I am grateful for this opportunity to reply to such a stimulating set of critical reflections. As the commentaries are quite distinct, I address each separately, beginning with Matthias Koenig’s commentary because it provides the broadest perspective on the book. Koenig raises in a clear and cogent manner three important methodological questions. The first concerns the methodological status of mechanisms. Koenig correctly notes that the strategy of analytical disaggregation pursued in the book—especially in the first chapter on difference and inequality—has an affinity with mechanism-based theorizing, notably with attempts like that of Tilly (Reference Tilly1998) to specify what Koenig calls “genuinely interactive or relational mechanisms.” As Koenig observes, the first chapter critically engages Tilly’s analysis of the mechanisms of exploitation and opportunity hoarding and proposes three alternative mechanisms or, in the less mechanistic if more anodyne language I prefer, processes: the allocation of persons to positions, the social production of persons, and the social definition of positions and their rewards. But Koenig goes on to note, quite correctly, that my account of these alternative processes remains too abstract to help explain variations in “concrete historically situated configurations of cultural differences.” To specify how categorical differences can generate or sustain social inequality, Koenig rightly notes, is not to explain when, why, and whether they do so. His suggestions for advancing a more distinctively explanatory agenda by (1) specifying contextual conditions that trigger, activate, or make relevant general processes or mechanisms and (2) specifying the concatenation or “coalescence” of mechanisms into “robust self-reinforcing social processes” are well taken (and are well illustrated by his own work [Connor and Koenig Reference Connor and Koenig2013] on religious differences and inequality).

Koenig’s second point concerns the analytical trade-offs entailed by the kind of cross-domain comparisons undertaken in chapters 1, 3, and 4 of Grounds for Difference and other work (Brubaker Reference Brubaker2015b, Reference Brubaker2015c, Reference Brubaker2016; Brubaker and Fernández Reference Brubaker and Fernández2019). The strategy of cross-domain comparison emerged in part from my long-standing interest in the social life of categories: in how categories and systems and practices of classification work in a variety of domains, in the first instance race, ethnicity, nationalism, and religion, but also in law, citizenship, gender, deviance, medicine, and psychiatry. In this context, I found Tilly’s argument that categories work in essentially the same way across substantively different domains powerful and intriguing—good to think with, as Tilly almost always is—but also profoundly flattening. My relation to Fredrik Barth’s (Reference Barth and Barth1969) famous injunction to focus on the nature and dynamics of ethnic boundaries, rather than on what Barth somewhat dismissively (and to his later regret) called the “cultural stuff” the boundaries enclose, has been similarly ambivalent. Barth, like Tilly, has profoundly influenced the way I think about ethnicity and about categories of difference more generally; yet I’ve found myself resisting the suggestion that the cultural (and social-organizational) “stuff” matters less than boundaries. Cross-domain comparison offered a way to theorize citizenship, gender, and ethnicity (chapter 1), religion and language (chapter 3), and gender and race (Brubaker Reference Brubaker2016) as “different differences” (Epstein Reference Epstein2007: 255), yet not as sui generis or incommensurable: as similar enough, in certain respects and for certain theoretical purposes, to make comparison possible, yet different enough to make comparison interesting.

Cross-domain comparison also suggested itself in the context of my emerging interest in the politics of religious difference. For half a century, social theory has been grappling with the politics of difference. Yet it has done so in a strikingly uneven way. Large and sophisticated bodies of work—both empirical and normative—address race, ethnicity, and nationalism, on the one hand, and gender and sexuality, on the other. But social theory has been much slower to address the politics of religious difference. Cross-domain comparison seemed a promising way to integrate religion more fully into the theoretical and empirical study of the politics of difference: it enabled me to stake out a middle ground between particularizing stances that treat religious identities and religious conflict as sui generis and generalizing stances that subsume the politics of religious difference under the rubric of politicized ethnicity (Brubaker Reference Brubaker2015c).

Koenig appreciates the potential gains from cross-domain comparison, but he observes that it risks obscuring the historicity of the domains and their constitutive categories. I don’t think this is necessarily true: cross-domain comparison can even highlight the historicity of the domains being compared. As I note, the fact that religion and language are today what I call categorically differentiated domains of cultural practice—understood by participants, observers, and major social institutions as partitioned into discrete categories, rather than as a continuous spectrum of variation—is not intrinsic to religion and language as ahistorical categories; it is “a product of history and politics, not least a history and politics of objectification, individuation, and boundary-drawing that has carved out distinct ‘languages’ from dialect continua, and constructed and institutionalized distinct ‘religions’ from fluid and varying sets of practices” (Brubaker Reference Brubaker2015a: 87). The historical constitution of the fields of religious and linguistic difference is amenable in principle to cross-domain comparative analysis, even if that historical project is not pursued in this book.

