The dialectics of unity and difference in the constitution of wage-labour: On internal relations and working-class formation

Introduction

Imagine that the left-wing debate between the politics of identity and those of class were recast in dialectical terms. Imagine, in other words, that the social relations of race, gender and sexuality, among others, were understood to be internally constitutive of class – rather than as radically external to it. To engage the thought experiment is, of course, to recognise how foreign dialectics are to our current intellectual and political moment. The flourishing of dialectical thought is invariably bound up with moments of mass insurgency and dramatic leaps in the self-organisation of the oppressed, such as the period 1917-23, which saw the theoretical and political innovations represented by Gramsci and the early Lukács, or the great revival of dialectical Marxism during the decade of upheavals associated with 1968. Periods of pronounced retreat for emancipatory politics, on the other hand, have their counterpart in the prevalence of mechanistic modes of thought within the left. If the debate between personal identity and class has been conducted in pre-dialectical terms, then this comprises an index of our de-radicalised historical moment. Yet, precisely in such periods the task of nourishing the vital resources of dialectics takes on a special urgency. For only the mole-like work of burrowing underground during the dark days can sustain those critical-dialectical energies which, as in Benjamin’s ‘secret heliotropism’ (Benjamin 1968: 255), bend toward the future, anticipating the coming ‘sunburst which, in one flash, illuminates the features of the new world’ struggling for actuality (Hegel 1977: 7).

It is in this spirit that I offer the current article. It is not my purpose to return to the texts that have framed the left debate between personal identity and class. ¹ Instead, I seek to overcome the impasse of this debate by way of a dialectical retrieval of key Hegelian-Marxist insights. This retrieval is thus futural in nature: it seeks to recover past resources whose time is yet to come, theoretical intimations of the not-yet that are indispensable to revolutionary thought and praxis (Bloch). Seeking to transcend the mechanistic thinking that has frequently inhabited debates over identity and class, I direct my challenges largely at the self-defined Marxist camp. Not only is it this camp that ought to be capable of superseding pre-dialectical thinking; but it is also the camp for which the stakes are highest, consisting as they do in the imperative to understand the world the better to change it. ² With these stakes in view, I urge that there are more powerful, more theoretically and politically robust ways in which we ought to represent the claims of class. I argue that the (abstract) universalism of bad class politics is simply the dialectical double of the bad (abstract) particularism of personal-identity theorists. I undertake therefore to dialectise the terms of debate – by theorising the reciprocal inter-constitution of the one and the many (and of identity and difference) that makes possible the concrete universality of wage-labour as a class – in order to transcend the bad universalism of class, which derives, I shall show, from a propensity to imagine wage-labour in the image of capital.

It is with the latter point that my exposition begins. Starting from Hegel’s critique of essence-thinking, which takes the inner form of things to be fundamentally opposed to their outer modes of appearance, I trace the way in which essence-categories of this sort are ‘true’ of the inverted world of capital. In the social universe of capital, in other words, concrete particulars are (violently) abstracted so as to be subsumed under the bad universality of value/capital. Nevertheless, the opposition between particulars (use values) and universals (exchange values/value) in fact manifests a unity in estrangement. For a whole series of positions within the left, the universality of class is imagined in the same terms – as one that overrides and suppresses differences rather than being constituted in and through them. The result is an oppositional positioning of class and particularised identity that conceals a deeper unity, albeit a unity-in-estrangement, as each side participates in a bad abstractionism that presupposes (and thus requires) its opposite. Of such opposites, Hegel notes, ‘Ordinary thinking … forgets their negative unity and so retains them as merely “differents” in general’ (Hegel 1969: 441). So it is with the debate over personal identity and class, as all too frequently one side valorises parts (particularised identities) that are atomised and abstracted from their systematic relations with other parts in a organic whole, while the other side posits the whole as a ‘false totality’ in opposition to the parts that comprise it (Kosik: 27). In the case we are examining, notions of class counterposed to differentiated identities are pseudo-concrete precisely because they posit class as devoid of all the concrete particularities and determinations that constitute living social groups. By stripping ‘class’ of the multiple determinations in and through which it is composed, such approaches render it a purely empty and formal universal bereft of the dynamic content of a living, organic relation. In Hegelian terms, the counter-position of universality (class) to particularity (personal-identity) is characteristic of the irresolvable alternations between categories of being and essence that comprise the standpoint of bourgeois thought – or, as Hegel prefers, of the Understanding – in contrast with the standpoint of Reason, which grasps the constitution of universals in and through their differences, i.e. as concrete universals.

