Race,Capitalism,and the Necessity/Contingency Debate

Abstract

Interest in the relationship between race and the expanded reproduction of capitalism has exploded across the social sciences and humanities over the past several years. Despite this widespread appreciation and interest, profound disagreement, debate, and analytical impression persists, not least regarding the relationship between race and the necessary ‘laws of motion’ of capitalist society. This article begins by tracing the core approaches to the race and capitalism conversation, paying particular attention to their understanding of the necessity/contingency distinction. It then proceeds to make the case for race as a contingent – which, emphatically, does not mean local or insignificant – relatively autonomous, and historically path-dependent terrain of struggle in capitalist society, which has largely functioned to maintain capital’s necessary disequilibrium between the value form and its value relations, but need not do so. It closes by exploring the implications of this claim in relation to recent historical-geographical research on post-1898 US imperialism.

Few academic concepts have seen their stock rise over the past ten years quite like ‘racial capitalism’. Which is not, of course, to say that the origins of ‘racial capitalism’ are only a decade old. As Peter James Hudson recalls in a recent intervention, we can trace the term’s movement from a 1979 essay, published anonymously under the heading ‘Neo-Marxism and the Bogus Theory of “Racial Capitalism”’, through to Cedric Robinson’s 1983 magnum opus Black Marxism, and beyond (see Hudson, 2018); and we might even go further back still, to the ‘old history of capitalism’ – as Jenkins and Leroy (2021) call it – and suggest that the work of scholars like WEB Du Bois, Eric Williams, Oliver Cromwell Cox, and Walter Rodney theorized racial capitalism avant la lettre. Still, it was surely not until the 2000s – and the aftermath of the financial crisis of 2008 in particular – that ‘racial capitalism’ truly took off within and across the humanities and social sciences. In the years since, the term has become ubiquitous on the American left, as has the fundamental claim behind its usage. That capitalism is racialized seems to require little in the way of theoretical or empirical verification; it is increasingly common sense, legitimated with gestures toward well-worn slogans: after all, ‘capitalism requires inequality and racism enshrines it’; ‘race is the modality through which class is lived’; and so on. Indeed, Robinson himself is often invoked in these contexts to point out – as if it were a self-evident matter of fact – that capitalism is best conceived as racial capitalism precisely because capitalist class formation was/is predicated upon forms of racialist thinking and distinction, which ostensibly find their root within European feudal society and a Western racialist metaphysic.

But is the meaning and significance of racial capitalism really so straightforward? As a number of skeptics have recently begun to point out, there are more than a few open questions and tensions in the racial capitalism discourse (see Go, 2021; Ralph and Singhal, 2019). Here, we might simply offer a few questions, among many, that are seemingly still up for debate, even among otherwise sympathetic interlocutors: What exactly do we mean by ‘race’ in the context of racial capitalism, and what are the parameters of ‘racialist thinking’? What do we mean by ‘capitalism’, and when, where, and how did it begin? Why should we refer to capitalism as racial capitalism if race is merely one of the modes of social differentiation – alongside gender, sexuality, and so on – that structures the dynamics of accumulation? What are the historical and geographical parameters of racial capitalism, and how does the concept travel outside of the United States? Does racial capitalism conceptually let capitalism pure and simple off the hook, by shifting the blame for its brutality onto an ostensibly exogenous factor (namely, racial oppression)? Are our existing theories of capitalism so deficient that we need a ‘new’ theorization, under the mantle of a new heading? In short: what is racial capitalism and what does the term provide, conceptually and politically?

The objective of this intervention is surely not to offer a response to each of these concerns (some of which require a much more extended discussion than is permitted here). What I have in view is both more modest and more meta-theoretical. My intention – to begin – is to establish the ‘state of the field’ on the problematique of race and capitalism; to draw out and clarify the dominant strands of thinking on that relation (particularly in the US academy); and, more specifically, to sketch their positions vis-à-vis what might be called the ‘strong program’ on racial capitalism – or the claim that capitalism is necessarily racialized. (Throughout this text, I presuppose that necessary relations are those in which ‘the nature of the relata depends on the relation’ (e.g. ‘the institution of marriage is a necessary condition for the occurrence of weddings’). Contingent relations, by contrast, can logically exist independent of each other (see, for context, Sayer, 1985: 49).) Put otherwise, I am concerned to ask: how does the broad, critical literature on race and capitalism understand the distinction between necessity and contingency within capitalist society – and on what basis? In brief, I will suggest that this broad literature – including that which makes explicit claims to the contrary – remains plagued by a persistent failure to adequately establish what exactly is contingent and what is necessary in capitalist society, or to identify how race functions once we do establish the contours of that distinction (see, for context, Go, 2021).

