I am grateful to the Archives européennes de sociologie for organizing this symposium and to my three critics for providing sharp arguments and fresh vistas in response to my warnings about “The Trap of Racial Capitalism” [Wacquant Reference Wacquant2023], this issue. This gives me an opportunity to clarify the key points I made in that article and the spirit in which I made them. In nuce, I am sympathetic to both the scientific and the civic thrust of the construct; only I find that its proponents have so far failed to make a case for its theoretical robustness and also failed to deliver the novel empirical goods promised.

This is a hot (political) topic and so it is important that cool (analytical) heads prevail. As W. E. B. Du Bois [(1901) Reference Du Bois, Green and Driver1978: 253] intimated,

In this text, I respond to each rejoinder in turn before suggesting an agenda that is in my view a necessary prolegomenon to a theory like “racial capitalism,” which is to recognize and theorize the varieties of regimes of racial domination in history, much as comparative political economists have taught us to acknowledge and dig into the varieties of capitalism. I will then sketch a bare-bones schema to decouple capitalism and race analytically so that their historical conjunction may be studied empirically.

Bhambra and Holmwood [2023] give a surprisingly dour and harsh reading of my position. My answer to the question of the viability of racial capitalism as an analytical construct is not so simple and “forthright” as an outright rejection, as they claim. The epigraph by Wittgenstein with which I open my article does leave open the possibility that a problematical word can be cleaned and, as the Austrian-born philosopher phrased it, “put back into circulation.” Mine is a modest contribution to the “organized skepticism” that Robert Merton [Reference Merton1973] showed is at the core of social science qua science. I do not casually jettison the concept; I voice concerns, albeit serious; I point at ambiguities, indeed numerous; I sound warnings, made all the more urgent by the parallels that I see between the scholarly bandwagon of “racial capitalism” and that of the “underclass” forty years ago when students of race and poverty in the metropolis were convinced that they had forged a novel concept—it was not novel—and discovered a new group—it was a rhetorical illusion, a figment of the class and racial imagination, lay and scholarly [Wacquant Reference Wacquant2022a].

I invite the researchers who use the notion to devote refreshed efforts to clarifying and articulating it, because I find it direly wanting on grounds of semantics, logics, and heuristics. Semantics: the two words “racial” and “capitalism” are among the most capacious and polysemous in sociology and history, and their coupling doubly so. Logics: the construct is not free of inconsistencies, ambiguities, and contradictions; it is dangerously underspecified and overstretched—like many epochal concepts, mind you, such as “neoliberalism,” “cosmopolitanism,” or “populism” [Centeno and Cohen Reference Centeno and Cohen2012; Lins Reference Lins Ribeiro2014; Tuğal Reference Tuğal2021]. Heuristics: the value of a concept is not decreed; rather, it has to be demonstrated by stimulating theoretical advances and producing new empirical objects, that is, descriptions, interpretations, and explanations of social phenomena that we could not have fashioned without it. A priori, nothing precludes these conditions being met; but they have not as of this writing, notwithstanding the sprawling literature and debates the concept has spawned. Too many of its users are advocates of “racial capitalism” because it resonates with the particular sociopolitical moment (of American academic and street politics). Their use of the notion is mostly rhetorical, sometimes metaphorical, and all too rarely analytical.Footnote ¹

Bhambra and Holmwood [2023: 165] take me to task for failing to see the elephant in the room of history, namely, what they call “modern capitalism [, which] arises and develops within the global structures of European colonialism.” I find their argument curiously Eurocentric. They write of “modern capitalism” as if Western capitalism were the only one on Planet Earth, and as if capitalism were a single homogenous entity across the European span and through the centuries. But, as Sanjay Subrahmanyam’s [Reference Subrahmanyam2023: infa] contribution indicates, the idea of a unified European civilization is a scholastic concept. Moreover, in their recapitulation of the stages of development of modern capitalism, from “colonialism through private property” to “colonialism as a national project” and, presumably, to contemporary neoliberal or financial capitalism (possibly exploiting the so-called global South), Bhambra and Holmwood commit the fallacy of historical continuity, which consists in assuming that an observed trajectory was the only possible one, thus eternalizing the linkage of capitalism and colonialism.Footnote ²

