Competition over collective victimhood recognition: When perceived lack of recognition for past victimization is associated with negative attitudes towards another victimized group

In the last decades, Western societies have witnessed a growing tendency of minority groups to profile themselves as victims in order to obtain more societal recognition (Moscovici & Pérez, 2009). Members of these minorities have publicly expressed negative attitudes towards other minorities, although the latter were not responsible for their past victimization. For example, Khalid Muhammad, from the Nation of Islam, stated that “The black Holocaust was 100 times worse than the so-called Jew Holocaust” (Muhammad, 1994, cited by Benn Michaels, 2006, p. 290), and “I say you call yourself Goldstein, Silverstein, and Rubinstein because you're stealing all the gold and silver and rubies all over the earth” (Baltimore, 1994, cited by Anti-Defamation League, 2013). Dieudonné, a French humorist of African descent, declared that the recognition devoted to Jews for the Holocaust prevented him from denouncing the victimization of Blacks during slavery and colonialism (2005, February 17). He was recently convicted for anti-Semitism in Belgium (Wauters, 2015). This phenomenon was described and analyzed by sociologists (e.g., Chaumont, 1997), philosophers (Ricoeur, 2007), and philologists (Rothberg, 2009; Todorov, 1996, 1998), who framed it in terms of competition over symbolic recognition. So far, this phenomenon has not been systematically researched by social psychologists.

Social psychological research (e.g., Bar-Tal & Antebi, 1992; Wohl & Branscombe, 2008) has shown that sharing a sense of collective victimhood can negatively impact intergroup relations. Moreover, group members can experience competitive victimhood, defined as “a belief in having suffered more than the out-group” (Noor, Brown, & Prentice, 2008, p. 481), which impedes post-conflict intergroup forgiveness. However, so far, this research has mainly focused on relations between former enemies, or between former victims and their perpetrators. And competitive victimhood has mainly been understood as bearing on the severity of their respective sufferings. The situation described earlier does not fit this description. In this paper, we argue that groups can compete over their respective victimhood even when they are not held responsible for each other's victimization. However, in such situations, the competition bears on the recognition of their victim status, over and above the severity of their respective sufferings. In turn, this competition over collective victimhood recognition can be associated with negative intergroup attitudes. Finally, in order to understand these societal situations, they should be framed as involving at least three entities: the two groups that compete over the recognition of their victimhood, and a third entity—for example, “society,” the government, or the international community—that has the power of granting or denying recognition.

Competitive Victimhood

The sense of collective victimhood results from events affecting individuals because of their membership in a social group, even if all members did not personally experience them (Bar-Tal, Chernyak-Hai, Schori, & Gundar, 2009). Previous studies showed that it can lead to competitive victimhood claims, that is, the assertion by a group that it has been subjected to greater suffering than the adversarial group (Noor, Shnabel, Halabi, & Nadler, 2012). In turn, competitive victimhood is associated with negative intergroup attitudes, contributes to conflicts escalation, and impedes their peaceful resolution (Noor et al., 2012). Indeed, Noor, Brown, and Prentice (2008) showed, in the context of the Northern Irish conflict, that competitive victimhood was associated with a lack of trust and empathy. Noor, Brown, González, Manzi, and Lewis (2008) also found that competitive victimhood was associated with reduced willingness to forgive the out-group. Research conducted in Israel also showed that group members who engaged in competitive victimhood were unwilling to let go of their resentment and to envision reconciliation (Shnabel, Halabi, & Noor, 2013). Finally, research conducted in Central Africa showed that competitive victimhood predicted intolerance for diversity, social distance from out-groups, and mistrust towards them (Vollhardt & Bilali, 2015).

However, these studies focused on the competition between groups formerly or currently in conflict. Yet, over the last decades, members of minority groups have publicly expressed negative attitudes towards other minorities that were not responsible for their past victimization (e.g., the Dieudonné affair or the discourse of the Nation of Islam cited earlier). These situations have been analyzed in other disciplines (Chaumont, 1997; Rothberg, 2009; Todorov, 1998), but, to our knowledge, their social psychological underpinnings were never investigated through systematic empirical research. So far, it seems that only one social psychological study examined a situation in which a competition over the victimhood status emerged between groups that were not directly responsible for their respective sufferings. Bilewicz and Stefaniak (2013) found that Polish participants who felt their group was strongly victimized during WWII, or were victimized more than Jews during this war, displayed more negative attitudes towards Jews. We argue that, in such situations, the competition primarily bears on the societal recognition of sufferings. In turn, this “competition over collective victimhood recognition” can be associated with negative intergroup attitudes.

