright dislike (Pettigrew & Tropp, 2011). Just-world reasoning (Lerner, 1980) or “blaming the victim” (Ryan, 1976) can lead to perceiving the impoverished as deserving of their low status, in effect having brought their disadvantages on themselves. In this way, the residue of past discrimination can sustain and even exacerbate discrimination in a “vicious circle” (Myrdal, 1944). Sociologists appeal to norms, which are widely shared understandings of what constitutes acceptable social behavior. Both formal norms (e.g., a posted 65-mph automobile speed limit) 5 Norton et al.’s (2004) Studies 3–6 found that the majority of simulated college admissions choices by their (mostly White) subjects favored Black over White applicants. These decisions were clearly not ingroup-favoring. Nevertheless, Norton et al.’s findings agreed with the other studies in this collection by finding that subjects were highly flexible in their weighting of qualification criteria, with this flexibility serving to justify outgroup-favoring biased decisions based on knowledge of applicants’ racial categories. In non-laboratory settings in which those who make judgments do not expect that their judgments will be monitored by others, illusory individuated judgments may more consistently favor ingroup members. “With malice toward none . . .” Draft of 9 November 2013 -20- and informal norms (keeping to the right if driving below the speed limit) powerfully guide behavior (Pettigrew, 1991, 2011). Also important are rewards and punishments that may independently shape the norms (e.g., an enforced 75-mph limit on a highway with a posted 65- mph limit). Norms shape intergroup interaction and provide a common meaning to all participants in the interaction. Moreover, discriminatory norms are typically cumulative. That is, discriminatory norms build on themselves and tend to reinforce each other across societal realms. Thus, America’s extreme residential segregation by race shaped and continues to maintain racial discrimination in employment, schools, home mortgages, and civic services generally (Pettigrew, 1975; Pettigrew & Taylor, 2002). Norms are also self-perpetuating. They come in time to be unquestioned, to be accepted simply because “that’s just the way things are done” (Pettigrew, 1991, 1998). The persistence of norms means that their discriminatory effects can outlive the initiating past causes of discriminatory practices. Those who initiated the norms may have been motivated by hostile prejudice; but later generations can adhere to norms that benefit their ingroup without harboring the animosity felt by the norms’ creators. The result is that norms are likely to remain unchanged even while attitudes are shifting markedly. At present, in the midst of rapid formal change in intergroup relations in Northern Ireland, South Africa, and the southern United States, old norms are slowly receding while new equalitarian norms have yet to develop fully. In all three societies, reticence and awkwardness characterize intergroup interaction, often accompanied by intergroup avoidance and informal discrimination that occurs without hostile intent (see especially Dovidio & Gaertner’s analyses of aversive racism—e.g., Dovidio & Gaertner, 2004; also Pettigrew, 1991, 2011). Norms can be so unquestioned that people think and behave in conflicting ways in different social contexts, while remaining unaware of the inconsistency. At an Indiana steel mill in the 1950s, Blacks and Whites were members of the same racially desegregated union and worked well together (Reitzes, 1953; for a similar example, see Minard, 1952). Only 12% of the White workers reported low acceptance of African Americans on the job. Those most involved in the union were the strongest supporters of the union’s pro-desegregation norms. Yet these same White workers also lived in racially segregated, all-White neighborhoods and many belonged to activist pro-segregation neighborhood groups. Indeed, 84% of those who accepted African “With malice toward none . . .” Draft of 9 November 2013 -21- Americans at work were highly resistant to Blacks living in their neighborhoods. Those most involved with their neighborhoods were the most resistant to having Blacks as neighbors. The behavior in each setting favored the setting’s prominent ingroup—union solidarity at work, racial solidarity at home. Allport (1954) considered conformity an essential concept for understanding prejudice, devoting a full chapter of his classic volume to the subject. In a later article, Allport (1962) wrote that “Conformity is the missing link that explains why and how societal forces eventuate into patterns of acceptance or discrimination . . .” (p. 132). Thus, conformity research also supports our thesis. Relatively unprejudiced Americans typically follow their ingroup’s norms. If, as is typical, these norms demand preferential treatment of the ingroup, most people—like the Indiana steel workers—will conform without personal animus toward the outgroup. Method Limitations Discrimination has been investigated with a wide range of measures, including behavioral interactions (overt and nonverbal), behavioral intentions, interpersonal judgments, self-reported attitudes, and implicit attitudes—all of which we consider in this section. We will explain why most research procedures used in studies of prejudice are inadequate to distinguish ingroup favoritism from outgroup hostility as causes of discrimination. The consequence is that relatively few studies provide data optimal for evaluating the role of ingroup favoritism in discrimination. To distinguish ingroup favoritism from outgroup hostility as a cause of discrimination, a study must meet two requirements. First, the study’s measures must distinguish favorableness from hostility in thoughts, feelings, or behavior toward others. To do this a measure must have an unambiguous neutral point—a value that is neither favorable nor hostile. Second, the study must use a design that permits comparison between behavior toward ingroup members and behavior toward outgroup members. The first requirement is needed to distinguish favorable from hostile behavior. The second requirement is needed to assess whether discrimination has occurred. Only a small fraction of the many existing studies of intergroup behavior meet even one of these two criteria, and many lack both. Most of this section analyzes methods regarding “With malice toward none . . .” Draft of 9 November 2013 -22- the first requirement. Shortcomings regarding the second requirement are more easily and briefly described at the end of this section. Overt Behavior Measures Numerous experiments have investigated discrimination by using unobtrusive assessments of overt helping behavior (see the review by Crosby et al., 1980). These studies succeeded in unambiguously identifying favorable behavior, by scoring subjects simply as helping or as not helping (inaction). Inaction is neutral behavior and helping is positive. The studies have no hostile behavior option. Given the widespread understanding of prejudice as hostile behavior, one might expect that there must be many studies in which discrimination has been observed in the form of overtly hostile behavior toward members of an outgroup. In searching, we could find just seven laboratory studies in which