The social psychology of intergroup tolerance and intolerance

ABSTRACT

Living with cultural, religious, and ideological diversity inevitably creates situations where people are confronted with beliefs, values, or practices that they disapprove of. In such situations, people may consider reasons to nevertheless tolerate what they disapprove of (i.e., forbearance tolerance), or reject these (i.e., intolerance). In the current review, we consider empirical research on the processes of intergroup tolerance and consider different manifestations of (in)tolerance. Then we discuss recent work on the dual process model of tolerance, which considers intuitive versus deliberative objections to cultural, religious, and ideological differences, and how these can be overridden to produce tolerance. We then review literature on the boundaries of tolerance including the role of moralization and moral values, before considering individual difference factors and situations of threat that can hamper intergroup tolerance. We conclude by considering five areas ripe for future exploration on the social psychology of intergroup tolerance and intolerance.

KEYWORDS:

• Tolerance
• Intolerance
• Intergroup Relations
• Diversity

People live in an increasingly diverse world with substantial group differences in their convictions, beliefs, and practices. Such differences can be difficult to accept when they conflict with one’s own convictions and way of life because each person tends to believe in the rightness and sensibilities of their own values, beliefs, and practices. Thus, not all beliefs, values, and practices can be simultaneously affirmed and confirmed because of their “propositional content that implies a distinction between true and false, right and wrong, beautiful and ugly” (Joppke, 2004, p. 242).

However, maintaining plural societies depends on people’s willingness to allow others to live the life that they want. As a key approach to achieving this, intergroup tolerance in its classical sense implies that people put up with outgroup beliefs and lifestyles that they disapprove of because these are meaningfully different and dissenting, and sometimes even antithetical and incompatible with ingroup convictions and worldviews (Cohen, 2004; King, 2012). Importantly, tolerance is not the opposite of prejudice (Gibson, 2006; Verkuyten et al., 2020a), and it is not the same as valuing diversity since people tolerate what they disapprove of or object to. Tolerance reconciles critical judgment with the protection and permitting of dissenting outgroup beliefs and practices. This makes tolerance a key ingredient for a successful diverse, equal, and open society: tolerance is “indispensable for any decent society – or at least for societies encompassing deeply divergent ways of life” (Oberdiek, 2001, p. 23; see also Furedi, 2011; Popper, 1945; Walzer, 1997).

In recent years, social psychologists have shown a growing interest in understanding and examining intergroup tolerance (e.g., Simon et al., 2019; Verkuyten & Yogeeswaran, 2017), and our own program of research is based on the “forbearance model of intergroup tolerance” (see Figure 1; Verkuyten et al., 2020a, Verkuyten et al., 2022a). In this review article, we discuss this model and the corresponding research that examines when and why people are tolerant, and how people define the boundaries of tolerance. In elaborating on this model, we will first propose that tolerance is a distinctive orientation that combines disapproval about outgroup beliefs and practices, with a behavioural intention to nevertheless accept – in the sense of not interfering with – these beliefs and practices. Subsequently, the model proposes that tolerance involves the weighing of reasons for disapproval with reasons for acceptance (first moderation pathway in Figure 1). Additionally, the model considers where people draw the boundaries of tolerance by prioritizing reasons to reject disapproved of beliefs and behaviours over reasons to accept these beliefs and behaviours (second moderation pathway in Figure 1). Finally, we discuss the role of individual differences in tolerance and situational conditions that trigger processes of threat.

Figure 1. The forbearance model of tolerance; adapted from Verkuyten et al. (2020a), open access.

Our research, primarily conducted in the Netherlands and Germany, predominantly focuses on perceived religious differences in relation to Muslim minorities. In relatively secular European countries, religion is often viewed unfavourably and Muslim minority beliefs, practices, and behaviours¹ tend to receive broad public disapproval (see Verkuyten, 2021) making debates around religion especially relevant for cases of intergroup toleration. Our empirical work mostly involves large national majority group samples from both countries and utilizes survey questions (7-point scales) and well powered survey-embedded experiments. These between-subjects experiments have the advantage of combining representative samples and random assignment making it possible to draw more ecologically valid and generalizable conclusions, compared to laboratory experiments (Sniderman, 2018; see also Schlueter & Schmidt, 2010).

In several of our studies, we have used four items that tap into the personal predisposition to endorse tolerance in general (e.g., “I accept it when other people do things that I wholeheartedly disapprove of”, “Even people that I detest have the right to live their own life”). This general individual difference measure is reliable and valid, and empirically distinct from the endorsement of cultural relativist beliefs (“One can only judge the norms and values of another culture from that culture itself”) and indifference (“I do not care if people want to live differently than me”). However, similar to the difference between the endorsement of principles such as freedom and equality and their practical policy implications (Dixon et al., 2017), general support of tolerance does not simply translate into actual tolerance. For the evaluation of concrete situations and cases, not only is a person’s predisposition relevant, but also various other considerations and concerns. There is as always a person by context interaction and the endorsement of general principles is something else than the evaluation of specific cases, and this is also true for tolerance. It is one thing to agree with the general notion that people should have the freedom to live the life that they want, but another to tolerate, for example, animal slaughter, abortion, apostasy, anti-vaccination, or religious symbols and clothing in public schools. In this article, we focus on these sorts of concrete issues that put diversity to the test and highlight the need for toleration.

Forbearance tolerance

The concept of tolerance goes back to classical antiquity in ancient India, and later also developed in 17^th century Europe, as an approach for managing religious conflicts and disputes about beliefs, convictions, and practices. Thus, tolerance pertains to specific outgroup beliefs, practices, and behaviours, rather than to categorical distinctions based, for example, on demographic characteristics such as race or gender. Tolerance does not address intergroup biases that result from categorical distinctions between the ingroup (“us”) and outgroup (“them”), or forms of racism, sexism or group-based prejudices more generally.² A categorical distinction between ingroup (“us”) and outgroup (“them”) does not make the outgroup, by itself, a subject for tolerance. Rather, tolerance signifies an approach towards an outgroup “insofar as it pertains to beliefs and opinions and forms of behaviour linked to them” (Furedi, 2011, p. 11). This is critically important for a culturally diverse and peaceful society, but also means that the concept of tolerance should not be inappropriately extended to categories of people as people. Tolerance is appropriate for specific outgroup beliefs, views, and practices, but not for addressing antipathy and hatred toward social categories or groups of people per se (Forst, 2013; Verkuyten et al., 2020a). A racist or a misogynist who does not act upon their racist or sexist feelings and beliefs does not tend to demonstrate tolerance, and a more racist person showing forbearance is not a more tolerant person. The possibility of being tolerant does not depend on being a bigot, and bigotry should not be addressed by stimulating tolerance, but rather by anti-racism and prejudice reduction strategies (Verkuyten et al., 2020a). Groups that are bound together by religion, belief, or ideology can be tolerated, but in those cases, tolerance refers to the beliefs and practices of members of those groups, or what the group stands for. Furthermore, people can also tolerate (or not) dissenting practices of ingroup members (e.g., burning of the national flag) or of their friends (e.g., smoking) and family members (e.g., gambling).

Our social psychological “Forbearance Model of Tolerance” is based on three aspects that collectively form the distinct and crucial concept of tolerance, and each relates to different psychological processes (Figure 1). First, tolerance involves objection or disapproval of specific outgroup beliefs and practices implying it is judgmental; “one cannot tolerate ideas of which one approves” (Gibson, 2006, p. 22). Tolerance is not neutrality, or an unwillingness to judge. Rather, tolerance involves not interfering with others’ beliefs or behaviors that are evaluated negatively, such as when outgroup members engage in cultural practices or voice ideological views that are unappealing, unsettling, or uninviting. The disapproval can be rather intuitive, for example, drawn from moral emotions, or it may be drawn from reflective thinking about the meaning and implications of unfamiliar or opposing values, beliefs, or practices (Verkuyten et al., 2022a).

Second, tolerance implies having value-based reasons for accepting what one is negative about. The acceptance in the sense of non-interference requires self-constraint that is not coerced or based on fear, social disapproval, possible sanctions, or concerns over self-image (Cohen, 2004; King, 2012). This makes tolerance different from, for example, expressing non-prejudicial views for reasons of conforming to ingroup norms and secure ingroup acceptance (Crandall & Eshleman, 2003), or from the external motivation not to respond with prejudice or the suppression of prejudice for reputational reasons (Plant & Devine, 1998). People are tolerant when they have internally motivated and value-based reasons (e.g., our concepts of equality or freedom) to refrain from negatively interfering with disapproved of outgroup beliefs and behaviours (moderating path “1” in Figure 1). In other words, there is a process of moderation whereby the intuitive or more reflective disapproval are overridden by deliberative thinking about value-based reasons to not interfere (Verkuyten et al., 2022a).

Third, tolerance differs from indifference (“who cares”) and relativism (“anything goes”) since there are boundaries to what can and should be accepted (Cohen, 2004; Forst, 2013). There are always beliefs or practices that are considered intolerable, for example, because these are considered morally unacceptable or as undermining cultural continuity and social cohesion. Children as young as 5 years old recognize nonrelativistic criteria for evaluating beliefs that reflect disagreements (Wainryb et al., 1998) and people continue to view their own moral convictions as absolute and universalistic (Skitka et al., 2021). Thus, there can be additional reasons not to tolerate a certain behaviour (e.g., harm, oppression, violence) that take priority over the reasons for accepting what one disapproves of (e.g., religious freedom), making the particular belief, practice or behaviour intolerable (moderation path “2” in Figure 1).