Capital and essence-categories

For Hegel, the standpoint of essence emerges when thinking has moved beyond immediate notions of being derived from sense-experience and perception. The essence-view realises that ostensible particulars are irreducibly universal. So, when I say of the large, multi-branched, green-leafed entity outside my house that ‘this thing I see right now is a tree’, my very terms of expression immerse me in universals, like the concepts of this, tree and now, all of which can pertain beyond this immediate moment. The language necessary to my description inherently links a particular, the tree outside my house, to universal categories that transcend my immediate experience. Indeed, without doing so my putative experience would be incommunicable (Hegel 1977: ch. 1).

In its break from immediacy, therefore, thought proceeds to develop essence-categories. It grasps that particulars invariably embody and participate in universal relations. But because these universals are not immediately accessible to sense-perception, particular phenomena are taken to be outward appearances that conceal their governing essences. While this is a necessary move for thought, it produces another partial (and hence partially untrue) mode of conceptualisation. For the move toward essences as the truths that lurk behind appearances devalues the necessarily rich and diverse forms of experience, and it generates abstracted notions of essence from which all particular determinations have been removed. Hegel considers this standpoint to be represented paradigmatically by Newtonian physics and Kantian metaphysics. Reviewing his critique of these shapes of thought should clarify the fundamental shortcomings of essence-thinking.

Perhaps Hegel’s most famous treatment of this ‘shape of consciousness’ occurs in the third chapter of the Phenomenology of Spirit, entitled ‘Force and Understanding’. Modern scientific thinking (the Understanding) realises, says Hegel, that the universal forces at work in the natural world – such as gravitation – cannot be derived directly from perception and sense-experience. In seeking out these universal forces, the Understanding thus disregards the actual content of our experiences – e.g. specific objects and their many properties, or what Hegel refers to as ‘matters’ – in search of the essential laws apparently animating them. Objects of experience (‘matters’) are rendered inert, as if they merely respond to the impact of external forces. The world is thus understood in fundamentally mechanistic terms, as a lifeless machine whose animating principle comes to it from the outside. Consequently, the universal (Force) is conceptualised as outside and independent of actual objects of experience, as if essences (natural laws) pertained to a ‘supersensible world’ persisting ‘above the sensible world’ (Hegel 1977: 87). Force takes the form therefore of an abstract One in opposition to the multiplicity and diversity of objects of experience; ‘essentially it has the form of the supersession of the existing “matters”, or is essentially a One’ (Hegel 1977: 83). The standpoint of the Understanding is thus rife with dualisms – between form and content, the supersensible and the sensible, the One and the many, the very antinomies which define Kantian metaphysics. While Kant may go beyond Newton in arguing that the supersensible realm of universal laws subsists on the side of (transcendental) consciousness, he remains trapped by the dualisms of mechanistic essence-thinking. Yet the price of these dualisms is that our ostensibly universal scientific laws become ‘more and more superficial’ (Hegel 1977: 91). Having turned away from the diversity, multiplicity and dynamism of experience, scientific thought empties itself of the actual content of life, just as Kant’s transcendental structures (pure forms) of perception and cognition are said to be prior to (and hence emptied of) all empirical experience of the world. Consequently, the Understanding in its Newtonian and Kantian guises constructs an abstract, mechanistic universe in which dead things are moved and determined by all-pervasive forces independent of them. Whatever this might be, Hegel insists, it is not our world; it is not the world in which we live our daily lives. That world is alive with dynamic interactions. Humans act in and upon the world, transforming the objects around us which, in turn, change us by changing our environment and field of action. Our natural and social world is thus always in motion, constituting a field of reciprocally-determining transformations, not external connections between inanimate bits of matter. Yet having generated a supersensible realm whose unity ‘is an abstraction’, modern scientific consciousness has estranged itself from the real content of the world of living beings (Hegel 1977: 100). As a result, the Understanding collapses amidst its own contradictions: seeking the universal content of experience it has instead trapped itself within a deracinated system of ‘scientific’ laws incapable of engaging the complex structure of any and all actual experiences. Seeking to cognise the world of sensible experience, science has ended up confining itself within an empty, formalistic and tautological realm of the ‘supersensible’.