Consider the case of slavery and settler colonialism in the British colonies of north America in the 17th century. When human bondage was established there, it was a means of recruiting and managing labor akin to serfdom and a novel institution in need of codification, for it did not exist in England. A series of nested decisions had to be made, among them whether slavery would apply for a set number of years or durante vita; whether the status would be inheritable and through what filiation; and what rights and obligations the slaves and the masters would have. There never emerged a single body of “slave law” applying uniformly across jurisdictions [Shute Reference Shute, Drescher and Engerman1998] and, for over half a century, the conditions of African slaves and European indentured servants were virtually indistinguishable [Kolchin Reference Kolchin1990].

More decisively still, the merging of enslavement and ethnoracial division was not inexorable, for the latter did not precede or cause the former. Rather it was slavery that spawned race, when it became strategic for the masters to divide their workforce and prevent rebellions by the “giddy multitude” [Morgan Reference Morgan1975]. It took until the revolutionary period for the construct of blackness as social and moral inferiority, transmitted through strict hypodescent, to diffuse and solidify [Fields Reference Fields1990]—a construct virtually unique in the world for its symbolic rigidity and social consequentiality. So there was nothing ineluctable in the association of blackness and human bondage.Footnote ³ Had the British not established naval supremacy in the 1670s and had the material condition of the English working class not improved at that time, the transatlantic slave trade would not have boomed and English indentured servants would have remained the more profitable and thus the main source of labor in the North American colonies, changing the historical cast of their economy, social structure, and culture. No iron law of capitalism mandated its association with enslavement.

The same applies to the linkage of race and capitalism in the Jim Crow South. Caste terrorism was not the ineluctable outcome of abolition but the contingent result of patterned struggles aiming to redraw social and symbolic space in the post–Civil War South. In the aftermath of abolition, the former slaves seized on and actualized new historical possibilities in both family, religion, work, unionism, and politics. As Du Bois shows in Black Reconstruction, 1860–1880 [(1935) Reference Du Bois2017], the outcome of the “three-cornered battle” between planters, ex-slaves, and poor whites was not preordained.Footnote ⁴ Moreover, Reconstruction under federal authorities could have been extended and entrenched, creating a different racial playing field. The fifteen years it lasted were indeed, in the words of historian Eric Foner [Reference Foner1988], an “unfinished revolution.”

So much to say that racial capitalism and kindred concepts erase historical variations, interludes, and contingencies to replace them with monolithic depiction and mechanical necessity. Moreover, we cannot assume that racial division, colonial or metropolitan, is functional to capitalism across all lands and epochs. We should take Churchill’s quip about the benefits of colonization for Britain, cited by Bhambra and Holmwood [Reference Bhambra and Holmwood2023: 170], with a grain of salt, for it is not so clear that industrial capitalism at the core benefited from colonial predation at all times and in all places. In France too, the postwar political and economic elites were convinced that France’s imperial holdings in Africa and Asia were essential to the material welfare of the nation, but in fact, the opposite was true, as economic historian Jacques Marseille shows in Empire colonial et capitalisme français. Histoire d’un divorce [Reference Marseille2015]. Until World War I, France’s colonies were a benefit to the metropolitan economy by providing cheap raw materials and an outlet for its commercial products. But, from the late 1920s to the late 1950s, the colonies were a drain on French capitalism and it took decolonization for the country to modernize its industrial sectors and compete effectively on the European and world markets. Much as slavery and then Jim Crow rule arrested industrialization and urbanization in the South of the United States, crimping US capitalist development in the process,Footnote ⁵ France’s empire was first a boon and then a bane. Moreover, after European powers had relinquished their colonies, capitalism across the continent thrived. Does this not invalidate the proposition that “the history of capitalism is determined by colonial processes and practices” [Bhambra and Holmwood Reference Bhambra and Holmwood2023: 175], without further qualification and periodization?

