Rogers Brubaker is Professor of Sociology at ucla. He has written many path-breaking books and articles on citizenship, ethnicity, and nationalism, including Citizenship and Nationhood in France and Germany, ^{Footnote 1} Nationalism Reframed. Nationhood and the National Question in the New Europe, ^{Footnote 2} Ethnicity without Groups, ^{Footnote 3} and Nationalist Politics and Everyday Ethnicity in a Transylvanian Town. ^{Footnote 4} Grounds for Difference, a collection of seven essays (four of which were previously published in earlier versions as journal articles), is his latest volume. In this book, the level of analysis is steadfastly “macroanalytic” [2] and the main focus is on three recent intellectual and social trends that are currently reshaping the contemporary politics of difference: the return of inequality as a key public concern (Chapter 1); the return of “biological objectivism” [54] in the study of race—in contrast with the “jurisdictional monopoly” (id.) over that subject enjoyed by the social sciences in the last decades of the twentieth century (Chapter 2); and the return of religion as a terrain of intense public conflict (Chapters 3 and 4). The three remaining chapters address various issues pertaining to the relation between nationalism and globalization. The common thread running throughout the book is Brubaker’s cognitivist stance, ^{Footnote 5} according to which ethnicity, race, nation, and religion are conceived as “principles of vision and division” [81] of the social universe, “perspectives on and constructions of the world” rather than “things in the world” [48; author’s emphasis].

In Chapter 1 Brubaker’s goal is to help “reconnect structural sources of inequality with cultural dimensions of difference” (p. 2) by identifying the specific mechanisms through which categories such as ethnicity, race, ^{Footnote 6} gender, and citizenship are involved in the generation and reproduction of inequality. In contrast with Charles Tilly’s more holistic approach, ^{Footnote 7} the operating assumption here is that such mechanisms will vary across categories and forms of difference: “To understand the relation between difference and inequality, […] it is helpful to begin with different kinds of difference” [18, author’s emphasis]. To take but one example, while de facto segregation—whether or not it is externally imposed—is one of the most central inequality-sustaining factors as far as ethnicity and race are concerned, social interdependence, as reflected in the domestic division of labor, is a key element in the perpetuation of gender-based inequality (p. 35). Beyond the specifics of any given case, Brubaker ends up synthetizing the outcomes of his disaggregating strategy through the following tripartition of interacting, difference-linked and inequality-generating processes, put forward as an alternative to Tilly’s emphasis on “exploitation” ^{Footnote 8} and “opportunity hoarding”: ^{Footnote 9} “the allocation of persons to (or their exclusion from) reward-bearing positions; the social production of persons unequally disposed and equipped to pursue desirable positions; and the structuring of positions and their rewards” [42]. He also emphasizes the anomalous resilience of citizenship—externally defined by the state—as a generally unchallenged basis of formal, legally sanctioned categorical exclusion from access to crucial rights and benefits, in contradistinction to the overall delegitimization of direct discrimination based on other ascribed identities. ^{Footnote 10} This, he suggests, stands as the most salient exception to the dominant trend, namely the (slow and uneven) decorrelation of inequality and categorical difference.