discrimination was assessed using measures that appeared unambiguously to involve outgroup-directed hostile behavior. The behaviors were electric shock administration in six (Baron, 1979; Donnerstein & Donnerstein, 1972; Genthner & Taylor, 1973; Griffin & Rogers, 1970; Prentice–Dunn & Rogers, 1980; Rogers & Prentice–Dunn, 1981; Wilson & Rogers, 1975) and aversive noise administration in the seventh (Mummendey et al., 1992). The shock administration studies incorporated electric shock either as an outcome to be administered to another player in a competitive game or as an experimental stimulus to be presented for the ostensible purpose of increasing another subject’s heart rate. In most conditions of these studies, White subjects administered more shocks to Whites (i.e., to presumed ingroup members) than to Blacks.6 Experimenters interpreted this in terms of egalitarian norms and concerns about appearing prejudiced. In only one condition of one of these experiments did White subjects administer more shocks to another presumed subject who was Black than to one who was White. Contrary to researchers’ expectations, both in experiments with aversive shock and experiments with aversive noise stimuli, fairness (i.e., ingroup and outgroup receiving the same outcomes) prevailed in distributing the aversive outcomes. Accordingly, no evidence of outgroup hostility has been reported in experiments using unambiguously aversive stimuli. 6 Although the Black subjects of the Wilson and Rogers (1975) experiment chose higher intensity shocks to administer to (outgroup) White than to (ingroup) Black confederates, the product of intensity and shock duration measures revealed that they had chosen somewhat greater shocks for Black (ingroup) confederates. “With malice toward none . . .” Draft of 9 November 2013 -23- A famous early field experiment on prejudice also found no evidence for discrimination in a situation in which, had discrimination occurred, it would have had to take an overtly hostile form. LaPiere (1934) and two Chinese traveling companions toured the Southwestern United States, seeking housing and dining accommodations at 251 establishments. Face-to-face denial of service to potential customers is undeniably a hostile act. LaPiere and the Chinese couple were refused service only once. This result was in stark contrast with the finding that, in response to a subsequent mailed questionnaire, more than 90% of these same establishments reported that they would not accommodate “members of the Chinese race.” Nonverbal Behavior Measures One might expect that nonverbal measures, such as facial expressions, body orientation, and voice tone, can easily be classified into positive and negative categories. However, nonverbal measures rarely afford a clear neutral point. As one example, when the subject can position his or her chair at variable distances from a fellow participant, even though smaller distance translates unambiguously to greater positivity toward the other participant (cf. Amodio & Devine, 2006), it is not possible to identify a specific distance that can be scored as “neutral.” Measures of speaking time are similarly unambiguous in direction (more conversation is more favorable) but equally lack specifiable neutral values. Likewise, when measures of facial affect are being obtained (cf. McConnell & Leibold, 2001), it is difficult to score them so as to identify a neutral point. Problems come in combining multiple responses for a subject. Clearly, smiles cannot all be counted as equally positive nor frowns equally negative, and some of each might even be intended to communicate the opposite.7 Behavioral Intention Measures The procedure introduced by Tajfel et al., (1971) identifies an unambiguous neutral point when the measures involve distribution of payments to others. Neutral behavior takes the form of allocating a payment equal to the per-person average of available points or funds. Favoring (positive) and disfavoring (negative) behavior then takes the form of payment that falls, respectively, above or below that average level. LaPiere’s (1934) mailed questionnaire provides 7 The difficulty of characterizing nonverbal intergroup behavior unambiguously as positive or negative can be seen in some innovative studies in field settings that have demonstrated discrimination based on sexual orientation ( Hebl, Foster, Mannix, & Dovidio, 2002 ) and obesity ( King, Spiro, Hebl, Singletary, & Turner, 2006 ). “With malice toward none . . .” Draft of 9 November 2013 -24- another illustration of a behavioral intention measure—willingness or refusal to accept Chinese guests—that could be scored unambiguously as positive or negative.8 8 In the context of assessing favoritism, behavioral intention measures compare interestingly with overt behavior measures. Crosby et al. (1980) found that outgroup-directed helping was greater when White potential helpers were in face-to-face interaction with Black potential help recipients, compared to ‘remote’ situations in which the two were not face-to-face. Whites’ opposition to helping minorities expressed on surveys may be an analog of the low rate of helping found in the remote conditions of the unobtrusive-measure helping studies reviewed by Crosby et al. “With malice toward none . . .” Draft of 9 November 2013 -25- Interpersonal Judgment Measures Consider the judgment task of recommending a jail sentence for a convicted defendant for whom the available evidence indicates both guilt and mitigating circumstances. If the permissible sentence range is from a low of 6 months to a high of 5 years, it is clear that the shortest sentence is favorable and the longest sentence is unfavorable. But how can one identify an intermediate sentence term that is neutral—neither favorable nor hostile to the defendant? Furthermore, if the dependent measure is obtained as (say) a 7-point Likert-format judgment of endorsement of the maximum sentence, what level of agreement can be assumed to be neither favorable nor unfavorable to the defendant? This limitation can be overcome by providing the subject with information about an average sentence for the defendant’s circumstances, then requesting endorsement of either a shorter, equal or longer sentence. Self-report Attitude Measures Self-report attitude measures are easily constructed with neutral points, as with thermometer scales that have end anchors of warm (positive) and cold (negative) and a middle anchor of “neutral.” A useful alternative recommended by Gaertner et al. (1997) is to have separate rating scales for positive and negative traits, with the low anchor indicating absence of the trait. Using this method, Gaertner and McLaughlin (1983) and Dovidio et al. (1989) found that White subjects did not discriminate against Blacks on negative-trait scales; they rated Whites and Blacks equally on these. In contrast, they did discriminate on positive-trait scales (rating Whites higher than Blacks). This strong finding uncovers discriminatory judgment in the form of favoring the ingroup rather that disfavoring the outgroup. Implicit Attitude Measures For the most widely used implicit attitude measures, subjects make rapid classification responses to both valenced word stimuli and to images, words, or names that represent two contrasted categories, which can be an ingroup and