Social Mechanisms and Causal Processes

At the outset of Grounds for Difference, Brubaker claims that, unlike in his previous work, he is “less focused on conceptual critique than on theoretical and empirical analysis, and less concerned with analytical disaggregation than with analytical synthesis” (Brubaker Reference Brubaker2015: 2). Indeed, each single chapter advances sophisticated theoretical arguments and provides important historical, comparative, and empirical insights on reconfigurations of cultural difference in modern nation-states. Upon greater scrutiny, however, the book turns out to contain instructive conceptual critiques as well—notably in chapter 2 on the return of biology, and in chapter 4 on religion and nationalism. In fact, quite often analytical disaggregation seems to be Brubaker’s driving intellectual motif. A particularly striking example is chapter 1, where he explicitly calls for a “more differentiated and disaggregated strategy” (ibid.: 18). In what follows, I discuss this programmatic chapter in greater detail, assessing its contributions to the literature on cultural pathways to social inequality as well as its relation to broader sociological controversies on the status of social mechanisms in sociological explanation.

Extending conventional accounts of social stratification, cultural sociologists have recently started scrutinizing how socio-psychological micro-mechanisms of cognitive stereotyping, as well as broader cultural processes of categorization and stigmatization, feed into the reproduction of social inequality (e.g., Lamont et al. Reference Lamont, Beljean and Clair2014). An early example of that line of reasoning is Tilly’s (Reference Tilly1998) analysis of exploitation and opportunity hoarding as two mechanisms linking (horizontal) categorical differences to (vertical) social inequalities. Brubaker’s powerful criticism of Tilly’s analysis centers on its failure to differentiate among distinct dimensions of difference (Brubaker Reference Brubaker2015: 18). He proposes an alternative analytical strategy that highlights distinctive ways through which categorical distinctions based on citizenship (ibid.: 19), gender (ibid.: 22), race or religion (ibid.: 28) generate or sustain inequalities. Illustrating his argument with examples largely taken from the United States and Western Europe, he emphasizes that these modes of categorization vary along several analytical axes—formal versus informal, external versus internal, degrees of embeddedness, and so forth—and that this these variations crucially matter for the production and reproduction of social inequality. For instance, whereas citizenship perpetuates inequalities by externally assigning persons to formal-legal status, religion can—qua belief and practice, values and networks—also (re-)produce social inequalities “from within” (ibid.: 33). On a more abstract level, Brubaker perceptively theorizes three general processes that underlie categorically caused or inflected inequalities: the allocation of persons to social positions, the social production of persons, and, often overlooked in conventional social stratification research, the social definition of positions and their rewards (ibid.: 35–41).

This disaggregated mode of analysis bears close resemblance to recent scholarship on social mechanisms (for a review, see Gorski Reference Gorski, Hedström and Wittrock2009). That scholarship, to recall, starts from the premise that sociological theories of the middle range, instead of pursuing a futile search for constant conjunction, covering laws, or statistical correlations, should strive to identify recurrent mechanisms in the social world. There is much disagreement regarding the explanatory logic and ontological status of social mechanisms; scholars notably disagree whether one should formulate (or even formalize) them based on methodological individualism and a rationalist model of the actor (Hedström Reference Hedström2005), or whether one should cast a wider net by including genuinely interactive or relational mechanisms—such as exploitation or opportunity hoarding (Tilly Reference Tilly1998: 13–14; Tilly Reference Tilly2001). There is general agreement, however, that social mechanisms open the black box between causes and effects, pay tribute to a more processual conception of causation, and allow for more fine-grained sociological explanations.