Need for Recognition and Consequences on Intergroup Relations

According to Barkan (2001), the individuals' need for recognition is insatiable. Staub (2008) even sees it as a “desperate need.” According to Honneth's (1996) struggle for recognition theory, recognition is a fundamental need because it is constitutive of identity: Human beings only fully exist on the condition that they are being recognized. When a lack of recognition is experienced, it can motivate a struggle to obtain it (Bachmann & Simon, 2014; Licata, Sanchez-Mazas, & Green, 2011). Honneth's theory is a general theory of social recognition dynamics. It does not specifically bear on victimhood, and it encompasses both interindividual and intergroup recognition, but it provides a relevant framework for addressing collective victimhood issues. In contemporary Western societies, an important part of the demands for recognition concern past collective victimhood (among Armenians, formerly colonized peoples, African Americans, etc.; Moscovici & Pérez, 2009). The victim status is highly coveted because it tends to empower victimized groups, which are perceived as morally superior, entitled to sympathy, consideration, and protection against criticism (Bar-Tal et al., 2009).

In line with Honneth's (1996) struggle for recognition theory, recognition can be defined as a case in which “A takes B as X,” where A is a person or a group, B is another person or group, and X is the property that A recognizes to B (Ikäheimo & Laitinen, 2007). In Honneth's theory, recognition encompasses the properties of love, respect, and esteem, but it can also be analyzed in terms of status, such as in regard to collective victimhood. The concept of recognition is germane, although distinct, from that of acknowledgement: “whereas only normative entities can be acknowledged, there is a sense of the word ‘recognition’ in which only persons (and possibly collectivities of persons) can be recognized” (Ikäheimo & Laitinen, 2007, p. 36). When acknowledgement of victimhood is at stake, the “normative entity” is a claim of having been victimized. This claim bears on facts and can be acknowledged if it is judged as valid by a third party—that is, A (a person or a group) takes B (a claim of having been victimized) as X (valid). Recognition is about the status of individuals or groups with regard to these victimizing events—that is, A (a person or a group) takes B (another person or group) as X (victimhood status). In other words, the identity of B is directly at stake in recognition dynamics (Licata et al., 2011), whereas it is only indirectly in acknowledgement dynamics. It is worth noting that A (the recognizing party) is not necessarily the party that was responsible for the victimizing event. In this paper, we focus on cases in which two groups compete for the recognition of their victim status by a third party: A is (perceived as) recognizing that an out-group (C) is a victim, whereas it denies this recognition to the in-group (B).

Research on victimhood acknowledgement is nonetheless highly relevant when victimhood recognition is at stake, as acknowledging that harm was done to a social group can be seen as a way of recognizing its victimhood status. To our knowledge, three studies have examined victimhood acknowledgment in intergroup contexts. David and Choi (2009) showed that former political prisoners expressed less desire for revenge when their personal sufferings were acknowledged. Alarcon-Henriquez et al. (2010) showed that intergroup attitudes between linguistic communities in Belgium improved after in-group members read that the out-group considered that in-group sufferings were as important as their own. Vollhardt, Mazur, and Lemahieu (2014) examined effects of acknowledgment in contexts of mass violence on psychological well-being and attitudes towards the former perpetrator. They showed that acknowledgment (vs. lack of acknowledgment) significantly increased victimized group members' well-being as well as their willingness for reconciliation. These studies provide evidence that acknowledgement of collective victimization can positively influence intergroup relations. However, they examined the impact of acknowledgment by groups casted as former perpetrators on intergroup relations with their victims.

A recent study by Simantov-Nachlieli, Shnabel, and Halabi (2015) is, to our knowledge, the only one that explicitly focused on the granting or denying of recognition of the victim status by a third party. They found that Palestinians and Israeli Jews who were informed that their in-group was granted the victim status by a third party (scientists) expressed more forgiving and conciliatory attitudes towards the out-group and held less pessimistic views of the conflict. However, this study also involved groups in conflict. None of the aforementioned studies examined the effects of recognition of victimhood, or lack thereof, on relations between groups that did not harm each other in the past. The present studies aim to fill this gap.