outgroup, such as White and Black. These implicit measures generate relative-attitude scores that index greater favorability to one group than to the other (e.g., Fazio, Sanbonmatsu, Powell & Kardes, 1986; Greenwald, McGhee & Schwartz, 1998). These measures often reveal evaluative separation between the two groups, but they do not unambiguously locate either group relative to a neutral point (i.e., a score that “With malice toward none . . .” Draft of 9 November 2013 -26- indicates neither positive nor negative evaluation). Some implicit measures may do a better job than others in distinguishing favorability from unfavorability in an absolute sense (e.g., Nosek & Banaji, 2001; Sriram & Greenwald, 2009), but even these measures are not established as having neutral zero points. Their zero points more confidently indicate indifference between the groups, which could mean that both groups are regarded equally positively or that they are regarded equally negatively. The “Second Requirement” Our treatment of method has to this point considered only the first requirement that we stated, which is to use measures of behavior or judgment that have an unambiguous neutral point. The second requirement is that the study must demonstrate discrimination by comparing behavior or judgments toward ingroup members versus outgroup members. To meet the second requirement at the individual-subject level requires a within-subjects design, in which each subject provides a measure for both ingroup and outgroup. Designs in which ingroup vs. outgroup is a between-subjects factor also permit assessment of discrimination, but not for individual subjects. For either type of design, determination of whether discrimination takes the form of ingroup favoritism or outgroup hostility can be had only if the first requirement (neutral point of dependent measure) is also met. The minimal group paradigm and studies that use unobtrusive measures of helping are the rare paradigms for which more than an occasional study meets both the neutral-point requirement and the ingroup–outgroup comparison requirement. Three Conclusions about Methods Methods Used in Most Studies of Discrimination Have Limited Capabilities Few studies in the voluminous research literature on prejudice and discrimination have used methods that can distinguish the relative roles of ingroup-favorable and outgroup-hostile behavior in producing discrimination. This could well be a consequence of the widespread (but, to us, incorrect) belief that discrimination most often occurs in the form of outgroup-directed hostility. When one makes this assumption, it might well appear unnecessary to investigate relative contributions of favoritism and hostility to discrimination. Most studies of discrimination address one of two other questions: (a) determining whether various experimental manipulations increase or reduce discrimination, or (b) determining whether various individual “With malice toward none . . .” Draft of 9 November 2013 -27- difference measures of prejudice successfully predict individual differences in discriminatory behavior. These aims do not require either of the two requirements for determining whether observed discrimination has resulted from ingroup favoritism or outgroup hostility. It is remarkable that relatively few studies have used dependent measures that assessed unequivocally hostile behavior. This paucity cannot be explained simply in terms of experimenters’ benevolence, because a paradigm involving aggression via (presumed) administration of electric shocks was widely used in the 1960s and 1970s (cf. Buss & Brock, 1963). The use of this well-known method in only six studies of race discrimination, during the same era in which more than 30 studies had investigated race discrimination in unobtrusive helping studies (Crosby et al. 1980), suggests that it may be much easier to observe discrimination in studies using unobtrusive measures of helping behavior than in studies that observe shock administration or other unequivocally hostile behavior. Balance of Findings with Existing Methods More studies have demonstrated discrimination resulting from ingroup favoritism than from outgroup hostility—an unexpected observation in light of the prevailing wisdom that discrimination typically occurs in the form of hostility directed toward outgroups. For whatever reason, it is apparently easier to demonstrate discrimination in the form of differential favoritism than in the form of differential hostility. The most plausible explanation is that, indeed, discrimination more often takes the form of ingroup favoritism than outgroup hostility. However, some portion of this imbalance in findings may also be due to the greater ease of meeting ethical research standards in laboratory studies using measures of benign behavior than in ones investigating hostile behavior. New Methods are Needed Prejudice researchers need to add methods to their toolboxes—methods that: (a) distinguish ingroup-favorable from outgroup-hostile subject behavior, (b) provide a comparison of outcomes to ingroup and outgroup members, and (c) are easy to administer in standardized form. Among existing laboratory methods, L. Gaertner and Insko’s (2001) multiple alternative matrices procedure for the minimal group paradigm comes closest to meeting this combination of requirements, but is not adapted to investigating face-to-face intergroup interactions. The most “With malice toward none . . .” Draft of 9 November 2013 -28- important method recommendation is to use measures—whether nonverbal, behavioral, self- report, or implicit—that have unambiguous neutral points to designate responses that are neither favoring nor hostile. Discussion We conclude that ingroup favoritism is currently more potent than outgroup hostility as a cause of intergroup discrimination in the United States. The support for this conclusion comes from multiple, well-established empirical paradigms, including laboratory studies of minimal group and similarity–attraction paradigms, field experiments using unobtrusive observations of helping behavior, and field audit studies of police profiling and of treatment accorded to potential job seekers, apartment renters, and home buyers. “With malice toward none . . .” Draft of 9 November 2013 -29- Two caveats to our conclusion are necessitated by the available research literature. First, many of the studies we cited involved race. It is conceivable that outgroup hostility may play a greater role in non-racial discrimination. Second, most of the studies we reviewed were conducted in the United States. Although conclusions based on these studies may also hold in other nations, it is premature to assume that the same will be found elsewhere—especially in locations such as Northern Ireland and South Africa, which have centuries-old histories of intergroup hostility and discrimination. Tests of the ingroup favoritism thesis in those countries should be highly informative. Pending broader investigations in other countries and with other forms of discrimination, our conclusion that ingroup favoritism enables discrimination should be regarded as most strongly established for Black–White racial discrimination in the United States. The Nature of the Evidence An obvious question prompted by our main conclusion is: Why is the