Moreover, and in light of this lacuna, this paper will develop a distinctive approach to the necessity/contingency question, which (1) renders race as a logically contingent – which, emphatically, does not mean local or insignificant (see Sayer, 1991) – relatively autonomous, and historically path-dependent terrain of struggle in capitalist society, which has largely functioned to maintain capital’s necessary disequilibrium between the value form and its value relations in the longue durée (see Moore, 2015); and which (2) emphasizes the importance of the necessity/contingency distinction if we are to contend with the uneven and combined development of often highly racialized capitalist conjunctures, and the variegated articulations of racialization that we find across distinct capitalist social formations (see, for context, Makki, 2015; Peck, 2019). Finally, this article will close by sketching what such an approach illuminates in the context of concrete historical-geographical research and reading, drawing on recent work on the historical geographies of post-1898 US imperialism.

Mapping Race and Capitalism

Somewhat schematically, we might suggest that there are three prominent approaches to the relationship between race and capitalism operative on the American left, and in the American academy, today. ¹ The first tendency effectively rejects the claim that capitalism and racism are necessarily related. In this view racism and capitalism are largely presented as two discrete social systems (see, for context (if not agreement on that claim), Robbins, 2019; Walzer, 2020). At times these systems are said to intersect and to function in the interests of capital: racism can be used to justify super-exploitation, to break down working class solidarity, and/or to lower the general wage. But it need not, and often does not. Indeed, racism often works against the interests of capital – these theorists argue – disrupting the smooth circulation and realization of value. And it is this fact that helps to explain why particular forms of ‘anti-racism’ can operate as ‘bourgeois class politics’; why (manifestations of) that discourse can operate as a strategy of capital. Michael Walzer, for example, articulates nearly every aspect of this approach in a few short sentences. ‘[C]apitalism and racism have to be analyzed separately’, he maintains. ‘They overlap sometimes, as they do today in the United States. But the overlap is circumstantial, not necessary’ (Walzer, 2020: n.p., emphasis added).

Crucially, however, several of the key scholars that we might associate with this approach are far less direct, obscuring the ontological and epistemological foundations at the root of their thinking. For example, the political scientist Adolph Reed Jr. is perhaps most readily associated with this perspective. His current understanding of hegemonic ‘race reductionism’ (Reed, 2020) – which incentivizes a singular focus on racism at the expense of class inequalities, and to the benefit of certain segments of the ruling class – inevitably suggests that racism and capitalism can be logically and historically disaggregated; it suggests that contemporary anti-racism could conceivably lead to a non-racist capitalism – even if racism emerged from within capitalist society, and is spatially and historically coextensive with it (see also Benn Michaels and Reed, 2020). Reed Jr. has also given readers good reason to assume that, for him, ‘class’ is an ‘autonomous phenomenon’, even within highly racialized socio-spatial conjunctures (Singh and Clover, 2018). Put differently, Reed Jr. has seemingly presupposed what we might call an ontology of atomistic association – not unlike many theorists of ‘intersectionality’ (see, for context, Harvey, 1996, ch. 2; McNally, 2017) – such that ‘race and class [appear as] distinct’, and we might even say discrete ‘social phenomena that can merely be concatenated, like peanut butter and jelly’ (Singh and Clover, 2018: n.p.). And yet, Reed Jr. has also insisted that he maintains an antagonistic relation toward the kinds of theoretical abstraction that would allow for distinctions to be drawn between the necessary and the contingent in capitalist society, in the first instance – a position that he first developed two decades ago in the context of debate with the political Marxist Ellen Meiksins Wood (see Conroy, 2022a; Reed, 2002, 2018; Wood, 2002).

A second, highly influential approach to the problem of race and capitalism today bears the obvious marks of both Althusser and Gramsci. Broadly speaking, this strand attempts to make the social totality visible as a ‘complex – not a simple – unity’ (cf. Mitchell, 1966: 16); or, to demonstrate that racism and capitalism (or race and class) are not simply two separate spheres or systems, but rather mutually constituted distinctions within a shared unity. This approach follows closely the work of Stuart Hall (as it developed, in highly original and complex ways, in the 1980s), and it is often deployed to suggest that these ‘distinctions’ within the whole maintain relative autonomy (see, for context, Althusser, 1962; Chari, 2017; Haider, 2021; Hall, 1986a). Another way of putting this is to say that this literature builds on Hall’s rejection of studies that render society as a ‘simple expressive totality’ in ‘which every level of articulation corresponds to every other, and which is, from end to end, structurally transparent’ (Hall, 1986a: 10). And it argues that the social totality is comprised of many moments and contradictions – of many partial totalities – none of which is directly reducible to, or independent from, the others. One must attend to historical specificity in order to understand the conjunctural placement of racism vis-à-vis capitalism (and/or to grasp the uneven penetration of racism across all levels of the social formation: the ideological, political, and economic), even if each part of the whole is understood as reciprocally co-evolutionary (Hall, 1986a; see also Hart, 2007; Omstedt, 2021: 1211).