In the present paper, we focus on the competition that might arise between two previously unrelated victimized groups (i.e., that are neither generally portrayed as the perpetrator nor the victim of the out-group in the context of the victimizing events). We argue that, in such situations, the competition bears on the societal recognition of their respective sufferings, which can be granted (or not) by a third party. More precisely, we reason that a sense of collective victimhood can be associated with negative attitudes towards an out-group that was not involved in the victimizing events because this sense of collective victimhood is associated with the feeling that the in-group lacks social recognition of its victimhood status, and this feeling is itself associated with the belief that this lack of victimhood recognition is due to the greater recognition that is being granted to the victimhood status of the out-group. It is worth noting that a feeling of lack of recognition can be experienced without intergroup comparison, whereas the attribution of this lack of recognition to that granted to the out-group is inherently comparative. According to this explanation, this feeling of lack of recognition is rooted in perceptions of the relationships between the in-group and the recognizing party: A does not sufficiently grant X (recognition of victimhood) to B (in-group). This feeling then motivates in-group members to draw comparisons with C (another victimized group). The belief that the greater attention devoted to the out-group's victimhood impedes in-group recognition—A does not recognize B because it recognizes C—is then associated with negative attitudes towards this out-group. It is worth noting that this reasoning is valid to the extent that victimhood recognition is perceived as a limited resource; that is, that the recognizing party has a limited “amount” of recognition to distribute between the in-group and the out-group, as if recognition dynamics obeyed the rules of a zero-sum game (Różycka-Tran, Boski, & Wojciszke, 2015).

This model, based on social recognition concerns, should contribute to explaining intergroup attitudes independently of competitive victimhood over the severity of sufferings. Indeed, competitive victimhood over the severity of sufferings mainly involves relations between two—formerly or presently—conflicting groups (Groups B and C) that compete over a different X (severity of sufferings). Previous research showed that this phenomenon may have significant negative influence on intergroup relations. However, it should have less influence on intergroup attitudes when relations between groups that were not involved in each other's victimizing event, rather than between former enemies, are concerned. In the latter case, group members are often motivated to establish their victimhood status (and therefore to cast the out-group in the perpetrator role), which can be achieved by comparing facts (group sufferings). In the former case, group members are motivated to be recognized as a victimized group, which can be achieved by comparing the relative recognition that is being granted to the in-group and to the out-group by a third party. However, these two phenomena are not mutually exclusive.

We present three studies. The two first studies involved two different minority groups: Sub-Saharan Africans and Muslims living in Belgium. Jews were chosen as the out-group because their victimhood in the Holocaust is generally perceived as benefitting from some recognition in Western societies (Chaumont, 1997; Rothberg, 2009), and because they have often been targeted when claims of lack of recognition were expressed by members of other minorities. The third study experimentally tested the causal link between the (objective) lack of victimhood recognition and negative attitudes in a fictive situation of past victimization involving groups that had no known history of victimhood (psychology students and law students).

Study 1

Although it is the outcome of a long process, currently, the victimhood of Jews during the Holocaust tends to receive considerable official recognition throughout Europe. As stated by Lagrou (2010, p. 282):

If one had to identify one common commemorative initiative that has benefitted from a genuine commitment and investment by European governments at all levels, it would probably be the ‘Holocaust Remembrance Day’ celebrated each year on 27^th January, the anniversary of the liberation of Auschwitz.

In Belgium, the History of the Holocaust is taught in secondary schools, commemorated on the 27th of January, and through a museum and memorial, the Jewish Museum of Deportation and Resistance at the Dossin casern. In 2012, the Prime Minister publicly apologized for the help provided by the Belgian State in the deportation of Jews.

In contrast, until recently, the Royal Museum of Central Africa conveyed a rather indulgent representation of Belgian colonialism in the Congo, Rwanda, and Burundi (Vellut, 2005), and no public apologies were ever uttered for the atrocities committed against Congolese people during King Leopold II's rule (Licata & Klein, 2010). After decades of “amnesia,” more critical representations of Belgian colonialism started to be taught in secondary history education from the end of the 1990s, although not systematically (Van Nieuwenhuyse, 2015).

The Sub-Saharan immigration to Belgium began more than 50 years ago with the arrival, then settling, of students from the former colonies (Swyngedouw & Swyngedouw, 2009). This immigration then diversified in terms of countries of origin, legal status, and migratory trajectories (Grégoire & Mazzocchetti, 2013). Beyond this diversity, the African collective imaginaire is replete with memories of victimhood:

an age-old, extraordinary history is reduced, as if by waving a magic wand, to three tragic acts, ghostly experiences and phobic objects par excellence: slavery, colonization, and apartheid—to which an attempt is now being made to add globalization (Mbembé, 2001, p. 3).