ingroup-favoritism- enables-discrimination thesis not already generally accepted as a prime explanation of discrimination? If the evidence is so extensive and most of it has been available for at least a few decades, why have students of prejudice and discrimination not previously arrived at this conclusion? The answer has two parts. The more important part is that, through the history of prejudice’s scientific study, most researchers have defined and understood prejudice as an affectively negative outgroup-directed attitude that they expect to result in hostile acts of discrimination. That view was justified for a long time, but is now questionable given the societal transformations that have cumulatively produced dramatic reductions in both endorsements of negative attitudes toward minority groups and in hostile forms of discrimination. The second part is that—as explained in this article’s analysis of research methods—very few empirical studies of discrimination have used methods that can evaluate the relative contributions of ingroup favoritism and outgroup hostility to observed discrimination. “With malice toward none . . .” Draft of 9 November 2013 -30- Our conclusion about limitations of research methods makes a second question obvious: If methods to evaluate relative contributions of ingroup favoritism and outgroup hostility are so inadequate, how can we reach a strong conclusion about those relative contributions? The strategy of this article was to build its case by establishing three points. First, ingroup-directed positivity is pervasive and there is no comparable evidence for pervasiveness of outgroup- directed negativity. Second, discrimination frequently occurs in the form of differential favoring and there is no comparable evidence for discrimination occurring so frequently in the form of differential harming. Third, established theories offer multiple bases for understanding the development of positive regard for others—a point to which we next turn. Multiplicity of Explanations for Ingroup Favoritism “Societal factors... are distal causal factors in group relations. ...At the same time, the intervening factor of personality is ever the proximal cause of human conduct. . . . There are no good reasons for professional rivalry and backbiting among social scientists preferring one approach or the other. They can and should be blended in our outlook” (Allport, 1962, p. 132). This review has sought to establish that multiple theories, at both individual and societal levels, can explain the strength of ingroup favoritism. Allport’s observation about the synergy of theories applies not only to theories that fall on different sides of the disciplinary boundary between psychology and sociology, but equally to the sets of theories within each of those disciplines. Furthermore, the various theories most often do not conflict in their explanations. Rather, they offer multiple, complementary theoretical routes to the goal of understanding ingroup favoritism. Our focus on ingroup favoritism therefore affords a rapprochement among social psychology’s person-centered explanations and other social sciences’ social structure- centered explanations. We do not claim that hostile prejudice plays no role in discrimination. However, we do claim that much discrimination occurs without hostile intent; it occurs either as a consequence of social structures (such as the self-sustaining properties of segregation in schools, homes, workplaces, and institutional discrimination) or as a consequence of mental processes that lack animus (such as norms, similarity–attraction, and the judgment processes that we labeled illusory individuation). “With malice toward none . . .” Draft of 9 November 2013 -31- A common denominator in these discrimination-producing societal and mental processes is that, without engaging outgroup-directed hostility, they all tend to result in favoring already advantaged groups. In this way, discriminatory outcomes will often occur without the intergroup animus that, traditionally, has been a defining feature of prejudice. We do not suggest that prejudice should therefore be re-conceived without reference to hostility. That would be too radical a conclusion from our observations, especially because we are not inclined to claim that it has always been thus. The important, and perhaps no-less radical, conclusion is that in contemporary American society intergroup discrimination has a potent life that now can occur without intergroup hostility. Has there been a Decline of Malice? Although much societally significant discrimination continues to occur in hostile forms, it is even more apparent that hostile acts of race discrimination in the United States have steadily declined during the past century. Perhaps the most dramatic indication of decline is evident in data concerning lynchings. Lynchings—which were group killings of (mostly) Black Americans —declined from an average of 150 per year in the late 19th century to disappearance in the 1950s (“4,733 Mob Action Victims Since ’82, Tuskegee Reports”, 1959). A second compelling source of evidence is much more recent: In 1996, the Federal Bureau of Investigation (FBI) started compiling data on hate crimes.9 Hate crimes targeting Black victims declined steadily from an average of 4,071 per year in 1996–1998 to 2,762 per year in 2009–2011. As percentages of the United States Black population in 2000 and 2010, the percentages of Blacks who were victims of hate crimes had declined (between 1999–2001 and 2009–2011) by a third, from 0.030% to 0.020%. Implications Our strong conclusion is that, in present-day America, discrimination results more from helping ingroup members than from harming outgroup members. This conclusion has substantial implications for the conduct of research on discrimination, for the teaching of prejudice and discrimination, and for the design of programs to reduce discrimination. Because the implications for research and teaching both follow from the potential importance of the conclusion for practical application, this section focuses on the implications for practice. 9 Annual counts of multiple categories of hate crimes for 1996 through 2011 are available (at http://www.fbi.gov/about-us/cjis/ucr/hate-crime). “With malice toward none . . .” Draft of 9 November 2013 -32- Our conclusion adds force to the approach of Gaertner and Dovidio (2000), who emphasized the possible discrimination-reducing impact of forming “superordinate” identities, which extend ingroup boundaries and thereby increase the diversity of others who are encompassed within the ingroup fold. Our conclusion also suggests a quite different approach to discrimination reduction: Adopting policies of targeted outgroup-helping, in effect seeking to level the ingroup- favoritism playing field. This suggestion fits with affirmative action strategies that aim to increase benefits to disadvantaged minorities or to groups regarded as under-represented in workplaces and selective educational institutions. Relatedly, DiTomaso (2012) observed that resistance to affirmative action programs has at least a partial explanation in affirmative action’s disruption of routine forms of ingroup favoritism, which include seeking friends to fill job vacancies and admitting “legacy” applicants to elite educational