Bhambra and Holmwood, curiously, read my nationality into my text, speculating on “the distance that Wacquant wishes to put between north America and France in terms of current political realities.” First, I would make the same argument if I were American, Vietnamese, or Uzbek: displacing the United States from its Archimedean position in the global study of race is a matter of epistemic salubrity [Wacquant Reference Wacquant2022b]. Second, how I personally view the “current political realities” of France (about which my article contains not a single word) is irrelevant to my evaluation of a concept forged by South Africans that has traveled to the United States and back and forth [Levenson and Paret Reference Levenson and Paret2022] and is now spreading around the Anglophone world. Third, I mention Gérard Noiriel’s study The French Melting Pot, not because it is about France, but because it is a model case study of the mechanisms whereby working-class formation can submerge ethnoregional and ethnonational identities. I could just as well have invoked studies comparing Europe and the United States [Katznelson and Zolberg Reference Katznelson and Zolberg1986], or Germany and Italy [Dipper Reference Dipper2020]. The point here is that the historical record is full of periods and episodes when class trumps race as a principle of social vision and division. Not because there is some ontological primacy of the one over the other, but because that primacy is the outcome of symbolic struggles waged in social space and the field of power [Bourdieu Reference Bourdieu1984].

In his contribution to the symposium, Sanjay Subrahmanyam [2023] recaptures the historicity of racial denominations and economic formations that are glossed over by the racial capitalism approach as well as by Bhambra and Holmwood’s [2023] monolithic view of “modern” capitalism and colonialism. He rightly warns us against the presentist fallacy which consists in retroprojecting current ethnoracial classifications and capitalist constellations onto the distant past and portraying that past as a necessary prelude to the irrevocable singularity of our present. I am thrilled by his contribution because he makes two moves that are central to my book Racial Domination [Wacquant Reference Wacquant forthcoming]: first, to recover the architectured variability of both race and capitalism, against the structural teleology of the more orthodox advocates of racial capitalism (and which is encapsulated in the analysis of contemporary society by kindred notions such as “structural racism,” which is equally conflationary); second, to move beyond the Euro-Atlantic world to recover the forms of ethnoracial division that are current in the Asian-Pacific domain as well as operative among the colonized themselves.

However, much as I am impressed by Subrahmanyam’s command of the rhapsodic variety of capitalistic and racialized forms in and beyond the Western Atlantic domain, which is in full evidence in his dazzling book Europe’s India: Words, People, Empires, 1500–1800 [Subrahmanyam Reference Subrahmanyam2017], I am concerned that his historicist account, plucking facts across wide expanses and epochs, can easily slide into the ideographic affirmation of particulars. I find it indispensable at this juncture to inject nomothetic discipline by forging ideal types à la Weber [(1904) Reference Weber, Shils and Finch1949: 90], “mental constructs” (Gedankenbilder) that can guide historical comparison by helping us to “determine in each individual case, the extent to which this ideal construct approximates to or diverges from reality.” Here we can learn from the lively debates on the “varieties of capitalism” triggered by the germinal work of John Hall and David Soskice [1981], based on the thesis that the capitalist economy is not coordinated and governed everywhere in the same way,Footnote ⁶ and recognize the varieties of modes of racial domination in different societies and epochs, manifested by different articulations of naturalizing classification and stratification. This is demonstrated by Orlando Patterson in his little-known yet fundamental essay on “Four Modes of Ethno-Somatic Stratification,” which shows how, the same root population, the descendants of African slaves, were dispersed across the Atlantic and were differentially incorporated in the United States, Latin America, the Caribbean, and northern Europe, “each mode refer[ring] to a unique configuration of ethno-racial ideology, ethno-demographic mix, ethno-class stratification, and level of societal racialization” [Patterson Reference Patterson, Loury, Modood and Teles2005: 167].

