In recent years, concerns about immigration, increasing numbers of refugees and asylum seekers, and the growing visibility of ethnic and racial minorities have triggered an expanding debate about the consequences of diversity for community and social cohesion in liberal democracies (see, for example, Banting and Kymlicka, Reference Banting and Kymlicka2006; Putnam Reference Putnam2007). Journalists, policy makers and ideologues have repeatedly expressed their fears of increasingly complex and multiethnic societies and the impact they may have on democratic politics and social relations. Some have even gone as far as warning that the solidarity necessary to maintaining our modern universal welfare states will simply crumble when facing demands from many different ethnic groups (Goodhart, Reference Goodhart2004). In a recent referendum, the majority of Swiss citizens agreed with the campaign of the Swiss People's Party, which warned about the increasing spread and dominance of the Muslim religion and thus voted to ban the further building of minarets in the country.Footnote 1 Moreover, several European countries recently experienced civil unrest as racialized poverty exploded into protests and riots. These riots are largely viewed as a reaction to the mixture of social exclusion, discrimination and Islamophobia faced by ethno-racial minorities in these societies (Murray, Reference Murray2006). Indeed, the acceptance of cultural diversity is relatively limited across Europe and the rise of far right parties in some European countries has been partly fuelled by widespread anti-immigrant sentiments (Citrin and Sides, Reference Citrin and Sides2008; Sniderman et al., Reference Sniderman, Peri, Figueiredo and Piazza2000; Pettigrew, Reference Pettigrew1998b; Norris, Reference Norris2005; O'Connell, Reference O'Connell2005; Schain et al., Reference Schain, Zolberg and Hossay2002).
Concerns about the potentially negative impact of diversity have also emerged in academic debates. Levels of generalized trust, for example, are not just lower among minorities themselves, but they are also lower among majority populations living in diverse surroundings (Alesina and La Ferrara, Reference Alesina and La Ferrara2002; Banting et al., Reference Banting, Johnston, Soroka, Wallerstein, Bardhan and Bowles2006; Hero, Reference Hero2003; Rice and Steele, Reference Rice and Steele2001). Robert Putnam (Reference Putnam2007) has recently argued that in the short run racial diversity is likely to reduce various aspects of social capital, defined as the norms of trust and reciprocity that characterize healthy communities. His study finds that in racially diverse areas in the US, citizens tend to trust each other less and are less able to co-operate with one another to address shared problems. He even finds that trust in one's own group members (such as the trust Blacks have of other Blacks) is reduced when facing social diversity. Several other studies confirm these insights in the US context (Alesina and La Ferrara, Reference Alesina and La Ferrara2000; Hero, Reference Hero2003). This kind of research leaves the impression that changing demographic realities in the US are going to make democratic politics more difficult.
However, it is by no means self-evident that these phenomena are applicable across other democracies or that higher numbers of ethnic and racial minorities will necessarily result in social unrest and rising anti-immigrant sentiments. At first sight, the Canadian case seems to provide a unique contrast to other countries. Canada's federal nature, its selective immigration system, unique welfare system and history of both biculturalism and multiculturalism raise questions about the relationship between ethnic, racial and religious diversity and a perceived tide of anti-immigrant attitudes and conflictual intergroup relations that have been observed in other countries. In the comparative context, Canada is often viewed as a successful case of immigrant integration and social cohesion where pro-immigration attitudes have been consistently on the rise (Kymlicka, Reference Kymlicka1989 and Reference Kymlicka2010; Reitz and Banerjee, Reference Reitz, Banerjee, Banting, Courchene and Seidle2007; Harell, Reference Harell2010). Yet recent debates and research elucidate the limitations of Canada's success in managing diversity. The early 1990s, for example, saw an increasing backlash against what was viewed as “mosaic madness” (Bibby, Reference Bibby1990). Indeed, just as in the United States, in the most ethnically diverse localities in Canada, citizens trust each other less than in more homogenous neighbourhoods (Soroka et al., Reference Soroka, Helliwell, Johnston, Kay and Johnston2007). Moreover, Canadian immigrants still experience higher unemployment rates and lower salaries when compared to their native-born counterparts (Hum and Simpson, Reference Hum and Simpson2004). More recently, debates in Quebec over reasonable accommodation and in Ontario over the use of Shari'a law in family arbitration highlight how Canadian public opinion can be divided about what the democratic response to increasing diversity should be.