institutions. In regard to employment discrimination, the courts provide opportunities for discrimination- reduction that depend not on intergroup attitudes and behavior, but on judges’ interpretations of law. It is therefore relevant to ask whether establishment of the potency of ingroup favoritism as a source of discrimination might affect efforts to reduce discrimination via litigation. Krieger (1998) pointed out that federal courts’ interpretations of Title VII of the Civil Rights Act of 1964 made it difficult for discrimination suits to succeed on the argument that plaintiff employees in protected classes were adversely impacted by ingroup favoritism that benefited others. By way of contrast with the situation when Krieger was writing in 1998, a 2012 federal court decision permitted a case based on an ingroup favoritism theory of discrimination to proceed. If the 2012 decision presages a future legal environment in which discrimination suits appealing to ingroup favoritism will generally be allowed to proceed, this article’s conclusions may prove useful to the courts deciding those cases.10 10 The contrast of two federal court opinions in 1993 and 2012 signals a change of direction regarding ingroup favoritism. The 1993 opinion, in the case of E.E.O.C. v. Consolidated Service Systems , declared that if a hiring policy “produce[s] a work force whose racial or religious or ethnic or nation-origin or gender composition pleases the employer, this is not intentional discrimination. The motive is not a discriminatory one”. The 2012 opinion, in the case of McReynolds v. Merrill Lynch , concluded that the defendant company might have discriminated by allowing a form of company-wide ingroup favoritism involving teams of brokers that could choose to exclude African American brokers: “The teams . . . are little fraternities [in which] the brokers choose as team members people who are like themselves. If they are white, they, or some of them anyway, are more comfortable teaming with other white brokers.” Especially noteworthy regarding these two opinions is that they were written by the same federal judge. “With malice toward none . . .” Draft of 9 November 2013 -33- A provocative recent article by Dixon et al. (2012) started, as we did, from observing that current conceptions may incorrectly link discrimination primarily to negative intergroup attitudes. Dixon et al. proceeded to conclude that collective political action by historically disadvantaged groups might be more efficacious in ending discrimination than efforts directed at increasing positivity toward outgroups.11 In closing, we must counter any impression that we regard favoritism as the only cause of discrimination worthy of scholarly attention. Although hostile forms of discrimination have declined steadily during the period in which prejudice has been studied scientifically, hostile discrimination nevertheless continues to exist in many forms, including racial and ethnic slurs, hostile work environments, hate crimes, and terrorism. At the same time, legal, ethical, and normative constraints against hostile discrimination now widely prevail in the United States, and there are few parallel constraints against forms of favoritism that can generate extensive discrimination. As ethnic and racial minorities become increasingly represented in American work settings, it is plausible that opportunities for favoritism to produce significant discrimination may even be increasing. 11 See Pettigrew and Tropp, 2011, for an alternate perspectivc on the conclusions reached by Dixon et al. (2012) “With malice toward none . . .” Draft of 9 November 2013 -34- Acknowledgments For comments on an earlier version of this manuscript, the authors are grateful to Marc Bendick, Sapna Cheryan, Faye Crosby, Samuel Gaertner, Miles Hewstone, John Jost, Lee Jussim, Michael Norton, Laurie Rudman, Janice Sabin, Heather Smith, Ulrich Wagner, and Vincent Yzerbyt. “With malice toward none . . .” Draft of 9 November 2013 -35- References “4,733 Mob Action Victims Since ’82, Tuskegee Reports”, Montgomery Advertiser, April 26, 1959. Reprinted in Ginzburg, R. (1962). 100 Years of Lynchings. New York: Lancer Books, p. 244. Abelson, R. P., Aronson, E., McGuire, W. J., Newcomb, T. M., Rosenberg, M. J., & Tannenbaum, P. (Eds.). (1968). Theories of cognitive consistency: A sourcebook. Chicago: Rand-McNally. Abrams, D., & Hogg, M. A. (1988). Comments on the motivational status of self-esteem in social identity and intergroup discrimination. European Journal of Social Psychology, 18,317–334. Allport, G. W. (1954). The nature of prejudice. Reading, MA: Addison-Wesley. Allport, G. W. (1962). Prejudice: Is it societal or personal? Journal of Social Issues, 18, 120–134. Amiot, C. E., Terry, D. J., & Callan, V. J. (2007). Status, equity and social identification during an intergroup merger: A longitudinal study. British Journal of Social Psychology, 46, 557- 577. Amodio, D. M., & Devine, P. G. (2006). Stereotyping and evaluation in implicit race bias: Evidence for independent constructs and unique effects on behavior. Journal of Personality and Social Psychology, 91, 652–661. Aronson, E. (1988). The social animal (5th ed.) New York: Freeman. Aronson, E. Wilson, T. D., & Akert, R. M. (2012). Social psychology (8th ed.). Old Tappan, NJ: Pearson. Ayres, I., Vars, F. E., & Zakariya, N. (2005). To insure prejudice: Racial disparities in taxicab tipping. Yale Law Journal, 114, 1613–1674. Baron, R. A., & Byrne, D. (1974). Social psychology (7th ed.). Boston: Allyn and Bacon. Baron, R. A. (1979). Effects of victim's pain cues, victim's race, and level of prior instigation upon physical aggression. Journal of Applied Social Psychology, 9, 103–114. Baron, R. M., & Graziano, W. G. (1991). Social psychology. Fort Worth, TX: Holt, Rinehart and Winston. Bendick, M. (2004, June). Using paired-comparison testing to develop a social psychology of civil rights. Paper presented at the biennial conference of the Society for the Psychological Study of Social Issues. Bendick, M. (2007). Situation testing for employment discrimination in the United States of America. Horizons Stratégiques, 5, 17–39. Bendick, M., Rodriguez, R. E., & Jayaraman, S. (2010). Employment discrimination in upscale restaurants: Evidence from matched pair testing. The Social Science Journal, 47, 802–818. Billig, M., & Tajfel, H. (1973). Social categorization and similarity in intergroup behaviour. European Journal of Social Psychology, 3, 27–52. “With malice toward none . . .” Draft of 9 November 2013 -36- Bogardus, E. S. (1925). Measuring social distance. Journal of Applied Sociology, 9, 299–308. Brewer, M. B. (1999). The psychology of prejudice: Ingroup love or outgroup hate? Journal of Social Issues, 55, 429–444. Brewer, M. B., & Crano, W. D. (1994). Social psychology. Minneapolis, MN: West. Brigham, J. C. (1991). Social psychology (2nd ed.). New York: HarperCollins. Brown, R. (1995). Prejudice: Its social psychology. Oxford, U.K.: Blackwell. Buss, A. H., & Brock, T. C. (1963). Repression and guilt in relation to aggression. The Journal of Abnormal and Social Psychology, 66, 345–350. Byrne, D. (1961). Interpersonal attraction and attitude similarity. The Journal of Abnormal and Social Psychology, 62, 713–715. Byrne, D., Ervin, C. R., & Lamberth, J. (1970). Journal of