But how do we tame the potential proliferation of types of racial regimes that has beset the literature on varieties of capitalism? Inspired by the distinction that Moses Finley [Reference Finley, Sills and Merton1968] establishes between “genuine slave societies,” in which slavery is epicentral to the economy, social structure, and polity, and “societies with slaves,” in which human bondage exists but is marginal to the institutional order,Footnote ⁷ I submit that we need to make a germane distinction between “genuine race-divided societies” and “societies with race.” This allows us to differentiate profiles of racial rule along key dimensions—the organization of the economy, the structure of social space, the makeup of the state, the symbolic constitution of the public sphere and subjectivity—rather than lumping them together under an over-capacious and vague notion of a “racialized social system,” defined as “societies in which economic, political, social, and ideological levels are partially structured by the placement of actors in racial categories” [Bonilla-Silva Reference Bonilla-Silva1997: 469].

A social formation that embodies the “genuine race-divided” type in history is South Africa under apartheid, where no corner of economy, society, state, and self escaped race effects, and where the imposition of racial order was overt, blunt, and systematic [Seidman Reference Seidman1999]. Contemporary Norway hews close to the pure type of a “society with race,” notwithstanding the long-standing marginalization of the Sámi and the recent growth and differential treatment of a distinct population of non-Western immigrants and refugees [Midtbøen Reference Midtbøen2018; see Hübinette, Lundström, and Wikström Reference Hübinette, Lundström and Wikström2023 for a similar argument about Sweden]. Much like it obliterates fundamental distinctions between contemporary forms of capitalism, the concept of racial capitalism gives us no purchase on the structural, functional, and ideational differences between these two types of regimes of racial domination.

Instead of treating capitalism and race as two cohesive and unchanging entities that are necessarily intermeshed, then, we are better served by considering them as points along two continua that are analytically decoupled. The varieties of capitalism can be reconceptualized along the axis of commodification (of land, labor, money, and core public goods). The varieties of regimes of ethnoracial domination can be reconceptualized along the axis of racialization (the extent to which classification and stratification are naturalized, Wacquant [Reference Wacquantforthcoming], chapter 1). Particular historical social formations can then be plotted in the bidimensional space formed by varieties of capitalism and varieties of racial rule, as indicated in Figure 1. In the contemporary era, the United States and South Africa are two societies characterized by high degrees of both commodification and racialization; Norway stands in the opposite corner, with high regulation of the economy and high ethnic cohesion. Canada falls into the high racialization–low commodification quadrant, while Argentina belongs to the high commodification–low racialization box. A given country can also travel along each of these dimensions, for instance, by implementing policies of economic deregulation (neoliberalism) or fostering ethnic recognition, affirmative action, and reparations (multiculturalism).

Figure 1 The analytic space of commodification and racialization

Exponents of racial capitalism keep us locked in the top-right quadrant, assuming that all capitalistic societies are fully racialized and all fully racialized societies are capitalistic, missing out on all the variants that are historically and politically relevant. A social science committed to the civic good must be able to acknowledge and capture these variations, so as to identify the best practicable levers of action to reduce the societal distortions and social harm that capitalism and racial domination cause when they join in the particular manner that they do.

Genuine race-divided societies and societies with race: two contrastive ideal types for one analytical agenda. The politics of race demands that we reject that distinction. The analytics of race commands us to affirm it. This is where advocacy and scholarship must find a compromise without invalidating their distinctive missions, or part ways, as the case may be when the pressure of activism and political struggle require blanket concepts and accusatory rhetoric. In any scenario, the political history of knowledge teaches us that social science makes its greatest contribution to the struggle for racial justice when it follows its own epistemic rules and criteria, no holds barred.