This special volume of The Canadian Journal of Political Science draws on papers initially presented at a 2009 conference sponsored by the Canadian Opinion Research Archive (CORA), School of Policy Studies, Queen's University, with funding from SHRCC as well as the Centre for the Study of Democratic Citizenship (CSDC) at McGill University. The resulting papers explore the implications of ethnic, racial and religious diversity for the attitudes and behaviours of citizens in Western democracies. In doing so, the authors contribute to a better comprehension of the Canadian case vis-à-vis other diverse societies. More generally, this volume provides a deepened understanding of the ways in which the changing demographics of liberal democracies are affecting democratic politics, both for better and for worse. The challenges of governing multiethnic, multilingual and multireligious societies will be a defining characteristic of the twenty-first century. Providing greater insights into the complex ways that citizens from all walks of life navigate this reality is the goal of this special volume. In this introduction, we review the main theoretical interdisciplinary perspectives on the study about the consequences of diversity for various aspects of social cohesion and intergroup relations. We then identify important gaps in this literature and discuss how the studies in this special volume attempt to overcome some of the identified weaknesses in the literature.
Diversity and Democratic Politics: Theoretical Perspectives
Recent political events and emerging feelings of hostility in Western democracies toward immigrants and newcomers have partly prompted this newly emerging research agenda on the consequences of diversity. This research agenda is in its very essence an interdisciplinary one, driven by questions and puzzles in social psychology, intergroup relations, research on attitudes towards immigrants and racial relations, and increasingly by research on redistributive justice, welfare state solidarity and social capital. What all of these approaches have in common is that they strive to explain more generally whether and how diversity might shape our ability to create and maintain tolerant, trusting, solidaristic, co-operative and democratic societies.
Despite this interdisciplinary effort, most research on the consequences of diversity is framed in terms of the underlying assumptions of intergroup relations, which have stumbled into a seeming theoretical paradox with two competing frameworks making contradictory predictions about what we can expect from rising ethnic, religious and cultural diversity in our surroundings.
On the one hand, the so-called contact hypothesis suggests that when people from different backgrounds have the opportunity to interact with each other, intergroup hostilities should be reduced (Allport, Reference Allport1958; Pettigrew, Reference Pettigrew1998a; Pettigrew and Tropp, Reference Pettigrew, Tropp and Oskamp2000). Especially when contact occurs among equals and interaction promotes a shared goal, people from different backgrounds are expected to develop positive feelings towards each other and, by extension, to other members of the outgroup. Indeed, such positive contacts across lines of group difference are supposed to promote a larger, supra-ordinate identity (Gaertner and Dovidio, Reference Gaertner and Dovidio2000; Dovidio et al., Reference Dovidio, Kawakami and Gaertner2002). Indeed, the contact hypothesis has larger societal implications beyond the groups involved in the interaction. An additional mechanism, labelled secondary transfer effects, suggests that the reduction in prejudice directed at one outgroup extends to other groups not involved in the original contact (Pettigrew, Reference Pettigrew2009). In other words, meaningful contact with Asian immigrants, for example, might be sufficient for reducing prejudice against Muslim immigrants. Generally, the contact hypothesis and the empirical evidence that supports it highlight the potential importance of diverse social networks in the development of political behaviour and attitudes.
A similar idea has been picked up in the political science literature on social capital, where associations and informal networks are differentiated based on whether they bring together people from diverse and different backgrounds (bridging ties) or whether they are constituted of members who are alike (bonding groups) (Putnam, Reference Putnam2000; Warren, Reference Warren1999). Putnam (Reference Putnam2000) has expressed high hopes for bridging ties, implying that when interaction is promoted across salient social cleavages, positive outcomes are expected for democracy. Thus, being involved in an association or other interaction contexts with a relatively high proportion of immigrants would foster more socially and politically tolerant attitudes among majority group members. There is some evidence in support of this. For example, being part of associations (Cigler and Joslyn, Reference Cigler and Joslyn2002) and friendship networks (Harell, Reference Harellforthcoming) that expose individuals to people who are different from them is positively related to political tolerance. Moreover, active bridging contact with diverse neighbours has been shown to positively relate to generalized trust (Stolle et al., Reference Stolle, Soroka and Johnston2008).
On the other hand, there is also a large body of research that argues that when different ethnic or racial groups are brought together, the result is heightened intergroup conflict (Blumer, Reference Blumer1958; Giles and Buckner, Reference Giles and Buckner1993; Tolbert and Grummel, Reference Tolbert and Grummel2003). Conflict results in part because ethnic or racial groups might struggle over the same socio-economic resources or for cultural dominance. Thus majorities see outgroups as a threat to their own ingroup, and as a result both ingroup favouritism and outgroup prejudice rise (Tajfel and Turner, Reference Tajfel, Turner, Austin and Worchel1979; Sherif et al., Reference Sherif, Harvey, White, Hood and Sherif1961). The conflict or threat hypothesis suggests that as localities become increasingly diverse, outgroup hostilities are likely to emerge. This phenomenon is especially found in less privileged areas and in the absence of meaningful contact (Oliver and Mendelberg, Reference Oliver and Mendelberg2000; McLaren, Reference McLaren2003; Branton and Jones, Reference Branton and Jones2005). Although direct support for the rise in ingroup feelings is often missing, the theory has received a seeming boost from a variety of results on the negative consequences of diversity for racial attitudes, tolerance of immigrants, trust and welfare state solidarity.
