Comments on Roediger’s Class, Race and Marxism

by Charles Post

Over the past three decades, David Roediger’s work has fundamentally reshaped the study of race and racism, both in the US and internationally. Starting with his path breaking collection of essays, The Wages of Whiteness[1], Roediger (and his collaborator Elizabeth Esch) have illustrated how both capitalist and wage workers have participated in the creation of race and utilized the myth of intrinsic and unchangeable differences amongst humans to defend and advance their social positions in capitalist societies. Roediger has described both how fluid the social fiction of race is, plotting shifting “racial boundaries” across time; and how it persists throughout the history of capitalism. In his latest book, a collection of previously published essays, Roediger attempts to provide an explicitly Marxian theoretical account of the relationship of race and class under capitalism—as part of an intervention in the ongoing discussions on the US left of race and class.

The main targets of Roediger’s criticisms are those Marxists who purportedly privilege class and the dynamics of capitalist accumulation over race. Politically, he challenges African-American scholars such as Adolph Reed and Cedric Johnson who argue that “anti-racist” politics which focus solely on racial differences at the expense of class inequality have become a bedrock of neo-liberal ideology and politics in the US.[2] Theoretically, Roediger targets the so-called “political Marxists”—the trend associated with Robert Brenner and Ellen Meiksins Wood. For Roediger, the emphasis of Brenner, Wood and their co-thinkers on the centrality of capitalist social property relations leads them to be incapable of explaining the origins and persistence of racial oppression.[3] Unfortunately, formulations by Ellen Meiksins Wood, in particular, have lent credence to such claims:

At the very least, class equality means something different and requires different conditions from sexual or racial equality. In particular, the abolition of class inequality would by definition mean the end of capitalism. But is the same necessarily true about the abolition of sexual or racial inequality? Sexual and racial equality… are not in principle incompatible with capitalism. The disappearance of class inequalities, on the other hand, is by definition incompatible with capitalism. At the same time, although class exploitation is constitutive of capitalism as sexual or racial inequalities are not, capitalism subjects all social relations to its requirements. It can co-opt and reinforce inequalities and oppressions that it did not create and adapt them to the interests of class exploitation.[4]

As a “political Marxist”—I prefer “Capital-centric Marxist”—I find Wood’s formulation extremely problematic. Wood confuses the theoretical and historical preconditions of capitalist social property relations with the necessary outcomes of the reproduction of these social property relations.[5] Wood correctly high-lights two crucial realities. First, neither racial nor gender oppression is a necessary precondition for the establishment of capitalist social-property relations. Put another way, capitalism comes into existence once producers and non-producers have to reproduce themselves through market competition—through the operation of the law of value. Second, capitalism can successfully reduce racial and gender disparities within the capitalist and professional-managerial middle classes—even allowing an African-American to head the most powerful capitalist state in the world.[6] However, she errs when she claims that capitalist social property relations can be reproduced without creating and recreating racial and gender oppression within the working class—that “sexual and racial equality… are not in principle incompatible with capitalism.”[7]

As Richard Seymour has pointed out, Roediger actually agrees with Wood that race is produced and reproduced independently of capitalism.[8] Roediger does not theoretically account for how the dynamics of capitalism produce and reproduce race as a social category. Race appears to operate externally to, but in a functional relationship to capitalist accumulation and competition. Ultimately Roediger has to rely on notions of “intersectionality”—where a number of independent systems of oppression (class, gender, race) intersect to shape social relations.[9] In their more explicitly theoretical formulations, Roediger and Esch argue that race allows capitalists to differentiate workers in order to enhance capital’s ability to raise the rate of surplus-value—the rate of exploitation.[10]

An alternative account needs to begin with an analysis of how the emergence of capitalist social property relations in England in the seventeenth century necessitated the invention of race. As most historians and theorists of race acknowledge, the notion of race—that humanity was divided into distinct groups with unchangeable characteristics, making some groups superior and other inferior—emerges in the process of English capitalist colonization of Ireland and colonial Virginia.[11] The notion of race arose to explain and justify slavery and other forms of bondage in societies where legal freedom and equality was becoming the norm. In societies before capitalism, where exploitation takes place through non-economic coercion, inequality was assumed to be the ‘natural’ condition of humanity. Only with capitalism, where exploitation takes place through the ‘dull compulsions of the market,’ can the notion of legal-juridical freedom and equality become the ‘common sense’ of society. Put simply, it is only with the development of capitalism that race becomes a necessary means of explaining and justifying inequality.