Hegel returns to this question in the Logic, where he describes the ways in which essence-categories involve the stripping away of appearances, i.e. the particular forms in which essences manifest themselves. This leaves us, he observes, with essences in which ‘the indeterminate, simple unity from which what is determinate has been eliminated in an external manner’. The result is a system of essences that reduces the diverse material of experience to ‘empty oneness’. And this means that particular elements of experience have not been grasped as inherently universal, just as universals have not been comprehended as constituted in and through their differences. Instead, particulars/appearances and universals/essences have simply been separated from one another, assigned different ontological statuses, or consigned to different sides of a theoretical ledger, as in Kant’s radical separation of the sphere of transcendental knowledge from the realm of (unknowable) things-in-themselves. While particulars are thus left in their apparent immediacy, as simple, immediate differences, so are ‘inherently lifeless and empty’ essences posited which ‘lack all determinate character’. Since determinations are not posited within essences, but merely treated as foreign to them, these essences lack the living features that endow them with dynamism, the aspects of becoming and transformation that are inherent in real essential relations (Hegel 1969: 389-90). We are left therefore with a concept of essence as ‘something unaffected by, and subsisting in independence of, its concrete phenomenal embodiment’ (Hegel 1892: 210).

Many readers will recognise this (undialectical) counter-positioning of essence and appearance – and the attendant notion that one must strip away appearances to reveal essences – as widespread in readings of Marx’s critical project. Further, one finds such thinking in accounts of social class as a hidden essence in opposition to the immediate forms of everyday thought and experience. Yet, I will argue, this is to posit class as an abstract universal, an essence devoid of the living content of experience. However much he recasts it in historical-materialist terms, Marx shares Hegel’s critique of abstract universals. He too seeks a unitary account of essential relations and their necessary modes of appearance, one in which the specific forms of appearance are necessary to, and thus inseparable from, their essential forms. While Marx sees the system of capital as one which cannot overcome the antagonistic dichotomisation of form (universal) and content (particulars), he projects the self-emancipatory constitution of the global working class as a concretely universalising movement that genuinely integrates the One and the Many. Indeed only such a movement – and its reconciliation of particulars and universals, the many and the One – could represent a transition to the realm of freedom. We may better see what is involved here by more fully clarifying Hegel’s position and the conception of internal relations that it entails.

In essence-thinking, as we have seen, appearance and essence, particularity and universality cannot be truly integrated and harmonised. Each term is set in opposition to the other with truth being assigned to the side of essence. Instead of the particular determinations that comprise a phenomenon being grasped as self-determining aspects of a whole, the whole is posited by way of abstraction from these determinations, the latter then being subordinated to the former. In this conception, as we have seen, universal laws (forces) impress themselves on particulars from the outside. Considered as inert bits (e.g. atoms) particulars are treated as lifeless, at best externally related to one another by means of powers beyond them. ‘This is what constitutes the character of mechanism, namely, that whatever relation obtains between the things combined, this relation is one extraneous to them that does not concern their nature at all’ (Hegel 1969: 711). In turning its back on particular determinations, in abstracting from the world of diverse and multiple objects, essence-thinking loses sight of ‘Life as a process’ (Hegel 1977: 107); and to recover the latter it must recuperate the content of life, the dynamic and contradictory development of universal life forces in and through the complex differentiatedness of actual objects. This is to say that it must encounter the identity of universal and particular, by retrieving a world in which the universal ‘is self-identical, but is also in itself different’ (Hegel 1977: 99). In insisting on the identity of identity and difference, Hegel is setting out the philosophical programme of concrete universality. However, it is only when we get to the Concept, at which point Reason attains actuality, that we arrive at concrete universality, a living and reciprocal unity of part and whole. Particular determinations are then grasped as constitutive of the wholes of which they are parts, and wholes are understood to be empty and lifeless when conceived in abstraction from their parts. As a result, parts are brought back to life and allowed to express their internal relations with other parts in the generation of a totality (Ollman: ch. 3). Concrete universals are thus governed by relations of reciprocity in which appearances and essences, parts and wholes are co-determining (Hegel 1969: 569-71). It follows that determination has become self-determination, part of the self-development of living organisms, rather than the external determination of lifeless entities.

Only at this stage can we say that ‘differences are no longer “swallowed up” by the pole of unity, or unity is no longer unstable and constantly in danger of fragmenting, the twin dangers within essence structures’ (Smith: 14). And here Hegel reminds us of the problem of freedom. In essence-thinking, he writes, determinateness ‘is presented only as posited by essence itself; it is not free, but is present only as connected with its unity’ (Hegel 1977: 391). ³ In his political writings, Hegel will go further and describe this approach as a form of ‘fanaticism.’ The latter, he writes, ‘is simply the refusal to admit particular differences’. The intellectual or political fanatic, in other words, insists on simply finding ‘the whole in every particular’ (Hegel 1991: 304). This is the problem of all forms of ethical life in which universality exists ‘only as posited by essence itself’, rather than as something that develops in and through the independent life of its moments. It follows that genuine freedom is only accomplished in the self-determination of the parts (e.g. citizens) as identities in and through difference, i.e. as internally differentiated unities. And this requires a structure of individual and collective self-determination based on mutual recognition and genuine reciprocity among citizens.

