The Foundations of Distributive Justice: A Morphogenetic Analysis of Gomberg and Fraser

This article applies Margaret Archer’s morphogenetic analytical dualism framework to critically compare the sociological foundations underpinning the work of two key American political philosophers: Paul Gomberg and Nancy Fraser. First, I focus upon Gomberg’s ideas for reforms to the division of labour and their impact upon race relations in the USA. I argue that, while Gomberg offers a radical egalitarian vision, it is flawed by its weak understanding of the relationships between structure and agency and between culture and economy. Following this, I consider Fraser’s theory of two-dimensional participatory justice premised upon distributive and recognition justice. I argue that, due to a more sociologically nuanced approach, her theory avoids the conflationism and epiphenomenalism that Gomberg lapses into. I conclude by noting that these conceptual issues have pertinence beyond these two authors and have implications for theories of distributive justice more generally.


Introduction
How we understand the relationship between the key sociological categories of agency and structure, and of culture and economy, matters centrally to what political philosophy terms distributive justice; that is, to questions regarding how we should organise our economic and social institutions in order to distribute fairly the benefits and responsibilities of social co-operation (Olsaretti, 2018: 1). This is not simply a matter of academic concern. The rise of a right-wing cultural-identity populism across many advanced economies has been attributed in some part to the effects of growing economic inequalities (Antonio, 2019;Fraser, 2017;MacLeavy, 2019). In this article, I shall discuss the work of two important American political philosophers who have addressed themselves to these conceptual and political questions: Paul Gomberg and Nancy Fraser. I begin by considering the work of Gomberg, a philosopher with radical views on how reforms to the economic sphere (in particular, current divisions of labour) can address cultural antagonisms and act as an agent for progressive change with regard to race relations in the USA but also, by implication, beyond. In a previous article (Morrison, 2019) I have argued that Gomberg's arguments for the normative value of high-quality work provide a valuable framework through which sociologists may make explicit the normally implicit value judgements that their work rests upon. Certainly, these ideas are starting to see wider application within sociological studies (Calder, 2016;Mills et al., 2016) and this points to the significance of his thesis.
However, within this article I shall apply Margaret Archer's (1995) morphogenetic analytical dualism framework to argue that, despite the ambition of Gomberg's philosophical vision, it rests upon a sociologically dubious understanding of agency and structure, and of economy and culture. These shortcomings, in turn, raise questions about its potential to improve race relations. Ultimately, the shortcomings that I raise in Gomberg's thesis are illustrative of wider theoretical fault-lines that relate to our understanding of distributive justice. In the second part of the article, therefore, I go beyond critique of Gomberg to outline the theory of two-dimensional participatory justice developed by Nancy Fraser. Fraser's work is, perhaps, better known than Gomberg's and has attracted much critical attention (Lovell, 2007;Olson, 2008). This thesis, I argue, coheres better with Archer's (1995) analytical dualism which insists on an analytic separation of structure and agency and of culture and economy. Consequently, I conclude, Fraser's political philosophy is more sociologically grounded than that of Gomberg and thus has greater utility as theory of distributive social justice in relation to the harms of structural racism. The contribution of this article, therefore, is to apply a sociological analysis to two important theories of political philosophy, not hitherto directly compared in such a way, with the purpose of elucidating their sociological foundations. As I note in the conclusion, this analysis has implications beyond these two writers as all theories of distributive justice rest upon sociological suppositions.

Gomberg: Contributive Justice
In her discussion of distributive justice, Olsaretti (2018: 4) notes that it may be understood in two quite different ways. On the first view, distributive justice centres upon, 'the mechanisms and procedures that only allocate a given amount of goods, and only a subclass of distributable goods, namely distributable economic goods like income and wealth'. This narrow, moneyist conception of distributive justice is the traditional, dominant perspective within academic and policy circles. Olsaretti (2018: 5) goes on to observe, however, that a wider interpretation of distributive justice is available, which 'can also take as its concern the productive mechanisms that affect which and what amount of distributable goods there are in the first place, rather than merely focusing on the mechanisms for the allocation of pre-given goods'. Gomberg's theory of justice certainly meets this latter description. It is centrally concerned with societal productive mechanisms, in the form of the organisation of work, and with overcoming the problem of racialised patterns of unequal access to a finite number of skilled jobs. In this broad sense, therefore, it may be termed a theory of distributive justice, although it is a label that he himself resists, preferring the term 'contributive justice' to distance his thesis from the economistic overtones of distributive justice (Gomberg, 2018). I too have accepted Gomberg's distinction in previous discussions of his work (Morrison, 2019) but here I employ the term distributive justice in Olsaretti's (2018) second broader sense as a flexible umbrella concept that captures the character of both Gomberg's and Fraser's theories.