For example, in the European context, the threat hypothesis has been applied to help explain rising anti-immigrant sentiment. Sniderman and colleagues (Reference Sniderman, Peri, Figueiredo and Piazza2000) document rising anti-immigrant sentiments in Italy and relate these to increasing numbers of immigrants from non-European countries who are viewed as especially different because of visible markers of race and ethnicity. Consistent with the threat hypothesis, they demonstrate that cultural and economic threat has, in part, driven Italians to vote for right-wing parties. Across Europe, rising immigration levels have been met with increased prejudice, political opposition, racial discrimination and more restrictive views of citizenship requirements (Pettigrew, Reference Pettigrew1998b; Wright, Reference Wright2010). In line with the conflict hypothesis, other studies demonstrate that this prejudice is driven by economic and cultural threat, which increases with the number of immigrants (Quillian, Reference Quillian1995; see also, O'Connell Reference O'Connell2005). When applied directly to measures of social capital such as generalized trust, the results are less clear cut. In a study of 20 European countries, various measures of dynamic changes in immigrant populations and overall diversity were only weakly correlated with a lack of generalized trust across Europe (Hooghe et al., Reference Hooghe, Reeskens, Stolle and Trappers2009). However, in a study across a larger group of developed countries, ethnic fractionalization was more strongly related to lower levels of generalized trust (Anderson and Paskeviciute, Reference Anderson and Paskeviciute2006).
At first sight, these two theoretical approaches seem rather contradictory. However, when it comes to outgroup attitudes, some reconciliation between the two perspectives has been offered. As Stein and colleagues note, “Ironically … the very conditions that give rise to [W]hite hostility toward minority group members set in motion a corrective for this hostility: intergroup contact” (Reference Stein, Post and Rinden2000: 299). While a context of ethno-racial diversity in the absence of interaction has a negative impact on outgroup attitudes consistent with the threat hypothesis, the presence of larger numbers of outgroup members also makes interaction between groups more likely, which can mitigate feelings of threat (McLaren, Reference McLaren2003; Stein et al., Reference Stein, Post and Rinden2000; Wagner et al., Reference Wagner, van Dick, Pettigrew and Christ2003).
The problem, then, appears to be that most studies do not account for the interaction between intergroup contact and social context. Indeed, most studies on the negative effects of diversity do not measure intergroup contact at all. In these studies, diversity is tapped at the contextual level, in countries, regions, states or provinces, counties, cities and census tracts, and these geographical units are assumed to capture social interaction. While it is generally believed that diverse contexts bring about more intergroup contact (Joyner, Reference Joyner2000; McPherson et al., Reference McPherson, Smith-Lovin and Cook2001), the possibility exists that not all diverse areas score equally on this account. The puzzle about diversity may not be as much about the presence or absence of minority groups as such, but about the level of segregation in each area (Massey and Denton, Reference Massey and Denton1993; Uslaner, Reference Uslaner, Selle and Prakashforthcoming) and how relationships between these groups are structured. If the most diverse areas and neighbourhoods are also the most segregated, then the potential for intergroup contact to counteract group conflict is minimized.
Putnam (Reference Putnam2007) has further challenged the potential for reconciliation between these two competing perspectives. He found that neighbourhood ethnic diversity is not only related to lower levels of generalized and outgroup trust, but also of ingroup trust. In his thorough analysis of the Social Capital Benchmark survey, he shows that people living in diverse areas are not just more suspicious of people who do not look like them; they are also more reserved and distrusting towards their own kind and generally socialize and interact less with each other (Putnam, Reference Putnam2007). These results came, of course, as a surprise to someone who earlier expressed high hopes for the democratic value of bridging ties. In this view, ethnic and racial diversity leads generally to a “hunkering down” and withdrawal from social interactions and group life of all kinds.
Putnam's results seem to reject the causal mechanisms of both theoretical approaches: he does not find that increased diversity leads to less prejudice and more trust as the contact hypothesis would predict; and in addition, diversity also diminishes ingroup trust, which speaks against the mechanisms of the threat hypothesis. However, Putnam, like most studies cited here, does not measure contact directly and so, at the end, the contact hypothesis as such remains untested. The possibility exists that the most diverse cities in his sample contain extremely segregated neighbourhoods where intergroup contact is simply unlikely. Furthermore, as Fieldhouse and Cutts (Reference Fieldhouse and Cutts2010) argue, the measurement of diversity using the Herfindahl Index does not capture important differences between diverse communities. The authors also suggest that we might expect entirely different consequences of what they call co-ethnic density and thus should look at diversity in many different ways.
In sum, the verdict is still out as to what exactly the consequences of diversity are and how they can be explained theoretically. It is clear that a highly engaging but also potentially divisive research agenda has developed over the last decades. The interdisciplinary research in this area needs to address the intricate links and effects of diversity and its contextual character at various geographic and organizational levels, as well as by considering aspects of segregation and intergroup contact. Mapping the complex ways in which rising diversity in various geographic units shapes and influences people's social experiences in a comparative context, and in turn their politics, is the goal of this volume.