In Chapter 2, which is exclusively US-focused, Brubaker assesses the extent to which the resurgence of naturalism fostered by recent advances in human genetics has altered understandings of race and ethnicity in biomedical research, forensic investigation, and ancestry testing. In particular, the fact that some genetic variants “that affect disease susceptibility and drug response […] are found in differing frequencies across socially defined racial categories” [67] has fuelled the systematic inclusion of nonwhites in clinical trials and the breakdown of results by race and ethnicity in the name of equality and justice, ^{Footnote 11} but also newly respectable beliefs in the biological reality of race. According to Brubaker, however, the increased knowledge about the “biological correlates” of race does not transform “that social category into a biological one” [81-82; author's emphasis]. The basic tenet of the constructivist and subjectivist conception of race remains valid: race has come into existence and persisted over time only as a result of human beliefs and practices of categorization and classification generative of social meaning. Yet that conception ought to be broadened and refined in order to take the abovementioned developments into account. Instead of dismissing biological evidence as irrelevant to the social sciences, what is needed is “a biosocial constructivism [that] would attend to the social shaping of biological processes as well as to the biological shaping of social processes” [84]. Beyond this admittedly vague prescription, Brubaker does offer an insightful account of the ambivalent implications of advances in contemporary genetics for the racialization—or deracialization—of American society. On the one hand, by underlining the preponderance of within-group over between-group variation, the heterogeneity of all human collectivities, and the unique nature of every individual’s genetic make-up, such advances may help erode notions of group “purity” or integrity. Thus, the ultimate goal of personalized genomic medicine is to adjust treatment to the specific genome of every individual. It thereby renders the use of race as a proxy for the probability of possessing some medically relevant genetic variants obsolete. On the other hand, however, by seemingly “providing a natural foundation for social identities” [54], geneticization risks bolstering essentialist understandings of racial difference in the short and medium term. This is so even though in biomedical research “the avalanche of ethnoracially organized data generated by the subgroup comparison mandate has guaranteed that many statistically significant findings [of difference] would be generated by chance alone” [57], and in spite of the difficulty of distinguishing genetic causes of health outcomes from social or environmental ones when race happens to be correlated with both. Caution is therefore in order.

In Chapter 3, Brubaker attempts to explain why, secularization theory notwithstanding, religion has now overtaken language as the main site of contention in matters pertaining to the political accommodation of cultural difference in liberal democracies. He does so through an enumeration of the differences between these two purportedly primordial “domains of group-forming cultural practice” [5]. One of them is the fact that religion, unlike language, is not a pervasive, universally necessary medium of interaction in both the public and the private spheres. This might feed the expectation that religious divisions would be less politicized than linguistic ones. Yet, while immigration generates both religious and linguistic pluralism, the former is more likely to be self-reproducing and to persist across generations than the latter. This is so mostly because linguistic reproduction—the intergenerational transmission of minority languages—, unlike religious reproduction, necessitates an extensive educational apparatus provided by the state, a condition that has been met in a relatively small set of countries (Canada, Belgium, Spain, Switzerland, and so on) and only for national minorities (as opposed to immigrant groups). ^{Footnote 12} Moreover, in contrast with linguistic diversity, the bulk of religious diversity does not stem from immigration exclusively; it is also the product of endogenous developments, such as conversions, which can transform the identity of the individuals involved far beyond the usual side effects of language acquisition and in a way that may bolster the politicization of the religious. Last but not least, that politicization is made possible by the lack of homogenizing pressure by the liberal state induced by its constitutive commitment to the ideal of neutrality with respect to religion, while no similar commitment exists with respect to language: “language tests for citizenship are routine, but a religious test would be unthinkable in a liberal polity” [95]. This differential constraint leaves more space for potential conflicts around religion than around language.

In Chapter 4 the author continues his investigation of religion by exploring the relation between religion and nationalism. He identifies and provides illustrations of four potential ways of connecting those two concepts in academic scholarship: religion may be understood as a phenomenon analogous to nationalism; as part of the explanation for the rise, intensity, or distinctive features of nationalism in particular cases; as a component constitutive of nationalism (rather than causally related to it); or as specifying a kind of nationalism distinct from the secular one. This fourth option is critically examined and ultimately rejected by Brubaker. The conception of nationalism as an intrinsically secular phenomenon is defended, and so is the claim—stated in Chapters 4 and 6—that “the territorial nation-state remains the dominant political reality” and “the decisive locus of membership even in a globalizing world” characterized by transnational mobility [115, 143], repeated assertions of its demise notwithstanding.