Importantly, while many do not, some theorists working in this literature do make explicit claims regarding the non-necessity of racism to capitalist society (see Hall, 1986a, on levels of abstraction; see also, Hart, 2007). For Michael Dawson and Emily Katzenstein, for example, white supremacy and capitalism are ‘articulated systems of domination rather than [. . .] intrinsically linked’ (Dawson and Katzenstein, 2019: 264). Indeed, some argue that ‘no “necessary correspondence” or expressive homology can be assumed as given’ between racism and capitalism, even if particular combinations are preferred by capital, and ‘sedimented and solidified by real historical development over time’ (cf. Hall, 1980: 330). The task is to uncover why racism is available and amenable to particular articulations within distinctive conjunctures (given its logical non-necessity), while also underscoring that these concepts cannot be conceived of ‘in privileged isolation’ within those contexts (Dawson and Katzenstein, 2019: 264). Hall, following Lenin, quite helpfully points out that ‘there is no “necessary correspondence” between the development of a form of capitalism and the political forms of parliamentary democracy’, but that this ‘does not prevent us from arguing that the advent of capitalism has frequently (tendentially) been accompanied by the formations of bourgeois parliamentary democratic regimes’; in fact, those regimes might constitute the best ‘political shell’ for capitalist development (Hall, 1980: 330). And yet, distinctions between necessity and contingency receive relatively short shrift in this literature – even among those that recognize that distinction – with much more emphasis placed on the ‘uneven and contradictory ways’ in which racism and capitalism are articulated, and on the movement of distinctive ‘regimes of articulation’ across space and time (Dawson and Katzenstein, 2019: 264; see also, McClintock, 1995).

This brings us to the third tendency on the question of race and capitalism – which I referred to above as the ‘strong program’ on racial capitalism. While both approaches outlined above are often cited in the conversation on racial capitalism, it is only this one which maintains that the latter term necessarily implies the former. For this broad group of scholars, capitalism is understood to require hierarchization, and racialization is understood as integral – indeed, posited as necessary – to that process (for versions of this claim, see Federici, 2004; Mies, 2014). Crucially, however, what exactly it means to suggest that capitalism requires racialization remains up for debate. For some, this seems to simply suggest that capitalism depends upon racialization to ensure intra-class competition, which limits class solidarities and undercuts the cost of labor in general (see, for context, Taylor, 2016: ch. 7). For others, the claim is that racialization has been (and continues to be) integral in the formation and reproduction of labor markets – to the instantiation of the distinction between owners and producers, and distinct classes and/or ‘classes of labor’ (see, on ‘classes of labor’, Bernstein, 2010). And yet, for others still, racialization is understood as central to the history of capitalism in that it functions to crystallize the distinction between expropriable workers and ‘merely’ exploitable ones (see, for context, Fraser, 2019).

Of course, Cedric Robinson’s landmark Black Marxism would seem to fall squarely into this third tendency; he is the writer most responsible for the popularization of ‘racial capitalism’ in the United States, and the one most readily associated with the claim that capitalism is necessarily racist. And yet Black Marxism contains critical ambiguities, not unlike much of the work in this strand. Not only does Robinson largely fail to coherently theorize ‘race’ itself in that text (to adequately distinguish it from other forms of social ascription and hierarchization) (see Go, 2021, and endnote 1), but he also largely fails to specify if the relationship between race and capitalism is an enduring historical one or a ‘logical’ one. In Black Marxism we find a capitalist world – one which emerged from European feudal society – that is saturated from end to end by variegated forms of racism; but it is hard to say whether or not capital requires them. According to Robinson, ‘[a]s an enduring principle of European social order, the effects of racialism were bound to appear in the social expression of every strata of every European society no matter the structures upon which they were formed’ (Robinson, 2000: 28). Indeed, if we can intuit an answer to that question, it seems to be that capitalism is necessarily racialized because it is a manifestation and (geographical) extension of the tenets of the ‘European social order’; capitalism is nested at a level of generality below that level, and thus carries with it the racialism that attends to all European social orders, even if racism takes on a distinctive capitalist form. This claim is a difficult one insofar as it forces us to suggest that capitalism – regardless of its contemporary geography and historical roots – is a strictly European social order (cf. Moore, 2015; Wallerstein, 2004). It also remains largely unsubstantiated in Robinson’s text. Still, what is perhaps most important – politically and conceptually – is that it leads quite directly, as per Ollman’s work on levels of generality (Ollman, 2003: 93, emphasis added), to the notion that ‘the abolition of capitalism will not do away’ with racism per se, but “only with [its] capitalist forms’. ²

