SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

Past research has demonstrated a broad association between prejudice and essentialism. However, research has also shown that essentialism and prejudice are not always linked in the same way – sometimes essentialist thinking is associated with prejudice, but sometimes it is not. The aim of the present research was to explore experimentally how prejudice might relate to essentialist beliefs about race differently depending on how race is being used (inclusively or exclusively) and who is the implied target of such treatment (ethnic minorities or the white majority). Study 1 (N=178) demonstrated that, although prejudice among white Australians is typically related to essentialist beliefs about Aboriginal identity, this relationship disappeared when racial criteria were used to exclude someone for ‘being white’. Under these conditions, prejudiced participants expressed opposition to such treatment and de-essentialized race. Study 2 (N=198) broadly replicated this pattern in a British context and indicated that prejudiced participants' de-essentialism of race was due to a stronger emphasis on values of equality under the same conditions. These results demonstrate that prejudiced people endorse essentialism when it can be used to exclude others (who they want to exclude), but reject essentialism when it is used to exclude them.

Essentialism has been the focus of much recent research on intergroup perception (see Yzerbyt, Judd, & Corneille, 2004, for an overview). In this literature, essentialism is described as the tendency to ascribe some invisible shared essence to all members of a particular group or social category (Rothbart & Taylor, 1992). Although such essences may be difficult to define, even by those who ascribe them (Demoulin, Leyens, & Yzerbyt, 2006), they serve as powerful explanations for apparent similarities among members of one group and apparent differences between them and members of some other group. More than this, the perceived existence of an essence implies permanence to group-based characteristics across time and context. As such, group essences can be drawn on routinely to make inferences about individual group members irrespective of the specific situation (Yzerbyt, Rogier, & Fiske, 1998). The focus of research in this area has been on understanding the extent to which individuals hold essentialist beliefs about various social categories, and how such beliefs guide stereotyped perceptions and discriminatory behaviour.

Outside of psychology, discussions of essentialism emphasize the fluid, dynamic, and political nature of such beliefs. For example, some feminist scholars have noted that although essentialism often comes at a cost (e.g. by inviting differential treatment and discrimination), some degree of essentialism is necessary for social movements based around identity (see Schor & Weed, 1994, for discussions). Thus, minorities might need to engage with ‘strategic essentialism’ in order to mobilize and affect social change. Others have argued that essentialism is equally strategic when used by the majority. For example, in examining the role of race in colonial societies, Stoler (1997) argues that while race is always taken for granted as a category with which to differentiate people, the ways in which racial boundaries are drawn is constantly shifting. This shifting ground allows those with power to mobilize particular models of race in ways that maintain their positions of power. According to Stoler, the very power of essentialist notions of race stem from their apparent fixity but underlying fluidity.

The aim of the present research is to extend previous psychological work on essentialism by examining its variable nature. Our core argument is that essentialist beliefs are not stable, as implied by much of the psychologically oriented research in this area. Instead, essentialist beliefs are endorsed strategically in order to defend perceivers' preferred outcomes. Thus essentialism will be endorsed when such beliefs help to rationalize unfavourable outcomes for an out-group, but will be denounced when they could be used by others to rationalize unfavourable outcomes for the in-group. In short, essentialist beliefs can be motivated and designed to prosecute an intergroup agenda (Morton, Haslam, Postmes, & Ryan, 2006; see also Morton, Postmes, Haslam, & Hornsey, in press). In the current studies, we examine how people with differing levels of prejudice draw on essentialist notions to argue for and against race-based inclusion or exclusion depending on how that inclusion or exclusion was explained.

Essentialism and prejudice

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

Given the manner in which essences facilitate inferences based on group membership, it is unsurprising that research has linked essentialist beliefs to prejudice and stereotyping. For example, research has demonstrated that essentialist beliefs about specific group memberships, or about human nature as a whole, are reliably associated with category devaluation (Haslam, Rothschild, & Ernst, 2000), increased prejudice and stereotyping (Bastian & Haslam, 2006; Haslam, Bastian, Bain, & Kashima, 2006; Haslam, Rothschild, & Ernst, 2002; Keller, 2005; Levy & Dweck, 1999; Levy, Plaks, Hong, Chiu, & Dweck, 2001; Levy, Stroessner, & Dweck, 1998; Martin & Parker, 1995), and resistance to interventions designed to undermine stereotypes or to reduce intergroup bias (Hong et al., 2003; Plaks, Stroessner, Dweck, & Sherman, 2001). Similar associations are observed in research that has manipulated, rather than measured, essentialist beliefs (e.g. Brescoll & LaFrance, 2004; see also Dar-Nimrod & Heine, 2006). In short, accumulated psychological research suggests a fairly singular relationship between essentialist thinking on one hand, and negative intergroup attitudes on the other.

In their work, Haslam and colleagues (Haslam et al., 2000, 2002) argued that essentialist beliefs fall into two underlying dimensions: natural kind beliefs (related to beliefs about the naturalness, immutability, and clearly defined nature of category membership) and reification (related to beliefs about the identity determining, informative and homogenous nature of social categories). Different groups can be essentialized along either or both of these dimensions (i.e. naturalized or reified) and both forms of essentialism are associated with category devaluation (Haslam et al., 2000; see Demoulin et al., 2006, for an elaboration). In a similar fashion, Yzerbyt and colleagues (Yzerbyt, Corneille, & Estrada, 2001; Yzerbyt, Estrada, Corneille, Seron, & Demoulin, 2004) have drawn a distinction between essentialism and entitativity. However, these researchers argue that there is likely to be a dynamic relationship between perceptions of entitativity (i.e. surface-level similarity among group members) and the attribution of underlying essences (i.e. deep-level similarity). That is, surface similarities are likely to act as a cue to the presence of an underlying essence, the perception of which underscores subsequent judgments of that group and its members.

Although there is considerable agreement that essentialism/natural kindness and entitativity/reification characterize perceptions across different group memberships, research on perceptions of specific groups has revealed alternative structures. For example, research on beliefs about sexuality suggests three factors: beliefs that sexuality is biologically based; historically and culturally invariant; and that sexual orientation groups are distinct from each other (Haslam & Levy, 2006). Moreover, in contrast to other domains, essentialist beliefs about sexuality are not straightforwardly related to prejudice. Perceiving sexuality to be biologically based and universal is related to tolerance, whereas perceiving sexual groupings to be fundamentally different is related to prejudice (see also Haslam et al., 2002; Hegarty, 2002; Hegarty & Pratto, 2001). In response to the very different structure and consequences of essentialist beliefs about sexuality, Haslam and Levy argue that ‘in different contexts, categories, or domains, essentialist beliefs may have positive, negative, or ambivalent implications. It may be that there are few consistent patterns of association between particular essentialist beliefs and attitudes and, hence, that any unqualified critique of essentialist thinking about social categories is untenable.’ (p. 482). Others have also alluded to the possibility that essentialism might be dynamic, variable, and conditionally related to prejudice (e.g. Plaks, Levy, Dweck, & Stroessner, 2004; Yzerbyt et al., 2004). However, we still do not know much about what these conditional relationships might be.

Essentialism in use

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

Although past research has presented a fairly uniform picture of the causes, consequences, and structures of essentialist beliefs, more fine-grained investigations into how essentialism operates in relation to specific group memberships suggests that a more complex analysis may be required: essentialism is not always expressed in the same way and it does not always reflect prejudice. This may just reflect differences between social categories – for instance, that sexuality is simply different from race or gender (Haslam et al., 2002). Or, it may reveal something more fundamental about essentialism itself – namely, that essentialism can be put to variable uses. To reach a more complete understanding of essentialism, we would like to move research away from its current focus on identifying uniform structures of essentialism, and singular relationships with intergroup attitudes, and towards an appreciation of how essentialism is used differently across contexts. From this perspective, the question becomes: when do people essentialize group membership, and when do they not?