Differentiating tolerance from prejudice and discrimination

Intergroup tolerance implies disapproval of specific beliefs, practices, and behaviours and the disapproval should not simply reflect group-based prejudice. Hence, a first important step is to empirically show that disapproval of outgroup beliefs and behaviours can differ from prejudicial attitudes towards categories of people as people. Individuals with more negative attitudes toward outgroups are more likely to oppose particular outgroup beliefs and practices. For example, majority members with stronger anti-Muslim sentiments tend to strongly oppose the building of Mosques and minarets, Islamic schools, and the wearing of headscarves (e.g., Helbling, 2014; Saroglou et al., 2009; Van der Noll, 2014; Van der Noll & Saroglou, 2015). Thus, opposition to and rejection of these specific practices can reflect a general negativity toward Muslims and be used to justify anti-Muslim prejudices.

However, not all disapproval of specific outgroup beliefs and practices reflects outgroup prejudice, as it can be based on other considerations. For example, one can oppose the building of Islamic schools because they believe religion of any kind has no place in schooling. Similarly, beliefs about sexual morality and sanctity of life can underlie opposition towards abortion (Rodriguez & Ditto, 2019), and concern for procedural justice can be a genuine determinant of opposition to affirmative action (Bobocel et al., 1998). Thus, an objection to a specific outgroup practice or belief may represent an expression of one’s prejudicial attitude toward the group per se, an expression of disapproval of that particular practice or belief, or a combination of these two. For example, majority members can reject the wearing of a headscarf by Muslim public servants because they dislike Muslims, or because they believe that religion has no place in public institutions, or some combination of the two. Therefore, a first step in examining tolerance is to study the nature of the objections and to determine whether they meet the non-prejudicial criteria necessary to be considered components of forbearance tolerance of specific outgroup beliefs and practices. In our research, we have used four related strategies to examine the different possibilities that might drive objections: (1) statistically controlling for prejudice, (2) considering multiple minority practices, (3) examining double standards, and (4) using a multiple-actor-multiple-acts design.

Controlling for prejudice

One strategy for assessing the role of group-based prejudice in rejection of specific outgroup practices is to measure the endorsement of general beliefs and values and statistically control for prejudicial attitudes. For example, enlightenment values of personal autonomy and rationality (e.g., Gustavsson et al., 2016; Imhoff & Recker, 2012), secularism (Dangubic et al., 2021), and self-direction and universalism (Elchardus & Spruyt, 2014; Saroglou et al., 2009) have been found to predict disapproval of specific Muslim minority practices and beliefs, over and above the statistical effect of generalized prejudice, both explicit and implicit.

Furthermore, people can reject a specific practice (e.g., ritual slaughter of animals) of a group (Jews, Muslims) to whom they have neutral or even positive feelings (Hurwitz & Mondak, 2002; Sniderman et al., 1989). People who do not have prejudicial outgroup feelings can still be critical and disapproving of specific outgroup beliefs and practices. For example, Helbling (2014) found in six European countries that people with secular liberal values felt positively toward Muslims as a group, but also conflicted regarding the legislation of religious practices such as the wearing of the headscarf. In addition, among national samples in the UK, France, Germany, and the Netherlands, a substantial portion of people with a positive attitude toward Muslims supported a ban on headscarves (Van der Noll, 2010; see also Saroglou et al., 2009), and also rejected Islamic education and the building of Mosques (Van der Noll, 2014).

Furthermore, Sniderman and Hagendoorn (2007) employed a covert measure of prejudice (list technique³), and found that disapproval of some Muslim practices can be based on principled liberal concerns, rather than hidden dislike of Muslims as a group of people. Similarly, in a nationally representative sample of German and Dutch majority group members (N = 3,712), we used a person-centered approach for examining subgroups within the population (Dangubic et al., 2022a). A person-centered approach (Latent Profile Analysis; e.g., Osborne & Sibley, 2017) makes it possible to examine, for example, whether feelings toward Muslims as a group of people, support of Muslims equal expressive rights, and the perception of Muslim unequal gender values and practices are psychologically combined in different ways by different subgroups of individuals.

Our findings show that in addition to a subgroup of individuals that had anti-Muslim attitudes in every respect (19%), there was a substantial subgroup of people in both countries (around 33%) that felt positively towards Muslims as a group of people and endorsed Muslim expressive rights, but that at the same time were also critical about Muslim gendered practices.⁴ Further, on a covert measure of prejudice (list technique), this critical subgroup did not demonstrate any hidden prejudices (0%), but on another covert measure, 36% of this subgroup felt negatively about Muslims wearing veils in public. These findings support the assumption that these individuals are genuinely critical of gendered practices without being negative towards Muslims as a group of people and not respecting their equal expressive rights. This interpretation was further supported by the finding that this subgroup of individuals scored relatively higher on measures of unconditional respect, open-minded thinking, and the endorsement of civil liberties.

Considering multiple minority practices

In addition to statistically controlling for outgroup prejudice when examining disapproval of specific outgroup beliefs and practices, it is possible to examine majority members’ attitudes towards a range of minority group practices. This allows one to distinguish between those people who consistently accept or reject various practices, and people who do not consistently accept or reject the different practices. Cross-practice consistency can indicate a general negativity (or a general positivity) toward Muslim minorities and inconsistencies can indicate more principled considerations or social normative concerns about specific practices (Dangubic et al., 2021; Sniderman et al., 1989). Specifically, some people may display negative feelings toward Muslims as a group of people and disapprove of a range of practices (“anti-Muslim”), while others have positive feelings toward Muslims and consistently accept all practices (“pro-Muslims”). Additionally, people can accept some minority practices, but not others and this practice-based inconsistency suggests that people not only consider the type of the group they are asked to tolerate, but also the nature of the dissenting practices (e.g., Gibson & Gouws, 2003; Petersen et al., 2010). The consideration of multiple practices might demonstrate that people have objections to a particular practice (e.g., wearing of a headscarf) but not toward another practice (e.g., Islamic primary schools). A particular practice might raise specific moral concerns, such as the wearing of a headscarf which might evoke concerns about gender equality. However, the rejection of some Muslim practices might also result from social norms that make it socially acceptable to express one’s prejudiced feelings by rejecting these practices and not others (Crandall & Eshleman, 2003). Thus, some individuals may be practicing a form of principled disapproval that emerges from specific values and beliefs held about specific practices, while other individuals may be practicing a form of negativity that emerges from a prejudice toward the group per se that they may be unwilling to express.

Using data from five representative samples of the Dutch majority (N = 3,253), we again used a person-centered approach to empirically identify different subgroups of individuals that differ in their subjective constellations of general feelings toward Muslims as a group of people and disapproval of various Muslim practices mentioned earlier (Adelman & Verkuyten, 2020; Dangubic et al., 2021). The latent profile analyses of the different samples indicate that four main subgroups of majority members exist: those who either have consistent positive attitudes towards the group and the range of practices (around 30% of the national samples) or rather a profile of negative attitudes (around 20%), those who display more principled-based intolerance (around 20%; neutral to positive attitudes toward Muslims and selective disapproval of some practices) or prejudiced intolerance (around 30%; neutral to positive attitudes toward the group and cross-practice negativity) (see Figure 2 for an example). Furthermore, these subgroups of majority members differ in predictable ways on social dominance orientation (support of hierarchical group relations), authoritarianism (endorsement of normative conformity), status quo conservatism, cognitive sophistication, perceived outgroup threat, the tendency to engage in anti-discrimination activities, and prejudicial feelings towards other minority outgroups and the disapproval of non-Muslim minority practices. For example, individuals in the negative subgroup were not only more authoritarian and conservative, but also most negative towards refugees and Polish immigrants and disapproved more strongly of homosexual men kissing in public and gender-neutral practices than the other subgroups (Dangubic et al., 2021). Furthermore, individuals in the principled subgroup were relatively more likely to engage in cognitive sophisticated thought, to have lower social dominance orientation, and were more inclined to be involved in anti-discrimination activities (Adelman & Verkuyten, 2020).

Figure 2. Latent Profile Analysis results: Dutch majority members evaluations of two Muslim minority groups (of Turkish and Moroccan origin) and acceptance of five Muslim minority practices (Muslims enacting their own faith in public life; Muslim women wearing a headscarf anywhere in the Netherlands; Muslims in the Netherlands being able to celebrate their Islamic holidays at home, but also in public; Muslims building mosques in the Netherlands; Muslims establish Islamic schools). Taken from Adelman & Verkuyten (2020), open access.