Once capitalist social property relations become dominant, the production and reproduction of race is rooted in the dynamics of labor-market competition. Robert Brenner and Johanna Brenner argued in a 1981 essay, that “workers are not only collective producers with a common interest in taking collective control over social production. They are also individual sellers of labor power in conflict with each other over jobs, promotions, etc.” As competing sellers of labor power, workers are open to the appeal of politics that pit them against other workers—especially workers in a weaker social position:

It appears possible for the stronger sections of the working class to defend their positions by organizing on the basis of already existing ties against weaker, less-organized sections. They can take advantage of their positions as Americans over and against foreigners, as whites over and against blacks, as men over and against women, as employed over and against unemployed, etc. In so doing, working people may act initially only out of what they perceive to be their most immediate self-interest. But over time they inevitably feel the pressure to make sense of these actions and they adopt ideas which can make their actions reasonable and coherent. These ideas are, of course, the ideas of the right.[12]

In other words, the reproduction of capitalist social property relations through the “dull compulsions of the market” – through accumulation and labor-market competition—that creates the social matrix for the production and reproduction of race.

Roediger and others reject the notion that capitalist accumulation and competition directly create the conditions for race and racist ideology. Implicitly or explicitly, they argue that the operation of the law of value should homogenize labor— equalizing wages, conditions of work and the like. In the words of Roediger and Esch[13]:

From The Communist Manifesto forward, capitalism has for more than 160 years received credit from the mainstream of Marxism for introducing a ‘cosmopolitan character to production and consumption in every country.’ Marx later argued, ‘As against capital, labor is the merely abstract form, the mere possibility of value-positing activity, which exists only as a capacity, as a resource in the bodiliness of the worker.’ The body, so central to…management’s manipulation of racial difference, is not absent in Marx, but the emphasis lies on its particularities being overcome by reducing it to standardized movements, on its race yielding to class. Value arises from making labor abstract, not from accentuating differences among workers.

Unfortunately, this conception of Marx’s theory of value, accumulation and competition, while quite common among Marxists and non-Marxists, reflects neither Marx’s mature theory nor the actual history of capitalism. The work of Anwar Shaikh and Howard Botwinick[14], whose understanding of capitalism is similar to Brenner and Wood, argue that the reproduction of capitalism produces not homogeneity but constant differentiation among capitalists and workers. The operation the law of value—where the exchange-value of different commodities are expressed in the amount of socially average abstract labor time required to produce them—does not depend upon the homogenization of labor. Rather, it is capitalist competition and accumulation that allows the products of fundamentally different concrete human labor-processes to exchange as equivalents by abstracting from those concrete differences.[15]

The notions that accumulation and competition should homogenize conditions of production, labor-processes, wage rates and the like is ultimately derived from neo-classical economics idealized vision of competition. “Perfect competition,” which should produce uniform labor-processes, profit rates and wages, makes the existing economic order appear efficient and just. Real capitalist competition—from the birth of capitalism in English agriculture in the 16th century, through the industrial revolution of the 18th and 19th century to the emergence of the transnational corporations in the 20th century—has never corresponded to the dream world of “perfect competition.” Capitalist competition is fought through what Marx called the “heavy artillery of fixed capital”—constant technological innovation, taking the form of the increasing mechanization of production.