The Problem of Contingency and Necessity

If Robinson fails to precisely distinguish between what is necessary and what is contingent in capitalist society, he is surely not alone. This seems to be a failure that plagues each of the tendencies identified above; that is, despite strong assertions that imply a grasp of that distinction. In the context of the literature within the ‘strong program’ on racial capitalism, beyond Robinson, it is simply unclear how exactly we can claim that racism is necessary to capitalist society (either as a means of dividing the working class, enabling expropriation and dispossession, or for any other reason) (see, for context, Camp et al., 2019; Jenkins and Leroy, 2021; Melamed, 2015; Raine, 2019). Of course, it is certainly the case that the history of global capitalism has been deeply racialized – and it remains so today. Racism has quite plainly been instrumental in producing, justifying, and ‘legitimizing’ each of those procedures named above (see Táíwò, 2021). And yet, the claim that capitalism is necessarily racialized would seem to require some variation on the following line of argumentation: capitalism requires expropriation, dispossession, and/or working-class stratification, and race and racism (must) name the process of social ascription through which that occurs. This position hardly seems defensible: it is an obvious form of ex-post facto functionalism in which race is read solely in reference to its historical ‘role in securing the sociospatial conditions for capital accumulation’ (cf. Brenner, 2019: 61; see also Fraser, 1998, on functionalism in debates on redistribution, recognition, and heterosexism). Such an argument involves beginning with an understanding of capital and its requirements, and then reverse engineering a definition of race so as to make it conform to a particular functional role vis-à-vis accumulation.

And this kind of ambiguity and confusion is not only found in the literature that has embraced the language of racial capitalism, in various ways. In the case of work of a dual systems proclivity (‘tendency one’ in the outline above), we find an ostensibly clear sense of capitalism as a bounded social system with particular logics, distinct from racism and processes of racialization. But of course, if that is the case, one would presume that these authors provide a clear sketch of what exactly is distinctive about capitalism – a sketch of its core conceptual components and necessary features. In much of this writing, this is largely – if not entirely – unclear, perhaps due to the enduring conviction among its core proponents that we must reject ‘elaborate ideal-typical formulation’, tout court (see, for context, Reed, 2002). Meanwhile, in work that proceeds in an Althusserian and Gramscian vein, we find a different manifestation of the same theoretical problem. While this strand within the debate on race and capitalism provides a compelling commitment to conjunctural analysis, and to the complex combinations that might define any historical-geographical moment, we are left with either (1) a complete obfuscation of the necessity/contingency distinction (and thus, perhaps, with the sense that race and class are equally necessary to capitalist social formations); or (2) with the sense that racism is a non-necessity to capitalist society, but without a clear sense of the necessary ‘laws of motion of the capitalist mode of production’ – the ‘major goal of Marx’s investigative effort’ (Ollman, 2003: 163) – that mediate such articulations and combinations. (Hall (1986a) alludes to these laws but does not engage them directly.)

Of course, many would suggest that establishing such a distinction between the necessary and the contingent is futile, if not pedantic. It is often said by contemporary scholars of race and capitalism that ‘actually existing’ capitalism is, and has always been, racialized (in some form or another), rendering any attempt to disaggregate those two concepts as politically and/or intellectually misguided (see McNally, 2017). But this is a perplexing claim on several levels. For one, this kind of abstraction – the parsing of the distinction between the necessary and the contingent (see Sayer, 1981) – is absolutely central to the project of historical-geographical materialism. Abstraction is an indispensable – and inescapable – moment in social investigation, and its function is ‘precisely to differentiate the essential, primary, or necessary properties of a particular phenomenon from its superficial, secondary, or contingent elements’ (Brenner, 2019: 40). The ‘salient question’ is thus not whether an abstraction is a ‘partial, one-sided depiction of a constitutively multifaceted, overdetermined social world’, but rather, ‘whether the specific kinds of conceptual abstractions proposed . . . offer a “practically adequate” basis on which to demarcate, in analytically precise yet situated, historically determinate, politically informative terms, the constituent properties’ of the phenomenon under investigation (Brenner, 2019: 40–41). And perhaps even more importantly, this precise abstraction – the parsing of the relationship between racism and capitalism, specifically – is at the very core of many contemporary political struggles and debates, whether their participants are aware of it or not. It is at the very core of widespread and hotly contested discussions regarding, inter alia, what exactly constitutes an anti-capitalist and/or anti-racist politics (and to what extent those categories overlap); regarding whether or not a non-racist capitalism is possible; and regarding what a future post-capitalist society might look like (see, for context, Butler, 1998; Fraser, 1998; Hartmann, 1981; Mills, 2017; Walzer, 2020).

In pointing toward these issues I am, to an extent, echoing the recent work of Julian Go, which has raised the problem of necessity and contingency in the literature on racial capitalism (narrowly conceived in relation to tendency three outlined above). In his view, this literature obscures the distinction between a theory of capital – which offers a ‘formalized and abstract representation of the inner workings of capital, its accumulation, its contradictions, and its necessary demise through a series of central categories that capture the key elements of the capitalist system’; and a theory of capitalism, which ‘refers to capitalist development and dynamics in their empirical specificity’ and which is ‘meant to describe specific capitalist formations and developments as they really exist in the world’ (Go, 2021: 42). Crucially, according to Go:

A theory of capitalism might demonstrate that race has been historically necessary for capitalist accumulation by reference to empirical reality: historically, capitalism and race have always been intertwined. But the claim that race is a logical necessity to capitalism would have to derive from a theory of capital, not from the empirics alone [. . .] On this score, the arguments for the logical necessity of capitalism’s entanglements with race fall short. (Go, 2021: 43, emphasis added)