Some young Sub-Saharan African immigrants have denounced the lack of recognition of their group's sufferings in comparison with other groups: “Apartheid and the genocide in Germany have been recognized, we want a recognition too” (John, a Congolese interviewee, cited in Grégoire & Mazzocchetti, 2013, p. 106). For example, the demonstration “White King, Red Rubber, Black Dead” took place in Brussels in 2013 to claim for acknowledgment of the “Congolese genocides.” Joseph Salomon Mbeka, president of “Change in Congo,” stated that “Congolese suffered twice from mass crimes within a century, without any action or recognition from the international community” (Change in Congo, 2013). However, to our knowledge, the Jewish community was never publicly targeted by African activists in Belgium.

Nevertheless, based on the reasoning described earlier, we hypothesized that Sub-Saharan Africans' sense of collective victimhood would be associated with anti-Semitic attitudes (H1). Then, that this link would be mediated by their sense of lack of victimhood recognition, itself associated with the belief that the attention devoted to Jews' victimhood prevents the recognition of their own (competition over victimhood recognition, H2). Finally, we predicted that competitive victimhood over the severity of the in-group's sufferings would not mediate this link over and above these two mediators (H3). Indeed, it should have less influence given that Africans and Jews were not involved in each other's victimizing event.

Results

Preliminary results

Means, SDs, and correlations are presented in Table 1. It is worth noting that attitudes towards Jews were significantly higher (M = 61.22, SD = 22.04) than the midpoint of the scale, t(126) = 5.74, p < .001. Gender and political orientation were not significantly correlated with any of the variables of interest and will thus not be described in the following analyses. Identification with Africa was significantly correlated with collective victimhood, competitive victimhood, and secondary anti-Semitism, but not with lack of recognition and competition over recognition.

Table 1. Descriptive measures and correlations (Studies 1 and 2)

		M	SD	1	2	3	4	5	6	7	8	9	10	11	12
Note: SD, standard deviation. *** p < .001; ** p < .01; * p ≤ .05.
1. Collective victimhood	Study 1	5.86	0.80	1
	Study 2	5.48	1.32	1
2. Lack of recognition	Study 1	5.62	1.13	.37***	1
	Study 2	5.48	1.09	.35***	1
3. Competition over recognition	Study 1	4.81	1.97	.43***	.44***	1
	Study 2	5.42	1.71	.36***	.34***	1
4. Competitive victimhood	Study 1	4.27	1.76	.38***	.40***	.33***	1
	Study 2	3.93	1.79	.43***	.24**	.41***	1
5. Secondary anti-Semitism	Study 1	4.46	1.47	.23**	.30***	.46***	.28***	1
	Study 2	5.33	1.03	.26**	.47***	.52***	.34***	1
6. Primary anti-Semitism	Study 1	—	—	—	—	—	—	—	1
	Study 2	5.14	1.17	.47***	.32***	.50***	.53***	.35***	1
7. Thermometer: Jews	Study 1	61.22	22.04	−.16	.04	−.27**	−.03	−.24**	—	1
	Study 2	65.98	25.99	−.19*	−.04	−.21*	−.46***	−.18*	−.42***	1
8. Thermometer: Belgians	Study 1	66.46	20.93	−.22*	−.05	−.20*	−.03	.04	—	.55***	1
	Study 2	77.60	23.15	−.13	−.06	−.11	−.30***	−.11	−.21*	.76***	1
9. Thermometer: Africans	Study 1	—	—	—	—	—	—	—	—	—	—	1
	Study 2	79.05	19.72	.10	.07	.06	−.08	.01	−.11	.59***	.61***	1
10. Thermometer: Maghrebians	Study 1	60.45	23.17	−.04	−.04	−.10	.08	.01	—	.60***	.58***	—	1
	Study 2	—	—	—	—	—	—	—	—	—	—	—	1
11. Thermometer: Israelis	Study 1	—	—	—	—	—	—	—	—	—	—	—	—	1
	Study 2	37.44	30.87	−.17	−.14	−.26**	−.22*	−.27**	−.44***	.51***	.34***	.24**	—	1
12. Identification	Study 1	6.02	1.29	.26**	.10	.16	.26**	.18*	—	−.03	−.08	—	.07	—	1
	Study 2	6.59	.76	.30***	.24**	.05	.12	.13	.17	−.07	−.13	.12	—	−.07	1

As expected (H1), collective victimhood was positively correlated with secondary anti-Semitism, although not with attitudes towards Jews (feeling thermometer). It also correlated negatively with attitudes towards Belgians. Secondary anti-Semitism was also positively associated with lack of recognition, competitive victimhood, and competition over recognition. Further, as expected (H2), lack of recognition was correlated with competition over recognition, competitive victimhood, and secondary anti-Semitism. Finally, competitive victimhood was correlated with competition over recognition. Attitudes towards Maghrebians were not associated with any of these variables.