There are three main elements to Gomberg's thesis: (a) his argument for the importance of quality work as a key social good; (b) his prescriptions for how to make it more readily available through radical restructuring of current employment arrangements; (c) the claims he makes about how these reforms will address structural racism in the United States. In drawing upon Archer's (1995) analytical dualism framework, I shall argue that, despite its radical philosophical vision, the relationship between parts (b) and (c) of Gomberg's thesis is sociologically tenuous.
The founding element of Gomberg's (2007) theory is that high-quality, intrinsically meaningful and satisfying work is a key social good. The starting point for his thesis is that, while work is necessarily instrumental in being directed towards some exterior goal (provision of food, shelter, etc.), it is not entirely so. Rather, for Gomberg (2018: 514) what distinguishes work from other human activities is that it is a socially organised contribution to a larger group for which we gain recognition and which, ideally, provides some sense of personal meaning. In this sense, work has historically been both a natural and, in consequence, a normatively reinforced activity (Gomberg, 2007). The relatively recent rise of modern capitalism, with its specialised division of labour and proliferation of low-skilled routinised jobs, has disrupted the historically socially embedded and participatory character of work and alienated most workers from its true purpose (Gomberg, 2007(Gomberg, , 2018. Gomberg wants to reclaim the importance of work that is both intrinsically meaningful to the individual in its utilisation of their skills and which, by extension, develops both personal self-esteem and social prestige. Gomberg, in short, follows in a distinguished Aristotelian lineage of philosophers who argue for 'The Good Life' around the principle of meaningful labour. The theory of contributive justice distils Gomberg's own philosophy of The Good Life into three key theses that link the social goods of complexity, contribution and esteem: what we are able to contribute is key to both our own self-esteem and how others esteem us; the contribution of complex abilities wins us greater self-esteem and social esteem than the contribution of routine abilities; our scope to apply and develop complex abilities in everyday social labour has a direct impact upon other social activities and affects our overall sense of well-being (Gomberg, 2007: 66-67). Gomberg (2007) elaborates further on the relationship between these three theses through his discussion of 'norms of identity' and 'norms of prestige'. In terms of paid employment, norms of identity refer to the standards expected of our contribution within a job; in that respect, norms of identity are the norms that are internal to the work and to the skill content of that job. We may compare this with Alasdair MacIntyre's (1981: 188) better-known concept of 'internal goods': the sense of achievement we may obtain from doing a job or other task well in relation to standards within that field of practice. By contrast, norms of prestige relate to the relative social status that a particular job attracts, wherein higher skilled employment tends to enjoy higher social prestige; in this regard, norms of prestige are the norms that are external to any individual employee's actual performance within their job. Again, there is a close analogy with MacIntyre's (1981: 190) 'external goods': the prestige or money that result from performing an activity well but which, unlike internal goods, are competitively fought-over goods which do not have a necessary relationship to the activity itself.
The distinction between Gomberg's two types of norms is not watertight nor meant to be. For example, the norms of identity derived from a competent and valued contribution to a job, including a relatively low-skilled routinised one, may be a source of both selfesteem and social esteem (Gomberg, 1995(Gomberg, , 2007. Furthermore, while Gomberg (2007: 87) is correct to say that wealth is the key determinant of prestige in marketised societies and that low-skill employment is typically low-paid and therefore lacking in social recognition, he also argues that there exists a close correlation between occupations of 'complex mastery' and high pay. On this latter point, Gomberg requires more nuance as the two do not neatly map onto each other. Nevertheless, Gomberg (2007) makes a good general point: that in a highly competitive, socially hierarchical society such as the USA, different forms of employment attract very different levels of social prestige. Furthermore, the social competition of marketised societies means that there exists a closely dependent relationship between the two types of norms: waged labour is a key source of social prestige and because of the centrality of this norm within our social subjectivities, it also strongly shapes most individuals' self-esteem. Moreover, the division of labour that characterises modern capitalist economies and the resultant competition for a finite number of high-skilled, socially prestigious jobs means that both social prestige and, consequently, self-esteem become limited, fought-over social goods (Gomberg, 2007(Gomberg, , 2016. Finally, Gomberg (2007Gomberg ( , 2016 notes that in the USA this competition takes place over an extremely unlevel playing field: deeply embedded discriminatory practices support an enduring racialised black-white division of labour in which African-Americans cluster disproportionately within low-skilled, less prestigious employment. For Gomberg (2007), problems that are rooted deep within the political economy require suitably radical structural solutions, and here I turn to the second part of his theory of contributive justice.