Others have also argued for this shift in perspective. In a study of ethnic minority and majority group members' everyday discourse, Verkuyten (2003) found that both groups drew on essentialist ideas when discussing the link between culture and ethnicity. However, essentialism was not simply used to enact discrimination (on the part of the majority) and justify inequality (on the part of the minority). Instead, essentialist notions were variably endorsed and denied by both majorities and minorities depending on the rhetorical goals they pursued. For instance, ethnic minorities essentialized culture in response to talk of assimilation, but de-essentialized it in response to issues of discrimination. Conversely, ethnic majorities essentialized cultural differences to justify discrimination, but de-essentialized these to argue against multiculturalism. On the basis of this, Verkuyten has argued that essentialist beliefs about ethnicity can take on either progressive or oppressive meanings depending on the issues at hand and the status of the individual as a minority or majority group member (see also Verkuyten, 2006; Verkuyten & Brug, 2004). As such, it is important to supplement social - cognitive approaches to essentialism with an understanding of how essentialism is preformed in debates about equality.

Making a similar point in the domain of sexuality, Hegarty (2002) found that heterosexual students' beliefs about the immutability of sexuality predicted tolerant attitudes about homosexuality, but only among those who had on a separate occasion judged immutability beliefs to be typical of tolerant people. Based on these patterns, Hegarty suggests that the relationship between essentialist beliefs and intergroup attitudes might depend on how such beliefs are socially constructed rather than what they mean in some absolute sense. Taking this point further, the meaning of essentialism is likely to be more than an individual construction but is also likely to depend on the context in which those ideas are put forth. For instance, one reason why pro-gay people, and homosexuals themselves, might essentialize sexuality is because the very existence of this group is questioned in a way that is not true for other group memberships (e.g. race and gender). Essentialism in this context is a political argument, not just an individual belief.

In the context of gender, research has also revealed that relationships between prejudice and essentialist beliefs vary according to political objectives (Morton et al., in press). These studies found that prejudice and essentialism were positively related only among men who were first asked to consider the ways in which men and women were becoming more equal. By contrast, when men were asked to consider how women were still disadvantaged relative to men or when they were asked to consider the ways in women might now be relatively advantaged over men, the relationship between sexism and essentialist beliefs disappeared. This suggests that essentialist beliefs are a resource used by the prejudiced members of high status groups to guard against the threat of change. When there is no threat of change, or when change has already occurred, essentialist beliefs lose their utility and are no longer endorsed by the prejudiced individuals. Context determines what essentialist beliefs mean and thus how they are used and by whom (see Schmitt, Branscombe, & Kappen, 2003, for a similar argument).

Finally, a parallel body of research informed by attribution theory (Weiner, Perry, & Magnusson, 1988) has concluded that explanations of stigmas, which emphasize the immutable and uncontrollable nature of these should lead to less anger, more sympathy, and more positive attitudes towards stigmatized individuals (e.g. Corrigan, Markowitz, Watson, Rowan, & Kubiak, 2003; Crandall, 1994; Whitley, 1990). This seems contrary to the conclusions of research into essentialism by suggesting that when stigmas are perceived to be biologically determined (involuntary, stable, and uncontrollable) this will elicit the more positive responses from the non-stigmatized. However, in this domain recent research has also highlighted that the relationship between attributions of immutability and positive attitudes is not inevitable, and that the causal order may be reversed (Hegarty & Golden, 2008). That is, beliefs about the underlying nature of stigmas (causal attributions) are expressions of prejudice rather than being the stable foundation on which prejudice is based.

The present research

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

Despite evidence for a variable relationship between prejudice and essentialism, very little research has systematically investigated the conditions under which people essentialize group memberships, the conditions under which they do not, and how both essentialism and de-essentialism might relate to individual differences in prejudice. Accordingly, the aim of the current study was to experimentally investigate how associations between prejudiced attitudes and essentialist beliefs vary across context.

When thinking about when prejudice might be expressed through essentialism, and when it might not, previous qualitative research suggests that this is related to the different models of social relations that people are seeking to argue for or against (Verkuyten, 2003). In particular, this research suggests that ethnic minority and majority participants will express different notions of ethnic identity to argue for inclusion or exclusion of the minority culture in the mainstream (i.e. multiculturalist or assimilationist projects). In our research, we sought to extend this analysis by investigating how the use of race in a given setting (i.e. inclusive vs. exclusive treatment) might influence how prejudice is expressed through essentialist beliefs.

When thinking about the effect of inclusive versus exclusive treatment on support for essentialist notions of race, it seems reasonable to expect that this will depend on who is the actual or implied target of such treatment (the majority or the minority). People should be particularly motivated to reject negative treatment when this is perceived to be directed towards their own group. By contrast, people should be less responsive to treatment (positive or negative) when this does not relate to in-group identity. Consistent with this idea, recent perspectives on intergroup relations have emphasized that negative intergroup behaviours are often more to do with the in-group and how it is perceived, rather than simply a function of negative attitudes towards the out-group (i.e. in-group favouritism: Brewer, 1999; Livingstone & Haslam, 2008; Reicher, Cassidy, Wolpert, Hopkins, & Levine, 2005; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987; Yzerbyt, Castano, Leyens, & Paladino, 2000). For example, research has demonstrated that opposition to affirmative action policies by dominant group members is more pronounced when such policies are framed as harming the in-group rather than helping the out-group (Lowery, Knowles, & Unzueta, 2007; see also Lowery, Unzueta, Knowles, & Goff, 2006). Thus, when social policies, or other forms of social treatment, do not have consequences for the in-group, they are less likely to be experienced as problematic.

Along these lines, negative treatment (e.g. exclusion) should be more concerning to white participants when that treatment is somehow linked to in-group identity (implicating the self) rather than out-group identity (psychologically a less relevant context), or to more positive forms of treatment (e.g. inclusion). Furthermore, given that exclusive treatment linked to majority identity is likely to arouse the most opposition (Lowery et al., 2006), under these conditions it also seems reasonable to assume that people might become motivated to adopt belief systems that undermine or argue against such treatment (Verkuyten, 2003). Whereas essentialist notions might typically be used to justify discrimination against others, when that discrimination is perceived to be directed towards one's own group, essentialist notions should be avoided or downplayed. Finally, this pattern of variable essentialism should be most pronounced among those invested in racialized distinctions, that is those higher in prejudice. One implication of this reasoning is that prejudice should be associated with essentialist thinking about race, except when race is being used to exclude the majority. When exclusive treatment is linked to majority identity, prejudiced people should argue against such treatment and de-essentialize race accordingly.

Study 1

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

White Australian participants were presented with a scenario describing a competition for young indigenous artists. The scenario explained that the organizing committee of this competition had controversially decided to include or exclude one artist (varying inclusive or exclusive treatment). The artist was racially ambiguous – that is, the artist identified himself as Aboriginal, but this was not self-evident on the basis of appearance. Given the ambiguity of the artist, it was possible to explain their treatment in ways that implicated different identities. One half the participants read that this treatment occurred because/despite the artist being ‘not clearly Aboriginal’ (i.e. symbolically connecting treatment to minority identity) whereas the other half read that this treatment occurred because/despite the artist being ‘white’ (i.e. symbolically connecting treatment to majority identity).

