Racism and the sociological imagination


email: robert.carter@warwick.ac.uk


Our chief purpose in this article is to argue for a restoration of a strong notion of agency to sociological accounts of social relations, and particularly those concerned with group formation and conflict. We contend that much contemporary sociological writing on this topic continues to rely on the concepts of race and ethnicity as primary explanatory or descriptive devices. This has two important consequences: on the one hand it reproduces the powerful theoretical obfuscation associated with these concepts, whilst on the other it prompts the notion that human agency has only an illusory role as an intentional agent. Drawing on the intellectual resources of a Hegelian-inflected historical materialism and realism, we challenge both claims by arguing for a post-race, post-ethnicity sociology of group formation, one which allows a greater scope for agency in the determination of social life.


The work of Robert Park (1950) represents the first attempt to develop a sociological account of racism and ‘race relations’. He wrestled with the methodological consequences of regarding race relations as social relations whose significance to individuals derives from their symbolic force; they are effectual to the extent to which people believe them to be so (Carter 2000).

Central to his account were two important claims: that the term ‘race’ referred to socially constituted phenomena – there were no races in a significant biological sense, but many people understood there to be ‘races’– and that in so far as this was the case then it was possible and appropriate for sociologists to identify and analyse situations of ‘race relations’, that is those situations in which social relations between people were grasped in terms of ‘race’. We contend that there was no substantial development of these insights until the 1970s (and even in 2003, Park's perspective underpins the ASA statement, issued after several years of deliberation, on whether sociologists should continue to use the term, and conduct research into ‘race’; the document, tellingly entitled The Importance of Collecting Data and Doing Social Scientific Research on Race, concluded that they should). More importantly, apart from Cox's path breaking work, there was little attempt within mainstream sociology to resolve the two key questions arising from Park's approach. The first of these was how to account for the spatial and temporal formation of ‘racial groups’ (to move, in other words, from the claim that ‘race’ was a popular interpretative category of lay life to the empirical identification of ‘racial groups’); without such an account, group formation came to be naturalized, and ‘racial’‘groups’ seen as naturally occurring. A second question concerned the persistence of inequalities. Park's use of an evolutionary and ecological approach to social differentiation was a modified form of Social Darwinism with a sharp tang of ethnocentrism, and came close at times to naturalizing differences between cultures.

This points towards the relevance of a materialist approach. Of course, Cox was moving in this direction at roughly the same time as Park was developing his race relations approach, but Cox (whose Caste, Class and Race was published in 1948) was marginalized within academic sociology. Nevertheless, Cox fastened onto the two key weaknesses of Park's approach; his inconsistent use of the concept of race (‘Park does not even settle upon a working definition of a race’, noted Cox. ‘He assumes that there are fundamental color antipathies between whites, yellows, and blacks’ (Cox 1970: 465); and his relative neglect of the social relations of production and the role of class relations in shaping the changing forms of racism. It was not until the revival of Marxism following the events of 1968 that Cox's cogent summary of the Parkian approach as ‘weak, vacillating and misleading’ (Cox 1970: 474) began to resurface through the work of Robert Miles and others.

Robert Miles' (1982) philosophical starting point was his trenchant rejection of the liberal sociology of race relations paradigm, and especially its use of race as an analytical and descriptive concept. For Miles, the subject of study was not race nor relations between races but how and why parts of the human population came to be constructed and defined as members of different races with different levels of cultural endowment. By employing a conceptual distinction between essential and phenomenal relations characteristic of Marxist approaches, he attempted to sideline the ‘race versus class’ debate by claiming race and class occupied different analytical spaces. In particular, race was employed to refer to a social and historical construction, an effect of ideology masking real social relations based on class.

Substantively, Miles (1982; 1989; 1993) claimed the genesis of race-making was intimately entwined with the projects of Atlantic Slavery, colonialism and nationalism. Through these large-scale social processes, large parts of Africa, Asia and South America came to be economically underdeveloped at the expense of the economic development of the European nation-states of western Europe and later North America. It was against this backdrop that ideas about race began to have some analytic purchase as a way of ideologically rationalizing the rule of western capitalist elites, and, thereby, the process of capitalist accumulation.

Whilst slavery and colonial regimes were politically overthrown in a wave of nationalist-inspired revolutions (Nairn 1977), racism continued to be reproduced because of the continuing economic dominance exercised by the departing powers. Specifically, the continuing uneven development of the capitalist world economy meant that international labour migration to the former colonial powers became an essential element of the post-colonial world-system such that when the demand for labour could not be met within the confines of the national state, individual employers and the state secured labour from beyond its national boundaries. It was at this historical moment that racism came to be replenished, with international migration being politically refracted by the State through the historical and ideological lens of racism (and nationalism).