To address the social problems of work, Gomberg (2007) proposes dismantling the division between routine and complex labour, a suggestion that marks his thesis as particularly innovative and radical. He contends that the operative division between what is routine or complex may be found internal to most tasks and is not a naturally occurring division in the organisation of labour (Gomberg, 2007: 81). However, under contemporary employment specialisms, many people in wholly routinised work have little or no autonomy over their own labour processes and so will necessarily become alienated from them. The solution, therefore, is that all workers must share in the routine tasks of such workers, thus freeing this group to develop higher-level skills if they so wish (Gomberg, 2007: 81). This opens up the possibility of sharing out tasks of different skill-levels and interest equally across all workers so that high-quality employment does not become the preserve of particular social groups. For Gomberg (2007) the (at least potentially) unlimited opportunity to access high-skill work produced by these reforms is more radically egalitarian than liberal attempts to provide equal opportunities to obtain a limited number of good jobs within a competitive field. And this emphasis upon equality of outcome leads on to the third element of Gomberg's theory: the claims he makes about how these reforms will address structural racism, and particularly that of the black-white binary divide, in the United States.
The starting point for Gomberg's (2007: 101) ideas about race is that in a political economy based upon finite opportunities for skilled and meaningful work, scarcity has to be socially managed in order to avoid instability. In the USA, race and other categories perform the function of organising limited opportunity through practices of racism (Gomberg, 2007: 37). The solution to this is to make opportunity for self-esteem and social prestige available to all through the work innovations outlined above. Gomberg makes a number of claims for how this will help to tackle structural racism in the United States. On one level, he argues for its effects upon workplace-level discrimination: 'Racism in distribution of work requiring complex abilities is ended when it is not skewed away from black workers' (Gomberg, 2007: 131). On another level, he contends that the impact of work reforms will be felt far beyond the workplace itself and extend to intra-class relations: 'Sharing labor, we can transform relationships and eliminate racial conflict within the working class' (Gomberg, 2007: 129). However, Gomberg's vision is ultimately larger still and aims to reshape social subjectivities: Relationships are transformed. The social psychology of vanity, jealousy of others' accomplishments and, on a group level, racism, arises from competition for limited positions of prestige . . . When labor is shared, esteem earned from contribution of complex abilities is not scarce and others are no longer competitors for it. (Gomberg, 2018: 527) This ideal requires that we eschew moneyist distributive justice since payment for our labour encourages us to see work as a disutility and to labour for ourselves and not for others; in institutional terms, this utopia translates as a 'moneyless communist society' (Gomberg, 2018: 528). In the next section, I draw upon elements of Archer's (1995) morphogenetic analysis to argue that while Gomberg's vision may address the first of these levels -discriminatory workplace practices -its claims to tackle the wider harms of racism founder upon a number of problems with its sociological underpinnings.

Structure and Agency
A key tenet of Archer's (1995) approach is the concept of analytical dualism: an understanding that structure and agency are analytically separable entities distinguished through their distinct properties and temporalities. On this latter point, the notion of antecedence is important. As Archer (1995: 165) notes, 'Society is that which nobody wants, in the form in which they encounter it, for it is an unintended consequence.' In other words, we are all born involuntaristically into a pre-existing society, the structure and culture of which shape us, but which we, in turn, also shape. Different temporalities also produce different properties. Because we are all born into a society, the structural and cultural relations which we inherit and which help to form us are relatively autonomous from agency in the sense that they pre-date present-day social actors and therefore have a degree of free-floating 'objective' life outside of any individual's apprehension of them. Agency is crucial however since, as Archer (1995: 184) insists, socio-cultural conditioning is only able to apply its effects on people and is only effective by means of people, with the result that no prior structural or cultural influence functions deterministically as a 'hydraulic pressure', but rather may be open to reflective evaluation by agents. Archer (1995) elaborates further on the complexity of the structure-agency relationship through the three-stage morphogenetic cycle, whereby: (a) structural and cultural antecedent conditions develop as a consequence of past social actors' socio-cultural interactions, their temporality making them relatively autonomous from; (b) present-day social and cultural interaction where social actors (individual and collective) pursue interests under conditions obtained in stage (a); (c) structural and cultural elaboration where social actors and agents both consciously and unintentionally reproduce/transform the conditions at stage (a).
What, though, does this analytic tell us about Gomberg's own treatment of agency and structure? As I have noted, Gomberg's sociology is clearly structuralist in its attendance to processes of cultural inferiorisation (racism) through macro-economic reforms (work arrangements). However, Gomberg's view of the relationship between structure and agency falls into a form of 'downwards conflationism' whereby 'structure and agency are conflated because action is treated as fundamentally epiphenomenal' (Archer, 1995: 81). For Archer (1995: 83) downwards conflationists lack historicity because they do not accept that social structures have been created by the past actions of social agents -that social structures have social origins. Consequently, as Archer (1995: 84, emphasis in original) argues, they 'basically restrict their treatment of structure and agency to an examination of the impress of structure upon agency in the present'.