Moving Beyond Contact versus Context
While interdisciplinary research delivers a fairly mixed picture about the effects of diversity and its theoretical explanation, the current public discourse in Western societies is fuelled by stories and events that paint a rather pessimistic picture. The capacity of Western societies to integrate immigrants from different ethnic, racial and religious backgrounds has led to serious debates about what integration requires, both from the majority and the minority (Harell and Stolle, Reference Harell, Stolle and Hoogheforthcoming). At the heart of these debates is a question of how much, and what types of, diversity a political community can support while still maintaining democratic institutions and social cohesion. Clearly, there is a need to examine these issues in more detail.
This is especially the case because it seems that public and media debates around these issues have currently taken an overly pessimistic and one-sided view of the issue. This is clearly the case with recent events in the Canadian context. For example, Giasson and colleagues (2010) followed press coverage of the reasonable accommodation debates that occurred in Quebec in 2006 and 2007. They provide empirical evidence that the media disproportionately framed the debates around diversity and reasonable accommodation issues in a negative light (see also Bouchard and Taylor, Reference Bouchard and Taylor2008). In doing so, they argue that the media fuelled the perception of a crisis in Quebec and the image of Francophone Quebeckers as intolerant. We assume that such coverage is not unique to the reasonable accommodation debates, given the media's tendency to focus on sensational and negative news stories (for a review, see Graber, Reference Graber2004). There is a real need, then, to look beyond the media's coverage and to assess the extent to which a crisis around diversity exists and, equally importantly, the ways in which experiences with diverse others translate into political values and behaviours.
We propose four extensions of current work that will help to draw a more complete picture of this relationship. First, we argue for a broad comparative approach and particularly a closer look at Canada as a unique case; second, we highlight the need to better understand which types of policies might overcome, mediate or exacerbate any potential negative influence of diversity; third, we urge a broader consideration of the consequences of diversity on a wider variety of social cohesion measures beyond generalized trust; and finally, we suggest the need to consider how experiences in diverse settings shape selected groups of individuals differently.
The Comparative Advantage
The first issue involves whether the findings about the negative effects of diversity on trust, racial attitudes and other indicators of social cohesion in the US context might be generalizable to other countries. While some initial results show how the relationship between country-level diversity and social cohesion indicators travels to other continents, not much work has been done to understand the more local dynamics of diversity comparatively (however, see Lancee and Dronkers, Reference Lancee and Dronkers2008; Reeskens, Reference Reeskens2009). Developing a comparative research agenda provides important leverage in understanding how context matters for citizens' reactions to changing demographics. Three contextual factors may drive such comparisons: economic equality, histories of immigration and experiences with diversity, and the level and changes in diversity. For example, economic inequality has a strong impact on generalized trust (Rothstein and Uslaner, Reference Rothstein and Uslaner2005) and might exacerbate or even trump the effects of ethnic diversity (Branton and Jones, Reference Branton and Jones2005; Oliver and Mendelberg, Reference Oliver and Mendelberg2000; Blake, Reference Blake2003). As income inequality is much higher in the US than in Canada and in most European countries (Gottschalk and Smeeding, Reference Gottschalk and Smeeding1997), it is important to investigate the relation between localized diversity and civic attitudes, such as trust under conditions of greater economic equality.
Another criterion for comparison that might add nuance to our understanding of how diversity impacts citizens' attitudes and behaviours are differences in the history of immigration and experiences of diversity across liberal democracies. Here, the distinction between so-called settler societies and former colonial powers may be particularly useful (Stasiulis and Yuval-Davis, Reference Stasiulis and Yuval-Davis1995; Harell, Reference Harell2010). Does the accommodation of ethnic, racial and religious diversity differ in countries with long histories of immigration compared to European societies that have only recently become destinations for mass immigration? While settler societies deal with specific discourses about whiteness and belonging, the founding myths of these societies often include a narrative about immigration. In the US, the myth of the melting pot clearly suggests a community built on diversity (Hirschmann, Reference Hirschmann1983), while in Canada, its identity as a multicultural society, however contested, reflects a founding narrative based on immigration (Day, Reference Day2000; Dreiger and Shivalingappa, Reference Dreiger and Shivalingappa1999; Johnston et al., Reference Johnston, Banting, Kymicka and Soroka2010). In contrast, European states have only more recently dealt with mass immigration, and so ideas about citizenship have often been constructed largely around membership in the historic community, or “nation” (Smith, Reference Smith1986; see also Shulman, Reference Shulman2002).