However, it was precisely one of the strengths of Miles' framework – his objection to the employment of the concept of race in either description or analysis – that was, simultaneously the source of the most serious weakness in his work. Whilst conceding that individuals may be forced to organize against racism independently around racialized identities due to the racism of the white working-class, Miles remained unwilling to accommodate such anti-racism and anti-racist agency around black identity within his Marxian frame because of his concern that the continued use of the term ‘race’ only served to sustain the conditions for the reproduction of racism within society:

 . . . as a result of reification and the interplay between academic and common sense discourses, the ‘use’ of race as an analytical concept can incorporate into the discourse of anti-racism a notion which has been central to the evolution of racism. As a result, anti-racist activities then promote the idea that ‘races’ really exist as biological categories of people. Thus, while challenging the legitimacy of unequal treatment and stereotyping implicit and explicit in racism, the reproduction within anti-racist campaigns of the idea that there are real biological differences creating groups of human beings sustains in the public consciousness a notion which constitutes an ideological precondition for stereotyping and unequal treatment (Miles and Torres 1999: 26).

Miles' failure to accommodate anti-racist action constructed around the racialized identity of black in 1970s Britain created immense difficulties relating to political practice within his theoretical frame. In particular, in the context of the state racism unleashed against Britain's racialized minority populations in the 1970s and 1980s (see CCCS 1982), Miles was left advocating support for an idealized and unified class subjectivity, hoping this would evolve out of a shared class position in the process of production.

In summary, then, the work of Park helped to undermine a sociological commitment to biological notions of ‘race’ and pointed to the socially constituted nature of ‘race relations’; the work of Miles, on the other hand, helped to embed the socially constituted nature of ‘race relations’ within a materialist framework. However, by the end of the 1980s these theoretical frameworks were challenged by the ‘cultural turn’ in sociology. Key to this, within the sociology of racism, was the work of Stuart Hall and two other leading scholars – David Theo Goldberg and Paul Gilroy – who took up Hall's core ideas and developed them in a distinctive way.1

Stuart Hall, subjectivity and the dangers of interpellation

Whilst working within the same structuralist Marxist tradition as Miles, Stuart Hall (1980, 1996) was more alive to questions of subjectivity and identity formation as a result of E. P. Thompson's (1978) polemical attack on Althusser and his attempt to square structuralism with Marxism. For Thompson, this had resulted in the construction of a flawed theoretical apparatus – an ‘orrery of errors’– which had effectively banished the idea of human subjectivity from Marxism. Indeed Hall, because he engaged with Althusser's work seriously, was the object of Thompson's ire in a now in/famous debate held at Ruskin College Oxford in December 1979 (Samuel 1981: 375–408). Hall, whilst rejecting many of Thompson's substantive criticisms, nevertheless had independently begun to turn to Gramsci as a way of re-introducing the ‘historically concrete’ and human subjectivity into his theoretical frame. As he retrospectively acknowledged: ‘Gramsci is where I stopped in the headlong rush into structuralism and theoreticism. At a certain point, I stumbled over Gramsci, and I said, “Here and no further!” ’ (Hall 1988: 69).

The additional analytic purchase Hall achieved by introducing a notion of subjectivity that had hitherto been missing from his work enabled him to demonstrate how the ascription of racist identities could also be appropriated by the racialized and infused with a new ideology of resistance to counter racism and discrimination:

The racist interpellations can become themselves the sites and stake in the ideological struggle, occupied and redefined to become elementary forms of an oppositional formation – as where ‘white racism’ is vigorously contested through the symbolic inversions of ‘black power’. The ideologies of racism remain contradictory structures, which can function both as vehicles for the imposition of dominant ideologies, and as the elementary forms for the cultures of resistance. Any attempt to delineate politics and ideologies of racism which omit these continuing features of struggle and contradiction win an apparent adequacy of explanation only by operating a disabling reductionism (Hall 1980: 342).

However, his attempt to bring the subject back into history through the work of Gramsci was flawed because Gramsci was read through a structuralist–Marxist lens. In particular, the root of the problem lay in Hall's use of interpellation, a concept derived from Althusser and Laclau, and employed to denote the process by which individuals were constituted by ideologies, and so become subjects of ideology (Hall 1996). Two corollaries of this understanding were that interpellated individuals believed such subjectivities or identities were self-generated and so freely accepted, even embraced, their subjection, thereby contributing to the continuation of the capitalist system. And second, even when subjects did resist, they remained interpellated individuals.