Koenig registers his surprise that Grounds for Difference does not analyze the workings of and struggles over religion as a category of practice. I recognize, of course, the centrality of such struggles to the contemporary politics of religious difference, and the analysis of such struggles figures centrally in my larger project on the politics of religious and linguistic difference. (Struggles over language as a category of practice also figure centrally in that project, though with a key difference: struggles for recognition as a language—as opposed to a “mere” dialect or variety of another language—are central to the politics of linguistic difference, and to many forms of nationalist politics, just as struggles for recognition as a religion are central to the politics of religious difference; but struggles over the definition of language per se are nowhere near as important as struggles over the definition of religion.) In Grounds for Difference, however, I bracket these questions to develop an argument about the long arc of the politics of religious and linguistic difference in Western liberal settings. I seek to explain why religion has tended to displace language as the cutting edge of contestation over the political accommodation of cultural difference in recent decades, in a striking reversal of the longer-term process through which language had previously displaced religion as the primary focus of contention. But I acknowledge that the chapter on religion and language would have been enriched by a more reflexive stance toward its central categories of analysis.

Koenig’s final point concerns the tension between conceptualizing modernity as (on one level) a single interconnected global process and specifying and explaining the wide variation in the social and political organization of and contestation over cultural difference in the modern world. He rightly notes that the liberal democratic settings to which my analysis is largely confined represent only one strand of modernity, and he observes that a more comprehensive analysis would come to substantially different conclusions about matters such as the characteristic forms taken by the politics of religious difference or the persisting significance of strong forms of categorical inequality. Koenig generously suggests that the inventory of mechanisms presented in Grounds for Difference has the potential to help develop more sophisticated accounts of the variable trajectories of the politics of difference in the modern world.

Volker Schmidt’s contribution addresses chapter 1, “Difference and Inequality.” As noted previously, this chapter takes as its point of departure a critical engagement with Tilly’s theory of categorical inequality and seeks to specify in more differentiated terms the ways in which categories of difference are implicated in the production and reproduction of inequality. Schmidt reads the argument through a dual lens that is at once sociological, highlighting the logic of functional differentiation, and philosophical, highlighting the logic of egalitarian individualism. He then focuses on the few pages of the chapter that are devoted to citizenship and specifically on my argument that, while categorical exclusion on the basis of sex, race, ethnicity, religion, and sexual orientation has been massively delegitimized and even illegalized in liberal democratic settings in recent decades, categorical exclusion on the basis of citizenship remains largely unchallenged. Citizenship is thus “the great remaining bastion of strong categorical inequality” (Brubaker Reference Brubaker2015a: 45) in such settings. As an ascribed and inherited status that binds the vast majority of the world’s population to the state to which they have been assigned by the morally arbitrary accident of birth, citizenship serves to perpetuate and—to the extent that it continues to be largely taken for granted—to legitimize vast inequalities in life chances on a global scale.

Schmidt accepts this argument but probes its normative implications. The emergence of a world society, he suggests, has “rendered obsolete the nation-state-based model of citizenship.” World society requires global citizenship and global justice. And if exclusion based on nation-state-level citizenship is no longer justifiable, then “borders have to be torn down.” Xenophobia cannot be limited in its meaning to the ill-treatment of noncitizens within the territory of a state; it begins rather “with the very erection of borders, and the first act of xenophobia is to forcibly prevent anyone from crossing them.” Consistent adherence to the principles of egalitarian individualism requires granting “all of humankind virtually unlimited mobility.”