Real competition and accumulation through increasing the mechanization of production differentiates capital and labor in two ways. First, the process of mechanization in one branch of production leads to a portion of the workforce being made redundant from capital’s point of view. This constant replenishment of what Marx called the reserve army of labor, the mass of unemployed and underemployed, not only regulates wages within the boundaries of profitability, but create the possibility of heterogeneous labor-processes, profit-rates and wages between branches of industry. While the increasingly capital-intensive industries enjoy higher profits and the possibility of higher wages, the constant replenishing of the reserve army allows the constant reproduction of labor-intensive industries with lower profits and lower wages. Put simply, “sweated labor” under capitalism is not some atavistic hangover of earlier forms of production, but the necessary consequence of the continued, but necessarily uneven and combined mechanization of production.[16]

The constant generation of the reserve army, with workers experiencing different levels of precarity and desperation, produce workers who have little choice but to accept the worst jobs. In the presence of the reserve army, the mobility of capital and labor sets limits to, but cannot eliminate, overall wage differentiation. Put simply, low wage sectors can avoid raising wages by tapping into pools of desperate workers. Building on the work of Samuel Friedman,[17] Botwinick argues these low-wage industries often draw from specific labor-reserves—specific layers of unemployed and underemployed workers who experience different conditions of the reproduction of their labor-power—in order to maintain their profitability. A clear example of a distinct reserve army of labor is migrant workers. The physical separation of inter-generational reproduction in various parts of the global South and day-to-day reproduction in the global North, allows capitalists in low wage industries to pay wages below the costs of reproduction of labor-power in the global North.[18]

Second, competition within and between industries necessarily differentiates labor-processes, profits and wage rates. In the competitive “war of all against all”, firms with older investments in fixed capital, even if they no longer allow the firms to reduce unit costs and raise its profit margins and rates, cannot be abandoned immediately in favor of new and more efficient machinery. According to Botwinick:

Given the presence of fixed capital investment, however, new techniques cannot be immediately adopted by all firms in the industry. Because fixed capital generally requires prolonged turnover periods, new techniques will be adopted primarily by those capitals that are in the best position to do so. Thus, although new capitals will enter the industry with ‘state of the art’ equipment and other existing capitals will gradually begin to replenish and expand their productive facilities with the latest techniques, older, less efficient capitals will also tend to live on for many years. This is particularly true within prolonged periods of rapid growth… Rather than creating identical firms, competition therefore creates a continual redifferentiation of the conditions of production.[19]

The production and reproduction of race can be rooted the differentiation of labor-processes, profits and wages that necessarily result from capitalist accumulation and competition.[20] Put another way, race and class are co-constituted under capitalism.[21] Race is the necessary and unintended consequence of capitalist competition and accumulation. The “abstract” laws of motion of capital spontaneously produces and reproduce the notion that different groups of workers have unchangeable characteristics, making some inherently more or less “reliable” workers. Both capitalists and workers, especially when working class organizations like unions are weak,[22] utilize race as a way of ordering the “employment queue.” On the one hand, capital is confronted with a mass of workers, almost all of whom can perform almost any task required. Employers use fictional racial “characteristics” to determine who are the most “reliable” and “efficient” workers for different tasks—as Roediger and Esch describe brilliantly in The Production of Difference. On the other hand, workers attempt to socially construct themselves as “white” to protect themselves from the pressures of the reserve army of labor and the threat of being easily replaced as capital deskills labor through the fragmentation of tasks and mechanization.[23] It is this form of labor-market competition that fueled antebellum northern white skilled workers projection “onto Black workers what they still desired in terms of the imagined absence of alienation, even as they bridled at being treated as slaves or ‘white niggers.’[24] The result of the continual competition among workers—which mirrors the competition among capitalists—is the over-representation of workers of color among the reserve army of labor (consistently higher rates of unemployment, underemployment and poverty) and in the labor-intensive, low-age sectors of production.

White “advantage” is fundamentally rooted in this competition—lower levels of unemployment and poverty, access to more secure and better paying jobs, which allow white worker greater access to better housing, education for their children and the like. However, the continuous, spontaneous reproduction of race through capitalist accumulation and competition undermines the collective capacity of workers to resist capital’s demands, creating a down-ward spiral in wages and working conditions for all workers. Not surprisingly, the “wages of whiteness” have become rather meager in the age of neo-liberalism. Multi-racial working class unity will not be produced spontaneously– it will require the rebuilding of a culture and organization of solidarity.[25]   Clearly, struggles for universal, class wide demands—higher wages, greater job security, health care and pensions not tied to employment, etc.—will be crucial. But so will be race specific demands like plant and industry-wide seniority,[26] affirmative action in hiring and promotion, opposition to police brutality and the like. Put simply, effective class organization and politics—forging working class unity among a racially heterogeneous class—must include anti-racist organizing and demands.