Consequently, in Hall's conceptual framework, because the working class were always interpellated, the prospect of this class reaching ‘beyond ideology’ or lifting the veil of ideology and moving towards a higher form of (class) consciousness in explicit recognition of their objective, material interests was lost. Such an understanding of subjectivity and human agency was at odds with Gramsci's (and Marx's) theory of working-class self-emancipation and the understanding that the working class could, under definite social conditions, break free from such ideologies of domination. The outcome was that despite his well-intentioned attempt at rethinking Marxism, Hall ended up offering a portraiture of the white working class that like the Utopian Socialists before him (Marx and Engels 1977; Draper 1978), reduced this class to mere victims of the degradations inflicted by the capitalist system; a class with little capacity to resist the power of ideology in fragmenting and dissipating resistance to elite rule.

This is not to deny that Hall's approach, demonstrating the power of ideology in integrating the working class, and thereby fragmenting opposition to the capitalist state, was wholly inconsistent with Marx's approach and the emphasis he placed on ideologies shaping working-class conceptions of the world. After all, it was Marx (1987: 45) in The German Ideology who claimed famously that:

The ideas of the ruling class are in every epoch the ruling ideas, i.e. the class which is the ruling material force of society, is at the same time its ruling intellectual force. The class which has the means of material production at its disposal, has control at the same time over the means of mental production, so that thereby, generally speaking, the ideas of those who lack the means of mental production are subject to it.

Significantly, however, Marx – unlike Hall – explicitly juxtaposed such an understanding of ideology and its impact on working-class consciousness with a conception of this same class as the eventual ‘gravedigger of capitalism’ (Marx and Engels 1977). Indeed, it could be contended that what distinguished Marx's Marxism from other philosophical and theoretical traditions was not its emphasis on socially produced inequalities or political economy, nor even its focus on class struggle and capitalism, but instead, its conception that the working class was the universal class – the class whose own particularist interests, under given historical conditions, would synchronize with the transformation needed by society as a whole. It was in this sense, that Marx, ‘nominated the proletariat as the universal class . . . [and] hence, the agent of revolution’ (Draper 1978: 71).

Marx himself was acutely aware of the theoretical and practical dilemmas posed by what others have rather lazily interpreted as this ‘contradiction’ in his work. How then did Marx set about resolving the dilemma of a class that, on the one hand, was so thoroughly dehumanized in capitalist society and politically divided on the grounds of nationalism, racism, sexism and other ideologies of domination, with, on the other hand, a simultaneous conception that it was only this class that had the capacity to transform capitalist social relations and so release the full potential of humanity? Or as the young Sidney Hook (2002: 157) succinctly put it, ‘how is it possible for human beings conditioned by their cultural education and environment to succeed in changing that environment?’

Marx solved this dilemma by introducing the concept of (class) struggle, broadly understood as encompassing forms of collective working-class resistance to the multifarious forms of capitalist exploitation and oppression. For Marx, it was only through struggle that the working class could change politically and reject what he memorably termed the ‘old crap’ and thereby become fit to rule. That is, it was only through struggle that attachments to deeply-held ideological positions would become unsettled and open up a political as well as ideological space from within which those articulating an internationalist working-class standpoint could attract an audience and begin the process of manufacturing the necessary preconditions for the socialist transformation of society. Thus, for Marx, it was in the course of struggles against capitalist exploitation and oppression that the working class would begin to lose their attachments to long-held reactionary sentiments and thereby begin to transform themselves:

Both for the production on a mass scale of this communist consciousness, and for the success of the cause itself, an alteration of men on a mass scale is necessary, an alteration which can only take place in a practical movement, a revolution; the revolution is necessary, therefore, not only because the ruling class cannot be overthrown in any other way, but also because the class overthrowing it can only in a revolution succeed in ridding itself of the old crap and become fitted to found society anew (Marx cited in Draper 1978: 74).

This dialectical solution to the question of how those who are dehumanized by capitalist society are simultaneously the ones who are also most likely to transform it that Marx brings to bear on the question of social change is absent in the body of work produced by Hall. The outcome is that whilst Hall takes us beyond the economistic and reductionist Marxism of the 2nd and 3rd Internationals, it is at the cost of expunging the beating heart of Marxism – the conception of the working-class subject as the grave-digger of capitalist society. In doing so, Hall abandons, albeit unwittingly, the revolutionary standpoint that was central to Marx's life and thought.