Here Schmidt joins the debate in political theory on global justice (see also Schmidt [Reference Schmidt, Aarnio, Hoeren, Paulson, Schulte and Wyduckel2013]) and, more specifically, on the ethics of immigration control, aligning himself with the case made by Carens (Reference Carens1987) for open borders or at least with theorists like Seglow (Reference Seglow2005) or Risse (Reference Risse2008) who have argued for very substantial limits on the right of states to exclude would-be migrants. Although I recognize and indeed underscore the moral arbitrariness of exclusion on the basis of citizenship, I am not persuaded by these arguments. Or rather: I am persuaded by these arguments on their own abstract philosophical terms, but I am not persuaded that states should simply open their borders or admit vastly larger numbers of immigrants. This position seems to me to take too little account of the immense gap between intuitions—which I share—about the moral arbitrariness of excluding by citizenship, and beliefs—which I do not share, but which are widely held among citizens of prosperous liberal democratic states—that even present immigration levels are much too high and that borders are “out of control.” The centrality of these beliefs (and of more general anxieties about economic and cultural globalization) to the success of Brexit, Trump, and right-wing populism throughout Europe seems clear (Brubaker Reference Brubaker2017a, Reference Brubaker2017b). Dismissing or stigmatizing such anxieties only strengthens them, and it makes liberals seem even more out of touch. David Frum’s comment—”When liberals insist that only fascists will defend borders, then voters will hire fascists to do the job liberals won’t do”—may have been glib, but it is not entirely unfounded.Footnote ¹

Ann Morning’s contribution addresses chapter 2, “The Return of Biology.” This chapter explores the complex and ambivalent implications of the return of biology for the theory and practice of race, ethnicity, and nationhood in the domains of biomedicine, forensics, genetic genealogy, and identity politics. Morning’s comments focus on my introductory account of the return of biology and my concluding “constructivist reply” to the challenge posed by neo-naturalist and neo-objectivist accounts of race. Given Morning’s own important work in this area (Morning Reference Morning2011, Reference Morning2014a, Reference Morning2014b), I am gratified that her critique nuances and qualifies certain aspects of my account while accepting its basic premises and lines of analysis.

In commenting on my introductory account of the return of biology, Morning urges more sustained attention to the “field of power” in which the social and life sciences are embedded and in particular to the “uneven playing field” on which biology commands vastly greater economic resources and cultural power than the social sciences, especially sociology and anthropology. While I comment at various points on this striking disparity, I agree that it deserves more sustained attention. I also agree that my broad-brush sketch of the tacit “division of jurisdiction” between social scientists and biologists in the last decades of the twentieth century—which enabled social scientists, in effect, to claim exclusive jurisdiction over race as a social phenomenon with no foundation in biology—gives short shrift to differing views on race among social scientists and among biologists.

Morning’s second set of comments addresses what she calls my concluding “manifesto” for a constructivist response to the new objectivism and naturalism. “Manifesto” is probably too grand a characterization of my concluding remarks, but I’m again grateful for Morning’s sympathetic critique. Here Morning’s main point—and it is a very important one—is that the “broader public conversation cannot be overlooked” when social scientists address the ways in which biological processes shape social life. In a public context in which sweeping—and often recklessly misleading—claims are made about genes, social scientists must indeed be cautious, vigilant, and rigorous both in our research and when we join the public conversation.

Philip Gorski’s commentary addresses my chapter on “Language, Religion, and the Politics of Difference,” conceived as an homage to Aristide Zolberg. Unlike the lengthy and analytically dense chapters addressed by Schmidt and Morning, written specifically for this volume, this is a much shorter chapter that remains close to its original form as a public lecture. I would be the first to acknowledge that the argument reflects the limitations of that genre: it is insufficiently qualified and nuanced, and as I noted previously, it would have been strengthened by a more reflexive stance toward its central categories of analysis.

That said, I believe Gorski fundamentally misreads my argument. Most basically, he ignores the fact that the chapter concerns the politics of religious and linguistic pluralism, not the politics of religion and language in general. As a result, Gorski’s main critical comments—especially those concerning the power-laden nature of language and the politics of language in a broader sense—do not bear on my argument. The same holds for his discussion of broader Durkheimian and phenomenological understandings of religion and his observation that even the most avowedly secular polities cannot do without their forms of sacrality, transcendence, or collective effervescence. I fully agree with Gorski’s observations, but they are irrelevant to my argument. Gorski simply does not engage the chapter’s central argument that religious pluralism tends to be more robust than linguistic pluralism in contemporary liberal societies and polities, reflecting the differing ways in which religious and linguistic pluralism are generated, reproduced, and institutionalized in these settings, and that religious pluralism gives rise to difficult and intractable problems of “deep diversity” that are not occasioned by linguistic pluralism.