We hypothesized that white participants would generally be more concerned with exclusive rather than inclusive treatment. However, we expected that these concerns would be most evident when negative treatment implicated majority identity (i.e. when someone was excluded for being white) and among participants higher in prejudice (i.e. a three-way interaction involving treatment, identity label, and individual differences in prejudice). Importantly, we expected these patterns to be evident not only on explicit measures of support/opposition to the decision, but also on essentialist beliefs about race as a means to reinforce/undermine the validity of such decisions. Thus, we predicted that when race is implicitly being used to exclude their own racial group, prejudiced participants should oppose such treatment and downplay the meaning of race as a social category.

Method

Participants

Participants were 194 undergraduate psychology students at a large metropolitan Australian university. Participants who were not Australian citizens (N=12) or who self-identified as a member of an Australian indigenous community (N=4) were excluded prior to analysis. The final sample of 178 participants comprised 66 men and 112 women who ranged between 17 and 42 years of age (M=18.96, SD=2.80). Participants received course credits in return for their time and were fully debriefed on completion of the study.

Design

Before completing a questionnaire, participants read a paragraph describing a decision made by the organizing committee of a recent ‘Emerging Indigenous Art Award’. The details of the scenario were varied to manipulate key variables of interest. Specifically, the paragraph summarized that one artist had been controversially included in [excluded from] the competition despite him being [because he was] not clearly Aboriginal [white]. The paragraph was accompanied by a picture of the artist, who was racially ambiguous, posing with his art.1 Thus, we manipulated the treatment of the artist (inclusive vs. exclusive) and the identity label with which the artist was described (‘not clearly Aboriginal’ vs. ‘white’). A measure of prejudice towards indigenous Australians was included as an additional independent variable.

Measures
Immediate responses

Immediately after reading the scenario, on a single item the participants indicated whether or not they considered the target's race to be clearly Aboriginal (1, strongly disagree; 9, strongly agree). Then, the participants indicated whether they considered the committee's decision to be legitimate, justifiable, unfair, and racist (1, strongly disagree; 9, strongly agree). After appropriate reverse scoring, these four items were combined into a single index (α=.89) on which higher scores indicated stronger opposition to the committee's decision.

Beliefs about Aboriginal identity

Nine items measured participants' beliefs about the essential basis of Aboriginal identity. Five items addressed natural kind essentialism, for example, asking participants the extent to which they agreed that ‘Being Aboriginal or not is determined primarily by a person's biological make up’ and ‘There are certain physical features that a person must have to be Aboriginal’. A further four items measured reification essentialism, for example, by asking participants the extent to which they agreed that ‘Aboriginal people tend to be very similar to each other and have many things in common’ and ‘Knowing that a person is Aboriginal tells you a lot about them’.

Consistent with past research (e.g. Demoulin et al., 2006; Haslam et al., 2000, 2002), factor analysis (principal components with varimax rotation) suggested the existence of two underlying factors, one dominated by natural kind beliefs and the other by reification beliefs, which together accounted for 51.17% of the variance in individual items. Accordingly, we created a composite measure for each dimension of essentialism: natural kinds (α=.67) and reification (α=.69).

Prejudice

Finally, participants completed a measure of prejudice against Aboriginal Australians (adapted from Pedersen, Beven, Walker, & Griffiths, 2004). Eight items assessed the extent to which participants agreed with a number of common negative beliefs about Aboriginal people as well as perceptions of the legitimacy of the Aboriginal rights movements. These items formed a reliable scale and were combined and averaged into a single index on which higher scores indicated stronger prejudice against Aborigines (α=.89). The measure of prejudice was taken at the end of the study to avoid priming prejudiced beliefs prior to completion of the key dependent variables.2

Results

To examine how treatment of the target, identity label, and individual differences in prejudice were related to participants' immediate responses to the scenario and beliefs about Aboriginal identity more generally, we conducted a series of hierarchical multiple regression analyses. In these analyses, treatment condition (inclusion, 0; exclusion, 1), identity label (not Aboriginal, 0; white, 1), and measured prejudice (centred) were entered at the first step. At step 2, the three possible two-way interactions among these variables were tested, and the three-way interaction was tested at step 3. The creation of interaction terms and the probing of significant interactions were conducted via recommendations outlined in Aiken and West (1991) for analysing continuous and categorical data via regression.

Immediate responses

The analysis performed on the perceived race of the target (i.e. whether participants considered them to be clearly Aboriginal or not) revealed that only inclusion of the main effect terms at step 1 contributed significantly to variance explained, R2ch=.06, Fch(3, 174)=3.49, p=.02. Inspection of the regression coefficients at this step revealed that this was due to a significant main effect of prejudice alone, β=−0.23, t(174)=3.11, p=.002: prejudiced participants were less inclined to consider the target to be clearly Aboriginal. There were no effects of treatment or identity label on perceptions of the target's race, and no higher order interactions among the variables.3

In the analysis of opposition, inclusion of the main effect variables at step 1 produced a significant increment in variance explained, R2ch=.51, Fch(3, 174)=61.22, p<.001, which was due to a main effect of treatment alone, β=0.71, t(174)=13.50, p<.001. Participants were more inclined to oppose the decision to exclude rather than include the target. This main effect was, however, qualified by a significant three-way interaction among the variables at the final step, R2ch=.01, Fch(1, 170)=4.41, p=.04 (Figure 1). Further analysis revealed that this three-way interaction was due to a prejudice×identity label interaction that was significant when the target was excluded, β=0.26, t(170)=2.23, p=.03, but not significant when it was included, β=−0.06, t(170)=0.79, p=.43.

image

Figure 1. Study 1: The interaction among prejudice, treatment, and identity label on opposition to the committee's decision.

Download figure to PowerPoint

When the target was excluded and labelled ‘not clearly Aboriginal’, then prejudice was weakly negatively associated with opposition to the committee's decision, β=−0.21, t(170)=1.75, p=.08. However, when a target labelled ‘white’ was subjected to the same treatment this relationship was non-significant but positive, β=0.16, t(170)=1.40, p=.16. Put differently, participants higher in prejudice opposed the exclusion of a ‘white’ target more than the exclusion of a target labelled ‘not clearly Aboriginal’, β=0.29, t(170)=2.50, p=.02. Participants lower in prejudice were not influenced by the identity label when giving their opinions, β=−0.09, t(170)=0.80, p=.43.

Beliefs about Aboriginal identity

Next, we repeated this analysis on beliefs about the nature of Aboriginal identity along the two dimensions of essentialism: reification and natural kind beliefs. The analysis performed on reification revealed a significant effect of the variables included at step 1, R2ch=.24, F(3, 174)=18.05, p<.001. This was due to a significant main effect of prejudice, β=0.49, t(174)=7.30, p<.001, and a marginally significant effect of identity label, β=−0.13, t(174)=1.92, p=.057. Participants higher in prejudice were more likely to reify Aboriginal identity, as were those exposed to a target labelled ‘not clearly Aboriginal’ rather than ‘white’. There were no further main effects or interactions among the variables.

The same analysis performed on natural kind beliefs revealed a significant effect of the variables included at step 1, R2ch=.09, Fch(3, 174)=5.58, p=.001, which was due to a significant main effect of prejudice, β=0.27, t(174)=3.65, p<.001, and a marginally significant effect of treatment, β=0.13, t(174)=1.82, p=.07. Participants higher in prejudice were more likely to endorse a naturalized view of Aboriginal identity, as were those exposed to exclusion rather than inclusion of the target. These effects were, however, qualified by a significant three-way interaction among the variables at step 3, R2ch=.04, Fch(1, 170)=8.47, p=.004 (see Figure 2).

image

Figure 2. Study 1: The interaction among prejudice, treatment, and identity label on naturalization of Aboriginal identity.