Paul Gilroy, postmodernism and the retreat from class

Paul Gilroy (1982) initially tried to rethink historical materialism in such a way as to encompass questions of class subjectivity and identity formation that had been ignored by structuralist Marxism. Hence, the pressing analytic (and strategic) question was defined as that of establishing the processes by which racism could be challenged and the working class unified around a class subjectivity: ‘Our premise is the problem of relating ‘race’ to class, . . . for socialist politics’ (Gilroy 1982: 276). The key concept that Gilroy employed to analytically grasp this dynamic process of social change was class struggle, defined in such a way as to include ‘the relentless processes by which classes are constituted – organized and disorganized – in politics, as well as the struggles between them once formed’ (Gilroy 1982: 284). Here, Gilroy opened up a theoretical and political space by which to re-conceive autonomous black struggles against racism, in the community as well as within the workplace, as forms of class struggle. The theoretical implications of such a position are clear; if black struggles are class struggles, then these struggles contribute to a process of class formation whereby a consciousness of class becomes synchronized with a consciousness of ‘race’:

Though for the social analyst ‘race’ and class are necessarily abstractions at different levels, black consciousness of race and class cannot be empirically separated. The class character of black struggles is not the result of the fact that blacks are predominantly proletarian, though this is true. It is established in the fact that their struggles for civil rights, freedom from state harassment, or as waged workers, are instances of the process by which the class is constituted politically, organized in politics (Gilroy 1982: 302).

By locating the analysis of racism at the heart of processes of class reformation (and dissolution), the black working-class, far from being peripheral to working class politics was now brought centre stage and given a vanguard role that Marx had attributed to the working class as a whole:

In our view of class formation, the racist ideologies and practices of the white working class and the consequent differentiation of ‘the blacks’ are ways in which the class as a whole is disorganized. The struggles of black people to refuse and transform their subjugation are no simple antidote to class segmentation, but they are processes which attempt to constitute the class politically across racial divisions –‘that is which represent it against capitalism against racism’ . . . these struggles do not derive their meaning from the political failures of the classically conceived, white, male working class . . . it appears that autonomous organization has enabled blacks . . . to ‘leap-frog’ over their fellow workers into direct confrontations with the state in the interest of the class as a whole (Gilroy 1982: 304).

Remarkably, however, these influential statements on the workings of contemporary racism and emancipatory politics, rather than representing a key moment in the historical renewal of the materialist method, actually marked Gilroy's departure to more postmodern forms of social thought. Hence, just five years after the publication of the collectively-authored The Empire Strikes Back (CCCS 1982) where he began the intellectually fruitful task of rethinking the relationship between race ideas and class formation, Gilroy published There Ain't No Black in the Union Jack (1987) which saw him effecting a divorce between them. Hence, the re-conception of black struggles as class struggles, constituting one moment in the historical unification of the working class, was now rejected, with such struggles disengaged from any type of class analysis:

If these struggles (some of which are conducted in and through ‘race’) are to be called class struggles, then class analysis must itself be thoroughly overhauled. I am not sure whether the labour involved in doing so makes it either a possible or desirable task (Gilroy 1987: 245).

Instead, drawing on social movement theory emerging from western Europe (itself a product of working-class defeat), Gilroy moved to settle his account with Marxism by reconceiving black struggles against racism (or what remained of them by the late-1980s) as one of the burgeoning social movements alongside those of the feminist, ecology and youth movements. This breach with his previous Marxist approach was made explicit with his conclusion that:

The Proletariat of yesterday, classically conceived or otherwise, now has rather more to lose than its chains. The real gains which it has made have been achieved at the cost of a deep-seated accommodation with capital and the political institutions of corporatism. It's will . . . ‘is apt to be a reformist will’ (Gilroy 1987: 246).

Principally, there are two factors that help to understand the remarkable turnaround in Gilroy's theoretical and political position. First, was the decisive defeat suffered by anti-systemic movements that had their origins in the world revolution of 1968. The political exhaustion of the militant workers movement in western Europe, and the anti-war, anti-racist and feminist movements in Europe and the USA heralded the election of reactionary political forces such as Thatcher in Britain and Reagan in the USA. Coupled with the fall of the Stalinist bloc of eastern Europe and of the Soviet Union itself in 1991, the utopian sentiments that had been sparked in the late 1960s were extinguished. Against this political climate, Marxist approaches to racism and other phenomena, with their concern for making history as well as understanding it, found themselves marginalized in the academy for the more esoteric concerns of postmodernism and poststructuralism. It is clear that academics, even those as impressive as Gilroy, could not fail to be affected by such developments.