It seems obvious that Brubaker’s domain-comparative analysis aims at social mechanisms thus conceived. In revising Tilly’s earlier account, chapter 1 sketches an entire array of mechanisms that provide fine-grained accounts of how categories of difference produce or reproduce social inequalities. More specifically, chapter 1 identifies relational mechanisms that put emphasis on social interaction and interdependence, instead of starting with decontextualized individual actors. What seems less obvious, however, is how these relational mechanisms, which Brubaker so brilliantly dissects abstractly, help explain concrete historically situated configurations of cultural differences. Like other processual and mechanism-based models of boundary making (Wimmer Reference Wimmer2008: 1010), Brubaker’s inventory of relational mechanisms is empirically “void.” It explains how categorical differences can potentially generate or sustain social inequalities, but it does not explain whether and when they actually do so. That task would methodologically require addressing two further questions. The first question concerns specifying the contextual conditions that trigger any of the aforementioned mechanisms. For instance, a mechanism whereby participation in different religious groups generates forms of social capital that confer differential economic advantages (Brubaker Reference Brubaker2015: 34) depends on broader religious field characteristics, as transatlantic comparisons of religious effects upon occupational attainment among migrants suggest (Connor and Koenig Reference Connor and Koenig2013). The second question is how social mechanisms, which in principle could either amplify each other or cancel each other out, coalesce into robust self-reinforcing causal processes in a given historical moment. An example for such coalescence is the “discriminatory equilibrium” observed among Muslims on the French labor market (Adida et al. Reference Adida, Laitin and Valfort2016); here, informal categorical exclusion based on religious stereotypes prompts reactive religiosity, in turn reaffirming these stereotypes and thus perpetuating socio-economic disadvantages.

Both problems—the context dependence of social mechanisms and their coalescence into causal processes—merit further consideration in a fully elaborated research agenda on configurations of cultural difference. Only then can one utilize Brubaker’s inventory of social mechanisms for the comparative analysis of cultural differences and their interplay with social inequalities in concrete, spatially, and temporally bounded social settings.

Religious Difference and the Politics of Categorization

In debates on contemporary reconfigurations of cultural differences, the politics of religious difference have attracted particularly high attention. Indeed, whereas the research field of citizenship and immigrant integration, ethnicity, and nationalism tended to ignore religion for decades, its main protagonists have shown increasing interest in religious differences and their socio-economic as well as politico-legal implications, the central concern being evidently with Muslims in the West (e.g., Joppke Reference Joppke2015; Voas and Fleischmann Reference Voas and Fleischmann2012). Whereas the recent literature largely adopts a compartmentalized view of religion, treats it as just another dimension of ethnicity, or, by contrast, reproduces stereotypes of Islamic exceptionalism, Brubaker’s domain-comparative approach adds novel insights by analyzing similarities and differences between religion and other grounds of difference in generating social inequalities (chapter 1), political conflict (chapter 3), and nationalism (chapter 4). In the following, I discuss chapter 3, where Brubaker compares religion with language and explores their changing salience in modern politics of difference. My main focus will be on potential trade-offs between a domain-comparative approach and efforts to historicize domains of categorical difference.

In chapter 3, which is based on his 2012 Gellner Lecture, Brubaker elaborates an inventory of social mechanisms that generate, reproduce, and institutionalize religious and linguistic differences, respectively. At the same time, he advances a historical-sociological argument about how these mechanisms coalesce under the contextual conditions of modern nation-statehood. On a longue durée scale, he argues, language became more politicized, being a key medium through which expanding territorial state bureaucracies turned a diverse peasantry into a national citizenry. Religion, by contrast, became less politicized as territorial states turned to popular sovereignty as basis of legitimacy thus undergoing a process of secularization. Strikingly, these same contextual conditions have strengthened the social mechanisms that generate, reproduce, and institutionalize religious differences. State neutrality contributed to the “robustness of religious pluralism” (Brubaker Reference Brubaker2015: 91) by allowing for a self-sustaining reproduction of religious beliefs and practices relying on family- and organization-based transmission processes (for empirical evidence see Soehl Reference Soehl2017). On a shorter time scale, however, religion has paradoxically reemerged as a major site of political struggle over the recognition and accommodation of differences, notably in contexts of international migration, or so Brubaker argues.