Serial mediation analysis

Using the PROCESS macro developed by Hayes (2012, Model 6; Figure 1 and Table 2), we then tested a serial mediation model to test Hypotheses 2 and 3: Collective victimhood was the independent variable; lack of recognition was the first mediator; and competition over recognition was the second mediator. Competitive victimhood was entered as the third mediator. Secondary anti-Semitism was the dependent variable. Identification with Africa was entered as a covariate in order to control for its effects. This analysis was performed using bias-corrected 95% confidence intervals (CIs) with 1000 bootstrap samples. The global indirect effect proved significant, ab = 0.42, standard error (SE) = 0.11, CI [0.25, 0.66]. As expected (H2), the second indirect effect (collective victimhood to lack of recognition to competition over recognition to secondary anti-Semitism) was significant. The fifth indirect effect was also significant (collective victimhood to competition over recognition to secondary anti-Semitism). There was no significant difference between these two paths, b = −0.12, SE = 0.09, CI [−0.32, 0.06]. Moreover, as expected (H3), indirect effects including competitive victimhood were not significant. Thus, competition over the severity of in-group sufferings had no more effect on secondary anti-Semitism when lack of recognition and competition over recognition were entered in the model. Competitive victimhood also did not play a significant independent role when it was placed as a second mediator (before competition over recognition). The significant total effect of collective victimhood on secondary anti-Semitism dropped below the significance level when mediators (lack of recognition, competition over recognition, and competitive victimhood) were entered. The total effect also dropped below the significance level, b = −0.02, SE = 0.17, CI [−0.36, 0.32] when only lack of recognition and competition over recognition were included as mediators. Finally, the covariate (identification with Africa) had a significant positive effect only on competitive victimhood.

Table 2. Results for the mediation model explaining the effect of SCV on secondary anti-Semitism (Study 1)

	b	SE	95% CI
aNote: Multiple regression on secondary anti-Semitism (predictors: IV, mediators, and covariate), R2 = .26; effect size (f2), 0.33; power, 0.99 (without competitive victimhood, R2 = .24; effect size (f2), 0.31; power, 0.99). Results in bold type are significant. SE, standard error; CI, confidence interval; IV, independent variable; DV, dependent variable.
Total effect (R2 = .07)
Collective victimhood to secondary anti-Semitism (d1)	0.36	0.17	[0.03, 0.69]
Direct effect
Collective victimhood to secondary anti-Semitism (d′1)	−0.06	0.17	[−0.40, 0.28]
Path from IV to mediators
Collective victimhood to lack of recognition (a1)	0.53	0.12	[0.29, 0.77]
Collective victimhood to competitive over recognition (a2)	0.72	0.21	[0.30, 1.14]
Collective victimhood to competitive victimhood (a3)	0.43	0.20	[0.04, 0.83]
Path from mediator to mediators
Lack of recognition to competition over recognition (b1)	0.57	0.14	[0.28, 0.85]
Lack of recognition to competitive victimhood (b2)	0.42	0.14	[0.14, 0.69]
Competition over recognition to competitive victimhood (b3)	0.09	0.08	[−0.07, 0.26]
Path from mediators to DV
Lack of recognition to secondary anti-Semitism (c1)	0.11	0.12	[−0.13, 0.35]
Competition over recognition to secondary anti-Semitism (c2)	0.29	0.07	[0.15, 0.43]
competitive victimhood to secondary anti-Semitism (c3)	0.09	0.07	[−0.06, 0.24]
Indirect effects
Total: Collective victimhood to secondary anti-Semitism	0.42	0.11	[0.25, 0.66]
Indirect 1: Collective victimhood (SCV) to Lack of recognition to secondary anti-Semitism (SAS)	0.06	0.07	[−0.07, 0.23]
Indirect 2: SCV to Lack of recognition to competition over recognition to SAS	0.09	0.04	[0.03, 0.18]
Indirect 3: SCV to Lack of recognition to competitive victimhood (CV) to SAS	0.02	0.02	[−0.01, 0.08]
Indirect 4: SCV to Lack of recognition to competition over recognition to CV to SAS	0.00	0.01	[−0.002, 0.02]
Indirect 5: SCV to competition over recognition to SAS	0.21	0.08	[0.06, 0.39]
Indirect 6: SCV to competition over recognition to CV to SAS	0.01	0.01	[−0.004, 0.06]
Indirect 7: SCV to CV to SAS	0.04	0.05	[−0.02, 0.19]
Covariate effects
Identification to lack of recognition	0.003	0.08	[−0.15, 0.15]
Identification to competition over recognition	0.08	0.12	[−0.16, 0.32]
Identification to competitive victimhood	0.22	0.11	[0.00, 0.44]
Identification to secondary anti-Semitism	0.11	0.09	[−0.08, 0.29]