To be clear, the charge against Gomberg in this respect is not that his analysis of racism lacks a historical perspective per se. In various publications he has demonstrated a detailed knowledge of the deep historical roots of present-day white-black relations in the USA (Gomberg, 2007(Gomberg, , 2016(Gomberg, , 2017. The point, rather, is that Gomberg's historicity collapses under the burden of expectations that he applies to structural reform. The weight of cultural history and its semi-autonomous conditioning of social actors' perspectives are expected simply to be lifted away through changes to present-day structure in the form of the division of labour. In terms of Archer's (1995) three-stage morphogenetic cycle, it appears that Gomberg's claims for the positive impacts of labour reforms on structural racism move us from stage (b) to stage (c) without sufficient attention to the conditioning influences of stage (a). And this is an important omission: a voluminous body of literature illustrates the significance of antecedent cultural relations in shaping the enduring but always evolving character of white-black racism in the United States (Alexander, 2012;Omi and Winant, 2015;Wacquant, 2000). In fact, the absence of a sense of temporality in Gomberg's thesis accords structure both an overly substantial role (as being over-determining of agency) and a curiously insubstantial one (whereby centuries of racism dissolve).

Structure/Economy and Culture
Another important element of critical realism, as elaborated by Archer (1995), is the distinction between what she terms structural and cultural 'emergent properties' (EPs). For Archer (1995: 167), an emergent property (EP) essentially refers to society's deep sub-structures which are relatively enduring and which possess causal powers over social life. Again, the notion of antecedence is important here since they are formed through the past socio-cultural interactions of previous social actors but exert a conditioning (though never determining) influence upon present-day actors. Importantly, the term 'emergent' is not simply a synonym for 'combination'. An EP is a complex of different interlinking elements which are characterised by the 'natural necessity' of their relationship to each other and which are not reducible to individual parts (Archer, 1995: 167). Class relations and the positions that social groups occupy within them are a good example of an EP. For example, the relationship between income distribution and class location is a necessary one in the sense that one cannot be understood without reference to the other. Race is also an EP in that same sense because ideas about race are a complex amalgam of inherited cultural perspectives, interpreted by present-day social actors, which exist in some form of logical relation to other ideas with which they variously compete, concur or contradict: a belief in tolerance, equality or freedom (Carter, 1998: 7).
Class and race also exemplify Archer's (1995) distinction between structural and cultural EPs. Here, Archer (1995: 175) employs the term 'structural' to denote that certain types of social relations depend primarily upon material resources, both physical and human. Structural is not, therefore, coterminous with 'economic' within Archer's (1995) schema but, rather, the latter is subsumed within the former. And it is in this sense that I employ the term structural in my critique of Gomberg's approach to the relationship between economy and culture. For Archer (1995: 175), class relations are a clear example of a structural EP, being necessarily defined ultimately by their non-discursive, material relations. This is a point supported by Ray and Sayer (1999: 14) who also note that, by the same token, distributions of income or wealth have no necessary connection to any discursively ascribed characteristics of social groups such as racialised minorities. Another aspect of the core materiality of structural EPs such as class relations is that their existence may ultimately be sustained by extra-discursive means. Thus, although material relations (such as the distribution of income) may be maintained by ideas, they are not necessarily so and may, in the final instance, be enforced by coercion or manipulation, thereby eschewing the need for legitimation (Archer, 1995: 175).
In Archer's (1995: 180) schema, culture is distinguished from structure through its necessarily subjective, propositional properties; it refers to all 'intelligibilia' -items with the dispositional capacity of being understood and interpreted by individuals. The social construction of race is just such a cultural item. And, as a cultural item, ideas about race have an objective existence. This is not in the sense of being right or wrong; rather, it is by virtue of the fact that such ideas are a product of past social interactions and have now slipped free of their progenitors through relatively enduring representations in books, films, theories, discourses and so on which now condition present-day social actors' interpretations (Carter, 2000: 83). Again, however, the temporality of analytical dualism permits us to see agency as the key mediating factor here. This temporality is best expressed in Archer's (1995) distinction between 'system integration' and 'social integration'. The former refers to both structural and cultural antecedent conditions while the latter indicates present-day socio-cultural interactions (Archer, 1995: 183). The two forms of integration are by no means always in congruence and, in times of social friction, may be at odds. Consequently, as Archer (1995: 181) observes, while culture may inhibit, it also embodies new possibilities and thus agency is never simply a direct restatement of cultural structures.