Download figure to PowerPoint

Further analysis of this three-way interaction revealed that, when the target was included, there was only a main effect of prejudice, β=0.29, t(173)=3.08, p=.002, such that participants expressed more natural kind essentialism the more prejudiced they were. This effect was not moderated by identity label, β=0.22, t(170)=1.66, p=.10. In contrast, when the target was excluded, a significant prejudice×identity label interaction emerged, β=−0.39, t(170)=2.41, p=.02. When the target was excluded and labelled ‘not clearly Aboriginal’, prejudice was positively related to natural kind essentialism, β=0.50, t(170)=3.08, p=.002. However, when the target was labelled ‘white’ under the same conditions, there was no relationship between prejudice and natural kind essentialism, β=−0.05, t(170)=0.29, p=.78. In sum, prejudice was positively related to natural kind essentialism in all conditions, except when the target was excluded for ‘being white’. Indeed, an analysis testing the interaction between prejudice and this specific contrast (exclusion for ‘being white’, 3; all other conditions, −1) was significant, β=−0.17, t(170)=2.32, p=.02, confirming that the relationship between prejudice and natural kind essentialism was different in this condition compared with all others.

Discussion

The results of this study demonstrate support for our predictions. Although white participants were generally more opposed to race-based exclusion than inclusion, the extent of their opposition appeared to depend on the participants' level of prejudice and how exclusive treatment was explained. When the exclusion of a racially ambiguous person was explained as being due to them being ‘not clearly Aboriginal’, prejudiced participants displayed slightly less opposition to such a decision compared with less prejudiced participants. However, when exclusion was explained as being due to the artist being ‘white’, prejudiced participants were somewhat more opposing of that decision. In short, among prejudiced participants, feelings of opposition towards exclusive treatment were contingent on who was being excluded: someone ‘white’ (more opposition) or someone ‘not clearly Aboriginal’ (less opposition).

Consequently, although prejudice was generally associated with essentialism, this relationship varied across conditions. Specifically, when racial criteria were used to exclude someone for being ‘not clearly Aboriginal’, prejudiced participants were more inclined to endorse essentialist notions of Aboriginal identity (i.e. to see this as naturalized). However, when the artist was excluded for being ‘white’, prejudiced participants were not more inclined to endorse essentialism than less prejudiced participants. Thus, the manner in which exclusive treatment was explained influenced prejudiced participants' support for such treatment and their beliefs about Aboriginal identity. Identity labels did not affect the relationship between prejudice and essentialism in the context of inclusive treatment. These patterns suggest that, although essentialism might be the default position of prejudice, prejudice is not always expressed through essentialism. Indeed, when race is being used to implicitly exclude their own, prejudiced participants might appear more like social constructivists than essentialists.

This analysis highlights the way in which prejudiced participants invoke different notions of race to argue for or against specific instances of inclusion or exclusion. However, it can be argued that the changing relationship between prejudice and essentialism across conditions was partly driven by movement among less prejudiced participants. That is, less prejudiced participants were most inclined to endorse a naturalized view of Aboriginal identity when a ‘white’ candidate was excluded, and least inclined to do so when a ‘white’ candidate was included. Given this pattern, it is important to consider what essentialist beliefs might represent to less prejudiced participants. It could be that less prejudiced participants were focused on understanding, rather than challenging, the committee's decision. That is, to understand the inclusion of a racially ambiguous candidate in an indigenous art prize, it is necessary to adopt an inclusive view of race. Conversely, to understand the exclusion of such a candidate from competition, a narrower view of race is necessary. It is interesting that this pattern was most apparent when the ambiguous candidate was labelled white – that is, when inclusion and exclusion were tied to participants' own race.

The apparent movement of less prejudiced participants introduces some ambiguity into our analysis. However, given the unexpected nature of this finding, it is important to determine whether these findings are reliable. In addition, although the pattern displayed by more prejudiced participants conformed to expectations, it is important to consider the mechanisms that might be driving such responses (i.e. the mediating process). We conducted a second study to resolve some of this ambiguity.

Study 2

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

Although the results of Study 1 were in line with our predictions, it remains unclear precisely why essentialist beliefs are the arena in which prejudiced participants support or challenge exclusion of individuals on racial grounds. It could be that when categories like race are being used in ways that support preferred outcomes, it is useful to see the category itself as meaningful. However, when such categories are being used in ways that do not support our preferred outcomes, it is useful to deny the meaningfulness of the category as well as attacking those outcomes. In this way, ideologies about race are likely to shift according to their utility to those in power (Stoler, 1997). This suggests a strategic use of essentialism (i.e. one that is calculating and targeted). However, it is unlikely that ordinary people would admit that their behaviour is a strategic attempt to restore or maintain power. Instead, prejudiced people may be able to rationalize their responses to race through appeals to more generic values.

Theorists have argued that contemporary forms of prejudice are built on 1) an assumption that discrimination no longer exists and 2) cultural values that conflict with actions designed to reduce inequality (Katz & Hass, 1988). That is, by ignoring evidence for discrimination, prejudiced people find policies designed to improve the position of minorities in society as unfair and unjust because they do not apply to all members of society equally. As such, prejudiced people may experience their opposition to social policies that make use of race as part of a ‘principled objection’ rather than outright hostility (e.g. Crandall & Eshleman, 2003; Federico & Sidanius, 2002). Interestingly, other research suggests that principles of fairness and equality are likely to come to the fore when the in-group's interests are at stake (e.g. Lowery et al., 2006). That is, people are likely to be particularly concerned about fair treatment when their in-group might be the recipient of negative treatment, but are less concerned when an out-group is similarly disadvantaged (see also Duck & Fielding, 2003).

Along these lines, it seems reasonable to expect that a selective application of egalitarian values might underlie the differential constructions of race observed in Study 1. That is, exposure to exclusionary treatment that is linked to majority identity might activate egalitarian concerns among those higher prejudice. Once activated, concerns about equal treatment should trigger racial de-essentialism as the means to argue against exclusionary treatment. Thus, prejudiced participants' opposition to such treatment becomes part of a broader universalist claim – as if to say: ‘we are all equal, therefore, race is meaningless’. Conversely, when exclusionary treatment is linked to minority identity, egalitarian concerns should not be activated amongst those higher in prejudice, and consequently there should be no motivation to de-essentialize race.

In order to explore these possibilities, we conducted a second study. The aim of this study was twofold. First, we sought to examine the reliability of the pattern observed in Study 1. Second, we sought to explore how the activation of egalitarian concerns might motivate the pattern of variable essentialism observed in the previous study. This study made use of the same design, but this time in a British context and with reference to a different ethnic minority group: Asians.

Method

Participants

Participants were 223 students approached at various locations on a British university campus. Participants who identified as anything other than white British were excluded prior to analysis (N=25). The final sample of 198 participants comprised 71 men and 127 women who ranged between 17 and 30 years of age (M=19.65, SD=1.77). Participants received a chocolate bar in return for participation and were fully debriefed on completion of the study.

Design

The design was identical to that of the previous study with one change: in this study, participants read about a decision made by the organizers of a prize for writers of Asian origin, rather than an Aboriginal art prize. As in the previous study, participants were presented with the picture of a racially ambiguous target and were told that the committee had controversially included [excluded] this person in [from] competition. One half of the participants were told that this treatment occurred because [despite] him ‘being white’. The other half were told that this treatment occurred because [despite] him ‘not being Asian’.4 In both conditions, it was emphasized that the candidate identified himself as Asian, but that his race was not clear to the committee. Finally, a British version of the modern racism scale was included as an additional independent variable.