Second, however, such a transition away from a materialist conception of history was made easier because of some of the weaknesses associated with Gilroy's theoretical framework, in particular, his failure to conceptualize the concept of class struggle in its totality. Gilroy used it only to re-conceptualize the struggles of black workers as class struggles but not as an overarching concept that could help to uncover and interpret the struggles of white workers and their dialectical relationship with those of the racially demarcated black class fraction (see Virdee 2000). This results in an abstraction of racist and anti-racist struggles from the historical rhythms of the class struggle, so that a historically concrete assessment of its impact in shaping and changing white working-class consciousness and identities is entirely missing from his work. The outcome is that whilst Gilroy produces a dynamic analysis of anti-racist politics and black culture and its racialization, the portraiture of the white working class is static, ahistorical and generally shorn of any subjectivity across time and space.

The implications of re-connecting the racialized class struggles of black workers to the class struggles of white workers and analysing their significance in their totality are disastrous for Gilroy's theory of race and class as Virdee (2000) demonstrates with respect to 1970s Britain – the period that Gilroy uses to formulate his position. Rather than the white working-class being defined as a racist class faction devoid of any subjectivity across time and space, we actually find that under conditions of militant class struggle and major political and industrial unrest, the attachment to racist and nationalist identities becomes unsettled, creating a space for the emergence of a stronger class identity amongst parts of the organized working class which lead to the formation of a fragile but real class solidarity across ‘racial’ lines at this specific historical conjuncture. This was evidenced most clearly in the mass support provided by white workers in the dispute involving Asian women at the Grunwick film processing plant but also entailed significant working-class involvement in the mass anti-racist movements of Rock Against Racism, the Anti-Nazi League and the trades unions.

Substantively, Virdee's (2000) work demonstrates that such a movement towards an anti-racist solidarity was not inevitable; anti-racist internationalists and racist nationalists sought to compete for the soul and support of the white working class. Critical to the formation of united working-class action was direct human intervention that transcended the racist colour line in the form of black workers engaged in independent anti-racist action and black and white socialist activists who recognized that racism served to divide the working class. This was something the working class could ill afford when trying to defend their class interests against employer and state attacks. Hence, key factions of the working class were ideologically won over to anti-racist ideas because black and socialist activists from a multiplicity of political parties were able to successfully synchronize the struggle against racism with the struggle against employer attacks on the working class.

Yet, because of his problematic conception of the idea of race and its relationship to class at the level of theory, Gilroy misses out entirely on analytically capturing these important changes in the consciousness of white workers, including towards racism. Having written off the trade unions as irretrievably racist, he is unable to analyse the important process of anti-racist racialized black formation in trade unions throughout the 1980s and 1990s (see Virdee and Grint 1994) which has led some to predict that when we witness the ‘fire next time’ and the intensification of the class struggle, black people will be an indispensable element of the struggle of organized labour against capitalist exploitation, and that trade unions by virtue of the leading presence of black workers will play a key role in combating racism.

Gilroy's neglect of the class struggles of white organized labour and their implications for consciousness and the presumption of their unchanging racism across time and place, helps us to understand his attribution of a vanguard role to the black proletariat and the subsequent reconception of it as one movement amongst many social movements. However, with the decline of black politics from the mid-1980s, Gilroy loses even this diminished conception of human agency and is reduced to making an abstract appeal for a planetary humanism to counter the growing array of racist absolutisms in the global era (Gilroy 2000) – a demand almost wholly devoid of any sociological understanding of the inequalities produced by the capitalist world economy.

David Theo Goldberg, subjectivity and the Foucauldian cul-de-sac

Two distinct positions evolved from Hall's work. The first is that taken by Gilroy with its retreat from class struggles in favour of a heterocultural cosmopolitanism; the second is influenced directly by the work of Foucault. Foucault's ideas are famously elusive; some of those drawing upon his work have had to be more reckless. The result is sometimes a version of Foucault in which a strong notion of discourse combines with a structuralist account of subjectivity to all but eliminate social agency; at other times Foucault is used to assert the ‘materiality’ of discourse so that it becomes an active agent in its own right. The work of Goldberg (1987; 1993) illustrates both these tendencies.

Goldberg's (1987) starting point is the shortcomings of orthodox Marxism and in particular its reduction of discursive relations to materialist relations such that the former are regarded as merely superstructural expressions of the latter. Instead, Goldberg argues that racism is a ‘complex social object’ whose explanation must avoid reductionist simplification. The key to doing this is to define racism as ‘a field of discourse’, a theoretical construct which includes ‘all the various entities constitutive of racism: Expressions . . . but also their underlying conceptual structure, and more overtly, acts and their consequences, as well as institutions . . . ’ (Goldberg 1987: 59–60). Racism as a field of discourse is both an ideational phenomenon and a matter of social practice –‘acts and their consequences’.