In Chapter 5, Brubaker offers an incisive critique of “diaspora studies” ^{Footnote 13} and of their tendency to “appl[y] […] the term to an ever-broadening set of cases” [121], thus stretching their focal concept beyond its core constitutive elements (spatial dispersion, “the orientation to a real or imagined “homeland” as an authoritative source of value, identity, and loyalty” [123], and a concern for boundary maintenance allowing for the intergenerational preservation of an identity defined in contradistinction to the host society). This unbridled expansion of the field, he suggests, is made possible only by the unjustifiable combination of the “strong” definition mentioned above and of a “weak” one equating membership in a diaspora with ancestry, the attraction of which lies in the aggrandizement of the phenomenon that it entails. Yet “the “groupness” of putative diasporas […] is contingent and variable” [130]; it ought to be conceived as the outcome of political struggles, an outcome that analysts should avoid prejudging. This requires rejecting the use of ancestry as a proxy for diaspora membership. “Diaspora” is a (normatively loaded) “category of practice” […] used to make claims, articulate projects, formulate expectations, mobilize energies, and appeal to loyalties” [129], a category that is part and parcel of an attempt at transforming the social world rather than describing or making sense of it. The temptation to unreflectively convert it into a category of analysis—and/or to define populations as “diasporas” without any empirical inquiry as to the extent to which their members adhere to the diasporic enterprise—ought to be resisted. ^{Footnote 14}

Grounds for difference is an impressive and thought-provoking volume. All the outstanding qualities characteristic of Brubaker’s work are on display: the book is analytically rigorous, theoretically imaginative, grounded on an extensive and powerfully synthetized literature, and elegantly written. It is required reading for anyone interested in the topics under study. I will only venture one critical observation.

Despite a few references to works of political theory scattered across the text, there is a striking contrast between the inquiry’s remarkably large scope and its implicitly monodisciplinary orientation as reflected in the conceptual apparatus put forward. Of course, that a sociologist should focus on the sociological literature is more than understandable, given the hermetic character of disciplinary boundaries in the US academic field. Yet, in Chapter 1, this restriction arguably leads the author to advance a relatively complex and less than immediately transparent terminology in order to make distinctions that are less original than they seem. Thus, what Brubaker calls “categorical inequality in the […] processual sense”, namely “the allocation of categorically distinct bundles of rewards and opportunities on the basis of ascribed categorical identities” [45] is more commonly known as discrimination. Similarly, all or most of the “four types of processes [that] can produce categorical exclusion from or unequal representation in desirable social positions”, i.e. “formal categorical exclusion; informal yet strictly or largely categorical exclusion; categorically inflected selection; and categorically-neutral screening on category-correlated position-relevant characteristics” [36], go under more familiar names. “Formal categorical exclusion” is legally grounded direct discrimination. “Informal yet strictly or largely categorical exclusion” is direct discrimination with no legal basis. The same goes for the somewhat mysterious “categorically inflected selection processes”, of which there are two varieties: “First, gatekeepers may hold conscious beliefs—correct or incorrect—about average group differences in position-relevant characteristics. Selection processes are categorically inflected to the extent that gatekeepers’ decisions are based not only on their assessments of observed individual characteristics but also on their beliefs about average group characteristics, taken as a proxy for unobserved individual characteristics. Second, unconscious category-linked associations may bias gatekeepers’ assessments of individual characteristics” [37; author’s emphasis]. The first variety is strictly identical to “statistical discrimination”, a notion that has been one of the main building blocks of the theory of discrimination in economics since the early 1970s ^{Footnote 15} and around which philosophical ^{Footnote 16} and legal ^{Footnote 17} reflections have gravitated as well. Surprisingly, that notion does not appear in Grounds for Difference, despite the fact that it is used by some sociologists. ^{Footnote 18} The second variety is called “implicit bias” by social psychologists and is also the object of a sizeable literature, ^{Footnote 19} the existence of which remains unacknowledged. Finally, even beyond the set of discrimination-centered approaches, Brubaker does not mention philosopher Elizabeth Anderson’s masterful study, The Imperative of Integration, ^{Footnote 20} another major—and less abstract—attempt at disaggregating the reproduction of (in that case, racial) inequality by mapping out the interactions between three generative processes (segregation, stigmatization, and discrimination), an omission all the more startling as Anderson also uses Tilly’s work on “durable inequality” as one of her starting-points. None of this should be taken to suggest that Brubaker’s classification has no comparative advantages over Anderson’s alternative framework or a discrimination-focused typology. That the author does not attempt to demonstrate the existence of such advantages is regrettable nonetheless.