Measures

As in the previous study, we assessed responses to the specific scenario and beliefs about Asian identity. We also included a measure of egalitarianism, our proposed mediator. Finally, participants completed a British measure of modern prejudice. All responses were given on 9-point scales ranging from 1, strongly disagree to 9, strongly agree.

Immediate responses

Immediately after reading the scenario, on a single item participants indicated whether they disagreed or agreed that the target's race was clearly Asian. Then, on four items participants indicated whether they considered the committee's decision to be unjustified, unfair, motivated by special interests, or motivated by positive intentions. After appropriate reverse scoring, these items formed a reliable scale (α=.73) and were combined and averaged into a single index of opposition to the committee's decision.

Beliefs about Asian identity

Eight items were used to tap participants' beliefs about the essential basis of Asian identity – including four items measuring natural kind essentialism (α=.59)5 and four items measuring reification essentialism (α=.75). Factor analysis (principal components with varimax rotation) confirmed the presence of two underlying factors that together accounted for 54.01% of the variance in individual items.

Egalitarianism

Egalitarianism was measured using two items: ‘There should be equality for everyone – because we are all human beings’, and ‘Everyone should have an equal chance and an equal say in most things’. These items were combined and averaged to form a single index (r=.78, p<.001).

Prejudice

Finally, 14 items assessed individual differences in prejudice adapted to the British context (from Lepore & Brown, 1997). These 14 items formed a reliable scale and were combined and averaged into a single index on which higher scores indicated stronger prejudice against ethnic minorities (α=.88). As for the previous study, we assessed prejudice at the end of the questionnaire to avoid priming prejudiced beliefs prior to completion of the key dependent variables.6

Results

As in Study 1, treatment condition (inclusion, 0; exclusion, 1), identity label (not Asian=0, white=1), and measured prejudice (centred) were entered into a series of hierarchical moderated regression analyses following procedures outlined in Aiken and West (1991) for analysing continuous and categorical data via regression.

Immediate responses

Consistent with the emphasis on the target's ambiguity in the scenario, participants generally perceived the target's race to be unclear (M=3.88, SD=1.71; on a 9-point scale). Although inclusion of the main effect variables at step 1 did not contribute significantly to variance explained, R2ch=.03, Fch(3, 191)=1.87, p=.14, inspection of the regression coefficients at this step revealed a marginal effect of treatment, β=0.14, t(191)=1.90, p=.06, suggesting that the perceived ambiguity of the target's race was slightly more pronounced when the candidate was excluded rather than included. There were no further effects of identity label, participants' own levels of prejudice, or any higher-order interactions among the variables.7

Analysis of opposition revealed a significant effect of the variables included at step 1, R2ch=.20, Fch(3, 193)=16.37, p<.001. Inspection of the regression coefficients at this step revealed that this was due to significant main effects of treatment, β=0.41, t(193)=6.41, p<.001, and prejudice, β=0.19, t(193)=2.95, p=.004. Participants opposed the exclusive treatment more than the inclusive treatment, and participants higher in prejudice tended to express stronger opposition to the committee's decisions irrespective of their nature.

However, there was also a significant treatment×identity interaction at step 2, R2ch=.03, Fch(3, 190)=2.53, p=.059, β=0.28, t(190)=2.54, p=.01. This interaction is depicted in Figure 3. Further analysis revealed that participants expressed relatively more opposition to the decision to exclude a person for ‘being white’ than for being ‘not Asian’, β=0.19, t(192)=2.11, p=.04. When the committee's decision was to include the target, there was no effect of race label, β=−0.13, t(192)=1.51, p=.13.

image

Figure 3. Study 2: The interaction between treatment and identity label on opposition to the committee's decision.

Download figure to PowerPoint

Although these patterns demonstrate that white participants express most opposition to the decision to exclude a target for ‘being white’, and that participants higher in prejudice were most likely to express such opposition, the three-way interaction was not significant in this study, β=0.12, t(189)=1.09, p=.28.

Beliefs about Asian identity

The analysis of natural kind beliefs revealed a significant effect of the variables included at step 1, R2ch=.20, Fch(3, 193)=16.49, p<.001, which was due to a significant main effect of prejudice alone, β=0.45, t(193)=6.96, p<.001. Participants higher in prejudice were more likely to essentialize Asian identity in natural kind terms than those lower in prejudice. In contrast to the previous study, there were no further main or interactive effects of the variables.

When this analysis was repeated on reification beliefs, there was also a significant effect of the variables included at step 1, R2ch=.25, Fch(3, 194)=21.70, p<.001. This was due to significant main effects of prejudice, β=0.48, t(193)=7.72, p<.001, and treatment, β=−0.13, t(193)=2.13, p=.03. Participants who were higher in prejudice were more inclined to reify Asian identity as were those faced with inclusive rather than exclusive treatment. However, these effects were qualified by a significant three-way interaction among the variables at step 3, R2ch=.02, Fch(1, 190)=4.16, p=.04. As can be seen in Figure 4, prejudice was significantly associated with reification essentialism in all conditions, βs>0.50, ts(190)>4.12, ps<.001, except in the condition in which the candidate was excluded for ‘being white’, β=0.17, t(190)=1.18, p=.24. An analysis testing the interaction between prejudice and this specific contrast (exclusion for ‘being white’, 3; all other conditions, −1) was significant, β=−0.16, t(190)=2.52, p=.01, confirming that the relationship between prejudice and reification was different in this condition compared with all others.

image

Figure 4. Study 2: The interaction among prejudice, treatment, and identity label on reification of Asian identity.

Download figure to PowerPoint

Further analysis revealed that this three-way interaction was also partly due to a prejudice×treatment interaction that was significant in the white condition, β=0.38, t(190)=2.73, p=.007, but not significant in the Asian condition, β=0.01, t(190)=0.04, p=.97. This two-way interaction was, in-turn, driven by a main effect of treatment that was significant for high prejudice participants, β=−0.42, t(190)=3.25, p=.001, but not low prejudice participants, β=0.10, t(190)=79, p=.43. Prejudiced participants were less inclined to reify race when someone was excluded for ‘being white’ than when the same target was included.

Egalitarianism

Consistent with Study 1, the results suggest that prejudiced participants do not essentialize race when race is being used to exclude ‘them’. Given this, we explored how commitment to values of egalitarianism might explain this pattern. Hierarchical multiple regression on this variable revealed a significant main effect of the variables entered at step 1, R2ch=.21, Fch(3, 194)=17.66, p<.001. This was due to a significant main effect of prejudice, β=−0.46, t(194)=7.19, p<.001. Prejudiced participants were less inclined to endorse values of equality. However, this main effect was qualified by a significant three-way interaction among the variables at step 3, R2ch=.03, Fch(1, 190)=5.12, p=.03. As can be seen in Figure 5, the negative association between prejudice and endorsement of egalitarian values held in all conditions, βs>−0.39, ts(190)>3.21, ps<.002, except when the target was excluded for ‘being white’, β=−0.22, t(190)=1.53, p=.13. An analysis testing the interaction between prejudice and this specific contrast (exclusion for ‘being white’, 3; all other conditions, −1) was significant, β=−0.12, t(190)=1.94, p=.05, confirming that the relationship between prejudice and egalitarianism was different in this condition compared with all others.

image

Figure 5. Study 2: The interaction among prejudice, treatment, and identity label on endorsement of egalitarian values.