This intervention helpfully cuts through the noise. And I would add that it is clarifying not only in relation to work that self-identifies with the concept of racial capitalism, but in relation to the much broader conversation on the problematique of race and capitalism, which in general fails to establish which of its analytical concepts are devoid of ‘social content’ (and as such can be applied across capitalist historical geographies), and which are not (Go, 2021: 42). Nevertheless, it seems that we must go further still. That is, while Go quite forcefully submits that racism cannot be conceived as logically necessary in a theory of capital, he does not provide a complete account of how exactly we should theorize that relationship. For example, he suggests that we might more broadly claim that modes of social differentiation ‘of various types’ are necessary to capital. As he points out, such a capacious claim would allow us to suggest that ‘expropriation is logically necessary for exploitation, which is in turn necessary for capital accumulation, and expropriation requires differentiation among workers’ (Go, 2021: 44). But Go fails to fully develop this conceptualization, moving on to another possible theorization of the race/capital relation based on a reading of the work of Dipesh Chakrabarty. We are thus left with two observations, against the backdrop of work on race and capitalism, and Go’s recent intervention: first, we now know that this broad literature largely fails to coherently distinguish between the necessary and the contingent in its approach to the problematique; and second, that race and racism can hardly be conceived as (logically) necessary to capitalism, even if ascription and social differentiation might. What remains unanswered is how we should pursue the task of coherently integrating race and racism into a theory of capitalist society, attuned to the distinction between necessity and contingency at the highest level of abstraction.

As such, I want to suggest not only that race is a logically contingent phenomenon in capitalist society, but – more specifically – that it is a contingent, relatively autonomous, and highly path-dependent terrain of struggle (re)produced in dialectical relation to the abstract level dynamics and requirements of capital. ³ By abstract level dynamics, I am referring to those ‘general features’ that necessarily pertain to ‘capitalism as a mode of production and social system’; this is the highest level of abstraction that ‘provides the basis for examining, in general terms, a number of systemic processes that [by definition] underpin all capitalist social formations’ (Brenner, 2004: 20). Following the work of Banaji (2010: 60), such abstract level dynamics can be said to include: ‘the production and accumulation of surplus-value, the revolutionisation of the labour-process, the production of relative surplus-value on the basis of a capitalistically constituted labour-process, the compulsion to increase the productivity of labour’, and class struggle. And we would do well to add – perhaps most importantly for the conversation on race and racialization – that at this abstract level we can also identify capitalism’s need to maintain the ‘disequilibrium in the value relation of capitalization and appropriation’ (Moore, 2014: 257); or, the fact that, following Jason W. Moore, ‘value as abstract labor cannot be produced except through unpaid work/energy’, and that the ‘“commodification of everything” can only be sustained through the incessant revolutionizing, yes, of the forces of production, but also the relations of reproduction’, which ‘cut across the paid/unpaid work and human/extra-human boundaries’ (Moore, 2014: 262–3, emphasis in original). More simply, at this abstract – necessary – level we can identify capital’s need to maintain a ‘zone’ of extra-economic expropriation greater than its ‘zone’ of capitalization; its need for extra-economic processes that ‘identify, secure, and channel unpaid work [from] outside the commodity system into the circuit of capital’ (Moore, 2015: 17), in order to support the reproduction of the wage nexus.

Of course, in making such a – perhaps contentious – point on the relationship between race and capitalism, it is imperative to clarify what exactly I am saying, and what not. For one, I am suggesting that race is a contingent phenomenon in capitalist society. Race is not logically necessary to capitalism as a mode of production, which is itself a historically unique and contingently formed social structure whose ‘abstract central logic’ is defined by the temporal determination of value (see Postone, 1993; Sewell, 2008: 532). Second, in my use of ‘contingent’ I am not at all suggesting that race is a local, random, or insignificant factor in the history of capitalism. Rather, following Sayer (1991: 293, emphasis in original) – and the tradition of critical realism more broadly (Cox, 2013) – I maintain that ‘contingency and necessity need to be disassociated from scale differences’; there is no relation between contingency/locality or necessity/globality, per se. It might very well be the case that capitalism has only ever been racialized at the scale of the world-system, and thus that racism has been spatially and temporally coterminous with really-existing capitalism since the start. This is a matter of empirical inquiry, and (among other things) a matter of how one periodizes capitalism and its origins, as well as racism itself (see, for context, Arrighi, 2010; Fields, 1990). As Singh puts it (Singh, 2017: 58):

From medical experimentation to crime statistics, debt peonage, labor market manipulation, rent harvesting, infrastructural exclusion, and financial speculation, the racial differentiation that many have considered only tenuously related to the itinerary of capitalism during slavery and its afterlife, has directly produced capitalist value and contributed in an ongoing way to the technical development of capitalism on its alleged frontiers, and new harvests of people separated from land and resources consumed within the web of capital.