A field of discourse itself emerges from a ‘discursive formation’, understood by Goldberg (1987), following Foucault, as a ‘totality of ordered relations and correlations’. The ordering in a discursive formation is accomplished through the ‘rules of formation’, an array of pre- and proscriptive statements linking subjects to each other and to objects: ‘the good citizen is one who is prepared to die for their country’; ‘a homosexual cannot be a good Christian’. Such rules are constitutive of a discursive field, notes Goldberg, by ‘defining an object that can be spoken of and the mode in terms of which it can be analysed, its elements named and classified, its functions explained’ (Goldberg 1987: 60)

If racism is the effect of a field of discourse, which is itself the product of a discursive formation with its own rules for determining what may be said and what may be known, then the attenuation of any vital sense of social agency is clear. Attenuation of agency becomes the extinction of agency when Goldberg (1987) emphasizes the rules of formation governing the formation of objects and the formation of subject positions (see Fairclough 1993: 41–9 for a helpful discussion of Foucault's ‘rules of formation’). These rules are constitutive; that is to say, for Foucault discourses contribute to the production, transformation and reproduction of objects (thus the object race is constituted partly by the discourse of racism), but also to the production, transformation and reproduction of subject positions and the subjectivities associated with these (‘immigrants’ and ‘British’ might be regarded as two subject positions constituted by discourses of race and immigration in postwar Britain, for example).

These claims concerning the constitutive role of discourses can be taken in one of two ways. The first, weak, reading acknowledges the key role of discourse in mediating experience and in shaping our notions of social reality; the second, strong reading presses towards a view of subjectivity and reality in which who we understand ourselves to be, and how we take the world to be constituted, are wholly discursively produced: the realm of the extra-discursive and the ontologically prior become minute and irrelevant. Agency is limited to ‘how subjects internalize and interpellate discourses and act in terms of them’ (Goldberg 1987: 62); there is not a great deal of room here for the notion that discourses might be modified by how people interpret and act on them (as opposed to being compelled to ‘act in terms of them’).

Yet in Racist Culture (1993) Goldberg seems to restrict the limits of discourse. Whilst arguing, contra Marxism, that the ‘discourses of race and class, while intersecting in various important and codetermining ways, are at basis conceptually and effectively autonomous’, he retains a significant distinction between exclusion and exploitation: the discourse of racism operates as a mode of exclusion by invoking ‘questionable differentiations’ which ‘succeed only in so far as the terms they generate remain subjectively persuasive’, whilst exploitation ‘necessarily requires actual class differentiations, though not (necessarily) discursive rationalisation’ (Goldberg 1993: 106–107). In other words, racism seems to require ‘race-ists’; exploitation, on the other hand, rests on ‘actual’ class differences and is indifferent as to whether there are ‘class-ists’ or not. Indeed, Goldberg emphasizes, in this work at least, the ontological emptiness of the concept of race. Reaffirming Banton's observation that no-one has ever seen another person's race (Banton 1998), he points to the circularity involved in racial classification:

Observing racial differences between persons can only be successful if significant racial criteria have been presupposed, and so observation cannot be the grounds of the differentiation. And it is altogether incredible that the supposed significance of the differences lies simply in the observable ‘givens’ of nature (Goldberg 1993: 86).

Thus it is that ‘In the case of race . . . there is literally no object referred to, no ‘given’ phenomenon to be saved’ (Goldberg 1993: 88).

However, Goldberg's (1993) commitment to the strong reading of Foucault mentioned above leads him to insist that discourses interpellate subjects; they ‘. . . are the intermediary between self and society; they mediate the self as social subject’ (Goldberg 1993: 57). As we have seen, discourses, in this view, come to be constitutive of human subjectivities, imposing technologies of discipline and power and encouraging docility ‘. . . by reducing even social subjectivities, or at least some forms of social subjectivity, to physical dimensions and correlates’ (Goldberg 1993: 53). This more active role for discourse begins to loosen the materialist elements in Goldberg's approach and the relations between discourses and human agency become more exiguous. This tension is most apparent in Goldberg's use of the term race. Whilst recognizing that the political utility of the concept derives partly from its ontological vacuity, an important shift in the depiction of causality occurs in Goldberg's discussion. Rather than the concept of race being deployed by social actors seeking to dominate or subjugate others through the representation of particular forms of difference, it is the concept itself which engages in these projects; its

 . . . conceptual emptiness allows it parasitically to map its signification of naturalized differences onto prevailing social views and scientific theories that are readily acceptable and accepted, to articulate and extend racialized exclusions in the name of and legitimized by discourses seemingly neutral and impartial. Race has thus been able more or less continuously since its emergence to naturalize difference and to normalize exclusions.