How then does this framework help to cast a light on Gomberg's treatment of economy and culture in his discussions of race and the transformation of work practices? Gomberg's analysis is of a clearly Marxian kind: the origins and effects of racism are viewed through the lens of class oppression. Gomberg (2017) applies the classical Marxian Reserve Army of Labour (RAL) thesis to argue that high unemployment and a low-wage, vulnerable workforce is fundamentally functional to the efficient workings of capital accumulation within the USA. Furthermore, the black-white binary has historically been, and continues to be, the principal organising categorisation through which the RAL has been constituted and maintained (Gomberg, 2007(Gomberg, , 2017. Gomberg (2017) notes the enduring power of oppressive racist structures in the USA and their ability to adapt to resistance and to evolve new forms of domination: the shifts from chattel slavery, to Jim Crow segregation to more recent forms of exploitation via welfare, immigration and mass incarceration policies. This is an analysis he shares with other US commentators on race (Alexander, 2012;Omi and Winant, 2015;Wacquant, 2000). Where Gomberg differs from these writers, however, is in his insistence that, because racism is functional to capitalism, it is also a by-product of it that can only be removed by the ending of capitalism itself (Gomberg, 2017: 73). And it is here that race, in Gomberg's analysis, becomes collapsed into class. Gomberg (2007Gomberg ( , 2017 is at pains to emphasise that he does not dismiss the significance of the particular harms of racism in a society predicated upon race divisions. Nevertheless, it is clear that racism, as expressed in racialised divisions of labour and skill, is ultimately seen as a sub-category of wider class inequalities. He makes this point explicitly in arguing that, 'what black people experience is a severe and generalized version of what many workers of all "races" experience at work every day. Subordinate workers at work are not full persons' (Gomberg, 2007: 135). He elaborates on this point, perhaps controversially, with what might read as a rather careless conflation of degraded labour with formal chattel slavery: Explore the term 'wage slavery', used by nineteenth century socialists. A slave is a body under the will of another; the slave's will is impotent. A chattel slave is a slave all of the time. A wage slave is a slave for a period of hours. (Gomberg, 2007: 134) The sum of this analysis is that race is not accorded a special causal status -and the corollary to this is that the solution to racist work practices is not to focus on race per se, by addressing the conditions of African-Americans in the USA, but by aiming to eliminate the class inequalities that the capitalist division of labour inevitably produces: if it is wrong for black people to suffer from higher unemployment, why is it better for others to suffer from unemployment and the concomitant loss of status? If it is wrong for the routine jobs to be skewed toward black workers, why is it better if someone else must work those jobs? Equality requires sharing labor. (Gomberg, 2007: 136) There are potentially many things to unpick from Gomberg's arguments but, for the purposes of this present article, I shall apply critique by making further reference to Archer's (1995) distinction between the necessary and the merely contingent relations between social entities. As indicated previously, where the relations between the components of a social item serve to define what the entity is to the extent that its very existence depends on these, these may be characterised as necessary relations. By contrast, contingent relations between entities are characterised by a relationship that, although it may be felt experientially close, are not necessary because it is not impossible that one could exist without the other and the nature of either does not require them to exist together (Archer, 1995: 173).
Following this framework, I argue that Gomberg mistakes contingency for necessity in his assumptions of the relationship between culture (race/racism) and economy/structure (the division of labour) in that both could exist without the other. If we focus on the culture-economy side of the equation, there is no reason in principle why capital accumulation, even in the USA, could not take place in a differently racialised or even nonracialised social system (Carter, 1998(Carter, , 2000Ray and Sayer, 1999) -a point we may extend to other forms of social domination such as patriarchy (Sayer, 2005: 88). At this point, I want to echo Carter's (1998: 7) comment on much the same point: contingent here does not imply arbitrariness; race ideas and racist practices have played a central role in defining whose labour power should be sold and under what conditions, and nowhere is this more true of advanced capitalist nations than the USA. Capitalism, in other words, undoubtedly benefits from structural racism in its concrete practices. Nevertheless, the relationship is ultimately a contingent rather than a necessary one. However, the more pertinent question for my argument is the economy-culture side of the equation: would the abolition of capitalism (at least as presently constituted through current divisions of labour as per Gomberg) serve to eradicate structural racism in the USA? Or, to put it another way, can the latter exist without the former? Or, on a more conceptual plain, does culture have its own semi-autonomous causal powers outside of structural/economic conditions? I believe that Gomberg's philosophy pays insufficient attention to these fundamental sociological questions and falls short in consequence. Race is collapsed into a sub-category of class and Gomberg's radicalism is then restricted to that conceptual lens with the consequence that reforms that may address class relations directly would only engage tangentially with racism. Thus, the effect of Gomberg's (2007Gomberg's ( , 2017 changes to the division of labour is to restructure some of the necessary and internal properties of class relations. This approach is consistent with Gomberg's (2018: 518) focus upon workplace relations and his reforms may well serve to tackle discriminatory racism in the distribution of skilled and meaningful employment. However, as I have indicated, Gomberg goes beyond this claim to argue that contributive justice may address the wider cultural harms of racism. My critique here, though, is that employment relations are by no means the only site for the reproduction of racist practices and that too much is being asked of workplace reforms to effect wider cultural change. This, in turn is a consequence of Gomberg's treatment of culture as epiphenomenal to economy/structure in a way that fails to acknowledge the distinction between the two and the semi-autonomous causal powers of each from the other.