Moreover, because race functions within capitalist society (to paraphrase Stuart Hall from another context) on a terrain that has been structured historically (Hall, 1986b: 41), I am suggesting that its content is thus malleable but not entirely up for grabs. Within the really-existing history and present of capitalism there exists a set of clear ‘structural constraints’ on the meaning and significance of race (see Postone, 2004). On a planetary scale the meaning of race is deeply conditioned by the long historical ways in which it has been operationalized according to (and forged in dialectical relation with) the abstract level dynamics sketched above – especially capital’s need for expropriated work to maintain the disequilibrium between the value form and its necessarily more expansive value relations (see Fraser and Jaeggi, 2018; Moore, 2015). For some scholars this kind of observation suggests that contingencies ‘do not necessarily remain so’; or, that processes like racialization can get ‘reworked to become a necessary aspect of capital’s production relations’ (Cox, 2021: 11, emphasis added). I would not go so far. Rather I would simply underscore race and racism’s relative autonomy in capitalist society, and note that contingencies can be made into ‘internal elements of the encompassing social logic of capitalism’ at particular scales and in particular contexts (cf. Harvey and Scott, 1989: 226, quoted in Cox, 2021: 8).

Put otherwise, my contention is that race has historically been a critical means through which capital has identified and demarcated expropriable (as opposed to exploitable) work; it has long been – and continues to be – central to capital’s efforts to stave off crises, and to ensure that the zone of expropriated work continues to outpace the zone of commodification and exploitation (Conroy, 2022a, 2022b; Moore, 2015). (As Moore (2015) notes, without a vast reserve of expropriable work, capital invariably faces a crisis of overaccumulation, which can only be resolved by spatially reorganizing capitalist production and reproduction.) This conceptualization allows for us to avoid the kind of ex-post facto functionalism that attends to much work on racial capitalism, while also noting that the history of capitalism has created a broad, fairly generalizable, path-dependency. It allows us to acknowledge racism’s functional role in maintaining profits across a range of settings, while also noting that the meaning of race itself remains a variegated and ‘politically negotiated [. . .] coalescence’ (cf. Brenner, 2019: 61), which can shift in relation to historically embedded or inherited regulatory landscapes and geo-institutional configurations. This approach insists that the image of a seamlessly coordinated (and transhistorical) relationship between race and the logics of accumulation is misguided – an image that, oddly enough, prevails in much of the literature on racial capitalism, producing a paradoxical kind of economism – even if processes of racism and racialization have in the longue durée ‘generated significant, markedly patterned, cumulative effects’ (cf. Brenner, 2017: 164).

Indeed, a particular strength of the theorization sketched here is that it allows us to grasp these markedly patterned dynamics, without losing sight of the variegated and conjunctural articulations that might define the race and capitalism relationship within distinct social formations. It allows us to see that the necessary dynamics of capital accumulation – those dynamics established above, following Moore (2015) – take shape in and through historical-geographical conjunctures (cf. Castree, 2002; Hall, 1986a). The ‘violent, convulsive, and contested’ movement and territorialization of capital’s laws of motion proceed in and through variegated ‘historical and sociocultural “frictions”’ (Makki, 2015: 491). ⁴ And this is imperative because racism and racialization are, indeed, among those entrenched, embedded, and spatialized contingencies. Contingent patterns of racialization are thus the partial cumulative effect of previous rounds of capitalist metabolism; and they are made and remade – if only slightly – in and through subsequent rounds of accumulation and regulatory restructuring (albeit not in relation to those imperatives alone) (Massey, 1995). With this approach, we are now capable of identifying the dynamics of uneven and combined racialized capitalist development, in which the necessary and contingent elements of any capitalist social formation are reproduced as an ‘inconstant outcome of a moving and unequal matrix of articulations, (inter)relations, and unbalanced interdependencies’ (Peck, 2019: 2). We are able to underscore that the abstract dynamics established above collide with, transform, and emerge through contingent – and often highly racialized – historical-geographical conjunctures.

In saying as much, we are also able to move beyond some of the more tedious debates that plague discussions on race and capitalism. For one, it is now clear that ‘racial capitalism’ need not be removed from our social theoretical lexicon, despite arguments to the contrary (see, for context, Go, 2021; Walzer, 2020). Insofar as that phrase draws attention to the ways in which racism, as a concrete form of social domination, comes to shape and is shaped by the abstract determinations of capitalist society at particular scales, it is a reasonable heuristic. The problem is, simply, when ‘racial capitalism’ is asked to do more analytical work than it possibly can – to provide insight, absent any conjunctural research, onto specific combinations and articulations; or when it is used to suggest that capitalism is necessarily racialized, across all abstractions of extension (including potential future capitalisms). The above model also suggests quite clearly that the phrase ‘racial capitalism’ need not preclude (as is often alleged) other ways of modifying and describing capitalism, such as ‘hetero-patriarchal capitalism’ and the like. ‘Racial capitalism’ can simply operate as a way of attuning our focus to the particular – and, we would add, contingent (again, see Sayer, 1991) – force that race has played in mediating and structuring capitalist dynamics, if not a massively insightful one.