Personality and Social Psychology, 16, 157–165. Chanin, R., Rintels, J., et al. (2011). Restoring a national consensus: The need to end racial profiling in America. Washington, DC: The Leadership Conference on Civil and Human Rights. Cohen, J. (1977). Statistical power analysis for the behavioral sciences (rev. ed.). New York: Academic Press. Crosby, F., Bromley, S., & Saxe, L. (1980). Recent unobtrusive studies of Black and White discrimination and prejudice: A literature review. Psychological Bulletin, 87, 546–563. Darley, J. M., & Gross, P. H. (1983). A hypothesis-confirming bias in labeling effects. Journal of Personality and Social Psychology, 44, 20–33. DiTomaso, N. (2012). The American non-dilemma: Racial inequality without racism. New York: Russell Sage. Dixon, J., Levine, M., Reicher, S., & Durrheim, K. (2012). Beyond prejudice: Are negative evaluations the problem and is getting us to like one another more the solution? Behavioral and Brain Sciences, 35, 411–425. Donnerstein, E., Donnerstein, M., Simon, S., & Ditrichs, R. (1972). Variables in interracial aggression: Anonymity, expected retaliation, and a riot. Journal of Personality and Social Psychology, 22, 236–245. Dovidio, J. F., Mann, J., & Gaertner, S. L. (1989). Resistance to affirmative action: The implications of aversive racism. In F. Blanchard & F. Crosby (Eds.), Affirmative action in perspective. New York, NY: Springer-Verlag. Dovidio, J. F., & Gaertner, S. L. (2004). Aversive Racism. In M. P. Zanna (Ed.), Advances in experimental social psychology, (Vol. 36, pp. 1–52). “With malice toward none . . .” Draft of 9 November 2013 -37- Eagly, A. H., & Dickman, A. B. (2005). What is the problem? Prejudice as an attitude in context. In J. F. Dovidio, P. Glick, & L. A. Rudman (Eds.), On the nature of prejudice: Fifty years after Allport. Oxford, UK: Blackwell. Fazio, R. H., Sanbonmatsu, D. M., Powell, M. C., & Kardes, F. R. (1986). On the automatic activation of attitudes. Journal of Personality and Social Psychology, 50, 229–238. Feagin, J. R., & Feagin, C. B. (1996). Racial and ethnic relations. 5th Ed. Upper Saddle River, NJ: Prentice Hall. Festinger, L. (1957). A theory of cognitive dissonance. Palo Alto, CA: Stanford University Press. Finch, J. F, & Cialdini, R. B. (1989). Another indirect tactic of (self-) image management: Boosting. Personality and Social Psychology Bulletin, 15, 222–232. Fiske, S.T., & Neuberg, S.L. (1990). A continuum model of impression formation from category- based to individuating processes: Influences of information and motivation on attention and interpretation. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (Vol. 23, pp. 1–74). San Diego, CA : Academic Press. Franzoi, S. L. (1996). Social Psychology. Madison WI: Brown & Benchmark. Gaertner, S. L., & Dovidio, J. F. (2000). Reducing intergroup bias: The common ingroup identity model. Philadelphia, PA: Psychology Press. Gaertner. S. L., Dovidio, J. F., Banker, B., Rust, M., Nier, J., Mottola, G., & Ward, C. M. (1997). Does racism necessarily mean anti-Blackness? Aversive racism and pro-Whiteness. In M. Fine, L. Powell, L. Weis, & M. Wong (Eds.), Off white (pp. 167–178). London: Routledge. Gaertner, L. & Insko, C. A. (2001). On the measurement of social orientations in the minimal group paradigm: norms as moderators of the expression of intergroup bias. European Journal of Social Psychology, 31, 143–154. Gaertner, S. L., & McLaughlin, J. P. (1983). Racial stereotypes: Associations and ascriptions of positive and negative characteristics. Social Psychology Quarterly, 46, 23–30. Gaertner, S. L., & Bickman, L. (1971). Effects of race on the elicitation of helping behavior: The wrong number technique. Journal of Personality and Social Psychology, 20, 218–222. Gawronski, B.; & Strack, F. (Eds.) (2012). Cognitive consistency: A fundamental principle in social cognition. New York, NY: Guilford Press. Genthner, R. W., & Taylor, S. P. (1973). Physical aggression as a function of racial prejudice and the race of the target. Journal of Personality and Social Psychology, 27, 207–210. Greenwald, A. G., Banaji, M. R., Rudman, L. A., Farnham, S. D., Nosek, B. A., & Mellott, D. S. (2002). A unified theory of implicit attitudes, stereotypes, self-esteem, and self-concept. Psychological Review, 109, 3–25. Greenwald, A. G., McGhee, D. E., & Schwartz, J. K. L. (1998). Measuring individual differences in implicit cognition: The Implicit Association Test. Journal of Personality and Social Psychology, 74, 1464–1480. “With malice toward none . . .” Draft of 9 November 2013 -38- Griffin, B. Q., & Rogers, R. W. (1977). Reducing interracial aggression: Inhibiting effects of victim's suffering and power to retaliate. Journal of Psychology, 95, 151–157. Hebl, M. R., Foster, J. B., Mannix, L. M., & Dovidio, J. F. (2002). Formal and interpersonal discrimination: A field study of bias toward homosexual applications. Personality and Social Psychology Bulletin, 28, 815–825. Heider, F. (1958). The psychology of interpersonal relations. New York: Wiley. Heilman, M.E., & Haynes, M.C. (2008). Subjectivity in the appraisal process: A facilitator of gender bias in work settings. In E. Borgida & S.T. Fiske (Eds.), Beyond common sense: Psychological science in the courtroom. (pp. 127–156). Oxford: Blackwell Publishing. Hewstone, M., Rubin, M. & Willis, H. (2002). Intergroup bias. Annual Review of Psychology, 53, 575–604. Hodson, G., Dovidio, J. F., & Gaertner, S. L. (2002). Processes in racial discrimination: Differential weighting of conflicting information. Personality and Social Psychology Bulletin, 28, 460–471. Jost, J. T., & Banaji, M. R. (1994). The role of stereotyping in system-justification and the production of false consciousness. British Journal of Social Psychology, 33, 1–27. Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system justification theory: Accumulated evidence of conscious and unconscious bolstering of the status quo. Political Psychology, 25, 881–919. Jost, J. T., Pelham, B. W., & Carvallo, M. R. (2002). Non-conscious forms of system justification: Implicit and behavioral preferences for higher status groups. Journal of Experimental Social Psychology, 38, 586–602. King, E. B., Spiro, J. R., Hebl, M. R., Singletary, S. L., & Turner, S. (2006). The stigma of obesity in customer service: A mechanism for remediation and bottom-line consequences of interpersonal discrimination. Journal of Applied Psychology, 91, 579–593. Krieger, L. H. (1998). Civil rights Perestroika: Intergroup relations after affirmative action. California Law Review, 86, 1251–1333. Kunda, Z., & Thagard, P. (1996). Forming impressions from stereotypes, traits and behaviors: A parallel-constraint satisfaction theory. Psychological Review, 103, 284–308. Lamberth, J. (1994). Revised statistical analysis of the incidence of police stops and arrests of Black drivers/travelers on the New Jersey Turnpike between Exits or Interchanges 1 and 3 from the years 1988 through 1991. Unpublished report, Temple University. LaPiere, R. (1934). Attitudes versus actions. Social Forces, 13, 230–237. Lerner, M. J. (1980). The belief in a just world: A fundamental delusion. New York: Plenum. Lippa, R. A. (1994). Introduction to social psychology (2nd ed.). Pacific Grove, CA: Brooks/Cole. “With malice toward none . . .” Draft of 9 November 2013 -39- Locksley, A., Borgida, E., Brekke, N., & Hepburn, C. (1980). Sex stereotypes and social judgment. Journal of Personality and Social Psychology, 39, 821–831. Lynn, M., Sturman, M., Ganley, C., Adams, E., Douglas, M., & McNeil, J. (2006). Consumer racial discrimination in tipping: A replication and extension. Unpublished manuscript, Cornell University. Lynn, M., Sturman, M., Ganley, C., Adams, E., Douglas, M., & McNeil, J. (2008). Consumer racial discrimination in tipping: A replication and extension. Journal of Applied Social Psychology, 38, 1045-1060. McCarthy, J. M., Van Iddekinge, C. H., & Campion, M. A. (2010). Are highly structured job interviews resistant to demographic similarity effects? Personnel Psychology, 63, 325-359. McConnell, A. R., & Leibold, J. M. (2001). Relations among the Implicit Association Test, discriminatory behavior, and explicit measures of racial attitudes. Journal of Experimental Social Psychology, 37, 435–442. Meertens, R., & Pettigrew, T. F. (1997). Is subtle prejudice really prejudice? Public Opinion Quarterly, 61, 54–71. Minard, R. D. (1952). Race relations in the Pocahontas coal field. Journal of Social Issues, 8, 29–44. Mummendey, A., Simon, B., Dietze, C, Gmnert, M., Haeger. G., Kessler, S., Lettgen, S., & Schaferhoff, S. (1992). Categorization is not enough: Intergroup discrimination in negative outcome allocation. Journal of Experimental Social Psychology, 28, 125–144. Myers, D. G. (1995). Psychology. 4th Ed. New York: Worth. Myrdal, G. (1944). An American dilemma. New York: Harper & Row. Nelson, T. D. (2002). The psychology of prejudice. Boston: Allyn and Bacon. New York State Task Force on Police-on-Police Shootings. (2010). Reducing Inherent Danger: Report of the Task Force on Police-on-Police Shootings. Available at: http://www.hks.harvard.edu/var/ezp_site/storage/fckeditor/file/pdfs/centers- programs/programs/criminal-justice/Police-on-Police_Shootings.pdf Nier, J. A., & Gaertner, S. L. (Eds.) (2012). The reality of contemporary discrimination in the United States: The consequences of hidden bias in real world contexts. Journal of Social Issues, 68, 207–438. Norton, M. I., Vandello, J. A., & Darley, J. M. (2004). Casuistry and Social Category Bias. Journal of Personality and Social Psychology, 87, 817–831. Nosek, B. A., & Banaji, M. R. (2001). The Go/No-go Association Task. Social Cognition, 19, 625–664. Office of the Attorney General (1999). The New York City Police Department’s “stop & frisk” practices: A report to the people of the State of New York. http://www.hks.harvard.edu/var/ezp_site/storage/fckeditor/file/pdfs/centers-programs/programs/criminal-justice/Police-on-Police_Shootings.pdf http://www.hks.harvard.edu/var/ezp_site/storage/fckeditor/file/pdfs/centers-programs/programs/criminal-justice/Police-on-Police_Shootings.pdf “With malice toward none . . .” Draft of 9 November 2013 -40- Osgood, C. E., & Tannenbaum, P. H. (1955). The principle of congruity in the prediction of attitude change. Psychological Review, 62, 42–55. Pettigrew, T. F. (Ed.) (1975). Racial discrimination in the United States. New York: Harper & Row. Pettigrew, T. F. (1979). Racial change and social policy. Annals of the American Academy of Political and Social Science, 441, 114–131. Pettigrew, T. F. (1991). Normative theory in intergroup relations: Explaining both harmony and conflict. Psychology and Developing Societies, 3, 3–16. Pettigrew, T. F. (1998). Prejudice and discrimination on the college campus. In J. L. Eberhardt & S. T. Fiske (Eds.), Confronting racism: The problem and the response (pp. 263–279). Thousand Oaks, CA: Sage Publications. (Reprinted from the "Higher Education Extension Service Review"). Pettigrew, T. F. (2011). Toward sustainable psychological interventions for change. Peace and Conflict, 17, 179–192 Pettigrew, T. F., & Meertens, R. (1995) Subtle and blatant prejudice in Western Europe. European Journal of Social Psychology, 57, 57–75. Pettigrew, T. F., & Taylor, M. C. (2002). Discrimination. In N. J. Smelser & P. B. Baltes (Eds.), International Encyclopedia for the Social and Behavioral Sciences (pp. 3,762–3,766). Oxford, U.K.: Pergamon. Pettigrew, T. F., & Tropp, L. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90, 1-33. Pettigrew, T. F., & Tropp, L. R. (2011). When groups meet: The dynamics of intergroup contact. New York, NY: Psychology Press. Platow, M. J., & Van Knippenberg, D. (2001). A social identity analysis of leadership endorsement: The effects of leader ingroup prototypicality and distributive intergroup fairness. Personality and Social Psychology Bulletin, 27, 1508–1519. Prentice-Dunn, S., & Rogers, R. W. (1980). Effects of deindividuating situational cues and aggressive models on subjective deindividuation and aggression. Journal of Personality and Social Psychology, 39(1), 104-113. Reitzes, D. C. (1953). The role of organizational structures: Union versus neighborhood in a tension situation. Journal of Social Issues, 9, 37–44. Reskin, B. F. (1998). The realities of affirmative action in employment. Washington, DC: American Sociological Association. Reynolds, K. J., Turner, J. C., & Haslam, S. A. (2000). When are we better than them and they worse than us? A closer look at social discrimination in positive and negative domains. Journal of Personality and Social Psychology, 78, 64–80. Rivera, L. A. (2012). Hiring as cultural matching: The case of elite professional service firms. American Sociological Review, 76, 999–1022. http://csaweb110v.csa.com.oca.ucsc.edu/ids70/view_record.php?id=7&recnum=5&log=from_res&SID=8b5s3s7uou98l4ffq4jtvtilc4 http://csaweb110v.csa.com.oca.ucsc.edu/ids70/view_record.php?id=7&recnum=5&log=from_res&SID=8b5s3s7uou98l4ffq4jtvtilc4 “With malice toward none . . .” Draft of 9 November 2013 -41- Riordan CM. (2000). Relational demography within groups: Past developments, contradictions, and new directions. In Ferris GR (Ed.), Research in personnel and human resources management (vol. 19, pp. 131–173). Greenwich, CT: JAI Press. Rogers, R. W., & Prentice-Dunn, S. (1981). Deindividuation and anger-mediated interracial aggression: Unmasking regressive racism. Journal of Personality and Social Psychology, 41, 63–73. Rokeach, M. (Ed.) (1960). The open and closed mind. New York, NY: Basic Books. Rokeach, M., & Mezei, L. (1966). Race and shared belief as factors in social choice. Science, 151, 167-172. Ryan, William (1976). Blaming the victim. New York: Vintage. Sacco JM, Scheu CR, Ryan AM, Schmitt N. (2003). An investigation of race and sex similarity effects in interviews: A multilevel approach to relational demography. Journal of Applied Psychology, 88, 852–865. Saucier, D. A., Miller, C. T., & Doucet, N. (2005). Differences in Helping Whites and Blacks: A Meta-Analysis. Personality and Social Psychology Review, 9, 2–16. Schuman, H., Steeh, C., Bobo, L., & Krysan, M. (1997). Racial attitudes in America. Cambridge, MA: Harvard University. Secord, P. F., & Backman, C. W. (1964). Social psychology. New York: McGraw-Hill. Simpson, G. E., & Yinger, J. M. (1985). Racial and cultural minorities. 