And the literature on race in the USA offers powerful evidence to substantiate the need to maintain a binary economy/structure and culture analysis. There is not space here to do justice to the richness and complexity of the theoretical perspectives which have been applied to this area. In essence, though, while many prominent writers within the field recognise the central role of capital accumulation in creating the conditions for race relations in the USA, race itself cannot just be reduced to class (Alexander, 2012;Omi and Winant, 2015;Wacquant, 2000). There is simply too much else to explain about racism that cannot be understood through an exclusively class-focused conspectus. Empirical studies into race in the USA also support the contention that race ideas possess their own distinct causal powers and effects quite apart from those of class. Even where class (in the economistic sense) is held constant, race powerfully conditions where individuals live and, consequently, the resources to which they have access (Moore, 2008;Sampson, 2019). Similarly, recent studies employing a Critical Race Theory perspective have shone a powerful light upon the everyday microaggressions to which African-American pupils are subject and from which a middle-class economic status provides no refuge (Allen, 2012;Reynolds, 2010).

Disciplinary Boundaries?
Thus far I have discussed the theory of contributive justice as elaborated by the US political philosopher Paul Gomberg. I have applied some key elements of Archer's (1995) analytical dualism to argue that, despite the radicalism of Gomberg's philosophical vision, the claims he makes for his proposals in relation to racism lack sound sociological foundations. At this point, however, a critic may counter that contributive justice is the work of a political philosopher and not of a sociologist. While the work of a sociologist lies primarily in analysing what is, a political philosopher concerns themselves with what ought. Consequently, Gomberg's focus is necessarily normative as opposed to the more analytically descriptive work of sociology: it is upon the elaboration of a utopian future. The utopianism of contributive justice is a quality that Gomberg (2007: 158) strongly defends against possible charges that his thesis is overly idealistic. His utopianism, he argues, is of the type that floats ideas for debate and which in consequence is unobjectionable; and as a philosophical thought-experiment there may be little objection to the fact that his thesis places its focus upon the third stage of Archer's (1995) threestage cycle -structural and cultural change -while the first two stages are relatively neglected.
Nevertheless, if we continue in applying Archer's (1995) morphogenetic analysis to the theory of contributive justice, it is clear that Gomberg is making claims about the causal powers of his proposed reforms to social and cultural structures. That is, in asserting the transformative effects on racist social structures of his restructurings of the division of labour, Gomberg is making claims of an inescapably sociological type. It is appropriate, therefore, that a work of political philosophy that makes such contentions is held up to a measure of sociological scrutiny to evaluate its utility as a theory. And, on this point, I now turn to the work of Nancy Fraser.

Fraser: Two-Dimensional Participatory Justice
As a critical social philosopher, Fraser's work, like Gomberg's, is centrally concerned with the analysis of and remedies for social injustices under conditions of advanced capitalist development. For Fraser (1999), as with Gomberg, society is composed of real structural groupings that are based in relatively enduring socio-cultural orders of differentiation and hierarchy. Class, gender and race have been key areas of interest. However, a key point of departure between the two writers lies in what I argue to be Fraser's more sociologically sophisticated treatment of the relationship between economic and cultural forces and between agency and structure.

Economy and Culture
In adopting a 'dual-systems' approach, Fraser contends that inequalities are the product of the comingling of analytically separable economically rooted and culturally rooted forms of injustice. And, perhaps paradoxically, it is the very complex mixing of the two, an effect of developments in late modern capitalism, that requires such a dual analytical approach. Thus, processes of personal responsibilisation now mean that cultural activities have become increasingly, although not wholly, economically inflected; similarly, 'the economy is not a culture-free zone, but a culture-instrumentalizing and re-signifying one. Thus, what presents itself as "the economy" is always already permeated with cultural interpretations and norms' (Fraser, 1999: 44). The role of critical theory for Fraser is to problematise accepted substantive distinctions between the 'public' and the 'private', and the 'economic' and the 'cultural'. Her contribution in this respect is a 'twodimensional participatory parity' theory of social justice premised upon what Fraser (1999) terms 'perspectival dualism'.