Using multiple actors to identify double standards

The possibility that some people might disapprove of a specific practice of an outgroup rather than be prejudiced toward the outgroup per se can also be taken to imply that one does not apply a double standard and disapproves of the same practice regardless of who engages in it. This type of research moves beyond measuring prejudice to the assessment of discrimination of groups that are equally warranted in claiming their rights and benefits (Sniderman, 2018). For example, in a between-subjects experimental study, we found that Dutch adolescents were less tolerant toward Muslims exercising their right to establish their own schools and to demonstrate and protest, compared to non-religious groups exercising these rights (Verkuyten & Slooter, 2007). Further, the disapproval of particular religious practices (clothing, symbols) can either be motivated by more general anti-religious sentiments or rather discriminatory attitudes toward Muslims as a group (Mondak & Huruwitz, 1998; van der Noll & Saroglou, 2015). If people, for example, are equally opposed to Muslims and Christians engaging in the same act (actor consistent; no double standard), it is more likely that they have general reasons for doing so, or hold religion more generally in contempt. And if people apply a double standard and reject a particular act only for Muslims (actor inconsistent), it is more likely that prejudicial feelings towards Muslims are involved. For example, in a study in Quebec (Bilodeau et al., 2018), a distinction was made between individuals who favoured a ban of all religious symbols and those who favoured religious minority restrictions. While the former group of individuals were motivated by secular and liberal values, the latter were motivated by prejudice and feelings of cultural threat.

In experimental research, we compared majority members’ disapproval of specific practices of two target groups (Sleijpen et al., 2020; see also Verkuyten & Slooter, 2007). In a first study (N = 361), the practices were either performed by immigrant groups of Turkish Muslims or Turkish Christians, and we found that people were equally (in)tolerant of the various practices (e.g., wearing of religious symbols; organizing religious lessons in a community center) for both. However, people might reject Turks as an immigrant group, independently of whether they are Muslim or Christian (Spruyt & Elchardus, 2012; Strabac & Listhaug, 2008). Therefore, in experiment 2 (N = 381), we examined tolerance towards Muslims and native Orthodox Protestants as two numerically small religious communities in the Netherlands (both representing around 5% of the Dutch population) that both struggle to hold on to their religious values and beliefs in the increasingly secular context of the Netherlands (Fetzer & Soper, 2003; Ribberink et al., 2017). Data revealed that the disapproval of Muslim minority practices had less to do with Muslims per se, but rather with the perceived normative deviance of the practices, independently of the religious minority group (see Figure 3 for the findings of experiment 1).

Figure 3. Mean tolerance scores (with 95% error bars) for two target groups and four practices that differ in perceived normative deviance. “religious necklace”, allow a front desk clerk to visibly wear a necklace with a Christian cross/Islamic crescent moon; ‘religious lessons, organise religious lessons in a community centre during which Christian/Islamic values will be taught; “quiet room”, request to set up a quiet room at work; “anti-abortion speech” a Christian minister/Muslim imam wanting to give a speech at an educational college in which abortion is equated with murder. Taken from Sleijpen et al. (2020), open access.

In another experimental study (Havermans & Verkuyten, 2021), we asked Dutch majority members (N = 832) to indicate their positive (tolerate, learn from) and negative (aggress, avoid) behavioural intentions towards either immigrants or co-ethnics (“autochthonous Dutch”), “who compared to you have very different beliefs, norms and values and thus live very differently”. The findings showed no double standard, but rather a similar degree of toleration toward both target groups (M = 4.38, SD = 1.27, and M = 4.33, SD = 1.09, respectively; also similar scores for the other behavioural intentions). Furthermore, the possible moderating role of authoritarianism⁵ was considered and there was no difference in tolerance of the two target groups between respondents low or high in authoritarianism. One interpretation of these findings is that belief incongruency is unsettling and that people have a desire for maintaining their cultural worldview (M. Brandt & Crawford, 2020). As a result, they tend to react similarly towards both immigrants and co-ethnics whose values and norms conflict with their own norms and values, independently of individual differences in, for example, authoritarianism, openness to experience, conscientiousness, and religious fundamental beliefs (M.J. Brandt & Van Tongeren, 2017).

Using multiple actors and multiple acts

A fourth strategy is to combine the previous two approaches and simultaneously consider various practices and multiple outgroups simultaneously. People might reject both Christian and Muslim civil servants wearing religious symbols, but accept Christian primary schools while rejecting Islamic schools. Or, they might reject both Muslims and Christians across both practices.

Simultaneously considering multiple acts and actors makes a more detailed and nuanced understanding of people’s disapproval possible. Specifically, it allows us to determine the proportions of people as well as which individuals show equal disapproval or rather discriminatory disapproval by using a double standard in disapproving of Muslims, but not, for example, Christians or Jews practicing their religion (Hurwitz & Mondak, 2002). We used this strategy in two studies using national samples from different European countries.

In a first study, we analyzed data of national samples of majority members in five Western European countries (N = 1,580) who were asked four questions about the practices of Christians and Muslims (i.e., two target groups) focusing on two separate practices (wearing religious symbols and religious education in public schools; Dangubic et al., 2020). With these two actors and two acts, there are sixteen logically possible combinations of acceptance and rejection and nine of these combinations are presented in Table 1 (the remaining seven involve positive discrimination of Muslims which was very exceptional). The Table shows that there is empirical support for the existence of equal disapproval of two practices next to discriminatory disapproval of Muslims. Furthermore, logistic regression analyses shows that prejudice towards Muslims was associated with a higher likelihood of displaying discriminatory than equal disapproval, and that those endorsing secularism were more likely to show equal disapproval compared to discriminatory disapproval.

Table 1. Combinations of multiple acts (religious symbols and education) and multiple actors (Muslims and Christians) approach (without positive discrimination of Muslims). percentages of majority members from five West European countries.

Combinations	Symbol		Education
	Christian	Muslim	Christian	Muslim
C1.	Accept	Accept	Accept	Accept	Equal acceptance: 16.6%
C2.	Reject	Reject	Accept	Accept	Partial equal rejection: 28.9%
C3.	Accept	Accept	Reject	Reject
C4.	Reject	Reject	Reject	Reject	Equal rejection: 15.3%
C5.	Reject	Reject	Accept	Reject	Discriminatory rejection: 39.2%
C6.	Accept	Reject	Reject	Reject
C7.	Accept	Accept	Accept	Reject
C8.	Accept	Reject	Accept	Accept
C9.	Accept	Reject	Accept	Reject

Note: Table adapted from Dangubic et al. (2021). Open access

In a second study using national samples from Germany and the Netherlands (N = 3,703), majority members were asked about four different practices and three religious groups (Muslims, Christians and Jews; Dangubic et al., 2022b). With latent profile analysis, we found one subgroup that used a double standard and more strongly rejected Muslims than Christians and Jews engaging in the same practices (discriminatory rejection; 16%). The other four subgroups responded to the practices in different ways, but independently of the religious group and thus did not demonstrate double standards in their rejection or acceptance of the three groups’ practices (equal acceptance = 18.6%; equally moderate = 35%; equal rejection = 17.3%;, and equal rejection of specific practices, but not other practices = 13.1%). Thus, a substantial proportion of the population in both countries disapproved of specific practices for all three religious groups (rejections combined; 30.3%) and endorsement of secularism was found to be an important reason for this.

These findings and the different types of research that we discussed indicate that disapproval of specific outgroup beliefs and behaviours does not have to reflect outgroup prejudice. Varied beliefs and practices can be evaluated differently based on diverse considerations and concerns which makes the topic of forbearance tolerance relevant.

Practicing tolerance

With forbearance tolerance, there is a combination of more principled disapproval of particular outgroup beliefs and practices, while nonetheless holding a positive judgment of and behavioural intention toward accepting – in the sense of not interfering – these disapproved-of practices and beliefs. This inconsistency between the negative attitude (disapproval) and positive intention (acceptance) makes it a psychologically distinct orientation (Verkuyten & Yogeeswaran, 2017). It requires the recognition of relevant reasons for condoning other cultural, religious, and ideological beliefs and different modes of conduct to which one objects. In such cases, one’s spontaneous or more reflective disapproval of outgroup beliefs and practices (e.g., building of mosques) is overpowered by considerations to nonetheless allow those beliefs and practices in society (e.g., religious freedom), which is indicated in Figure 1 with the first moderation path.

Research has shown that people may be tolerant of beliefs and practices they disapprove of because they endorse liberal and democratic values (Gibson, 2006; Sullivan & Transue, 1999). For example, valuing civil liberties has been found to be associated with higher tolerance of headscarves in Germany (Van der Noll, 2014), and those who more strongly value individual freedom tend to be more accepting of Muslim religious symbols (Dangubic et al., 2020; Saroglou et al., 2009). However, beyond the importance of holding specific values such as civil liberties to the tolerance process, we have examined two additional factors that influence tolerance. First, drawing on the work of Forst (2013), we examined the roles of equality-based respect versus pragmatic considerations of coexistence for the tolerance process. Second, we examined the importance of recognizing and accepting relevant reasons for people’s disapproved of messages and behaviours, as part of determining what to tolerate and what not to tolerate.