The statistical power was very high (0.99), and the effect size of a multiple regression model including the independent variable, the mediators, and the covariate was consequent (R² = .26). It decreased only very slightly without the competitive victimhood variable (R² = .24). 2

Study 2

In order to address the limitations of Study 1, we conducted a second cross-sectional study. Study 2 first aimed at replicating the serial mediation model validated in Study 1 among members of another minority group. The second aim was to test it on a distinct dependent variable, less directly related with recognition of victimhood concerns: primary anti-Semitism.

We conducted Study 2 among Muslims living in Belgium. The choice of this population allowed us to address two limitations of Study 1. First, even though this should not be generalized, anti-Semitic attitudes and behaviors have been reported among members of this population, such as insults in schools, harassment, aggression, or the depredation of Jewish religious sites (Jikeli, 2015), and research revealed that a higher level of anti-Semitic attitudes was observed among Muslims than in other non-Jewish religious groups (Koopmans, 2015). Second, Belgium did not colonize any Muslim country, so that it should not be held responsible for the historical victimization of Muslims, or at least not as directly as in Study 1.

According to Modood (2003, p. 101), “Muslims do have the most extensive and developed discourses of unity, common circumstance and common victimhood among non-EU origin peoples in the EU.” Muslim victimhood discourses mainly focus on recent Western military interventions in Muslim countries (Iraq, Afghanistan, Somalia, Bosnia, or Kosovo), and on the Israeli–Palestinian conflict. In addition, radical Muslim discourses describe Muslim history as a continuous line of humiliation by the West, from the seventh century onward (Lara, 2003). Importantly, there is also a widespread feeling of religion-based discrimination among Muslim immigrants in Europe, especially since the 9/11 attacks (Jikeli, 2015).

Within Belgian society, Jews and Muslims can be viewed as two minority groups that are not mutually responsible for each other's victimization. However, at the international level, Jews and Muslims have a long history of intergroup relations, paved with instances of conflict. The ongoing Israeli–Palestinian conflict is very salient in both Jews' and Muslims' consciousness. Some Muslims may attribute the responsibility of their group's victimization to Jews, based on their perception of the role played by Israeli Jews in this conflict. This might lead to a more important role of competitive victimhood over the severity of sufferings in this study than in Study 1.

We hypothesized that Belgian Muslims' sense of collective victimhood would be associated with secondary anti-Semitism (H1a), as well as with primary anti-Semitism (H1b). Then, we postulated that these links would be mediated by lack of recognition, itself associated with competition over victimhood recognition (H2a and H2b). This causal chain should explain secondary and primary anti-Semitism independently of competitive victimhood over the respective severity of the groups' sufferings (H3a and H3b).

Study 3

The goal of Study 3 was to experimentally test the hypothesis that receiving less recognition than another group for a victimizing event can induce negative intergroup attitudes, even when this group is responsible neither for the harm done nor for denying recognition. We thus manipulated the perceived recognition of collective victimhood after a past instance of victimization of in-group and out-group members by members of a third party. We also wanted to test this relationship among groups with no previous history of collective victimization in order to avoid interference from collective memories of past intergroup conflicts. And we wanted to create a situation in which the recognizing party is clearly distinct from the perpetrator group. Thus, we designed a fictional situation of victimization involving three groups of students and manipulated the objective recognition of victimhood by a third party (the university authorities).

We hypothesized that members of a victimized group who perceive that their group's victimhood recognition was denied whereas it was granted to another, equally victimized, group will express negative attitudes towards this out-group. This should not happen in situations in which only the in-group received recognition, or in which the two groups' victimhood was equally recognized or denied.