While historical sociologists, sensitive to the religious-political legacies of Europe’s confessional age, may contest Brubaker’s argument of state-imposed linguistic homogenization and state-tolerated religious pluralization (see Gorski in this symposium), I do think that his domain-comparative approach might stimulate innovative and overdue research on the distinctiveness of linguistic and religious identity politics. That said, his domain-comparative approach also faces potential analytical trade-offs. As I shall argue, comparing the grounds of categorical difference—religion, language, and so forth—and analytically dissecting the social mechanisms associated with them may run the risk of ignoring the historicity of these very categories.

In developing my argument, I rely on Brubaker’s own earlier work, where he repeatedly followed Bourdieu’s admonition to engage in a radical epistemic break with preexisting knowledge of the social world and to critically reflect upon vernacular categories of practice. In Grounds for Difference, he engages in this radical break as well, for instance in chapter 5 when discussing the concept on diaspora (Brubaker Reference Brubaker2015: 129), or in chapter 6 when treating the nation-state with its assumed congruence of citizenship, peoplehood, and nationhood as a Geertzian “model of” and “model for” political organization (ibid.: 131). Against this background, it is quite surprising that he consciously decides not to reflect upon religion as a category of practice. This decision is even more surprising, given that he is fully aware of a growing literature in anthropology, history, and religious studies that has reconstructed the early modern genealogy of the concept of “religion” (Asad Reference Asad1993) or unveiled the scholarly as well as missionary invention of “world religions” in nineteenth-century colonial encounters (Masuzawa Reference Masuzawa2005).

Brubaker justifies his decision not to reflect upon the vernacular category of religion by stating that “the scope of [the] argument is limited to contemporary liberal polities” (Brubaker Reference Brubaker2015: 89). While restricting scope conditions is perfectly legitimate, the question does arise whether his decision comes with a loss even when analyzing the politics of religious difference in the West. These politics do indeed involve symbolic struggles over the very category of religion; they are, as it were, politics of categorization—as Brubaker acknowledges in passing (ibid.: 90). The jurisprudence of constitutional and international courts, arguably highly influential institutions providing legitimate representations of the social world, provides ample illustration of these struggles (see Koenig Reference Koenig, Bolden and Kymlicka2015). Repeatedly, these courts have had to decide whether headscarves, turbans, or crucifixes are religious, cultural, or political symbols. How courts categorize such symbols results in distinctive allocation of rights, resources, and recognition as enshrined in constitutional law or international human rights. French legislation outlawing the wearing of hijabs in public schools and burqas in public spaces required not only careful legal balancing of individual rights to religious freedom with constitutional principles of laïcité but it also required deciding whether these practices should legitimately count as “religion” in the first place. Similarly, the European Court of Human Rights in Lautsi v. Italy had to deliberate over whether crucifixes in public schoolrooms symbolized the state’s religious preferences or just the cultural heritage of the nation. Commenting on First Amendment cases in the United States, critical legal scholars such as Winnifred Sullivan (Reference Sullivan2005) have forcefully explored the paradoxes plaguing courts’ attempts to implement the kind of religious neutrality that Brubaker regards as at minimum approachable for liberal states.

Such politics of categorization are even more widespread if one broadens the scope of comparative analysis to modern nation-states beyond the West. For instance, the Alevi movement in Turkey has strategically postured as distinctive “religious” minority in judicial battles before the Turkish Constitutional Court and the European Court of Human Rights (Dressler Reference Dressler2013). Likewise, the Ahmadi movement has mobilized the category of “religion” to claim its distinctive identity against the Sunni majority in Pakistan (Saeed Reference Saeed2017). Many more examples come to mind in East Asian nation-states, where intense politics of religious difference have emerged from the superimposition of modern “religion” upon rather fluid communities of ritual or belief. Indeed, sociologists of religion have long argued that the global diffusion of the very category of religion is a key component in reconfigurations of religious difference (Beyer Reference Beyer2006).