Further still, the above formulation helps to cut through the most recent round of acrimonious political debate on the ‘race vs. class’ problematique. For one, it suggests that the whole debate as to whether anti-racism is anti-capitalist (and vice versa) is wrongheaded: this will clearly depend not only on the content of the political action under consideration, but also on how race is being deployed (or articulated) vis-à-vis the abstract laws of motion of capitalist society in a particular historical-geographical conjuncture. And the above formulation helps to reject the notion that there is an epistemological, political, or moral priority to address the abstract level determinations of capitalism over forms of racist domination. Given the historical malleability and relative autonomy of race, it is quite possible that racist domination will exceed the demands or imperatives of capital accumulation, and that it will demand forms of urgent political action that are more or less dissociated from anti-capitalist politics. In an Althusserian vein, we might suggest that the social totality – as a ‘structure articulated (combined) in dominance’ – can indeed be dominated by logics of racism and racialization, even if the contradiction between the forces and relations of production determines the ‘character of the social totality’ in the last instance (see, for context, Conroy, 2022c; Peet and Lyons, 1981: 194).

Historicizing Race and Capitalism

As should by now be clear, the above distinctions do many things in relation to ongoing debates on race and capitalism. But perhaps one of the most important among them is that they also demonstrate precisely how not to read the history of capitalism. After all, without a sense of the distinctions elaborated above – without a theorization of the relationship between the necessary and contingent in capitalist society – one might (wrongly) read any number of historical texts as evidence for any one of the three theoretical tendencies mapped at the outset, and subsequently superseded. We can turn to the recent scholarship on the historical geographies of US imperialism after 1898 to demonstrate this point. For example, one might read the recent literature on Fordlandia in the 1920s – on, that is, the Ford Motor Company’s effective colonization of the Amazon – as evidence that racism and capitalism are two entirely autonomous, or discrete, social systems; as proof that racism – as a distinctive sphere unto itself – simply functions at times in the interests of capital, and to impede its imperatives at others. After all, Henry Ford’s decision to establish Fordlandia as a veritable capitalist city in the Brazilian Amazon hinged on racialized (and anti-Semitic) fears of civilizational decline in the United States, and on particular fantasies regarding the redemption of an ‘American way of life’. It also proved economically disastrous. As Greg Grandin puts it, Fordlandia ‘had less to do with overcoming and dominating nature’ – less to do with profitability – ‘than it did with salvaging a vision of Americana that was slipping out of [Ford’s] grasp at home’ (Grandin, 2009: 8); Ford ‘deliberately rejected expert advice and set out to turn the Amazon into the Midwest of his imagination’ (Grandin, 2009: 16). Racialist thinking, as a discrete factor in the social totality, simply intersected with and got in the way of rational accumulation in that context – or so it might seem.

Further still, one might read recent work on the post-1898 imperial history of sugar as evidence for the claim that race and class are relatively autonomous concepts of equal necessity in capitalist social formations. After all, Merleaux’s (2015) recent, and highly compelling, work in Sugar and Civilization demonstrates that following the Spanish American War a robust debate emerged in the United States which pitted racist anti-imperialists – who favored strong tariffs to protect European-descended sugar beet capitalists in the West and Midwest – against monopolistic sugar refiners that demanded low duties on raw cane produced in America’s newly acquired overseas imperial territories. Both factions of sugar capital were staking an economic demand. But racial thinking – and racism – mediated their demands in radically different ways. In one case racism facilitated an exclusionist impulse (and a strong support for tariffs); in another it supported an imperialist and paternalist impulse interested in folding racialized ‘others’ into the production process. And thus, we might (albeit wrongly) read this conjuncture as demonstrating the variegated ways in which race and class – as relatively autonomous, and equally necessary dimensions across capitalist social formations – come to be articulated in the context of a complex whole.

Finally, without an adequate distinction between the necessary and contingent in capitalist social formations, one might even read recent work on the US imperial presence in Central America, particularly during the 1920s and 1930s, and find proof for the position that racial distinction functions necessarily as an instrument of cheapening – as a necessary mode of ethico-political devaluation in the service of profits in any and every capitalist conjuncture. After all, as Augustine Sedgewick has pointed out in his work on coffee, beginning in the 1880s in El Salvador a process of liberal land reform facilitated the expropriation of ‘“unproductive” communities’ and the private sale of their land; these ‘uprooted people’ were then ‘consolidated into a workforce for the coffee farms by vagrancy laws enforced by a new arm of the national police’ (Sedgewick, 2015: 320). This meant that by the end of the First World War – after US bankers had successfully ‘picked up’ Salvadoran coffee accounts – the coffee industry in the country was defined by a straightforward and highly racialized imperial relation. Plantation production was defined such that capital was American and white, and labor was neither. And this produced a situation defined by racist violence pursued in the name of accumulation. Starvation, as Sedgewick notes, came to structure the workday on coffee plantations, as planters maintained that hunger itself ‘gave energy’ to their (racialized) laborers, while ‘eating produced idleness’ (Sedgewick, 2015: 324).