General Discussion

Through this set of three studies, we examined the link between a sense of collective victimhood and negative attitudes towards an out-group that was not involved in the victimizing event. Conflicts over victimhood had already been investigated by social psychologists, who have described their negative effects on intergroup relations (Noor, Brown, González, et al., 2008; Noor, Brown, & Prentice, 2008; Shnabel et al., 2013). These studies investigated groups that were previously involved in the same conflict, either as enemies or as victims and perpetrators. And this competitive victimhood bore on the relative severity of the sufferings undergone by members of these two groups. However, some societal situations, in which group members have expressed negative attitudes towards an out-group in relation to their respective victimhood, do not fit this pattern. Some minority groups' claims for recognition of the victim status have been associated with the derogation of an out-group, allegedly because this group benefitted from more societal recognition of its historical victimhood than the in-group. Scholars from other disciplines have framed these societal phenomena in terms of competition over social recognition (Chaumont, 1997; Todorov, 1998). Further, Moscovici and Pérez (2009) have argued that the struggle for the victim status' recognition is an important, if not the main, strategy of minority influence in contemporary democratic societies. Inspired by these analyses, we reasoned that, in circumstances in which two social groups have not been formerly involved in each other's historical episode of collective victimization, group members can develop negative attitudes towards the out-group to the extent that their sense of collective victimhood is associated with a sense of lack of recognition, and that this lack of recognition tends to be attributed to the greater recognition that is being granted to this out-group's victimhood. In such situations, competitive victimhood over the severity of groups' sufferings should be less relevant than what we named “Competition over collective victimhood recognition.”

Studies 1 and 2 investigated these processes among members of two minority groups (Sub-Saharan African immigrants and Muslims) focusing on their attitudes towards another minority group (Jews). In these two studies, the expected association between sense of collective victimhood and negative attitudes—secondary anti-Semitism in both studies and primary anti-Semitism only in Study 2—towards an out-group that was not involved in the historical victimization of the in-group was obtained. Further, these studies showed that this association was explained through a path involving a sense of lack of societal recognition for in-group victimhood, associated with the attribution of this lack of in-group recognition to out-group recognition. Competitive victimhood (over the severity of groups' sufferings) was positively associated with all the variables of interest in both studies and with both primary and secondary anti-Semitism. However, in Study 1, and in Study 2 when secondary anti-Semitism was measured, it did not contribute to mediate this link over and above these two variables bearing on recognition. Yet, in Study 2, when primary anti-Semitism was measured among Muslim participants, competitive victimhood proved to be a better mediator than the “recognition” causal path. However, this latter path, as well as other paths involving recognition variables, still significantly and independently mediated the effect. This suggests that the competition bore on the societal recognition of in-group victimhood rather than on the severity of the suffering itself. Moreover, these effects were obtained while controlling for the effect of in-group identification, and only for the out-group perceived as benefitting from more victimhood recognition.

In Study 3, we experimentally tested the causal link between the objective lack of recognition and negative attitudes towards another victimized group in a fictional situation of past victimization. It involved members of real in-group and out-group that had no known history of collective victimhood (reciprocal or else): students in psychology and in law. We manipulated the level of recognition granted, by the university authorities, to the two groups independently of their actual victimization by a third group (business students), which was described as equally severe. In line with our hypothesis, attitudes towards the other victimized group were significantly more negative in the condition in which the out-group had been granted recognition, whereas the in-group had not, compared with the other conditions. This is clear evidence that perceiving inequality in the distribution of recognition for collective victimhood can trigger intergroup animosity. Interestingly, attitudes towards the perpetrator group did not follow the same pattern, which ascertains that this manipulation specifically affected attitudes towards the other victim group. In addition, the recognizing party (university authorities) was judged more negatively when it failed to grant victimhood recognition, either to the in-group or to the out-group.

As acknowledged earlier, Study 3 is distinct from Studies 1 and 2 because it does not imply the same level of emotional involvement. However, the three situations had common features that were relevant for testing our hypotheses. Indeed, the three situations involved three parties: a victimized in-group, a victimized out-group, and a recognizing party; and they focused on the recognition of collective victimhood rather than on the experience of victimization. In Studies 1 and 2, our prediction that concerns over the recognition of collective victimhood should, at least partly, explain attitudes towards another victimized group was tested by including measures that clearly addressed the present-day recognition of collective victimhood by the host society, rather than the experience of victimization itself. In Study 3, victimhood recognition by a third party was manipulated independently of the experience of victimization, which was kept stable across all conditions.

Overall, these three studies provide evidence that concerns over the societal recognition of collective victimhood can be associated with intergroup animosity. In some societal contexts, in which formerly victimized groups that have no common history of intergroup conflict live in the same society and do not compete for other reasons, competition over collective victimhood recognition can be the main, or even the sole, source of intergroup tension. In other settings, it can be one factor among others, or play no role, such as in dyadic post-conflict situations.