(Goldberg 1993: 210).

This claim provides the opportunity for some barnstorming conceptual inflation: if race is capable of this protean adaptability, then its ambitions become limitless; it can shift from being a questionable and discredited concept associated with nineteenth century science to being the basis of contemporary governance and this is precisely what it comes to do in Goldberg's later notion of ‘the racial state’ (Goldberg 2002).

In Goldberg's account, race is a product of modernity and its need to account for, to know and to control ‘Otherness’; in modernity ‘what is invested with racial meaning, what becomes increasingly racially conceived, is the threat, the external, the unknown, the outside’ (Goldberg 2002: 23). Race is also a form of crisis management, a means of containing the threat of the diverse and the different:

The racial state, the state's definition in racial terms, thus becomes the racial characterization of the apparatus, the projects, the institutions for managing this threat, for keeping it out or ultimately containing it. . . .  (Goldberg 2002: 34)

This depiction rests on a common-sense understanding of race ideas, and as such reproduces them in a reified form; but it also conceives of race as a practice, as something done to some people by others. (It is always possible to do this with race ideas precisely because their meaning is so indeterminate). A conception of ‘race as practice’ gives rise to a confused view of social causation, with crippling political consequences. To begin with, the racial state becomes refractory to significant political intervention or modification:

In states that are racially conceived, ordered, administered, and regulated, the racial state could be said to be everywhere. And simultaneously seen nowhere. It (invisibly) defines almost every relation, contours virtually all intercourse. It fashions not just the said and the sayable, the done and doable, possibilities and impermissibilities, but penetrates equally the scope and quality, content and character of social silences and presumptions. The state in its racial reach and expression is thus at once super-visible, in form and force and thoroughly invisible in its osmotic infusion into the everyday, its penetration into common sense, its pervasion (not to mention perversion) of the warp and weave of the social fabric (Goldberg 2002: 98).

From this view, as Goldberg acknowledges, the racial state is as much ‘a state or condition of being as it is a state of governance’. Such a state is the normal form of the modern state, according to Goldberg, and leads to the conclusion that the current global arrangements may be identified as a racist world order (Goldberg 2002: 104). Here the term ‘race’ (and its cognates ‘racial’, ‘racism’) has lost specific meaning; once race is everywhere (with the racial state) it is, of course, nowhere since it is impossible to identify. The political consequences of this position are dire: anyone foolhardy enough to challenge this ‘super-visible . . . and thoroughly invisible’ mammoth, must count their chances of succeeding meagre.

These pessimistic conclusions follow precisely because challenging the ‘super concept’ that race has become necessarily requires a ‘super-agent’ to challenge it, and history has not been generous to the search for super-agency. Clearly, an account of racism and ethnicity which considers the interplay between structure and agency in less extravagant terms is required and in the concluding section we identify some elements of such an account.


As with earlier generations of sociologists who attempted to settle their accounts with Marxism (e.g. Weber 1993; Dahrendorf 1959; Giddens 1981), historical materialism has proved difficult to silence. Despite the dark days of the 1990s, there appears to be a growing interest in historical materialist accounts of society as evidenced by the establishment of a number of new journals like Cultural Logic and Historical Materialism but also the publication of original works focusing on Marxist theory (e.g. Nimtz 2000; Lih 2005). More substantively, the present paper and other works (e.g. Virdee 2000; Meyerson 2001, Dardar and Torres 2004) demonstrate that there is a process of intellectual renewal of historical materialist accounts of racism under way in the early twenty first century. Through a critical engagement with sociological and postcolonial theory, concerns about identity formation, subjectivity and human agency are being addressed by drawing on the intellectual resources of Hegelian-inflected Marxism, as well as realism. We have considered some of the key features of this renewal by examining different traditions of sociological thought and their efforts to develop a sociological account of racism and race thinking.

In reviewing the debates about the relationship between racism and class relations, about the meanings of terms such as ‘race’ and ‘ethnicity’ and about the efforts of social researchers to contribute to a public understanding of social processes with a view to the amelioration of inequities and injustices, it is difficult not to concur with Turner's recent comments concerning the failure of British sociology to develop ‘. . . any sustained, macro-sociological analysis of the historical decline of Britain as a world power in the twentieth century’ (Turner 2006: 169). In a similar vein, Rojek and Turner have insisted that an ‘adequate cultural sociology would have to be driven by an empirical research agenda, embrace an historical and comparative framework, and have a genuinely sociological focus, that is, a focus on the changing balance of power in Western capitalism’ (Rojek and Turner 2000: 629). Their charge – that the cultural turn in Anglo–American sociology has accomplished an aestheticization of culture that has severely diluted any capacity for political critique – seems to us well founded. Even Park had some conception of reform, believing that the purpose of sociological effort was the shifting of public opinion in order to achieve ‘racial equality’ within the confines of liberal capitalism; now it seems that even reform is doomed to failure.