Perspectival dualism offers a distinction between injustices associated with cultural inferiorisation that originate in society's cultural-valuational order (what Fraser terms 'misrecognition') and those that have their roots in unequal material-economic arrangements (which Fraser (1999) calls 'maldistribution'). The relevant forms of remedy are 'recognition justice' and 'distributive justice' (Fraser, 1999). These distinctions are themselves based upon Fraser's (1999) division between 'economy' and 'culture'. Fraser (1999) is at pains to emphasise that the distinctions she makes between these categories are analytical and not substantive or ontological. Economy and culture (and their attendant injustices of misrecognition and maldistribution) are 'social processes and social relations' that have emerged as a consequence of the fact that, in advanced capitalist societies, class (as understood in the economistic sense) and status (in the Weberian sense) no longer neatly map onto each other (Fraser, 1999: 40). In real life, all social practices involve a concrete intertwining of both economic and cultural dimensions (although not always in equal degrees). For Fraser (1999), though, the value of the analytical separation she makes between the two forms of injustice, and their associated remedies, is that it permits us to focus upon the cultural aspects of what may normally be regarded as economic policies and to highlight the economic dimensions of what are typically seen to be cultural processes (Fraser, 1999: 45).
The comparisons between Fraser's schema and Archer's distinction between structural and cultural EPs are apparent. In employing a dual-systems approach, Fraser is able to accord both economic and cultural forces causal powers in a way that Gomberg's sociology cannot. Thus, her thesis avoids the pitfall, which I have argued that Gomberg enters into, of treating culture as mere epiphenomenon to economy and of mistaking contingency for necessity in the relationship between race and divisions of labour. On a Fraserian reading, racism is a product of a complex concrete intertwining of cultural forces (race ideas) and economic forces. Race itself is, therefore, a key example of what Fraser (1995: 74) terms a 'bi-valent' category: one that is rooted simultaneously in the economic structure and in the cultural status order of society. And, this form of perspectival dualism requires an appropriately flexible normative theory of justice.
Fraser's perspectival dualism translates into a critically normative theory of justice that addresses itself to economically rooted distributive justice and culturally rooted recognition justice: a twin approach summed up by the concept of 'parity of participation'. (A later development of this theory incorporated a third dimension of representation. For the purposes of this article, however, I shall limit discussion to Fraser's original two dimensions.) By parity of participation, Fraser (1999: 50n, emphases in original) means, 'the condition of being a peer, of being on a par with others, of standing on an equal footing'. There are three key criteria for this, all of which address themselves to the workings of social structures. First, the theory stipulates legal equality which refers to full recognition of rights before the law. The second criterion requires that the distribution of material resources be of a reasonably equitable level so as to ensure the potential for all social actors' 'voice' and independence. These two requirements are what Fraser (1999) terms the 'objective' preconditions of participatory parity. The third precondition is 'intersubjective parity'. This criterion demands that the 'cultural patterns of interpretation and evaluation' expressive of social institutions confer equal levels of respect upon all social actors and thus provide them with equal opportunities for gaining social esteem (Fraser, 1999: 37). Both types of preconditions need to be met for individuals to achieve full participatory parity, as neither alone is adequate.
It will be evident from the above that two different treatments of the categories of economy and culture have produced two quite distinct theories of distributive justice, if we accept Olsaretti's (2018) second more generous view of the concept. Thus, Gomberg's particular take on Marxian structuralism has led him to propose economic reform in order to effect cultural change, the idiosyncrasy of his vision lying in the rejection of financial redistribution in favour of redistribution of the social goods of meaningful work and social recognition. By contrast, Fraser's two-dimensional theory incorporates a traditional narrowly economistic view of distributive justice, but this is complemented by an analytically parallel concept of recognition justice aimed at the more equal distribution of cultural recognition. Furthermore, differences in the two writers' treatments of structure and agency become apparent if we consider Fraser's discussions of her proposed remedies for social injustices.

Structure and Agency
Fraser shares Gomberg's radical egalitarianism, although her proposals are quite different. She argues that the solution to deeply rooted structural injustices (both economically based and culturally based) is through what she terms 'transformative remedies', that is, 'remedies aimed at correcting inequitable outcomes precisely by restructuring the underlying generative framework' (Fraser, 1995: 82). A good example of this approach can be found in Fraser's (2013: 134) vision of a 'Universal Caregiver' welfare state (for the USA) in which all jobs would be designed with a much shorter working week than present-day full-time work to meet the needs of workers who are caregivers; domestic caregiving itself would receive substantial public financial support and be incorporated into a single social insurance system on a par with paid work. Fraser's purpose here is to effect distributive justice via a mass reallocation of state financial resources towards the unpaid work of domestic care that capitalism relies upon. Equally, though, Fraser (2013) seeks to achieve radical recognition justice: state financial support for caregiving aims to dismantle the androcentric norms that sustain the highly gendered binary divide between paid work and unpaid domestic labour and to share out the latter equally. And the form of political economy that Fraser (1995: 84) most closely associates with these types of remedies is socialism; however, it is a socialism that accepts tightly regulated market relations where these can offer the potential to counter extra-economic forms of domination (Fraser, 2013: 232).