While this points to a useful comparison between historically immigrant-receiving countries like Canada, the United States and Australia, on the one hand, and European democracies on the other, it should be noted that expanding research outside the US context may also highlight the implications of more nuanced differences in the nature of diversity within settler societies. For example, the form of ethnic diversity experienced in US society is a unique mixture of an older cleavage based on a history of slavery (African–Americans and Whites) and more recent types of diversity resulting from various waves of immigration, for example, from Latin America and Asia (Sniderman and Piazza, Reference Sniderman and Piazza1993; Schildkraut, Reference Schildkraut2007; Waters, Reference Waters2001).
Empirical studies on the effects of diversity on trust outside the United States are scarce and usually limited to single-case studies that reveal contradictory results.Footnote 2 For example, in Canada the percentage of visible minorities in a neighbourhood has a negative effect on generalized trust among the majority, and to a lesser extent, the minority populations (Soroka et al., Reference Soroka, Helliwell, Johnston, Kay and Johnston2007 Stolle et al., Reference Stolle, Soroka and Johnston2008), but recent studies in Australia and the United Kingdom are less conclusive on this matter. In Australian neighbourhoods, ethnic fractionalization is not related to generalized trust, while linguistic fractionalization shows only a modest negative effect (Leigh, Reference Leigh2006). A recent study in the United Kingdom shows that the negative effects of diversity disappear when taking into account the level of socio-economic resources that are available within the community (Letki, Reference Letki2008). Another null finding of local diversity on trust was reported in the context of Flanders (see Reeskens, Reference Reeskens2009); however, a Dutch study that tested a broader array of social cohesion indicators found a negative consequence of diversity on most of them (Lancee and Dronkers, Reference Lancee and Dronkers2008).
Given these mixed results and the potential advantages of a comparative framework, this special issue addresses whether the US findings on the negative relationship between social diversity and generalized trust (and other social norms) hold in other Western democracies at the country and more localized levels (Fieldhouse and Cutts, Reference Fieldhouse and Cutts2010; Kesler and Bloemraad, Reference Kesler and Bloemraad2010). At first sight, some of the articles in this special volume seem to confirm the negative findings of previous work. Fieldhouse and Cutts provide one of the few country comparisons of the effects of localized diversity on neighbourhood norms and participation in the UK and US. While they find a negative effect of diversity in both cases, they are able to qualify their results by making distinctions between majorities and minorities. Kesler and Bloemraad show that a rise in immigration negatively influences social trust, organizational membership and political engagement in 19 countries, but their work does not stop there. They document a complex interaction between both the institutional context and the way diversity affects various indicators of social capital, pointing to the importance of linking diversity to the institutional context in which it is embedded.
Linking Diversity to Institutions
By focusing on just one country, research is unable to address the question of whether different integration regimes or other political institutions might be able to mitigate or overcome the assumed negative effects of diversity. Yet, from a policy perspective it is critical to assess not only the impact of diversity but also how different policy platforms shape citizens' experiences with the shifting demographic make-up of their countries.
While most Western democracies are confronted with increasing ethnic and cultural diversity, they certainly have developed different policies and institutions to address it (Joppke and Lukes, Reference Joppke and Lukes1999; Bloemraad, Reference Bloemraad2006; Freedman, Reference Freedman2004). Canada, Sweden, and to a degree the Netherlands and UK, for example, have adopted a multicultural approach (see Kymlicka, Reference Kymlicka2010; Banting and Kymlicka, Reference Banting and Kymlicka2006; Kymlicka, Reference Kymlicka1995), where national governments recognize the rights of ethnic minorities in various arenas of culture, education, politics and religion (for example, multiculturalism in school curricula, representation of ethnic minorities in politics and in the media, funding of ethnic group activities, the funding of bilingual education and so on). Contrary to this, the policies in countries such as France and the US focus much more on integrating newcomers into the mainstream culture of the receiving society (Koopmans et al., Reference Koopmans, Statham, Giugni and Passy2005; Koopmans and Statham, Reference Koopmans and Statham2001; Gordon, Reference Gordon1964). The question becomes which approaches are more successful in mitigating any potential negative effects of diversity, while creating conditions for social inclusion among various communities (Harell and Stolle, Reference Harell, Stolle and Hoogheforthcoming).
Views are divided and empirical research is limited on these issues. Some claim that multicultural policies create divisions in society by promoting attachment to ethnic and cultural identities to the detriment of a unifying national identity (Bissoondath, Reference Bissoondath1994; Barry, Reference Barry2001). This leads to cultural isolation among new ethnic groups, fewer majority–minority social interactions, a deepening of social cleavages, indifference and distrust (Barry, Reference Barry2001). By de-emphasizing other loyalties and ethnic backgrounds, “melting pot” and strongly integrationist policies are considered by some to be fundamental to ensuring that newcomers integrate well into the host society (Koopmans, Reference Koopmans2010). Koopmans in particular argues that assimilationist regimes or other integration regimes with limited welfare states are more beneficial for immigrants in particular, as they force them to adopt the value system of majorities as well as to acquire the language skills that they need for employment. In countries with multicultural policies and extensive welfare states (as in the Netherlands, Sweden and Belgium), by contrast, immigrants are able to survive on welfare support without making such adjustments and, this argument suggests, end up being unemployed (Reference Koopmans2010).