I would like to thank Richard Seymour, Todd Gordon, Tithi Bhattacharya, Bill Mullen, Jr., Kim Moody and Howard Botwinick for their comments on an earlier version of this essay.

[1] London: Verso, 1991.

[2] A. Reed, “Bernie Sanders and the New Class Politics” Jacobin (August 8, 2016) []; Cedric Johnson, “An Open Letter to Ta-Nehisi Coates and the Liberals Who Love Him” Jacobin (February 3, 2016) []

[3] Roediger, Class, Race and Marxism, pp. 25-27.For a similar argument see A. Anievas and K. Nisanciouglu, “The Poverty of Political Marxism,” International Socialist Review 94 (Fall 2014) []

[4] Wood, Democracy Against Capitalism: Renewing Historical Materialism (New York: Cambridge University Press, 1995), p. 259.

[5] C. Arruzza, “Remarks on Gender” Viewpoint Magazine (September 2, 2014) [] and “Logic or History? The Political Stakes of Marxist-Feminist Theory” Viewpoint Magazine (June 23, 2015) [] makes similar points.

[6] The integration of people of color and women into the capitalist and middle classes is the basis for the dominance of “identity politics”—a politics that emphasizes “representation” and “diversity” without challenging the structural inequalities of capitalism—among neo-liberal politicians and ideologues. Identity politics need to be clearly distinguished from anti-oppression politics that target the structural basis of class, race and gender inequalities in capitalist societies. For a useful discussion of the incompatibility of “identity” and anti-oppression politics, see Kelton Sears, “A Marxist Critiques Identity Politics: An Interview with Asad Haider” Seattle Weekly (April 26, 2017) []

[7] The work of Lise Vogel and other “social reproduction” theorists have argued that capitalism’s creation and maintenance of the separation between increasingly the socialized production of most commodities in factories, offices and stores; and the privatized reproduction of labor-power in the household/family creates the matrix for the production and reproduction of gender oppression. L. Vogel, Marxism and the Oppression of Women: Toward a Unitary Theory (Chicago, IL: Haymarket Books, 2014). See also J. Brenner, Women and the Politics of Class (New York: Monthly Review Press, 2000) and the forthcoming collection T. Bhattacharya (ed.) Social Reproduction Theory: Remapping Class, Recentering Oppression (London: Pluto Press, 2017).

[8] “Does David Roediger Disagree With Ellen Meiksins Wood?” Verso Blog (July 24, 2017) []

[9] For a critique of “intersectionality” that preserves the descriptive insights of the approach see S. Ferguson, “Intersectionality and Social-Reproduction Theory: Toward An Integrative Ontology” Historical Materialism 24:2 (2016)

[10] The Production of Difference: Race and Management of Labor in US History (New York: Oxford University Press, 2012), Introduction.

[11] Roediger Race, Class and Marxism, pp. 183-184 relies on the work of Theodore Allen, The Invention of the White Race, Volume 1: Racial Oppression and Social Control (London: Verso, 1994) and Volume 2: The Origin of Racial Oppression in Anglo-America (London: Verso, 1997) and Edmund Morgan, American Slavery, American Freedom: The Ordeal of Colonial Virginia (New York: W.W. Norton & Co., 1975). For an excellent summary of this historical material by a critic of Roediger see Barbara J. Fields, “Slavery, Race and Ideology in the United States of America” New Left Review I/181 (May-June1990), pp. 95-118.

One of the least discussed problems with the much lauded work of Cedric Robinson, Black Marxism: The Making of the Black Radical Tradition (Chapel Hill, NC: University of North Carolina Press, 1983) is his contention that racism preexisted capitalism in Europe, with a continuous history from classical antiquity. This claim fundamentally confuses forms of differentiating people before capitalism like religion (“heathens and believers”) and kinship-community (“strangers and neighbors/kin”) which tended to be highly flexible and changeable (through conversion, adoption, etc.) and the notion of race in which the characteristics that distinguish humans are unchangeable.