Download figure to PowerPoint

Further analysis revealed that the three-way interaction was also partly due to a prejudice×treatment interaction that was significant in the ‘white’ condition, β=−0.43, t(190)=3.00, p=.003, but not significant in the ‘Asian’ condition, β=−0.00, t(190)=0.02, p=.99. When contemplating a target labelled white, prejudiced participants endorsed egalitarian values more strongly when the target was excluded rather than included, β=−0.41, t(190)=3.07, p=.002. Less prejudiced participants endorsed egalitarian values regardless of the target's treatment, β=0.18, t(190)=1.39, p=.17.

Mediation

The above analyses reveal a parallel pattern of effects on reification essentialism and egalitarianism. Contemplating the exclusion of another for ‘being white’ activated egalitarian concerns among prejudiced participants and these participants also de-essentialized race under the same conditions. Thus, it was possible that egalitarian values might mediate the responses of prejudiced participants.

Given that the parallel patterns emerged specifically within the condition in which participants contemplated the ‘white’ target, we tested for mediation within this half of the sample. This analysis revealed a significant negative relationship between egalitarianism and reification, β=−0.27, t(98)=2.64, p=.01. Moreover, the prejudice×treatment interaction was weakened on the introduction of the mediator, β=0.35, t(98)=2.77, p=.007, to β=0.25, t(98)=1.93, p=.06. This suggests that prejudiced participants' de-essentialism of race in response to the exclusion of someone for ‘being white’ was mediated by their increased commitment to egalitarian values under the same conditions, an interpretation that was supported by a significant Sobel test, (z=2.12, p=.04). In the ‘Asian’ condition, there was no relationship between egalitarian values and reification, β=0.11, t(90)=1.11, p=.27, and accordingly no evidence for mediation, (z=.02, p=.99).

Discussion

The pattern observed in this study broadly replicates that observed in Study 1. Again, prejudice was associated with racial essentialism, except when race was used to exclude someone for ‘being white’. Under these conditions, the relationship between prejudice and essentialism disappeared. Moreover, the results of this study suggest that prejudiced participants' de-essentialism of race in response to the exclusion of another for being white is based on to the activation of egalitarian values. In this study, prejudiced participants not only opposed such treatment, but they also emphasized the value of equality and downplayed the meaningfulness of race under the same conditions. Thus, their responses are likely to be experienced as a principled objection to the use of race rather than as an expression of racial antipathy – indeed, patterns of essentialism were mediated through endorsement of egalitarian values.

Although this replicates Study 1, the results of this study differ in one aspect. Namely, the pattern described in Study 1 emerged on a measure of natural kind essentialism (i.e. perceiving race in biophysical terms), whereas the one observed in Study 2 emerged on a measure of reification essentialism (i.e. perceiving race as a meaningful and informative construct). Although both naturalization and reification are forms of essentialism (Demoulin et al., 2006; Haslam et al., 2000, 2002), they are also clearly distinct (Yzerbyt et al., 2001, 2004).

One reason for this difference may be the backdrop against which these studies were conducted. Aborigines (the target of participants' beliefs in Study 1) are the indigenous people of Australia, whereas white Australians (those making the judgments) are the products of migration. This situation is reversed in Study 2, with the indigenous population (white British) making judgments about the identity of a migrant population (Asians). This difference may map on to a distinction between forced and chosen categories that has been identified in recent research on essentialism. Specifically, Demoulin et al. (2006) demonstrated that forced social categories are more likely to be essentialized in natural kind terms, whereas chosen social categories are more likely to be essentilized through reification. As the indigenous population of Australia, Aborigines' presence in Australian society was not chosen. Asian populations, at some point, have chosen to migrate to Britain. This may explain why effects emerged on naturalization with respect to Aboriginal identity and reification with respect to Asian identity. Of course, there are many other distinctions between Aborigines and Asians that might explain the different patterns. Principally, Aborigines continue to be the most marginalized community in Australia, whereas Asians constitute a larger and more powerful group within the British context. Given the differences between these cultural contexts, it is even more striking that a similar pattern of variable essentialism emerged across the two studies.

General discussion

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

The aim of the present research was to extend past research into the relationship between prejudice and essentialist beliefs (e.g. Bastian & Haslam, 2006; Haslam et al., 2002, 2006; Keller, 2005) by considering how this relationship might be variable rather than singular (Hegarty, 2002; Morton et al., 2006; Verkuyten, 2003). Specifically, we examined how concerns triggered by inclusive versus exclusive treatment, and how such treatment is explained with reference to minority or majority identity, might influence the extent to which prejudiced individuals endorse essentialist beliefs about race. The basic point that emerges across two studies, conducted within two different cultural contexts, is that white participants were generally more concerned about race-based exclusion when that exclusion was linked to majority identity. Participants in both studies expressed the most opposition to a decision to exclude someone for being ‘white’ relative to conditions in which exclusion occurred because of questioned minority status or conditions in which targets were included despite racial criteria. These patterns of objection were most apparent among participants higher in prejudice, as evidenced by a three-way interaction among the factors in Study 1, and additive effect of prejudice in Study 2.

Beyond their opposition, prejudiced participants also appeared to challenge the meaningfulness of race when race was being used to exclude someone for being white (in terms of naturalness in Study 1 and reification in Study 2). Together, these patterns show that although essentialism might be the default position of prejudice, prejudice is not always associated with essentialism. Prejudiced participants will question the physical or metaphysical reality of race when race is implicitly being used to exclude members of their own group.

Our interpretation of these findings suggests a strategic dimension to essentialism. That is, prejudiced participants will invoke and deny essentialist ideas flexibly to argue for or against particular forms of social treatment (Verkuyten, 2003; see also Stoler, 1997). Consistent with this interpretation, Study 2 revealed that this pattern of variable essentialism among prejudiced participants also emerged on commitment to values of equality. That is, although prejudiced participants were typically less inclined to endorse egalitarian ideals, the exclusion of someone for ‘being white’ activated a commitment to equality among the prejudiced. Moreover, this increased commitment to egalitarian ideals mediated the variability in essentialism – at least when the target was described as ‘white’. This supports our contention that essentialist beliefs can be motivated constructions that are driven by concerns for equal treatment of the in-group as much as they are by desires to justify unequal treatment of out-groups.

These patterns also allude to a broader point. That is, in addition to the expression of prejudice through essentialism being variable and slippery, the taken for granted values to which such expressions appeal are also slippery. In this way, prejudice can be expressed in contradictory ways across contexts. Without paying attention to the role of context, it is difficult to pinpoint exactly what prejudice should look like vis-à-vis things like beliefs about race, but also broader ideological or value positions, such as commitments to equality (see also Schmitt et al., 2003). Similar to some critiques of research on stereotyping, we would argue that in the domain of essentialism ‘any approach which abstracts these from their argumentative context is likely both to obscure our understanding of the phenomenon and to reify particular political dispensations’ (Reicher & Hopkins, 2001, p. 104). For example, given the general correlation between essentialism and prejudice, it may be tempting to conclude that essentialism is the problem, and that dismantling essentialist notions of social categories is the path to tolerance (e.g. Keller, 2005). However, many progressive political movements rest on essentialist notions of identity (e.g. feminism, multiculturalism, gay liberation), and denying such essences can equally serve an oppressive agenda (Verkuyten, 2003).

Limitations and future research

The studies reported here represent an initial attempt to explore the dynamics of essentialism and prejudice in relation to race. Although the data provide some support for our core contention, three additional questions remain unanswered.

