The social life of prejudice

Abstract

ABSTRACTThis article considers a particular explanation (offered in Chapter 7 of Begby 2021) for the persistence of prejudicial stereotypes: that pluralistic ignorance can motivate individuals to act according to the roles they prescribe – even if no individual in a community either believes or endorses the stereotype – and moreover this can make it rational for subsequent generations to acquire prejudiced beliefs. I begin by surveying a few different ways that ‘vestigial social practices’ can persist despite being privately disavowed by most or all members of a community. Noting that many of them are transparently compatible with not believing that the persistent practice is appropriate, I argue that rational consideration of relevant alternative explanations precludes treating others’ behaviour as a kind of testimonial evidence for such prejudicial beliefs. But while it is doubtful that social dynamics provide grounds for rationally acquiring prejudice, it is likely that they explain actual acquisition of prejudice. So when evaluating whether a society is prejudiced, Begby is right that we must look beyond the private thoughts of its individual members. We should attend to the stabilising forces of social expectations, as well as how past prejudice shaped our material environment to reproduce stereotype-conforming social outcomes.KEYWORDS: Stereotypessocial normsrationality of prejudice Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1 Haslanger (Citation2019, 11–13), drawing on McGeer (Citation2007) and others, suggests that the social aspects of agency in fact go much deeper, providing the foundational scaffolding for human cognition and learning. Rather than considering individual agency as prior to social interaction and predicting how others will act, McGeer argues that when we attribute beliefs or predict behaviour we are partly ‘giving ourselves over to the task of producing comprehensible patterns of well-behaved agency in ourselves and others’ (Citation2007, 149). On this picture, social coordination has a regulative and shaping role on individual agency.2 In some cases, participants are perfectly aware that few people privately endorse the norm, but still rationally expect to face second-order enforcement – penalties for either violating the norm or failing to penalize others’ violations – and so the behavioural pattern remains stable until they can be credibly assured safety from penalization. Very likely Begby is right that many vestigial norms actually persist in part through pluralistic ignorance, precisely because we are actually still invested in many of the norms we publicly disavow – including many gender norms. For excellent discussion of how the interconnectedness of a wide array of social scripts can make it very difficult to in fact cognitively move on from socially embedded role-based expectations, see Bicchieri and McNally (Citation2016).3 For Bicchieri, this is the feature that distinguishes a social norm from a convention: norms are supported by an empirical expectation that others will act a certain way, together with a normative expectation that they believe we ought to behave in this way and will sanction departures. Conventions, by contrast, are supported solely by empirical expectations (Bicchieri Citation2006).4 As she explains, left-side driving remained the norm until changed by an official decree and months-long publicity campaign in 1972 (Ullmann-Margalit Citation1977, 88–89).5 In an early study of this form, Katz and Allport (Citation1931, 152–157) found this pattern among undergraduate fraternities. In analysing responses to a 1968 survey, O’Gorman (Citation1975) found that whites who neither strongly favoured desegregation nor segregation supported whichever policy they believed the majority of whites supported, and ‘in 1968 most white Americans grossly exaggerated the extent to which other whites supported racial segregation.’6 Medendian, Gambhir, and Gailes (Citation2021) found that 81% of metropolitan regions in the United States with more than 200,000 residents were more racially segregated in 2019 than they were in 1990.7 In the United States, median home prices increased at four times the rate of household incomes since 1960, while rents have increased at twice the rate.8 For discussion, see generally Rothstein (Citation2017, Chpt 11) and Anderson (Citation2010).