In contrast to this view, other scholars argue that policies that recognize cultural diversity actively promote tolerance, acceptance of otherness and reduce tensions between groups in society. For example, some research suggests that policies that promote an inclusive identity are associated with improved intercultural relations and attitudes toward immigrants (Berry, Reference Berry, Dacyl and Weston2000; Billiet et al., Reference Billiet, Maddens and Beerten2003). In this vein, Weldon (Reference Weldon2006) shows that a country's citizenship regime, defined as how institutionalized a dominant ethnic tradition is within a country, is strongly related to tolerance of ethnic minorities in Europe.
Kymlicka assesses the evidence in support of what he calls the “liberal multiculturalist hypothesis” (Reference Kymlicka2010), namely that countries which adopt multicultural policies recognizing the attachment of minorities to their culture and identity do not experience an erosion of liberal democratic values. In a broad review of the literature to date, he determines that there is little evidence that the multicultural policies that have emerged to accommodate increasing ethnocultural diversity have been detrimental to core liberal democratic values (see also Harell, Reference Harell2010). In his assessment of current research, he suggests that multicultural policies actually have a positive impact on civil liberties, employment equality, and solidarity. While recognizing that more research is needed, Kymlicka provides a compelling challenge to the current debate on the effects of ethnocultural diversity on democratic politics.
Drawing on data from 19 Western democracies over a twenty-year period, Kesler and Bloemraad (Reference Kesler and Bloemraad2010) take these considerations to heart in a more stringent empirical test. While they also find that immigrant-driven diversity and measures of social capital such as trust, organizational membership and political engagement are negatively correlated, they go a step further and show in a longitudinal model that in societies characterized by greater economic equality and multicultural policies the negative relationships can be mitigated. They conclude, therefore, that increasing diversity is not universally damaging to social capital, when counter-acted by the right policies. Their results provide an important test of the overall relationship between multicultural and other government policies and the social capital of majorities, yet future research needs to address more explicitly how various integration regimes shape gaps in a broad array of majority and minority attitudes and behaviours.
Beyond Trust
While work on the consequences of diversity originally focused on intergroup attitudes and prejudice, current political science research on the relationship between diversity and social cohesion has put its emphasis on generalized trust, one important measures of social cohesion (Crepaz, Reference Crepaz2007). However, we need to expand the study of social cohesion to other equally salient aspects of cohesive and co-operative societies. Certainly, generalized trust taps an abstract attitude toward strangers and people one does not personally know. It thus measures an important dimension of co-operative potential in societies (Uslaner, Reference Uslaner2002; Nannestad, Reference Nannestad2008). High levels of generalized trust characterize societies that are able to successfully solve collective action and corruption problems, as well as achieve high levels of political accountability (Putnam, Reference Putnam1993; Uslaner, Reference Uslaner2008; Rothstein, Reference Rothstein1998). However, as Arneil argues quite convincingly (Reference Arneil2010), generalized trust is not the most expected outcome in societies which have seen the struggles of various minority groups and movements to get equal recognition, acceptance and representation in Western democracies. These struggles reflect legitimate fights waged by marginalized actors for basic respect, and indeed are motivated and pushed by rising distrust toward established norms, institutions and groups. In short, generalized trust might be impossible to achieve in certain periods of struggle for citizens' rights that necessarily lead to confrontation with the established norms and practices.
Yet other aspects of social cohesion often go unnoticed. Social cohesion has a multitude of definitions but generally refers to belonging to a political community, both in terms of socio-economic inclusion as well as attachment and involvement within the larger society and its values (see, for example, Jenson, Reference Jenson1998; Harell and Stolle, Reference Harell, Stolle and Hoogheforthcoming). Along with a focus on ethnocentrism and intergroup attitudes (Citrin and Sides, Reference Citrin and Sides2008), an examination of the impact of diversity on social cohesion needs to be linked more directly to democratic rights, definitions of citizenship and attitudes about democratic institutions (Wright, Reference Wright2010). A new focus on democratic attitudes, including values of fairness, social solidarity and redistributive justice, seems to be a potentially fruitful avenue for research on the consequences of diversity and must be the contribution of the discipline of political science to this discourse. While intergroup relations are important in their own right, as Kymlicka notes (Reference Kymlicka2010), we should be equally, if not more, concerned about how such relations impact the pillars of liberal democratic regimes.