First, the scenarios we used to manipulate the inclusion and exclusion of particular targets made use of the existence of specific prize competitions for members of minority groups. It should be noted that the existence of such prizes, irrespective of who is included or excluded, is controversial. Participants may have been reacting against the very idea of such a competition as much as to the treatment of specific candidates. Thus, the broader controversy may have contributed to participants' responses. However, this does not explain the specific pattern of results and why responses varied according to our manipulations of treatment (inclusive vs. exclusive) and the identity to which treatment was linked (the minority or the majority). Indeed, our results suggest that one reason why opposition is directed towards such prizes is because prejudiced members of the majority see the prizes as designed to exclude them rather than promoting the minority (see also Lowery et al., 2006, 2007). Nonetheless, it remains to be seen how the present results would generalize to alternative forms of inclusive or exclusive treatment and to other social categories.

A second issue is that we have focused exclusively on majority group members' tendencies to essentialize minority identity. That is, although we asked for white Australians' beliefs about the nature of Aboriginal identity (Study 1) and for white British participants' beliefs about Asian identity (Study 2), we never asked about the nature of whiteness itself. It seems likely that how people construe whiteness might also shift according to context, and that beliefs about the essential nature of white identity might be an important guide to the treatment of racial minorities in society. However, the limited research that has differentiated between minority group and majority group essentialism suggests that essentialist beliefs about minority identity are a stronger predictor of attitudes about multiculturalism than essentialist beliefs about the majority (Verkuyten & Brug, 2004).

Finally, although our analysis suggests that essentialist beliefs are strategic constructions, alternative explanations exist. In particular, it could be suggested that the emphasis on egalitarian values and de-essentialism of race by prejudiced participants in the ‘white exclusion’ condition was not strategic but genuine. That is, experiencing racial exclusion directed towards their in-group might have opened prejudiced participants' eyes to the reality of discrimination. This experience may have caused them to actually become less prejudiced, and thus truly more egalitarian and less essentialist in their thinking. Without examining participants' beliefs across time and contexts, it is impossible to say whether the racial de-essentialism observed in these studies is of the context dependent and strategic kind that we suggest, or reflective of a genuine and enduring change in beliefs. However, we do think that the latter possibility is unlikely, and our data speak against this alternative explanation. In both studies, we measured prejudice after the manipulations and dependent measures to avoid priming prejudice at the outset. In Study 1, labelling the target as ‘white’ marginally increased prejudice, irrespective of treatment, and in Study 2 neither identity label nor treatment affected prejudice. Thus, we can say that the experience of racial exclusion did not reduce white participants' levels of prejudice.

The related question of whether prejudiced participants experience their variable commitment to egalitarian values and expression of essentialist ideas as genuine or calculating remains unanswered by our data. In this respect, the strategic aspect of essentialism is revealed more in the variability across contexts, and the in-group serving nature of this, rather than in the explicitly strategic motivations of individual actors. To resolve this issue, future research would need to explore how conscious individuals are of this variability, and what motives they ascribe to their own behaviour.

Conclusions

Essentialist beliefs are typically treated as an extension of prejudiced thinking. In contrast to this static view of essentialism, the present findings demonstrate that prejudiced people do not always view race as self-evident or essential. Instead, how people conceptualize race is contingent on the manner in which race is being used and what this means for self-identity. Although the dynamic potential of essentialism has been alluded to by some theorists (Yzerbyt et al., 2004; Plaks et al., 2004; see also Haslam, Bastian, & Bissett, 2004), attention to this dynamic has been relatively limited in psychological research on essentialism.

The present research shows that prejudiced people ‘see’ race in essentialist terms only when it can be used to exclude ‘them’ (i.e. the minority). When ‘we’ (the majority) are excluded on racial terms, it becomes more important to play down the essential features of race. Thus, prejudice, commitment to inequality, and essentialist ideas can be fused together in order to pursue particular lines of racist argument, but can equally be defused when alternative lines of argumentation are necessary. The studies presented here demonstrate that insights into the psychological aspects of essentialism can be enriched by understanding how essentialist beliefs are used in specific contexts – that is, by asking the question of who is essentializing whom and to what end?

Acknowledgements

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References

Thanks to Sarah Esposo and Daniel Healy for their assistance with data collection and data entry. This research was supported by an ESRC grant (RES-062-23-0108) to the first and third authors.