In short, Brubaker’s domain-comparative research agenda would benefit from a more radical break, or at least greater reflexivity, with respect to historically embedded categories of difference. Evidently, one cannot deny that some definition of religion is necessary for spelling out distinctive generative mechanisms of (re-)producing categorical differences. Yet, the contextual conditions under which these mechanisms operate also involve symbolic struggles over the very category of religion—struggles that should figure more prominently in a domain-comparative approach as proposed in Grounds for Difference.

Variable Configurations of Modernity

In his argument on the (de-)politicization of religious and linguistic differences and elsewhere in his book, Brubaker (Reference Brubaker2015: 92) operates with an ideal-typical distinction of “premodern and modern liberal societies.” This raises the question how the set of liberal democracies, to which he restricts some of his historical-sociological arguments, compares to the full set of modern polities. Are they the same, or is the former a subset of the latter? Answering this question requires entering into controversies regarding how to define the conceptual core and how to analyze the empirical variability of modernity. In the following, I critically discuss Brubaker’s defense of “single modernity” perspective in light of his substantive historical analyses as well as of his methodological commitment to dissect social mechanisms that (re-)produce configurations of cultural difference.

Controversies over the conceptual core of modernity have long occupied the discipline of sociology. Historical sociologists have made a particularly important contribution in criticizing grand narratives of convergent, teleological, and multidimensional social change inherited from nineteenth-century ideologies of progress and underlying postwar theories of modernization. They emphasized variable paths and patterns of modernization (Anderson, Bendix, etc.); brought back in state formation and class relations (Skocpol, Tilly); and, in the most recent wave of scholarship, emphasized contingency, culture, agency, and (post-)colonial entanglements in the (re-)making of modernity (Adams et al. Reference Adams, Clemens and Orloff2005). Brubaker has been a prominent contributor to this “third wave” of historical sociology, forcefully rejecting simplistic theories of modernization.

In Grounds for Difference, however, he seems to defend some of their more nuanced versions. In chapter 4 on religion and nationalism, for instance, he claims that the differentiation of religion and politics remained the valid core of classical secularization theory (Brubaker Reference Brubaker2015: 118). Reviewing a rich literature that has contested the secularist bias in nationalism studies, he perceptively discusses attempts to regard religion and nationalism as functional analogies, identify religious causes for modern nationalism, or dissect religious forms and content of nationalism. The upshot of his conceptual critique is a defense of the “modernist” position, emphasizing the distinctively secular character of nationalism as evinced by its immanent ontology or imaginary that divides the world horizontally into various nations (ibid.: 116). A precondition for this imaginary was secularization, understood as a grand process of differentiation in which social spheres acquired autonomy or Eigenlogik, to use Weber’s terminology.

But how do such grand processes of secularization, differentiation, and modernization relate to the undeniable variability of cultural, institutional, and political configurations of modernity? Brubaker addresses this question up front in chapter 7 where he criticizes the notion of “multiple modernities” that the late Shmuel N. Eisenstadt developed as part of his comparative civilizational analysis. Logically, Brubaker stresses that some minimal definition of modernity was conceptually necessary to compare modern civilizations or societies. Sociologically, he stresses the global scope and interconnectedness of those transformative processes—he cites social mobilization (Deutsch), division of labor (Gellner), and the “integrative revolution” (Geertz)—that resulted in the unmatched hegemony of nation-statehood as model for political organization. To explain the striking politicization of ethnicity worldwide, or so Brubaker argues, requires conceiving modernity as a “singular historical phenomenon” (ibid.: 147).

Brubaker’s two arguments in defense of the notion of a “single modernity” have their incontestable merits. They find strong support in neo-institutional world polity studies (Meyer Reference Meyer2010) as well as in recent global historical sociology (Go and Lawson Reference Go and Lawson2017), and they underline what Peter Wagner (Reference Wagner2008) has described as the inescapability of modernity in social theory. Incidentally, they would have readily been conceded by Eisenstadt, for whom modernity was after all the first truly global civilization, albeit refracted through different civilizational traditions. Yet, the two arguments leave unanswered the question how the “single modernity” perspective relates to the domain-comparative research agenda on configurations of cultural difference that Brubaker advances throughout his book. I would like to make two critical observations in this respect.