Essence-thinking is thus incompatible with freedom and concrete universality. Yet where Marx decisively supersedes Hegel is in his argument that essence-categories correspond to the relational forms of capital. ⁴ Consequently, only the definitive overcoming of capital (and its relations of civil society) can provide a basis for human freedom. It follows that the forms of thought that define essence categories are fundamentally incompatible with the project of working-class self-emancipation.

Capital and the antinomies of bourgeois thought

In one of the most famous Marxist interventions in philosophy, Georg Lukács argued that the contradictions we have identified in essence-thinking – between sensible and supersensible, form and content, appearance and reality, particular and universal – ought to be understood as ‘antinomies of bourgeois thought’. Modern critical philosophy, whose greatest thinker is Kant, systematically reproduces the ‘reified consciousness’ characteristic of a society dominated by the commodity form of social reproduction, he suggests (Lukács: 110-28). But what is it about bourgeois society that generates this antinomical structure of thought?

Notwithstanding some prominent shortcomings, the trajectory of Lukács’s argument traces key conceptual innovations of Marx’s Capital. ⁵ Building in particular on Marx’s famous analysis of the ‘fetishism of commodities’, in which the capitalist value form entails that human activity (labour) and relations among people should appear as relations among things (commodities), Lukács outlines the ways in which the dualities of Kantian philosophy are grounded in the commodity form. Kant’s notion of the unknowability of things-in-themselves – which always exist at an irreducible distance from human thought – mirrors, says Lukács, the reification of social relations into a thing-like ‘second nature’ outside human control (Lukács: 86-110). In capitalism, in other words, social relations assume the form of an estranged natural world that obeys its own laws, which appear to be independent of human influence. We see here a reproduction of the dilemmas of Understanding in Hegel’s Phenomenology, where natural laws are imagined as operating in a sphere radically separated from (and unaffected by) the phenomena of everyday life. In fact, Lukács’s argument can be pushed ever further by way of a closer interrogation of Marx’s analysis of the dual-form of the commodity. As Marx reveals, the exchange-value of the commodity cannot be derived from its use value (otherwise water would have a considerably higher exchange value than diamonds). The exchange-value of a commodity emerges, therefore, by way of ‘abstraction from its use-value’ – which occurs via the real abstraction of the market system – and from the material properties that make it a humanly useful object. It follows that as an exchange-value, ‘all its sensuous properties are extinguished’. The objectivity of commodities as (exchange) values in the capitalist market economy is therefore ‘phantom-like’, deriving as it does from something not immediately accessible to the senses (Marx 1976: 128). Value thus exists as a ‘supra-natural property’ of a commodity (Marx 1976: 149).

Marx proceeds to reveal that value derives from ‘human labour in the abstract’ – not the concrete labour that enters into a commodity’s production, but the socially-necessary quantum of labour required for its production in a system of generalised exchange. This quantum of labour is thus something separate from the actual living labour that entered into its production: it is a general social datum that exists over and against the concrete labour expended on its material production. Put differently, the duality of commodities as use-values and exchange-values is reproduced in the duality of the labour that creates them as both material things and values. Of course, the latter (value based on abstract labour) ontologically requires the former (concrete labour). But, just as use-values are merely ‘bearers’ of the values that matter in the inverted world of capital, so concrete labour merely bears the abstract labour that governs the values and movements (purchases and sales) of commodities. In this inverted world, ‘concrete labour becomes the form of manifestation of its opposite, abstract human labour’ (Marx 1976: 150).

We observe here the symmetry between the dual character of the commodity (and the labour that produces it) and the dualities of Kantian metaphysics. As if anticipating Lukács, Marx declares that commodities are ‘sensuous things that are at the same time supra-sensible’ (Marx 1976: 165), while emphasising that the system of generalised commodity exchange subordinates the sensuous to the supra-sensuous. In the capitalist mode of production, an abstract universal – value (abstract human labour expressed through money) – dominates the concrete particulars that make it possible (specific acts of labour, discrete goods and services), while imposing its totalising logic upon them. This is why Marx can refer to capitalist value as ‘an automatic subject’ (Marx 1976: 255). Nature, specific products of social activity, and human labour itself are all subsumed to capital’s circuit of self-expansion. This, of course, is structurally similar to the Newtonian relationship between diverse entities and the universal forces that govern them. However, Marx indicates his distance from essence-thinking by repeatedly underlining the fact that the hidden forces of capital assume quite specific and necessary forms of appearance. So, for instance, he describes ‘exchange-value as the necessary mode of expression, or form of appearance of value’; and he similarly refers to money as ‘the necessary form of appearance’ of socially-necessary labour-time (Marx 1976: 128, 188). Rather than mere epiphenomena that obscure essential relations, appearance forms are in fact intrinsic dimensions of – i.e. internally related to – the phenomena under investigation. Instead of disregarding appearances to get at essences, therefore, scientific criticism must demonstrate the fundamental relations that require certain ‘forces’, such as the law of value, to manifest themselves by means of appearances, such as money, prices, and wages, that are intrinsic or necessary aspects of the essential structures in question.