References

  1. Top of page
  2. Abstract
  3. Essentialism and prejudice
  4. Essentialism in use
  5. The present research
  6. Study 1
  7. Study 2
  8. General discussion
  9. Acknowledgements
  10. References
  • Aiken, L. S. West, S. G. Multiple regression: Testing and interpreting interactions Newbury Park, CA Sage 1991.
  • Bastian, B. Haslam, N. Psychological essentialism and stereotype endorsement Journal of Experimental Social Psychology 42 228235 2006.
  • Brescoll, V. LaFrance, M. The correlates and consequences of newspaper reports of research on sex difference Psychological Science 15 515520 2004.
    Direct Link:
  • Brewer, M. The psychology of prejudice: Ingroup love or outgroup hate? Journal of Social Issues 55 429444 1999.
  • Corrigan, P. W. Markowitz, F. E. Watson, A. Rowan, D. Kubiak, M. A. An attribution model of public discrimination towards persons with mental illness Journal of Health and Social Behavior 44 162179 2003.
  • Crandall, C. S. Prejudice against fat people: Ideology and self-interest Journal of Personality and Social Psychology 66 882894 1994.
  • Crandall, C. S. Eshleman, A. A justification-suppression model of the expression and experience of prejudice Psychological Bulletin 129 414446 2003.
  • Dar-Nimrod, I. Heine, S. J. Exposure to scientific theories affect women's math performance Science 314 435 2006.
  • Demoulin, S. Leyens, J.-Ph Yzerbyt, V. Lay theories of essentialism Group Processes and Intergroup Relations 9 2542 2006.
  • Duck, J. M. Fielding, K. S. Leaders and their treatment of subgroups: Implications for evaluations of the leader and the superordinate group European Journal of Social Psychology 33 387401 2003.
  • Federico, C. M. Sidanius, J. Racism, ideology, and affirmative action, revisited: The antecedents and consequences of principled objections to affirmative action Journal of Personality and Social Psychology 82 488502 2002.
  • Haslam, N. Bastian, B. Bain, P. Kashima, Y. Psychological essentialism, implicit theories, and intergroup relations Group Processes and Intergroup Relations 9 6376 2006.
  • Haslam, N. Bastian, B. Bissett, M. Essentialist belief about personality and their implications Personality and Social Psychology Bulletin 30 16611673 2004.
  • Haslam, N. Levy, S. Essentialist beliefs about homosexuality: Structure and implications for prejudice Personality and Social Psychology Bulletin 32 471485 2006.
  • Haslam, N. Rothschild, L. Ernst, D. Essentialist beliefs about social categories British Journal of Social Psychology 39 113127 2000.
  • Haslam, N. Rothschild, L. Ernst, D. Are essentialist beliefs associated with prejudice? British Journal of Social Psychology 41 87100 2002.
  • Hegarty, P. It's not a choice, it's the way we're built: Symbolic beliefs about sexual orientation in the US and Britain Journal of Community and Applied Social Psychology 12 153166 2002.
  • Hegarty, P., & Golden, A. M. (2008). Attributions about the controllability of stigmatized traits: Antecedents or justifications of prejudice? Journal of Applied Social Psychology, 38, 10231044.
  • Hegarty, P. Pratto, F. Sexual orientation beliefs: Their relationship to antigay attitudes and biological determinist arguments Journal of Homosexuality 41 121135 2001.
  • Hong, Y. Chan, G. Chiu, C. Wong, R. Hansen, I. Lee, S. How are social identities linked to self-conception and intergroup orientation? The moderating effect of implicit theories Journal of Personality and Social Psychology 85 11471160 2003.
  • Katz, I. Hass, R. G. Racial ambivalence and American value conflict: Correlational and priming studies of dual cognitive structures Journal of Personality and Social Psychology 55 893905 1988.
  • Keller, J. In genes we trust: The biological component of psychological essentialism and its relationship to mechanisms of motivated social cognition Journal of Personality and Social Psychology 88 686702 2005.
  • Lepore, L. Brown, R. Category and stereotype activation: Is prejudice inevitable? Journal of Personality and Social Psychology 72 275287 1997.
  • Levy, S. R. Dweck, C. S. The impact of children's static vs. dynamic conceptions of people on stereotype formation Child Development 70 11631180 1999.
  • Levy, S. Plaks, J. Hong, Y. Chiu, C. Dweck, C. Static versus dynamic theories and the perception of groups: Different routes to different destinations Personality and Social Psychology Review 5 156168 2001.
  • Levy, S. R. Stroessner, S. J. Dweck, C. S. Stereotype formation and endorsement: The role of implicit theories Journal of Personality and Social Psychology 74 14211436 1998.
  • Livingstone, A., & Haslam, S. A. (2008). The importance of social identity content in a setting of chronic social conflict: Understanding intergroup relations in Northern Ireland. British Journal of Social Psychology, 47, 121.
  • Lowery, B. S. Knowles, E. D. Unzueta, M. M. Framing inequity safely: The motivated denial of White privilege Personality and Social Psychology Bulletin 33 12371250 2007.
  • Lowery, B. S. Unzueta, M. M. Knowles, E. D. Goff, P. A. Concern for the ingroup and opposition to affirmative action Journal of Personality and Social Psychology 90 961974 2006.
  • Martin, C. Parker, S. Folk theories about sex and race differences Personality and Social Psychology Bulletin 21 4557 1995.
  • Morton, T. A. Haslam, S. A. Postmes, T. Ryan, M. K. We value what values us: The appeal of identity-affirming science Political Psychology 27 823838 2006.
  • Morton, T. A., Postmes, T., Haslam, S. A., & Hornsey, M. J. (in press). Theorizing gender in the face of social change: Is there anything essential about essentialism? Journal of Personality and Social Psychology.
  • Pedersen, A. Beven, J. Walker, I. Griffiths, B. Attitudes toward indigenous Australians: The role of empathy and guilt Journal of Community and Applied Social Psychology 14 233249 2004.
  • Plaks, J. E. Levy, S. R. Dweck, C. S. Stroessner, S. J. In the eye of the beholder: Lay theories and the perception of group entitativity, variability, and essence Yzerbyt, V. JuddC. M. CorneilleO. The psychology of group perception: Perceived variability, entitativity, and essentialism 127146 New York Psychology Press 2004.
  • Plaks, J. Stroessner, S. Dweck, C. Sherman, J. Person theories and attention allocation: Preferences for stereotypic vs. counterstereotypic information Journal of Personality and Social Psychology 80 876893 2001.
  • Reicher, S. Cassidy, C. Wolpert, I. Hopkins, N. Levine, M. Saving Bulgaria's jews: An analysis of social identity and the mobilisation of social solidarity European Journal of Social Psychology 36 4972 2005.
  • Reicher, S. Hopkins, N. Self and nation London Sage 2001.
  • Rothbart, M. Taylor, M. Category labels and social reality: Do we view social categories as natural kinds? Semin, G. R. FiedlerK. Language and social cognition 1136 London Sage 1992.
  • Schmitt, M. T. Branscombe, N. R. Kappen, D. M. Attitudes toward group-based inequality: Social dominance or social identity? British Journal of Social Psychology 42 161186 2003.
  • Schor, N. Weed, E. The essential difference Bloomington and Indianapolis Indiana University Press 1994.
  • Stoler, A. L. On political and psychological essentialisms Ethos 25 101106 1997.
  • Turner, J. C. Hogg, M. A. Oakes, P. J. Reicher, S. D. Wetherall, M. Rediscovering the social group: A self-categorization theory Oxford Blackwell 1987.
  • Verkuyten, M. Discourses about ethnic group (de-)essentialism: Oppressive and progressive aspects British Journal of Social Psychology 42 371391 2003.
  • Verkuyten, M. Multicultural recognition and ethnic minority rights: A social identity perspective European Review of Social Psychology 17 148184 2006.
  • Verkuyten, M. Brug, P. Multiculturalism and group status: The role of ethnic identification, group essentialism and protestant ethic European Journal of Social Psychology 34 647661 2004.
  • Weiner, B. Perry, R. P. Magnusson, J. An attributional analysis of reactions to stigmas Journal of Personality and Social Psychology 55 738748 1988.
  • Whitley, B. E. The relationship of heterosexuals' attributions for the causes of homosexuality to attitudes toward lesbians and gay men Personality and Social Psychology Bulletin 16 369377 1990.
  • Yzerbyt, V. Y. Castano, E. Leyens, J.-Ph Paladino, P. The primacy of the ingroup: The interplay of entitativity and identification European Review of Social Psychology 11 257295 2000.
  • Yzerbyt, V. Y. Corneille, O. Estrada, C. The interplay of subjective essentialism and entitativity in the formation of stereotypes Personality and Social Psychology Review 5 141155 2001.
  • Yzerbyt, V. Y. Estrada, C. Corneille, O. Seron, E. Demoulin, S. Subjective essentialism in action: Self-anchoring and social control as consequences of fundamental social divides Yzerbyt, V. Y. JuddC. M. CorneilleO. The psychology of group perception: Perceived variability, entitativity, and essentialism 101126 London Psychology Press 2004.
  • Yzerbyt, V. Y. Judd, C. M. Corneille, O. The psychology of group perception: Perceived variability, entitativity, and essentialism London Psychology Press 2004.
  • Yzerbyt, V. Y. Rogier, A. Fiske, S. Group entitativity and social attribution: On translating situational constraints into stereotypes Personality and Social Psychology Bulletin 24 10901104 1998.
Footnotes
  1. 1 We also manipulated whether the art was traditional or contemporary (i.e. stereotypically Aboriginal or not). However, this variable did not interact with any of the other independent variables and so is not discussed further.

  2. 2 Preliminary checks revealed that there was some association between the manipulation of identity label and levels of racism, F(1, 174)=3.46, p=.07, η2=.02. Although this could pose a problem to the use of racism as an independent variable, this was not considered an issue for two reasons: First, the effect of the manipulation was weak and second, the analyses were conducted via multiple regression that controls for correlations among predictor variables.

  3. 3 Controlling for this variable did not affect the significance of the results reported below. Accordingly, results are reported without the inclusion of this variable.

  4. 4 In Study 1, our manipulation of identity label qualified Aboriginal identity (‘not clearly Aboriginal’) in a way that was not true for white identity (‘white’). To remove this potential confound, we avoided the use of qualifiers in this study and simply referred to the target as either ‘not Asian’ or ‘white’.

  5. 5 By removing one item from this scale, the alpha-level was substantially improved (α=.73). However, improved reliability did not change the pattern of results.

  6. 6 In this study there was no association between the manipulations and the measure of prejudice.

  7. 7 Controlling for this variable did not affect the significance of the results reported below. Accordingly, results are reported without the inclusion of this variable.