5th Ed. New York: Plenum. Smeaton, G., Byrne, D., & Murnen, S. K. (1989). The repulsion hypothesis revisited: Similarity irrelevance or dissimilarity bias? Journal of Personality and Social Psychology, 56(1), 54-59. Smith, E. R. (1993). Social identity and social emotions: Toward new conceptions of prejudice. In D. M. Mackie & D. L. Hamilton (Eds.), Affect, cognition, and stereotyping (pp. 297–315). San Diego, CA: Academic Press. Smith, E. R., & Mackie, D. M. (2005). Aggression, hatred, and other emotions. In J. F. Dovidio, P. Glick, & L. A. Rudman (Eds.), On the nature of prejudice: Fifty years after Allport (pp. 361–376). Malden: Blackwell Publishing. Smith, H., Pettigrew, T. F., Pippin, G., & Bialosiewicz, S. (2012). Relative deprivation: A theoretical and meta-analytic critique. Personality and Social Psychology Review, 16(3), 203- 232. Sriram, N., & Greenwald, A. G. (2009). The Brief Implicit Association Test. Experimental Psychology, 56, 283–294. Stangor, C. (Ed.). (2000). Stereotypes and prejudice: Essential readings. Philadelphia, PA: Psychology Press. Stein, D. D., Hardyck, J. A., & Smith, M. B. (1965). Race and belief: An open and shut case. Journal of Personality and Social Psychology, 1, 281-290. “With malice toward none . . .” Draft of 9 November 2013 -42- Stephan, W. G., & Stephan, C. W. (1993). Improving intergroup relations. Thousand Oaks, CA: Sage. Tajel, H. (1970) Experiments in intergroup discrimination. Scientific American, 223, 96–102. Tajfel, H., Billig, M. G., Bundy, R. P., & Flament, C. (1971). Social categorization and intergroup behaviour. European Journal of Social Psychology, 1, 149–178. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In S. Worchel & W. G. Austin (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey, CA: Brooks/Cole. Taylor, S. E., Peplau, L. A., & Sears, D. O. (1994). Social psychology (8th ed.). Englewood Cliffs, NJ: Prentice Hall. Terry, D. J., & O’Brien, A. T. (2001). Status, legitimacy, and ingroup bias in the context of an organizational merger. Group Processes & Intergroup Relations, 4, 271-289. Thurstone, L. L. (1928). An experimental study of nationality preferences. Journal of General Psychology, 1, 405–425. Triandis, H. C. (1961). A note on Rokeach’s theory of prejudice. Journal of Abnormal and Social Psychology, 62, 184-186. Triandis, H. C., & Davis, E. E. (1965). Race and belief as determinants of behavioral intentions. Journal of Personality and Social Psychology, 2, 715-725. Turner, M. A., Ross, S. L., Galster, G. C., & Yinger, J. (2002). Discrimination in metropolitan housing markets: National results from Phase I HDS 2000. Washington, DC: Urban Institute. 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: Blackwell. Uhlmann, E. L., & Cohen, G. L. (2005). Constructed Criteria: Redefining Merit to Justify Discrimination. Psychological Science, 16, 474–480. Verniero, P., & Zoubek, P. H. (1999). Interim report of the state police review team regarding allegations of racial profiling. Paper presented at the Race, Police and the Community Conference sponsored by the Criminal Justice Institute of Harvard Law School, December 7– 9, 2000, Cambridge, MA. Wagner, U., Christ, O., & Pettigrew, T. F. (2008). Prejudice and group-related behavior in Germany. Journal of Social Issues, 64 (2), 403–416. Weiss, A., & Rosenbaum, D. P. (2009). Illinois traffic stops statistics study: 2008 annual report. Center for Research in Law and Justice, University of Illinois at Chicago. Downloaded from www.dot.state.il.us/travelstats/ITSS 2008 Annual Report.pdf. Wilson, L., & Rogers, R. W. (1975). The fire this time: Effects of race of target, insult, and potential retaliation on black aggression. Journal of Personality and Social Psychology, 1975, 32, 857–864. “With malice toward none . . .” Draft of 9 November 2013 -43- Yzerbyt, V. Y., Schadron, G., Leyens, J.-P., & Rocher, S. (1994). Social judgeability: The impact of meta-informational cues on the use of stereotypes. Journal of Personality and Social Psychology, 66, 48–55. Ziegler, S. A., Kirby, T. A., Xu, K., & Greenwald, A. G. (2013). Implicit race attitudes predict vote in the 2012 Presidential election. Poster presented at Political Psychology Preconference of the Annual Meeting of the Society of Personality of Social Psychology, New Orleans, LA. “With malice toward none . . .” Draft of 9 November 2013 -44- APPENDIX A: DEFINTIONS OF PREJUDICE Citation Definition Dollard et al. (1939, p. 152) Race prejudice, according to the present view, is a form of aggression Allport (1954, p. 9) an antipathy based upon a faulty and inflexible generalization Secord & Backman (1964, p. 413) an attitude that predisposes a person to think, perceive, feel, and act in favorable or unfavorable ways toward a group or its individual members Jones (1972, p. 61) an unjustified negative attitude towards an individual based solely on that individual’s membership in a group Baron & Byrne (1974, p. 218) Prejudice refers to a special type of attitude—generally a negative one— toward the members of some social group Tajfel (1982, p. 3) a favorable or unfavorable predisposition toward any member of the category in question Simpson & Yinger (1985, p. 21) an emotional, rigid attitude (a predisposition to respond to a certain stimulus in a certain way) toward a group of people Aronson (1988. p.231) a hostile or negative attitude toward a distinguishable group based on generalizations derived from faulty or incomplete information Baron & Graziano (1991, p 526) negative attitudes toward members of social groups Brigham (1991, p. 459) a negative attitude that is considered to be unjustified by an observer Smith (1993, p. 304) a social emotion experienced with respect to one’s social identity as a group member, with an outgroup as a target Stephan & Stephan (1993, p. 125) negative evaluations of social groups Brewer & Crano (1994, p. 464) negative affect directed toward all members of a specific social category Taylor, Peplau, & Sears (1994, p. 216) negative evaluations toward the outgroup Lippa (1994, p. 272) negative attitude that is based on another person’s membership in a social group Brown (1995, p.8) the holding of derogatory social attitudes or cognitive beliefs, the expression of negative affect or the display of hostile or discriminatory behaviour towards members of a group on account of their membership of that group Myers (1995, p. G– 10) an unjustifiable (and usually negative) attitude toward a group and its members [involving] stereotyped beliefs, negative feelings, and a predisposition to discriminatory action Smith & Mackie (1995, p. 170) a positive or negative evaluation of a social group and its members Feagin & Feagin (1996, p. 504) an antipathy, felt or expressed, based upon a faulty generalization and directed toward a group as a whole or toward individual members of a group “With malice toward none . . .” Draft of 9 November 2013 -45- Franzoi (1996, p. 386) a negative attitude directed toward people simply because they are members of a specific social group Stangor (2000, p. 1) a negative feeling or attitude toward the members of a group Nelson (2002, p. 11) an evaluation (positive or negative) [and] a biased perception of a group . . . based on the real or imagined characteristics of the group Eagly & Diekman (2005, p. 31) the relative devaluation in specific role contexts of members of a particular group compared to equivalent members of other groups Aronson, Wilson, & Akert (2012, p. 362) a hostile or negative attitude toward people in a distinguishable group, based solely on their membership in that group