Respect and coexistence

People can have different reasons for being tolerant of minority practices, beliefs, and way of life (Hjerm et al., 2020; Klein & Zick, 2013). Two main considerations for tolerance are, first, a respect tolerance based on respecting the equal standing and rights of others, and, second, a coexistence tolerance based on valuing and maintaining social harmony and peace. With the former, tolerance is based on the principled belief that minority members are equal, autonomous fellow citizens with the same dignity, rights and civil liberties. By contrast, the latter implies a more pragmatic acceptance of minority practices in order to avoid conflicts and to find and maintain peaceful cohabitation (Kirchner et al., 2011). These two considerations have been put forward in historical, philosophical and social scientific analyses of tolerance, with the latter preceding the former historically (Forst, 2013; Furedi, 2011). These reasons also correspond to the distinction between pragmatic and principled ways in which people reason about societal issues (e.g., Colombo, 2021) and the distinction between instrumentality-based (for the functioning of society) and respect-based cultural diversity beliefs (Kauff et al., 2020).

First, research indicates that respect can function as a basis for tolerance. Respect for outgroup members as equals has been found to foster tolerance in survey, longitudinal and experimental research among different groups and in different countries by Simon and colleagues (e.g., Simon et al., 2019; Zitzmann et al., 2021). Similarly, a stronger tendency to unconditionally respect others as human beings with intrinsic worth has been found to be associated with stronger positive action tendencies and weaker negative action tendencies toward ethnic and religious (Muslim) minority outgroups (Lalljee et al., 2009). The respect of others as equals is grounded in a common higher level group membership (fellow citizens, fellow humans) and does not preclude the possibilities to disapprove what they believe and how they live their lives. Tolerating other citizens’ divergent beliefs and practices (e.g., “respecting their equal civic right to hold particular beliefs and act according to it”) is not the same as considering these beliefs and practices as equally valid (e.g., “valuing what they believe and practice”). Tolerance involves an act of judgment, but this does not serve to censor or negatively interfere with “wrong belief” because tolerance demands respect for the equal right of others to hold and express their own beliefs.

Our research also shows that respect based on equal citizenship (Velthuis et al., 2021) and respect of others in virtue of being human (Dangubic et al., 2022a) are associated with higher tolerance of Muslim minority practices. In one survey study among ethnic Dutch adolescents (N = 926), we examined the endorsement of the equality of cultural groups in relation to tolerance of four specific Muslim minority practices (e.g., wearing of a headscarf, Islamic public schools) together with the feelings toward the respective Muslim actors (Thijs et al., 2021). A structural equation model revealed that there was a significant direct effect (β = .38, p < .001) of cultural equality on tolerance. The endorsement of the equal status of cultural minority groups was associated with higher tolerance of the different practices, independently of the Muslim outgroup feelings.

Second, in addition to tolerance from an equality-based principled respect (Simon, 2020), there can be coexistence-based pragmatic reasons for tolerance (Kirchner et al., 2011). For example, experimental research has found that an emphasis on the positive societal implications of minority practices can increase acceptance (Van Londen et al., 2010) Additionally, experimental research has found that thinking about possible negative societal implications of minority practices can reduce tolerance (Chong, 1994; Kuklinski et al., 1991).

In three studies involving four national samples of Dutch majority members (N = 2,491), we used confirmatory factor analyses to show that respect and coexistence, although positively associated (r ~ .56), are two empirically distinguishable reasons for being tolerant toward different immigrant target groups (Velthuis et al., 2021). The findings of all three studies further showed that respect was negatively related with prejudice towards immigrants (−.40 > r < −.50), and positively associated with tolerance of specific minority practices (r = .55). Endorsement of coexistence-based reasons was also correlated with lower prejudice (−.23 > r < −.25) and with higher tolerance (r = .37), but was not an independent predictor of tolerance when respect was also included in a multiple regression model. The more instrumental nature of peaceful coexistence might make it less morally imperative than respect-based tolerance, and therefore more contextual in nature, resulting in a weaker independent relationship with acceptance of concrete minority practices. This suggests that pragmatic reasons for tolerance could more strongly shift back and forth depending on what the situational context and pragmatic motives are.

Deliberation and recognizing reasons for disapproved behaviours

Discussions on forbearance tolerance focus on the need for reasoned objections (rather than group-based prejudice), and on tolerance being dependent on deliberative thinking about the objectionable behaviours that one disapproves of. Considered thought can lead to increased tolerance (Sniderman et al., 1989) by overriding feelings of disapproval (Pennycook et al., 2015). Dual process theories propose that these feelings can involve negative gut reactions of indignation, disgust, and anger toward outgroup conduct, or through deliberate thinking that results in negative reactions. Negative gut feelings can be triggered by repugnant acts, or cultural values, beliefs, and practices that are seen as morally misguided and wrong without necessarily reflecting a prejudicial attitude toward the outgroup per se. The deliberative process involved in tolerance does not change these negative gut feelings, but makes one refrain from acting upon them and negatively interfere with the freedom and rights of others (first moderation path Figure 1). In terms of dual process theories, tolerance can be based on detecting the need to override intuitive negative responses with deliberative processes (Verkuyten et al., 2022a). It requires self-restraint in which the first negative impulse is overridden by reflective thinking, and refraining from negative interference is considered morally right, even if intuitively it does not feel that way.

By contrast, one can also have negative reactions to outgroup beliefs or practices based on deliberate considerations of reasons to reject such beliefs, practices, or worldviews. In such cases, tolerance is based on the balancing of reasons to reject against reasons to nevertheless allow that which one objects to. Specifically, tolerance can also result from the wish to consider dilemmatic and novel situations extensively (Baron et al., 2015). Reasoning not only plays a post-hoc role in moral justification (Haidt, 2001), but also in deliberative thinking about real-world moral issues that people debate (e.g., gay marriage, apostasy, the building of minarets, offensive speech, or wet markets). There does not have to be immediate gut reactions on whether these practices are right or wrong, which makes reflective reasoning about values and principles critical (Cushman et al., 2006; Pizarro & Bloom, 2003). The implication is that in thinking about unfamiliar and dissenting outgroup practices and beliefs, contrasting values and norms that one subscribes to may be involved which requires balanced reasoning or weighing in whether to tolerate or not tolerate particular practices and beliefs.

We tested the role of dual process theories in tolerance across multiple studies. Using an experimental design to manipulate intuitive versus deliberative thinking (Capraro et al., 2019; Levine et al., 2018), we examined the prediction that deliberative thinking leads to higher tolerance of minority group practices. For example, in one study (N = 816; Verkuyten et al., 2022b), we focused on political tolerance of immigrant-origin individuals with dual citizenship (accepting that they, e.g., “hold public demonstrations”, “give public speeches”, and “found their own political organizations”). A growing number of nations permit dual citizenship, but continued worries about divided loyalties and political trustworthiness of dual citizens frequently boil up, leading to forms of political intolerance. As expected, the findings indicated that participants in the deliberative condition were more tolerant (M = 3.91, SD = 1.85) than participants in the intuitive (M = 3.57, SD = 1.75) and the control (M = 3.44, SD = 1.67) conditions. Further, the means of tolerance in the control condition did not differ significantly from those in the intuitive condition which is in line with the notion that in online surveys participants tend to respond to Likert-type questions based on their gut reactions and without much reflection (Krosnick, 1999). These findings were robust across demographic factors such as gender, age, religiosity, educational level, and also across ideological factors such as political orientation, and authoritarianism.

In other research (Verkuyten et al., 2022c), in a first study (N = 728) we focused on tolerance of Muslim minority practices and of gender debated issues (e.g. installing gender-neutral toilets) and again found that asking people to think carefully about the issue made them more tolerant (M = 3.67, SD = 1.70, and M = 3.15, SD = 1.39, respectively))compared to a condition in which they were asked to respond intuitively (M = 3.32, SD = 1.71, and M = 2.80, SD = 1.45, respectively) and a control condition (M = 3.24, SD = 1.51, and M = 2.79, SD = 1.34, respectively). Additionally, in two further studies (Ns = 419 and 628), we found that people took relatively more time in answering the questions in the deliberative condition and that deliberative thinking increased tolerance. This higher tolerance occurred regardless of whether people were asked to deliberate over more pragmatic or more principled reasons for acceptance (see Figure 4). Furthermore, the findings were similar across a range of minority practices and again robust across participants’ gender, age, educational level and political orientation.

Figure 4. Mean tolerance levels toward minority gender and Muslim practices across four experimental conditions. standard error bars included. taken from Verkuyten et al. (2022c). open access.

In three further studies using national samples of Dutch majority group members, we investigated the theoretical assumption that recognizing relevant arguments for supporting something people disapprove of increases tolerance of a range of perceived controversial practices (Verkuyten et al., 2021). In a first study (N = 851), we examined the central prediction that self-reported reasons for accepting negatively evaluated Muslim minority practices increases tolerance of these practices. Study 2 (N = 876) examined the same prediction with a native religious minority target group of orthodox Protestants, and in Study 3 (N = 659), we used an experimental design and focused on the perceived reasons of others rather than self-reported reasons for further testing the central theoretical prediction.