It is our belief that, if sociology is to provide a more relevant account of the phenomena of racism and ethnicity, it must engage seriously with the theoretical perspectives of realism and an agent-centred historical materialism. Two themes in particular seem to us to be pertinent to such a renewal. First, sociologists need to bring the subject back into history, that is, to restore to sociological analysis a view of social relations as the always incomplete outcome of people's efforts to refashion a world not of their own making; put more formally this is a view of structure and agency which recognizes their interplay. Much work in the classical tradition of sociology and beyond had such a view, albeit with varying emphases on the relative weight to be given to either structure or agency. At the core of this view, we would suggest, is an incorrigible commitment to the claim that history is the outcome of conscious, active beings seeking to realize their purposes and intentions in contexts not of their own making or choosing: people make choices but they do not choose the choices open to them (Pawson and Tilley 1997). The language best suited to describing such a world is that of purpose and agency. This is especially so when we seek to analyse and explain aspects of the social world, such as racism, ethnicism and discrimination, which we are interested in changing.

As we have argued throughout this paper, two important methodological consequences follow from the commitment to examine the interplay between structure and agency, context and action. The first is the necessity of taking seriously the perspective of ordinary life; among other things, this means examining and explaining people's motives and values as motives and values and not as something that can always be reduced to something else (ideology, say, or discourse or false consciousness or ‘selfish genes’, as though sociologists were the only authentically motivated and properly evaluative section of humankind). The second is the necessity of developing a materialist account of the contexts shaping social interaction – by thwarting intentions, generating a restricted repertoire of choices, prompting re-evaluations of what is worth striving for and so on – and the crucial mechanisms whose combination sustains or transforms these contexts.

Such a view raises directly the question of politics: whether one finds the world a congenial place or not, and therefore whether one has an interest in keeping it that way or changing it, is largely a question of who gets what. Rather than the aestheticized politics of the cultural turn, where politics is often reduced to the deconstruction of a text or a tradition or a perspective, we would urge the development of a properly materialist politics. This would emphasize the structurally persistent inequities of the distribution of wealth, property and cultural goods and resources and encourage political, rather than cultural, accounts of group formation. (For a demonstration of the form such concrete sociological and historical analysis of group formation might take see the work of Mann 2005; MacDonald 2006; Mamdani 2004.)

This discussion of inequality and politics brings us to our second theme, namely our contention that sociologists need to bring an emancipatory working-class subject (one that is ‘white’ but also increasingly ‘black’ and ‘brown’ in the core of the capitalist world-economy) back into their accounts of racism and anti-racism. It is precisely this standpoint position that has enabled our analytic frame to integrate and theorize the historical and contemporary anti-racist activism of the European working class (alongside their more often reported racism) – something the structuralists and poststructuralists have singularly failed to do over the past three decades. Following in the footsteps of Marx and the praxis philosophers of classical Marxism, we also wish to highlight the singular importance of the intensification of the class struggle in understanding transformations in the consciousness of the working class towards a more progressive, anti-racist politics. That is, history shows that it is only through struggle that attachments to deeply-held ideological positions become unsettled and open up a political as well as ideological space from within which those social activists and collectivities articulating an internationalist and anti-racist working-class standpoint can attract an audience (see Virdee 2000 for a historically concrete illustration of this claim in 1970s Britain).

This type of critical engagement with realist and historical materialist thought opens up the prospect of posing important questions that have been ignored or silenced by structuralists and poststructuralists alike. This in turn should aid historical social scientists in constructing a more holistic account of the part played by racism and anti-racism in shaping modernity, including recovering important aspects of the neglected history of white working-class rejection of racism and re-connecting the study of racism and anti-racism to the rhythms of the class struggle and historical capitalism.

This call for a paradigm shift is of course predicated on some sort of claim for the epistemic authority of social science to provide certain sorts of explanations of the social world. Of course, some will charge us with elitism. However, in our view, to insist that social researchers can reasonably claim to know something more about an issue such as racism, anti-racism, migration, discrimination than those who have not researched it is neither elitist nor immodest. It does not make researchers morally or intellectually superior or justify claims for special treatment or privileged conditions, nor does it dismiss the insights and self-understandings of social actors. It is merely to assert that our horizons of understanding, our knowledge of the social world, can only be broadened on the basis of social scientific knowledge; a ‘decorative sociology’, as Rojek and Turner have pointed out, will not do.