First, Brubaker’s defense of “single modernity” brings into sharper profile the restricted scope conditions adopted in the aforementioned substantive chapters. Conceiving modernity in the singular logically requires a highly abstract definition that goes beyond simplistic lists of concrete societal features (market economy, democracy, secular law, etc.) on which earlier modernization theorists have relied. Eisenstadt defines modernity abstractly as a cultural program characterized by reflexivity and a drive to social reconstruction, thus capturing both pluralist and totalitarian tendencies, both bright and dark sides of modernity, as it were (Eisenstadt Reference Eisenstadt2000). In similarly abstract manner, Brubaker (Reference Brubaker2015: 152) regards modernity as characterized by a “broad type of polity,” namely the model of territorial statehood legitimated by popular sovereignty. Thus defined, however, the set of modern polities is clearly broader than the set of liberal democracies to which Brubaker restricted the scope of his disaggregated analyses of categorical differences, social inequality, and political accommodation. His statement that “citizenship is the great remaining bastion of strong categorical inequality in the modern world” (ibid.: 45), for instance, is certainly accurate when considering the firm enshrinement of nondiscrimination norms in international human rights law. Yet, when considering various examples of ethnic exclusion, religious discrimination, and gender inequality in some postcolonial settings, it should be clear that other forms of categorical inequality do persist in modern nation-states. Brubaker’s domain-comparative research agenda would gain from scrutinizing more carefully the variable, liberal as well as nonliberal, configurations of cultural differences that modernity in all its varieties, including the imperial legacies that Brubaker rightly acknowledges (ibid.: 136–37), has generated.

At the same time, second, Brubaker’s defense of a “single modernity” perspective brings into sharper profile the strong potential of a mechanism-based mode of sociological explanation. Consider the question how to account for local variations that, in Brubaker’s view, characterize the global diffusion of models of nation-statehood (ibid.: 152). Neo-institutional world-polity theory, although cognizant that mimetic isomorphism comes with strong decoupling of ceremony and substance, is ill prepared to explain local adaptation, nonadoption, or even resistance against global scripts such as nondiscrimination or minority rights (but see Pope and Meyer Reference Pope and Meyer2016). Eisenstadt resorts to cultural visions of cosmic and social order embedded in Axial Age civilizations to explain variable modern constructions of collective identity, but thereby runs the risk of overly deterministic assumptions of historical path dependence. Brubaker’s inventory of social mechanisms (re-)producing categorical differences and social inequalities offers analytical tools to forge more sophisticated accounts for the variable political dynamics set in motion by the global diffusion of territorial statehood and nationalism—and, one may incidentally add, of “religion.” Recent Weberian attempts to reconstruct grand theories of modernity as constellations of social mechanisms (Aakvaa Reference Aakvaa2013) may offer guidance for dissecting the differential coalescence of mechanisms that generate variable configurations of cultural differences in modern nation-states that the burgeoning comparative literature on idioms of nationhood, citizenship regimes, or varieties of secularism aptly describes.

To conclude, Grounds for Difference affirms Brubaker’s stature as an agenda-setting theorist in the research field of ethnicity, nationhood, and citizenship. Grounded in an admirable command of literatures across diverse disciplines and subdisciplines, his domain-comparative approach opens novel vistas for studying the reconfigurations of ethnic, religious, linguistic, and other differences in modern nation-states. In my discussion of Brubaker’s book, I have raised some methodological questions concerning the status of social mechanisms in sociological explanation, the historicity of categories of difference, and the local variability of globally diffusing models of modern nation-statehood. I have done so in the hope that Brubaker’s research agenda, perhaps refined methodologically, will stimulate in-depth historical and comparative research. In short, I read this book as a precursor and exhortation to many sociological studies to come.