The findings of Studies 1–2 demonstrated that when participants recognized arguments (e.g., religious freedom, equal rights) to support minority practices of Muslims or Orthodox Protestants they themselves disapproved of, they showed increased tolerance toward such practices in society compared to those who did not recognize relevant arguments (Study 1, M = 3.23, SD = 1.23; vs. M = 2.61, SD = 1.18; Study 2, M = 3.59, SD = 1.65; vs. M = 2.03, SD = 1.1). In Study 3, the importance of considerations was experimentally extended by demonstrating that perceiving objectionable behaviour as being based on reasonable (vs unreasonable) considerations significantly increased tolerance (M = 4.57, SD = 1.34; vs. M = 4.32, SD = 1.39).

Collectively, these different studies demonstrate the importance of engaging in and perceiving reasonable considerations to enhance tolerance of disapproved of beliefs and practices.

Boundaries of tolerance

Beginning with Stouffer’s (1955) landmark study on people’s support for the expression of nonconformist views, empirical research has typically shown practice-related variance in tolerance. Tolerance as forbearance involves a process in which the decision whether or not to tolerate specific practices depends on the nature of each outgroup practice and the related moral values that are considered to be at stake (e.g., Chanley, 1994; Gibson & Gouws 2001). Research has found, for example, that the intercorrelations of tolerance of different Muslim minority practices are relatively weak (r ~ 0.25: Adelman et al., 2021b; Van der Noll, 2014; Verkuyten & Slooter, 2007), which indicates that people’s responses depend on the type of Muslim practice one is asked to evaluate.

The practice-related variance demonstrates that tolerance is not only a matter of equality-based respect for outgroup members or pragmatic considerations related to peaceful coexistence, but that the nature of the outgroup beliefs and practices matter. As discussed earlier, people can have reasons for being tolerant of specific outgroup practices, but not of other practices. Different outgroup practices can raise distinct moral and normative concerns. For instance, wearing a burqa might evoke the issue of gender equality for some, and religious education in public schools can evoke concerns about the secular nature of the state for others (Moss et al., 2019), leading to intolerance of both.

Perceived normative deviance and the related concern about ingroup cultural continuity can underlie prejudicial feelings and can be used to justify outgroup prejudice (e.g., Smeekes & Verkuyten, 2015). However, there can also be genuine concerns about illiberal practices (e.g., arranged marriage, sexual mores, rejection of apostasy) that challenge the liberal nature of society. Invoking liberal values as a normative standard in a liberal culture does not have to be prejudicial. For example, it would be paradoxical to expect a commitment to gender equality, while also accepting unequal gender arrangements among some minority groups (Dangubic et al., 2022a). Experimental research in the UK (Helbling & Traunmüller, 2020) demonstrates that people’s intolerance towards practices and beliefs of Muslim minorities are more driven by concerns about strict forms of religiosity that are seen as incompatible with western liberal norms and values, than about negative attitudes towards Muslims as a group.

Thus, some minority group practices can be perceived as undermining the continuity of the liberal culture of society. Tolerance is not without its boundaries and it appears to be more difficult to tolerate certain minority practices when the continuity of the national identity is considered to be at stake, including for majority members who endorse the general notion of equality-based respect (Velthuis et al., 2020).

Moralization

This heightened sensitivity to what is tolerable is also reflected in the process of moralization, and the development of specific moral values as possible boundary conditions for tolerance. Moralization refers to the process by which moral significance is attached to specific actions and behaviours, and preferences are converted into values (Rhee et al., 2019; Rozin, 1999). Because people tend to consider matters of morality as absolute and beyond compromise (Skitka et al., 2021), more strongly moralized outgroup activities and behaviours are likely to lead to greater avoidance and intolerance. In general, people are less accepting of divergent beliefs and behaviours that are viewed as moral issues, especially when these beliefs and behaviour and seen as violations of objective moral truths (e.g., Cole Wright et al., 2008; Skitka et al., 2021).

We have examined the role of moralization for intolerance in different ways across various studies. In one study, we first presented a sample of Dutch adolescents (N = 180) with four different Muslim minority practices that were expected to differ in the extent to which these elicit moral concerns (Gieling et al., 2010). Using both relative and absolute criteria, it was found that an imam’s speech against homosexuality was considered the most morally wrong, followed by the refusal to shake hands with someone of the opposite sex, the founding of Islamic schools, and then the wearing of a headscarf which was considered more a personal choice rather than a moral issue. Subsequently, in a further study, a large sample of adolescents (N = 970) were asked whether the responsible authorities (municipality, school) should tolerate or not tolerate the specific practice (4-point scales). The focus on authorities was deliberate because tolerance involves situations in which one is able to interfere (Cohen, 2004). The findings show that tolerance was highest for the wearing of the headscarf (M = 2.33, SD = .98) which had been considered a personal rather than moral issue, and lowest for the imam’s speech (M = 1.99, SD = .85) which had been moralized the most. In between these two was the tolerance of the two other practices that did not differ from each other (M = 2.16, SD = .94, and M = 2.20, SD = .95, respectively).

We also found this moralization-intolerance link in another study among a sample of ethnic Dutch adults (N = 404; Study 1; Adelman et al., 2021a). Muslim minority practices that were considered more controversial (e.g., “the refusal of some Muslims to shake hands with people of the opposite gender”, and “arranged marriages for Muslim women”) were moralized more strongly and tolerated less, and increased moralization mediated the statistical effect of normative dissent on intolerance. These findings emerged when controlling for prejudicial attitudes toward Muslims, indicating that the differences in moralization and intolerance occurred independently of prejudice.

Furthermore, the importance of moralization carried over across both active and passive forms of tolerance. Tolerance can not only take a more minimal (“passive”) form of not interfering with the disapproved of conduct of others, but can also take a more maximal (“active”) form in which one actively supports and defends the rights of others to make their case and practice their beliefs. These two forms of tolerance are potentially quite different because psychologically, passivity is less demanding and often less risky and blameworthy than activity (Keinan & Bereby-Meyer, 2017; Kordes-de Vaal, 1996), although less capable of protecting minority practices from the intolerance of others. We examined among Dutch majority members (N = 815) the role of moralization for both active and passive tolerance of a range of Muslim minority practices (e.g., building a mosque, Islamic schools) that differ in their required social accommodation and in whether they have a direct impact on one’s personal life (Study 1) or on society as a whole (Study 2) (see Adelman et al., 2021b).

We found that while a substantial minority of people engaged in passive tolerance toward various disapproved of outgroup practices (between 19.1–33.5%, depending on the practice), very few people were willing to engage in active tolerance (between 0.5–5.9%) by pro-actively supporting the disapproved of practices. For the participants who had a neutral or positive attitude toward the practice, a similar pattern of preference for passive rather than active support emerged which suggests that this difference is at least partially driven by a general preference for inaction over action. Importantly, however, moralization had a similar effect on both forms of tolerance and in relation to the various practices (B_s = −.09 to −.23, p_s = .004–.001). People who had stronger moral concerns about particular Muslim minority practices were less willing to engage in either active or passive tolerance for most of the different practices and across the personal and societal context.

In a further experimental study, we used the multiple-act-multiple-actors approach described earlier (Hirsch et al., 2019). In addition to different religious practices that were found in a post-hoc test on a national sample (N = 218) to differ in how strongly they were moralized, these practices were performed by either a culturally dissimilar (Muslims) or a more similar (orthodox Protestant) minority group. Data (N = 1,225) revealed that ethnic Dutch participants expressed discriminatory intolerance towards Muslims compared to orthodox Protestants, but not when the practices elicited high degrees of moral concern. This again indicates that intolerance of specific practices is not just a question of outgroup prejudice, but can involve moral disapproval of specific practices independently of the religious minority group engaging in it (Cole Wright et al., 2008). Thus, intolerance is more likely if outgroup (or ingroup) practices raise moral concerns, for example, by being perceived as causing harm to others (e.g., gender inequality, hate speech) or as mistreating or threatening others’ freedoms and civic rights (e.g., of homosexuals, apostates).

Moral values

Moralization can lead to intolerance, but moralized acts can also be viewed either as unacceptable or acceptable depending on the particular moral value that is at stake. Values tend to be inherently comparative and competitive (e.g., Tetlock, 1986), and the trade-off among multiple values that are simultaneously relevant affects cognition and behaviour. For tolerance and intolerance, not only does the degree of moralization matter, but also the specifics of the moralization process: the prioritizing of contrasting moral values that are used to evaluate minority beliefs and behaviours (Nelson et al., 1997; Peffley et al., 2001). For example, for some majority members, tolerance of minority beliefs and practices may follow from the freedoms that characterize a liberal society. For others, the rejection of these beliefs and practices might stem from concerns about communitarian values of societal cohesion. Thus, tolerance of minority practices can be weighed, on the one hand, by the value placed on individual liberties and religious freedom that are supported by overwhelming proportions of citizens in western democracies (Wike & Simmons, 2015), and, on the other hand, by whether such practices undermine communitarian values of maintaining social cohesion and a unified society, which are also endorsed by most citizens (Silver, 2018). These two values can conflict with each other making it important to consider the relative trade-off that people make between them for understanding their tolerance (i.e., freedom trumps cohesion) or intolerance (i.e., cohesion trumps freedom) of different minority practices (see 2^nd moderation pathway in Figure 1). For example, the founding of Islamic primary schools can either be understood as primarily a matter of religious freedom, which could lead to greater tolerance, or primarily as a matter of group segregation that undermines social cohesion, which could lead to greater intolerance.