These situations often involve cultural minorities, be they indigenous or immigrant. As Craig and Richeson (2016) demonstrated, relations between minority groups can lead to coalitions, when a superordinate identity as minorities or as discriminated groups is salient (see also Hindriks, Verkuyten, & Coenders, 2014). Indeed, groups can create superordinate identities based on an inclusive interpretation of their victimhood (Vollhardt, 2012). Yet they can also evolve towards derogation. Competition over collective victimhood recognition is a potential cause of mutual derogation between minority groups. However, the minority status is not a necessary condition: members of majority groups can also experience lack of recognition for historical victimhood, bolstering out-group derogation. As a case in point, Bilewicz and Stefaniak (2013) showed that Polish majority members' sense of collective victimhood for WWII was associated with competitive victimhood, then negative attitudes, towards Jews. This phenomenon could be observed in any group inasmuch as its members perceive it as having been victimized, when this perception is associated with a feeling of lack of recognition, and when this lack of recognition is attributed to an out-group's recognition. Similarly, under these conditions, any out-group could be the target of animosity inspired by competition over collective victimhood recognition, as Study 3 suggests. These studies were not designed to provide an extensive explanation of anti-Semitism, which is multifactorial (Bilewicz, Winiewski, Kofta, & Wojcik, 2013). Nevertheless, Studies 1 and 2 suggest that concerns over collective victimhood recognition can account for a significant part of the variance of anti-Semitic attitudes, especially among the Muslim sample.

Another prerequisite is ideological: Competition over collective victimhood recognition should appear only in societal contexts in which victimhood is a desirable status. It is the case in most contemporary Western democratic societies (Barkan, 2001), but not in other cultural and/or political contexts. Cross-cultural research is needed for testing this moderation hypothesis.

Social recognition is usually described as a symbolic resource (Fraser & Honneth, 2003). Unlike material resources, symbolic resources are usually considered as unlimited (Tyler, Smith, Gilbert, Fiske, & Lindzey, 1998). Yet recognition concerns are not always independent from material motivations, as claims for victimhood recognition are often accompanied by claims for reparation (Barkan, 2001). In Study 3, admittedly, recognition was not purely symbolic since recognized students were given the possibility to re-take the exam. Interestingly, in the present studies, participants tended to treat “recognition” as if it was a limited resource; they acted as if they believed that, if one group is being recognized, then the other one cannot, as if recognition was a zero-sum game (Różycka-Tran et al., 2015). Further research is needed to ascertain that this tendency persists when the prospect of material benefits is unambiguously absent or duly controlled for.

In this paper, we reasoned that experiencing a sense of lack of recognition can motivate in-group members to search for explanations. Drawing intergroup comparisons, they attribute their lack of recognition to the greater recognition granted to an out-group. Accordingly, we measured lack of recognition without mentioning this more recognized out-group. However, this does not exclude the possibility that perceiving unequal distribution of victimhood recognition could induce—or increase—a sense of lack of recognition. This alternative path was less relevant in the present studies owing to the sequential ordering of measures (the out-group was mentioned only after lack of recognition), but it could be envisioned in future research.

Future research could also examine other relevant mediators. In this paper, the out-group was not described as being actively responsible for the in-group's lack of recognition. A step further in that reasoning would be accusing this out-group of being intentionally responsible for this lack of recognition. For example, Dieudonné accused the “Zionists” of lobbying against the recognition of Africans' victimhood in France (Paris, March 2013). This belief should, obviously, further fuel intergroup animosity.

Importantly, these societal situations should be framed as triangular rather than dyadic. In his re-analysis of Sherif's Robbers cave experiment (Sherif & Harvey, 1961), Billig (1976) proposed that the intergroup rivalry observed in this famous study could be best understood by including the experimenters in the picture. After all, they had created the social conditions that ignited the conflict, and then those which led to reconciliation. Similarly, dynamics of recognition always involve groups that struggle for recognition, and an entity which has the power to grant, or to deny, this recognition (Honneth, 1996). This points to the important responsibility of authorities in managing social recognition of collective victimhood among groups. When this recognition is inequitably distributed across groups that perceive themselves as victims, the potential for intergroup conflict is high, and the authorities are at risk of losing their legitimacy.

Acknowledgements

We would like to thank Zoraïda Drese for collecting data for Study 2.

This research was conducted within the framework of COST Action IS1205 “Social psychological dynamics of historical representations in the enlarged European Union”. The authors declare that there are no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Appendix Appendix