Thus, Fraser, like Gomberg, offers a far-reaching philosophy for change. However, where Fraser differs from Gomberg is in her more sophisticated acknowledgement of the conditioning effects of antecedent economic and cultural structures and their impact upon the potential for change. In different articles, Fraser has drawn upon the socio-historical frameworks offered by (among others) Polanyi (Fraser, 2013) and Gramsci (Fraser, 2017) to more fully historicise her normative ideas. Fraser recognises, therefore, that structure can be enabling but also constraining -a perspective that sits well with Archer's (1995) critical realist treatment of structural and cultural relations. Consequently, Fraser is realistic about the extent to which her proposed remedies may be realisable. She notes, for example, that although socialism is 'cognitively compelling', for most people it is 'experientially remote', while a transformative politics of recognition, 'could turn out to be too negative and reactive, i.e. too deconstructive, to inspire struggles on behalf of subordinated collectivities attached to their existing identities' (Fraser, 1995: 91, n46, emphasis in original).
As I have previously indicated, Gomberg's discussions on race are by no means without a sense of history; my argument has been that history collapses under the weight of expectation attached to structural reform and agency becomes sidelined as epiphenomenal to structure. And here I believe that Fraser has a stronger understanding than Gomberg of the importance of agency in mediating structure to effect change. To bridge the gap between the ambition of her transformative remedies (located at stage three of Archer's morphogenetic cycle) and the conditioning effects of prior structural and cultural relations (stage one of Archer's cycle), Fraser accords agency (stage two of Archer's cycle) a stronger role than appears within Gomberg's thesis. This is well illustrated by Fraser's powerful commitment to public dialogue as a means to determine the public needs that transformative remedies may deliver upon. In line with her dual-systems approach that sees the economic in the cultural and vice versa, Fraser is dedicated to fostering a public debate that actively troubles accepted borders between the two spheres. As Fraser (1997: 88) contends, social inequalities become legitimated and entrenched when the discourse of 'economic privacy' attempts to partition some interests or issues from the sphere of public debate by 'economizing' them and framing them as simple private or technocratic matters pertaining to impersonal market forces, the result here being to 'enclave' social problems to the disadvantage of subordinated groups (Fraser, 1997: 88).

The Sociology of Distributive Justice
This article has made a study of the work of two political philosophers not previously directly compared. I have applied Archer's (1995) morphogenetic analysis to evaluate the assumptions on the relationship between structure and agency, and between economy and culture, that guide their theories. I have considered how the cogency of both writers' ideas, as theories of distributive justice in relation to racism, rests in large part upon their different understandings of these foundational sociological concepts. And I now conclude with some further comments on the role that sociology may play in relation to such theories.
In her summary view of the field, Olsaretti (2018: 3) notes some features of theories of distributive justice. The first pertains to their preconditions: the circumstances that must exist for questions of distributive justice to be relevant at all. Here, Olsaretti (2018: 3) observes that a consensus has coalesced around the view that questions of distributive justice emerge only when there is a relative scarcity of goods -neither great plenitude nor an extreme want. Under such conditions, questions of identity and conflicts of interest arise that require principles through which to reach an equitable solution. The second feature is the primary subject of distributive justice, which refers to where the injustices (and attendant remedies) should be located: in individual acts, all social practices or just particular institutions (Olsaretti, 2018: 4). Finally, for the purposes of this discussion, there is the object of distribution, which relates to what is to be distributed: money goods and/or other social goods such as social recognition or esteem.
If we consider the first of these features, we can see that race in the USA and beyond is certainly a site of relative scarcity of both material and non-material social goods. Having established such preconditions, a theory of distributive justice seeking to address the harms of racism will typically then be guided by underlying assumptions regarding its subject and its object. It has been the argument of this article that a cogent understanding of the subject of such a theory (the social locating of racism) requires an analytical dualism that permits us to see racism as both economically and culturally rooted and as a product of the dialectic between structure and agency. Analytical dualism of agency-structure is a necessary sociological frame for a theory of distributive justice of race because, although racism is powerfully conditioned by antecedent cultural and economic structures, it is also ever mutable and subject to the agency of socio-cultural interaction. Finally, following Olsaretti (2018), a theory of distributive justice directed at redressing racist injustices must then decide upon its object. However, where there is a theoretical lacuna with regard to the subject of the theory, this may follow through to its object. This has been a criticism that I have levelled at Gomberg's vision of effecting broad cultural change through economic levers: that the latter is insufficient to support the weight of the former. This, then, provides a clear argument for the importance of maintaining an understanding of the relatively autonomous causal powers of economic and cultural structures. Theories of distributive justice, whether focused upon race or other axes of social hierarchisation, are embedded in sociological assumptions; sociology, therefore, has a valuable role in inspecting the foundations of such theories.

Funding
The author received no financial support for the research, authorship and/or publication of this article.