Whether a particular value guides one’s actual judgement is not only dependent on the relative importance that one attaches to it, but also on the situation that makes competing values relatively less or more salient and relevant (Fazio, 1986; Feather, 1990). Most events and situations trigger multiple concerns and require a psychological balancing of reasons to accept or to reject them. For example, people may strongly endorse freedom of expression, but in order for this value to influence their judgment about a particular situation, it should be considered more important than, for example, the value of social peace. Nelson and colleagues (Nelson et al., 1997) found that when news regarding political actions of right-wing extremist groups was framed in terms of the importance of freedom of speech, participants had higher levels of tolerance for this group compared to a situation in which the importance of public order was emphasized (see also Zilli Ramirez & Verkuyten, 2011).

Assigning higher priority to individual freedom over social cohesion should make tolerance more likely, while attaching higher priority to cohesion over freedom should make tolerance of Muslim minority practices less likely. How this trade-off plays out is likely to depend on the extent to which a specific minority belief or behaviour is considered normatively dissenting. We examined the trade-off between the values of individual freedom and social cohesion, and how the prioritization of these two values is associated with tolerance of Muslim minority behaviours that differ in their degree of perceived normative dissent. Specifically, we tested the expectation that people will be less tolerant of more controversial practices (e.g., not shaking hands with someone of the opposite gender) when these are seen as a threat to the value of social cohesion over individual freedoms, but that people will be more tolerant of less controversial practices (e.g., the wearing of Islamic dress in closed meetings) when they consider the value of individual freedom more important than social cohesion. In two large scale studies in the Netherlands and Germany (N = 2,811; Adelman et al., 2021a), we found in both countries that intolerance was higher for those Muslim minority practices that are considered normatively dissenting, and therefore, perceived more negatively. More importantly, this increase in intolerance was related to changes in the relative importance of specific moral values and principles that formed the basis for people’s judgments. Specifically, as the practices increased in their perceived normative dissent, participants increasingly emphasized the importance of social cohesion over freedom (see, for example, Table 2, Study 2). Mediation analyses showed that this change in prioritization, in turn, increased intolerance towards those practices, all above and beyond the effects of prejudicial attitudes toward Muslims. In contrast, less controversial practices were tolerated more because the value of freedom was considered more important than social cohesion (see Table 2).

Table 2. Means and standard deviations for the manipulation check, moral trade-off, and intolerance across the three sets of non-normativity behaviours.

	Manipulation Check (attitudes)	Freedom over Cohesion	Intolerance
Normative dissent low	4.22a (1.00)	.33a (.84)	2.16a (.88)
Normative dissent: medium	3.35b (1.16)	.16b (.91)	2.81b (1.08)
Normative dissent high	2.22c (1.04)	−.32c (1.02)	3.72c (1.04)

Note. Different subscripts within each column indicate significant differences. The low normative dissent condition used practices that past research indicates are perceived as not very controversial, escalating upward to high normative dissent. Table from Adelman et al. (2021a), open access.

This research indicates that increasing perceived normative dissent is associated with higher intolerance of Muslim minority practices through moral prioritization. In subsequent work (Study 4 in Adelman et al., 2021a), we used an experimental design (N = 435) to examine the causal pathway between moralization and tolerance by testing whether moralizing a given action through framing it in terms of a specific moral value affects tolerance judgments. Specifically, we focused on one of the high level dissenting practices that raised relatively strong concerns about societal cohesion (not shaking hands with someone of the opposite gender) and examined whether emphasizing the importance of religious freedoms leads to higher tolerance. Following our reasoning about the role of moral values, we expected that this practice would be tolerated more when presented in terms of freedom of religion. We first found that the intolerance of this practice was similar in a control condition (M = 5.30, SD = 1.69) and one emphasizing the value of social cohesion (M = 5.26, SD = 1.77). This is consistent with the finding that the refusal to shake hands is considered highly dissenting and prompts prioritization of the moral value of social cohesion over freedom. However, when this practice was presented as a matter of religious freedom, people were found to be less intolerant towards this normatively dissenting practice (M = 4.74, SD = 1.83). This demonstrates that emphasizing a particular moral value can even lead to lower intolerance of a practice that is considered highly dissenting and challenging for societal cohesion.

Individual differences and situational factors in tolerance

Following the “forbearance model of tolerance” (Figure 1), we have discussed our research on the different aspects of the tolerance process. In addition to this research we have also considered individual difference variables and situational conditions that can make tolerance less or more likely. In this section, we will focus on the importance of some individual difference variables before considering the important situational condition of perceived outgroup threat in tolerance.

Our research reveals that social psychological constructs such as authoritarianism, social dominance orientation, and deprovincialization are important for tolerance (e.g., Adelman & Verkuyten, 2020; Dangubic et al., 2020; Velthuis et al., 2020). People who have a more authoritarian orientation in which cultural traditions and normative conformity are emphasized and people who more strongly endorse group based social hierarchies tend to be less tolerant of dissenting outgroup practices and beliefs, whereas people who have a less parochial ingroup view on the world tend to be more tolerant. In addition, individual differences in thinking dispositions (e.g., need for closure, open-mindedness, dialectic thinking, intellectual humility) can be expected to play a role in the tolerance process in which disapproved of outgroup beliefs and behaviours are condoned. For example, across several studies, we have found that open-mindedness as well as balanced thinking or the tendency to evaluate contrasting reasons (e.g., “I always try to weigh whether there are good reasons to accept or not accept differences”) are associated with higher tolerance of a range of minority group practices (e.g., Adelman & Verkuyten, 2020). In addition to these well-known constructs that are relatively well explored in the social psychological literature on cultural diversity and intergroup relations, we also examined the role of a new construct of slippery slope belief for tolerance (Adelman et al., 2021c).

Slippery slope belief

Slippery slope belief captures the idea that a non-problematic action, proposal or event is presented as the beginning of an inevitable process of cause and effect that will likely end in harmful or unacceptable results (Van der Burg, 1991; Volokh, 2003). Slippery-slope belief can be problematic because they might lead people to see single events as the first step on a road to a dangerous outcome. This might reduce people’s willingness to put up with outgroup practices and beliefs that they might otherwise see as tolerable. For example, supporters of the Swiss minaret ban in the 2009 Swiss referendum, argued that allowing the building of new minarets would be a further step on the inevitable road to the Islamization of the country (Miller, 2014).

However, individuals will differ in their general endorsement of slippery slope belief, with some individuals being more likely to think that there will be negative cascading outcomes to specific events than others. For example, when presented with a proposal for building a new minaret, some individuals might be more likely to believe that if Muslim organizations are permitted to express their religion (unobjectionable action), it will end up by Muslim taking over their neighbourhood and the country (harmful consequence). These individuals will be inclined to perceive this as a likely chain of events because they in general tend to believe that an unobjectionable or a small action is a first step on an inevitable road to disaster (e.g., Volokh, 2003).

Across six national samples from the Netherlands and Germany (N = 5,974), we developed and tested an individual difference measure of slippery slope belief and examined its importance for intolerance of outgroup freedoms and debated minority practices as well as agreement with real-world slippery slope examples across the political spectrum (Adelman et al., 2021c). We first developed and validated a reliable six item (7-point scales) measure of slippery slope belief (e.g., “If you give people one finger, they will eventually take the whole hand”, and “If you start to be okay with all sorts of things, you will slowly but surely end up on a slippery slope”) that in confirmatory factor analyses was found to be empirically distinct from generalized trust, sense of control, conspiratorial thinking, present-oriented focus, pessimism, optimism, authoritarianism, open-mindedness, and close-mindedness. Furthermore, tests of measurement equivalence showed evidence for the generalizability and stability of our measure of slippery slope beliefs across the different samples and the two countries with similar mean scores (M ~ 4.63, SD ~1.11).

Among the different samples and within both countries, multiple regression analyses showed that the measure of slippery slope belief was subsequently found to be an independent and relatively strong negative predictor (lowest β = −.29, p < .001) of general tolerance of immigrants, (“Immigrants must be able to preserve traditions and customs that are important to them”), Muslim minorities (“Muslims in the Netherlands must be able to show and experience their own faith in public life”), and also predicted debated practices not related to immigrant origin groups (e.g., the installation of gender-neutral toilets in public buildings; replacing words like “manpower” with gender-neutral terms; gay men kissing in public).

The different studies also demonstrated that the general inclination of slippery slope belief was positively related to a more right-wing political orientation and the related predispositions of status quo conservatism and normative conformity. However, slippery slope belief might be similarly used on both the political left and right when it comes to the evaluation of concrete situations that matter ideologically. Both the political left and right have been found to be susceptible to intolerance and biased ways of thinking (e.g., M. Brandt & Crawford, 2020; Ditto et al., 2019; but see Baron & Jost, 2019). Therefore, for testing whether slippery slope belief is relevant across the political spectrum, we examined (Study 4) the role of slippery slope belief for the acceptance of situations that are typically of concern among people on the left (“If we allow right-wing extremists to say what they want, fascists will eventually take over”) or for people on the right (“If immigrants retain their own language and culture, Dutch culture will eventually change beyond recognition”). To the extent that the propensity toward slippery slope belief reflects a general approach to expect small things or innocuous acts to lead to disastrous outcomes, we would expect to find that slippery slope belief predicts agreement with the concrete examples regardless of the political orientation of the participants and of whether these examples reflect politically left-wing or politically right-wing issues. This is indeed what we found with, for example, slippery slope belief significantly associated with lower acceptance of immigrants also among left-wing oriented people. Thus, the findings indicate that people who in general are inclined to think about small or innocuous events and acts as leading inevitably to dangerous consequences can be less tolerant of societal changes, even when these align with their ideological worldview.