But of course, all of the above readings would obscure the critical fact that in each of these historical-geographical moments, racism – as a logically contingent, relatively autonomous, and path-dependent phenomenon – came to be articulated with and transformed through the necessary features of capitalist society. More precisely, in each case we find – among other things – a capitalist terrain necessarily structured by capital’s need to maintain the relative disequilibrium between the value form and its value relations (Moore, 2015). This is simply the broad social matrix that structured the lives of sugar producers in 1910, Ford Motor Company employees in 1925, and coffee plantation owners in 1930. Whether or not these historical actors were aware of such abstract compulsions, their everyday lives – the profitability of their endeavors, the conditions of their labor, and so on – were mediated by them. ⁵ Meanwhile, racism was constituted through, shaped by, and (at times) exceeded the demands of these structural imperatives; racism’s function was to some degree malleable, supporting a range of purposes depending on, inter alia, inherited geo-institutional configurations, the dynamics of social struggle, and other multi-scalar determinations.

Moreover, while these histories demonstrate the relative autonomy of racism within the social totality, the historical geographies of post-1898 US imperialism also demonstrate quite clearly the patterned ways in which racialized subjects are integrated into circuits of capitalist valorization across contexts – as the theorization above would suggest. Indeed, this theorization helps us identify that racism and racial hierarchization have recurrently functioned to produce cheap labor and resources (and thus to guarantee capital’s necessary disequilibrium between the value form and its value relations). Take, for example, Jonathan Soluri’s path-breaking work on the social and environmental transformations wrought by intensive banana production on the north coast of Honduras after 1870. In that context, US fruit companies conspired with the Honduran government to receive railroad concessions that ‘provided the legal means by which [they] established control over vast quantities of resources’ (Soluri, 2006: 43). Not only did these concessions give US companies access to cheap ‘soil, timber, water, and mineral resources in addition to tax and duty exemptions’ (Soluri, 2006: 43), but they also forged a dynamic in which racialized dispossession, super-exploitation, and capital accumulation went hand in hand. And this racialized dynamic only becomes clearer if we attend to the problem of waste in that context. As Soluri has pointed out, the scourge of F. oxysporum – a fungus that significantly curtailed the productiveness of Honduran banana plantations – was exacerbated by the dense planting of Gros Michel monocrops throughout the 1920s. To address this crisis of ‘negative value’ (Moore, 2015), US plantation capitalists came to rely on both the opening up of new plantation geographies – and thus on renewed forms of highly racialized dispossession – and on the hardening and deepening of racialized exploitation within increasingly exhausted monocultural landscapes.

Of course, we can tell this very same story – in which racialized subjects and their resources are devalued and rendered expropriable and/or super-exploitable – many times over in the history of post-1898 US imperialism (see Conroy, 2022d). It is, in some ways, the story of US oil extraction in Saudi Arabia as well. As Jakes and Shokr remind us, building on the work of Vitalis (2007):

in the 1940s and 1950s, American oil companies reconstituted a system of segregated labor in ARAMCO’s oil camps in Saudi Arabia that divided workers according to a taxonomy of racial types. By replicating labor practices that had long existed in extractive industries in the southern United States, these companies developed a greater ability to reduce the costs they paid to mobilize a local workforce. As such, American oil companies depended on the cheapened work of racialized bodies to manage the United States’ most valued strategic asset overseas to produce what would, in the twentieth century, become the world’s most prized commodity. (Jakes and Shokr, 2017: 128)

Indeed, much of the recent historical literature on post-1898 US imperialism has focused precisely on these dynamics of cheapening, and on the functionality of racialization to capital in those terms – whether in the context of the Panama Canal (Greene, 2009), the US dust bowl (Holleman, 2018), the ‘first green revolution’ (Melillo, 2012), or the empire of cotton (Beckert, 2014).

Conclusion

There are many moving parts in this paper, but the synthesis is simple. Against the received literature on race and capitalism we must forge a new theoretical path. We should insist that race is a contingent – which, emphatically, does not mean local or insignificant – relatively autonomous, and historically path-dependent terrain of struggle in capitalist society, which has been deeply structured by capital’s necessary need for cheap inputs (see Moore, 2015). In other words, we would do well to underscore that capitalism’s necessary features invariably collide with, transform, and emerge through contingent historical-geographical conjunctures, which are quite often highly racialized. These contextually embedded norms of racialization – which are themselves the partial result of previous rounds of capitalist metabolism – are combined with capital’s abstract level demands in relatively predictable and functional ways (following Moore, 2015); but they need not be, and they can easily exceed the imperatives of accumulation. The analytical and political task within specific contexts is thus to attend to the relationship between these necessary and contingent elements – to how they are articulated and combined – and to the ways in which they might be undone. After all, lest we forget: even if capitalist society maintains certain necessary features, capitalism is not, in itself, an historical necessity. Both racism and capitalism are historical contingencies. They need not exist, and they can be unmade.