[12] “Reagan, the Right and the Working Class” Against the Current (Old Series) 1,2 (Winter 1981), p30.

[13] Production of Difference p. 6. Roediger and Esch’s understanding of value theory, accumulation and competition are drawn from Michael Lebowitz, “The Politics of Assumption, the Assumption of Politics” Historical Materialism, 14:2 (2006)

[14] Shaikh, Capitalism: Competition, Conflict, and Crises (New York: Oxford University Press, 2016); Botwinick, Persistent Inequalities: Wage Differentials Under Capitalist Competition (Princeton, NJ: Princeton University Press,1993. Botwinick’s seminal, but relatively unknown book, will be republished as part of the Historical Materialism book series in late 2017.

[15] A similar point, derived from Shaikh’s work, is made in Vivek Chibber, Post-Colonial Theory and the Specter of Capital (London: Verso, 2013), pp. 133-137, 145-147.

[16] Botwinick Persistent Inequalities, Chapter 3.

[17] “Structure, Process and the Labor Market” in William Darity, Jr., Labor Economics: Modern Views (Boston: Kluwer-Nijhoff Publishing, 1984), pp. 175-217.

[18] The original formulation of the physical separation of inter-generational and day-to-day reproduction of labor power as the basis of migratory labor system was in Michael Buroway, “The Functions and Reproduction of Migrant Labor: Comparative Material from Southern Africa and the United States,” American Journal of Sociology 81, 5 (March 1976), pp. 1050-1087. For a recent deployment of this argument in social reproduction theory see Susan Ferguson and David McNally, “Precarious Migrants: Gender, Race and the Social Reproduction of a Global Working Class,” in L. Panitch and G. Albo (eds.) Socialist Register 2015: Transforming Classes (New York: Monthly Review Press, 2014), pp. 1-23. Many thanks to Tithi Bhattacharya for sharing her unpublished paper on race, gender and social reproduction presented to the University of London conference “Capital 150: Marx’s Capital Today” September 1920, 2017.

[19] Botwinick, Persistent Inequalities, p.131. This argument should not be confused with ‘dual economy’ theories that posit a “core” with permanently higher profits and wages than the “peripheral” regions of the economy. See Botwinick, Persistent Inequalities, Chapters 5-7 for a detailed argument on how the “turbulent regulation” of profit rates, profit margins and wage rates through real capitalist competition prevent any branch of production or capital from retaining its “core” position.

[20] For a professional economists’ application of Botwinick’s work on wage differentials to race see Patrick L. Mason, “Race, Competition and Differential Wages,” Cambridge Journal of Economics 19 (1995), pp. 545-567.

[21] David McNally, “The Dialectics of Unity and Difference in the Constitution of Wage-Labour: On Internal Relations and Working-Class Formation” Capital & Class, 39:1 (2015), pp. 131-146.

[22] Contrary to some who claim that the rise of “identity politics” has eclipsed “class politics,” like David Harvey in Justice, Nature and the Geography of Difference (New York: Wiley, 1997), Chapter 12, I would argue that it is the decline of class organization and politics that makes “identity politics” the common sense of much of the left. Put another way, when class against class organizations, especially unions, are weak and competitive pressures among workers become dominant, organizing around racial and gender identities becomes a means of defending one’s declining position in the labor-market.

[23] Harry Braverman, Labor and Monopoly Capital: The Degradation of Labor in the 20th Century (New York: Monthly Review Press, 1974). Unfortunately most readers of Braverman’s masterpiece tends to equate deskilling with the homogenization of labor. Braverman himself was quite clear that the tendency to deskill work constantly differentiates workers.

[24] Roediger, Class, Race and Marxism, p. 68.

[25] See Vivek Chibber, “Rescuing Class from the Cultural Turn” Catalyst 1;1 (Spring 2017)

[26] Bruce Nelson’s Divided We Stand: American Workers and the Struggle for Black Equality (Princeton, NJ: Princeton University Press, 2001), Chapters 5-7 demonstrates how the CIO’s acceptance of departmental seniority set the stage for the reproduction of racial divisions among steel workers and other organized industrial workers in the post-war period.

If you like this article, please subscribe or donate.