Outgroup threat

Various situational and contextual factors are likely to affect the forbearance tolerance process. For example, intergroup contact, mere exposure, and increased familiarity might over time lead to higher tolerance of things that one disapproves of (e.g., same-sex marriage, gay rights; Boch, 2020; Janmaat & Keating, 2019). People are able to adapt psychologically to changing social standards and norms. This does not have to imply that they give up or compromise on their own convictions and beliefs, or do not continue to prefer their own over the conduct and belief of others; but they will be more likely to tolerate what they once thought was intolerable. There can be a process of ethical fading and demoralization where individuals increasingly reduce and withdraw their moral concerns about specific conduct (Tenbrunsel & Messick, 2004) so that things that used to be controversial and intolerable become included in people’s latitudes of acceptance (Prislin & Filson, 2009).

In addition to these gradual changes, there can be circumstances that make tolerance less likely such as political and public debates about insecurities and outgroup threats. Situations and events that are construed as threatening (e.g., symbolically, realistically, politically) will make tolerance more difficult. There is a substantial literature showing that perceived threats are associated with lower tolerance (e.g., Crawford & Pilanski, 2014; Gibson, 2006; Sullivan & Transue, 1999), and uncertainty leads to reduced tolerance when feeling threatened (Haas & Cunningham, 2014). Fear and anxiety reduce reflective thinking about one’s disapproval making tolerance less likely, irrespective of the values and considerations that one can have for being tolerant (Tenenbaum et al., 2018). In addition to the relatively large literature on the importance of different forms of threat for tolerance, we have focused on examining whether tolerance is lower when minority identity enactment is considered to have more threatening sociocultural consequences (Chanley, 1994; Kuklinski et al., 1991). We tested this proposition in two ways: by (1) examining the difference between tolerance of enactment and persuasion, and (2) by considering the distinction between private and public minority identity enactment.

First, we considered both people’s tolerance of Muslim minority practices (e.g., Muslim women wearing a headscarf) and tolerance of Muslims trying to persuade other Muslims to enact their religious identity in a similar way (to also start wearing a headscarf). It is one thing to tolerate religious minority identity enactment, but another to accept that other minority members are being persuaded to express their identity in a similar way. The former can be perceived as a matter of personal choice, while the latter can be considered as trying to influence others to do as one wishes, and thereby having more negative sociocultural implications. Trying to persuade others implies, for example, mobilizing fellow Muslims to engage in the same religious enactment practices, which some might consider as contributing to the “Islamization” of the country. Research has shown that the perception of minority group size is related to perceived threat (McLaren, 2003; Outten et al., 2012) and that “threat in numbers” predicts negative attitudes towards minorities (Earle & Hodson, 2019). We tested in different studies the expectation that tolerance of identity enactment (e.g., “Do you think it is OK that Muslims like Fatma enact their religious identity this way?’) will be higher than tolerance of persuasion (e.g., “Do you think it is OK if Muslims like Fatma try to persuade other Muslims to engage with their religion in the same way?’).

In four different survey studies (Ns ~ 180–1,139), we consistently found that Dutch adolescents were indeed more tolerant of Muslims involved in practices such as the wearing of religious attire (e.g., headscarf) and refusal to shake hands with someone of the opposite gender (i.e., tolerance of enactment), than of Muslims trying to persuade other Muslims to engage in these practices (i.e., tolerance of persuasion) (e.g., Gieling et al., 2010, 2012; Verkuyten & Slooter, 2008). For example in one study, 43% thought that the decision of a female teacher not to shake hands with men should be tolerated, but for the persuasion of fellow Muslims to do the same, this percentage was only 10%. And while 35% thought that the foundation an Islamic school should be accepted, only 18% found it acceptable that people try to persuade other Muslims to also found Islamic schools (Gieling et al., 2010). The difference between tolerance of enactment and tolerance of persuasion was quite consistent across age, gender, and educational level.

We further examined both forms of tolerance among a national sample of Dutch adults (N = 826) and found higher tolerance for Muslim identity enactment such as wearing a headscarf (M = 4.42, SD = 1.39) than for trying to persuade fellow Muslims to enact their identity in a similar way (M = 3.46, SD = 1.47). Additionally, lower tolerance for persuasion compared to enactment was found in different, contexts, including the private sphere (Velthuis et al., 2022). This suggests that trying to persuade fellow Muslims to enact their religious identity is perceived as increasing the number of Muslims in society who practice their religion, which is considered by many majority members as threatening (McLaren, 2003; Outten et al., 2012).

Second, we investigated tolerance of Muslim minority practices in different contexts, reasoning that Muslim identity enactment is likely to be considered more threatening in public contexts (street, work) as compared to private contexts. Furthermore, we examined whether the expected effect of context on tolerance is especially strong for majority group members who value ingroup culture maintenance. Religious expression in public contexts is a visible form of minority identity enactment, while identity enactment in the private sphere has less sociocultural consequences and therefore should be easier to tolerate. Research in the USA, for example, has shown that minority groups are construed as less American when expressing their minority identity publicly versus privately (Yogeeswaran et al., 2011), and that White American perceivers’ national identification elicited more negative explicit and implicit attitudes toward non-White ethnic groups when members embraced their ethnic heritage in public rather than in private (Yogeeswaran et al., 2014). Research has also shown that people are less tolerant when a particular practice is considered to have negative implications for themselves, their group, or society more generally (Chanley, 1994; Kuklinski et al., 1991).

Thus, the wearing of a headscarf in public may be tolerated less than wearing it in the private sphere. However, there are different sorts of “public” contexts and we therefore additionally examined tolerance of wearing the headscarf in the public context of the street compared to two work contexts (i.e., the general work context, and working as a civil servant). Although the wearing of a headscarf is visible in the street, the consequences for majority members are less obvious and direct, compared to the work context or in social interactions with civil servants. People may perceive the latter two settings as having to be “religious neutral” (i.e., “colour-blind”) situations in which religious identity enactment is less appropriate, and research has found that such a perception can lead to lower minority group acceptance (Dovidio et al., 2015). Thus, we expected that tolerance for wearing a headscarf in both work contexts is lower than in the street context.

Using a survey experiment involving a national sample of Dutch majority members (N = 826), we tested these expectations (Velthuis et al., 2022). Multivariate analyses demonstrated that there was a significant multivariate effect of context such that tolerance was significantly lower for religious expression in the street (M = 4.36, SD = 1.35) than in the private context (M = 5.10, SD = 1.26), and that religious expression at work (M = 4.16, SD = 1.37) and as a civil servant (M = 4.03, SD = 1.35) elicited significantly lower tolerance as compared to the street context (Velthuis et al., 2022). Additionally, religious expression as a civil servant did elicit similar levels of tolerance than in the more general work context. In support of a threat-based interpretation, these context-effects were found to be stronger among majority members who highly valued the maintenance of their ingroup culture, while people who had a low score on this measure were generally tolerant and did not seem to clearly consider the different sociocultural implications within the various contexts.

Future directions

In addition to reviewing our empirical work on forbearance tolerance, we want to briefly draw attention to five possible directions for further social psychological research to enrich our understanding of intergroup tolerance.

Domains of life

First, in our research, we have mainly focused on intergroup tolerance and did not systematically consider similarities and differences between, for example, tolerance in political, social, and moral domains of life (Vogt, 1997). However, these domains might raise differing concerns, and people’s tolerance in one domain might be relatively independent of their tolerance in the other domain (e.g., Mather & Tranby, 2014). Furthermore, tolerance of specific differences may trigger intolerance of other differences. While some people can display tolerance toward cultural, religious or sexual minorities, they may at the same time display intolerance toward ideological or viewpoint diversity. A plea for tolerance and recognition of marginalized groups can go together with intolerance (“de-platforming”) of those who think differently and are condemned (Costello et al., 2021). Applying these sorts of double standards that censor and silence opposite voices can be done for opportunistic and political reasons, but can also be understood in terms of our forbearance tolerance model. In weighing different values, people can come to prioritize justice and protection from harm (“speech is violence”) over the values of liberty and freedom of expression (Ceci & Williams, 2018), and future research into these tolerance trade-offs may give us a better understanding of these processes.