This is an area in which we know very little, although the evidence that is emerging paints a promising picture. Some of the articles in this volume reflect a broadening of the approach to include, for example, political tolerance and attitudes towards the welfare state into the analyses on consequences of diversity. For example, Harell (Reference Harell2010) examines political tolerance among young people in Canada and Belgium. This tolerance, defined as support for the speech rights of various disliked groups, has been seen as a fundamental principle of democratic politics, yet it has also become a key battleground in debates about accommodating ethnic and racial minorities. She demonstrates that many young people make an important distinction between speech they dislike and speech that specifically targets ethnic, racial and religious minorities. She argues this reflects the shifting discourse within liberal democracies over how to balance individual rights and concerns about social inclusion. In contrast to absolute notions of political tolerance, she argues this reflects a more multicultural political tolerance. Johnston and colleagues (Reference Johnston, Banting, Kymicka and Soroka2010) focus on another pillar of modern liberal democracies, namely social solidarity in the form of support for the welfare state. Drawing on the Canadian case, they examine the attitudes of native-born Whites in English Canada, a group that arguably represents the historic majority in the Canadian context. They argue that national identity among this group promotes sustained support for encompassing welfare state policies as well as reduces hostility toward immigrants. In a novel argument that crosses different policy domains of the welfare state, they highlight the important relationship between Canadian national identity and support for universal health care. In doing so, they make an important contribution to understanding how components of the welfare state are incorporated into national identity, and how the overarching attitude of nationalism, at least in the Canadian context, seems to extend to include newcomers. Their research underlines that, in the Canadian context, national identity maintains its inclusive character in the face of rising diversity.
Indeed, both these articles suggest that as research begins to move beyond questions of trust, there is little evidence as of yet that key democratic institutions and values will be uniformly challenged when facing a rise in ethnic and racial diversity. However, as the example of tolerance shows, they might be transformed to reflect the new realities of experiences with diverse others.
Differentiating Experiences
By broadening the research agenda to address a wider range of components of social cohesion, we see a fourth and final direction for this research agenda: a more nuanced approach to understanding how diversity affects various groups of people differently. As Arneil (Reference Arneil2010) makes clear, in lamenting the decline of social trust in the face of increasing diversity, we often overlook how struggles for social justice are creating spaces for marginalized communities. In this struggle for democratic rights, representation and recognition, ethnic and racial minorities might not just simply be less trusting of majority groups, but they might also react very differently to diversity around them. In short, changing levels of diversity may impact majorities and minorities differently, and even complicate our understanding of what majority and minority mean.
We see three ways, in particular, to differentiate experiences. The first is to recognize when our theoretical paradigms are majority-centric. Much of the research on how diversity shapes citizenship is really concerned with how increasing numbers of racialized minorities or immigrant groups impact the cultural majority in liberal democracies. As with Arneil (Reference Arneil2010), discussions of social trust often paint an idealized picture of community that overlooks the situation of marginalized members within the community (see also Arneil, Reference Arneil2006). While it is certainly a valid research agenda to examine majority group members, we urge researchers to recognize more explicitly the majority-centred view of this line of inquiry, as Johnston and colleagues have done (Reference Johnston, Banting, Kymicka and Soroka2010).
Moreover, changing demographic and social conditions may have a different meaning for or a different influence on members of minority groups (Stolle et al., Reference Stolle, Soroka and Johnston2008). Measures of diversity often simply focus on the size of the minority population. When heterogeneity is directly measured, little consideration is given to which groups are involved and their relative strength. For example, the widely used Herfindahl index does not make a distinction between whether a locality is entirely white or entirely black. The percentage of visible minorities, the preferred measure in the Canadian context (Soroka et al., Reference Soroka, Helliwell, Johnston, Kay and Johnston2007), does not necessarily distinguish between the heterogeneity of several groups versus the dominance of very few ethnic groups per locality. Moreover, only rarely do we know whether ethnic minorities in selected neighbourhoods are surrounded by other ethnic minorities of their own group or of a variety of other ethnic backgrounds (Bilodeau, Reference Bilodeau2009). Theoretically, such ethnic concentration effects should matter for how the role of diverse context plays out for minorities themselves; and they might also affect majorities differently. The assumption is that ethnic concentration might be beneficial for minorities, mostly because they allow for beneficial bonding ties and embedded resources (Portes and Sensenbrenner, Reference Portes and Sensenbrenner1993; McKenzie, Reference McKenzie2008), but questions arise as to how the existence of co-ethnic diversity might affect other aspects of social cohesion, such as democratic values, attitudes, engagement and trust. There is, therefore, a need to measure diversity more sensitively and focus on how social diversity affects minority communities, on their own and in relationship to cultural majorities.
Fieldhouse and Cutts (Reference Fieldhouse and Cutts2010) have risen to this challenge. In their case study of neighbourhood diversity in the US and Great Britain, they find that among majority group members in both countries, greater levels of diversity are negatively related to trust. However, they add nuance to these findings by pointing out that this effect is reduced, and even reversed, for minority group members. For example, while community participation among British minorities is lower than expected in diverse areas, attitudinal aspects such as neighborhood norms are substantially enhanced. Co-ethnic diversity also exerts additional effects for minorities.
The contribution by Hurwitz and Peffley (Reference Hurwitz and Peffley2010) also examines how minority and majority attitudes differ in the United States, a country whose traditional racial cleavages contrast the immigrant-driven diversity of the other countries in this volume. Focusing on attitudes toward the criminal justice system, they demonstrate convincingly that Blacks view the criminal justice system in the US as far less fair than their white counterparts do. What is particularly interesting here, and deserves far greater attention across multiple domains of public policy, is how the variation in attitudes is a reflection of the lived experiences of majorities and minorities. Hurwitz and Peffley convincingly argue that Blacks' views of the justice system are in part a reflection of an unjust system that discriminates against them. In doing so, they highlight the need to ensure that questions of discrimination play a central role in understanding not only how majorities react to minority communities but also in how individuals from minority groups experience their social context.
This leads to our third and final suggestion for differentiating experiences which emerges from the work by Abu-Laban and Couture (Reference Abu-Laban and Couture2010), namely to complicate our understanding of what minority means. They present a case study of how the changing demographic make-up of schools in Alberta has created conflicts over the institutional accommodation of Francophones in the Albertan education system. More specifically, they focus on how the minority Francophone community in Alberta is itself becoming more diverse through immigration. Contrary to conventional dichotomies between “Western” and “non-Western” immigrants and their relationship to secularism, they document how Muslim Francophone immigrants are struggling for a secular education within the Catholic Francophone school system which was historically designed to protect the French-speaking minority in the province. In doing so, they reflect an increasing interest among scholars in the experiences of “minorities within minorities” (Eisenberg and Spinner-Halev, Reference Eisenberg and Spinner-Halev2006). This is one area in which the study of the consequences of increasing diversity has been all but silent, yet as Abu-Laban and Couture (Reference Abu-Laban and Couture2010) document, it is, and will continue to be, an important component of understanding how democratic institutions are being used and shaped by various actors in contemporary liberal democracies.
Understanding Diversity and Democratic Politics in the Comparative Context
Clearly, demographic changes will continue to modify the social landscapes in liberal democracies. How these changes will affect citizens' attitudes and behaviours is of fundamental importance to understanding the politics of the twenty-first century. In this brief introduction, we have highlighted four ways to extend current research. First, we have argued that we need to take a comparative approach to understanding how diversity shapes democratic politics. Whether it is through case studies or comparative work, placing findings within the broader comparative context helps us to understand whether the apparent negative consequences of diversity observed for various aspects of social cohesion hold across a variety of contexts. Second, we have highlighted one contextual factor in particular: the policy environment in which diversity is accommodated and experienced. Clearly, how diversity is managed, as well as the larger rights and integration regime under which such policies fall (Smith Reference Smith2009), shape intergroup relations and the attitudes and behaviours of citizens more generally toward their political and social community. Third, we have pleaded for a broader examination of social cohesion that moves beyond trust. While trust is an important indicator, social cohesion is also about redistributive justice, solidarity, tolerance, respect and more generally about narrowing minority–majority gaps in economic and social resources. Finally, we have suggested a move away from the majority focus in research on the consequences of diversity, to understand better how rising diversity is experienced by specific groups of people and particularly minorities themselves.
The articles in this special volume rise to the challenge of addressing these shortcomings in various ways and, in doing so, provide a valuable contribution to our understanding of how ethnic, racial and religious diversity is understood and experienced by citizens, as well as its impact on their values and behaviours. At first sight, many of these articles seem to confirm the insights from US-dominated studies that a rise in immigration diversity might lead to a dampening of overall social cohesion measured by trust, neighbourhood norms and civic engagement. However, this volume has provided several nuances and qualifications. For one, multicultural policies, especially when combined with policies to create overall socio-economic equality, seem to be able to mitigate some of the negative effects observed in the literature. Moreover, by shifting our focus beyond trust, many of the authors in this volume have highlighted how diversity may shape, or at least interact with, core liberal democratic values like political tolerance and social solidarity, which adds nuance to the earlier results on trust. Finally, when taking differential experiences of majorities and minorities into account, some authors confirm how these groups are affected differently by rising diversity: minorities are often found to be less sensitive to the diverse context around them than majorities and, in some cases, even benefit from it. Understanding the ways in which ethnic, racial and religious diversity are affecting democratic politics clearly requires a much more serious consideration of the interplay of ethnic background, intergroup relations, diversity and the political institutions in which they are embedded. While future research should certainly expand upon these themes, we hope this volume begins to move the debate beyond the hysteria that has engulfed public debates in this domain to understand the issues in a more nuanced way.
Acknowledgments
We would like to thank all of the participants at the 2009 “Diversity and Democratic Politics” workshop for helpful comments on earlier drafts of the articles included in this issue. The workshop was held at Queen's University, May 7 and 8, 2009.