Persuasion and change

Second, future research can examine change in tolerance and intolerance, for example in terms of situational framing and persuasion, and in relation to the so-called asymmetry of (in)tolerance. This asymmetry refers to the notion that it is often easier to convince individuals to become less, rather than more, tolerant because intolerance is cognitively less demanding (Gibson, 2006; Petersen et al., 2010). With intolerance, the disapproval of the outgroup belief or practice agrees with rejecting it, whereas tolerance is more contradictory because it implies accepting what one disapproves of. For example, in one experimental study (Gieling et al., 2012; N = 1,139), we found that low national identifiers became more intolerant when they read arguments against tolerance of Muslim minority practices (e.g., social cohesion), whereas their acceptance of the disapproved of practices was not affected by arguments in favour of tolerance (e.g., religious freedom). Other research, however, has found that argument framing does not have a consistent effect on the level of tolerance (Hirsch et al., 2019), and calls for tolerance can help build support for public policies among the already tolerant, but simultaneously weaken the support among the intolerant (Djupe et al., 2015). In an EEG (electroencephalography) study (N = 172), we found that promoting tolerance increased anti-Muslim attitudes among political conservatives, but not among political liberals. The increased negativity was driven by changes in relative left-frontal asymmetry implying greater approach oriented anger or regulatory efforts to reduce cognitive conflict (Yogeeswaran et al., 2021). Future research can examine whether, when and why tolerance of outgroup practices depends, for example, on providing one-sided pro or contra, and two-sided arguments for condoning a dissenting practice (Xu & Petty, 2021). Furthermore, framing effects are particularly likely when people have somewhat ambiguous or relatively weak feelings about a particular issue. In contrast, strong feelings and attaching personal importance and moral value to an attitude make it resistant to change, with change more likely to occur with moral counter messages (e.g. freedom vs. social order) and not with nonmoral arguments (Luttrell et al., 2019).

Boundaries of tolerance

Third, future research could further examine how people define the boundaries of tolerance and when or why things are considered intolerable. For example, in addition to the general moral considerations of fairness, justice, and other’s welfare, there is the notion of reciprocity. This notion relates to the classical “paradox of tolerance”, which implies that one cannot tolerate those who are intolerant because this would destroy the very tolerance that one seeks (Popper, 1945). Being tolerant toward forces that fail to reciprocate undermines the benefits of civil liberties and equality, and therefore, cannot be tolerated. Experimental research has found that the willingness to be tolerant depends on whether the outgroup is perceived to be tolerant itself (Fairlamb & Cinnirella, 2021). Furthermore, research has demonstrated that tolerance is high and robust towards disliked or feared groups that observe liberal democratic rules, but not towards groups that are associated with violent and non-democratic practices (Petersen et al., 2010). A group’s right to live the life that they want should be reciprocated by its own commitment to tolerance of other groups. The norm of reciprocity might also be important for the asymmetry of tolerance because people might be more persuaded by arguments for being tolerant towards minority groups that accept the liberal democratic rights of others, while they might become intolerant towards groups that are perceived as not fully respecting these rights (Petersen et al., 2010).

Intergroup context

Fourth, tolerance involves the relation between those who tolerate and those who are tolerated. This means that the relevant intergroup context needs to be taken into account to understand how tolerance is experienced and practiced. Historically, the importance of tolerance was discussed for dealing with harmful effects of religious conflicts and most of our research has focused on tolerance of Muslim minorities. Some of our findings indicate that the tolerance process is similar toward other religious minorities (e.g., orthodox Protestants, Christians, immigrants) and sexual minorities, but our main focus has been on majority-minority contexts in which the majority tolerates dissenting minority groups to live according to their way of life. In a hierarchical situation, tolerance affirms the subordinate and dependent position of the minority and can also be used by majority members to define a positive ingroup distinction (‘we the tolerant vs the intolerant Muslims”, “we have been tolerant enough”; Verkuyten, 2013). Tolerance is likely to be different in a more horizontal relationship between groups that hold very different beliefs about the good life and have different cultural, religious, or ideological beliefs, norms, and practices (e.g., political groups with comparable levels of power). In such a situation, all groups are not only the objects of tolerance, but also the subjects of it, which creates unique circumstances to explore intergroup tolerance.

Country differences

Finally, our research has been mainly conducted in the Netherlands and Germany, and although the degree of tolerance tends to differ somewhat between these neighbouring countries (Dangubic et al., 2021; Erisen & Kentmen-Cin, 2017), it is likely that there are stronger differences between other national contexts (Weldon, 2006; Widmalm, 2016). National differences may be grounded in political, legal, economic and historical circumstances. For example, what is considered “freedom” or “harm” can be interpreted differently across cultural contexts, and what are considered cases involving tolerance of free speech or intolerance of harmful treatment can also differ (see Verkuyten et al., 2020b). Levels of tolerance have been found to be higher in more democratic societies that emphasize individual liberties (e.g., Widmalm, 2016). The political and institutional context can make tolerance more or less prevalent by either promoting tolerant norms and values or propagating authoritarian goals. Tolerance is more likely to spread in societies that allow political freedoms and are committed to basic liberal values. which are expressed in the educational system, and reflected in its institutions and legal regulations. People are shaped by the societies in which they are embedded and multilevel analyses indicate, for example, that country differences in citizenship regimes (laws governing the acquisition and expression of citizenship) affects people’s tolerance of ethnic minorities via their national identification and satisfaction with democracy (Weldon, 2006).

Examining country differences also raises the more general question of how to assess tolerance, and the need for reliable and valid measures that make it possible to obtain robust, reliable and comparable findings. Some measures for assessing individual differences in general tolerance have been developed and used in various countries (e.g., Hjerm et al., 2020; Velthuis et al., 2021), and in political science the ‘least-liked group technique’⁶ (Sullivan et al., 1979) is widely used for measuring political tolerance. However, these general measures are less useful for examining tolerance of specific dissenting practices and beliefs as the outcome of a psychological process of weighing the reasons for disapproval with reasons for acceptance. A focus on the forbearance tolerance process implies a focus on different considerations and how these play out in tolerating, or rather negatively interfering, with particular dissenting beliefs and practices. We have described various ways in which this process can be examined, and future studies might want to build further on these designs so that a common measurement approach for studying intergroup tolerance might develop.

Conclusion

A plural society implies group differences in beliefs, convictions, and practices that are not always easy to reconcile, and might involve ways of life colliding. Because of their propositional content, all religions, cultures, and ideologies are not genuinely considered by their adherents to have equal value, which means that they often do not coexist comfortably side-by-side. It is hard to value and celebrate diversity when one believes that certain practices interfere with the rights and liberties of others (e.g., abortion rights, right to apostasy), certain viewpoints are considered misguided (e.g., conservative or liberal), or certain ideologies or practices are considered oppressive (patriarchy). However, what is required in such situations is tolerance where people do not surrender their own convictions and ideals, but rather strive to overcome them by taking into account the reasons to nonetheless grant others their freedoms. Tolerance should not replace prejudice-reduction strategies or pro-diversity initiatives, but is an indispensable condition for living with difference and dissent. There are always substantial differences between groups of people which make tolerance one of those things that we “cannot not want”, even as we recognize its limitations and boundaries. In appreciating the necessity of contrasting views and conflicting beliefs for positive change and social development, tolerance constitutes a precondition for social justice, fairness, and freedom. Hence, it is important to understand why and when people develop tolerance towards outgroup beliefs, practices and behaviours they disapprove of, and how people draw the boundaries of what can be tolerated. As we have tried to show, social psychology can make an important empirical contribution to this understanding and thereby to finding and maintaining productive ways for managing divergent and conflicting beliefs and behaviours, and relieve intergroup tensions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This contribution was supported by a European Research Council (ERC) Advanced Grant under the European Union’s Horizon 2020 research and innovation programme (grant agreement No 740788; Horizon 2020 Framework Programme.

Notes

¹ Here we use the term “Muslim practices and beliefs” to refer to the way the public tends to perceive these practices and beliefs, without suggesting that these are necessarily part of Islam.

² In the social psychological literature on prejudice various conceptualizations, aspects and components are proposed. Here we use the term to refer to some negativity or antipathy “towards whole groups of people or towards individuals because of their membership in a particular group” (Brown, 2010, p. 4). There are many possible reasons for the group‐based antipathy, but what is shared is the focus on a category of people as people.

³ The list technique (also known as the item-count technique; Blair & Imai, 2012) entails randomly dividing participants into two experimental conditions. In the control condition, respondents are presented a list of items (groups or social issues) people might feel negative about. In the treatment condition, respondents are presented with the same list of items plus one “sensitive” item added. In both conditions, participants are asked to only indicate the number of items they are negative about, and not which items these are. The difference in means between the treatment and the control condition corresponds to the percentage of people who are negative about the sensitive item.

⁴ There was also a Moderately anti-Muslim profile (28%) consisting of people who were somewhat negative towards Muslims as a group of people and perceived that Muslims treat men and women differently, but did not reject all Muslim expressive rights. And there was an Indifferent profile (20%) consists of people who, on average centered around the neutral midpoint of the scales for group feelings, gender perception and endorsement of expressive rights.

⁵ Assessed with the extended version of the “child-rearing preference” measure (Feldman, 2020; Stenner, 2005). This measure is a trade-off between stimulating the value of social conformity versus autonomy in socializing children, and the items do not reference any social groups, events or actors, which means that the scale is not confounded with the attitudes towards minority groups and practices that one wants to explain.

⁶ People are first asked to indicate which group they like the least and subsequently whether they are willing to grant people of that group the full rights of citizenship.