It is certainly true that appearances frequently obscure the essential relations involved in a phenomenon. But this obscurity is not accidental; it is essential to certain kinds of phenomena that they should appear in mystifying ways. And this is especially the case for estranged social phenomena, which is why Marx can say that commodity fetishism expresses (alienated) human relations as what they are, ‘material relations between persons and social relations between things’ (Marx 1976: 166). Rather than a mere illusion, the commodity fetish is real; it expresses a truth about a reality that is false, i.e. inverted. The fetish does not simply reside on the side of consciousness, therefore; it is also an objective feature of an alienated world in which producers actually are dominated by the things they produce. As a result, fetishism cannot be overcome by the merely subjective effort of a consciousness – which would imply a form of idealism. Rather, ‘the veil’ that obscures capitalist relations can only be removed when social production ‘becomes production by freely associated men, and stands under their conscious and planned control’ (Marx 1976: 173).

It must be added that the ‘unity’ of essence and appearance that characterises the forms of capitalist relations is inherently unstable and unreconciled. As much as value alternately assumes the form of commodities and money while retaining its ‘identity with itself’ throughout its metamorphoses (Marx 1976: 255), its dualities are oppositions, as Lukács recognised. Its dual forms – use value/exchange value; concrete labour/abstract labour; commodity/money; wage-labour/capital – cannot be harmonised within its social framework. ⁶ As a result, one ‘moment’ of the process must drive to suppress another. This is what it means for use-value to be reduced to the role of a mere bearer of exchange-value, for instance; or for living labour to be dominated by dead labour in the form of capital. In circumstances of estrangement, one side of a doubled phenomenon dominates the other, even though the dominant element is in fact dependent on the dominated (as in Hegel’s dialectic of lordship and bondage). It follows that alienated doubles are unstable insofar as the two moments are at odds with each other (which is the social basis of class struggle). As a result, capital’s system is susceptible to crises, to convulsions in which ‘the antithesis between commodities and their value-form, money, is raised to the level of an absolute contradiction’. In such a situation, an ostensible mediator – in this case money, which is said to mediate the circulation of commodities – in fact emerges as an alien power and an end-in-itself. That is why, in a crisis, the bourgeois ‘pants his soul after money, the only wealth’ (Marx 1976: 236). ‘Money becomes an end-in-itself … wealth as One, in contrast to wealth diffused and fragmented in its commodity forms’ (Lebowitz: 97). Under conditions of political alienation, the same thing transpires with the state: posited as merely a mediator between citizens, it actually assumes the form of an overarching and external power unto itself. In each case, ‘forces’ locked into atomised relations of opposition – commodities in the capitalist market or citizens in a bourgeois state – require alienated third parties to manage their antagonistic interactions. But rather than integrating and harmonising the elements they are called upon to mediate, these externalised third parties ‘manage’ the contradictions involved only by way of reproducing them in new forms (Murray 1988: xvii). It follows that the unity of the diverse elements that compose capital cannot assume the form of a self-mediating totality. Instead, value drives to impose its unity, subjecting its determinations to an abstract universalisation that suppresses their concrete particularity. In this respect, capital is an essence-structure par excellence, and an obstacle to freedom and self-determination.

Spurious constructions of labour in the image of capital

I wish to suggest now that too often Marxist critics of the particularism at the heart of personal-identity politics have modelled their notions of working-class unity on the form of unification that characterises capital. As a consequence, they offer up an abstracted concept of class that is indifferent to the diverse forms of experience in capitalist society – and hence one whose experiential purchase is minimal. In so doing, they treat class as an essence-structure, in Hegel’s terms, that unifies labouring people from the outside – an approach that has had multiple political expressions in the practices of self-appointed external vanguards of the working class.

In theoretical terms, conceptualisation of class as an essence-structure is plainly represented by the so-called ‘structural definition of class’ advanced by G. A. Cohen. In Cohen’s view, ‘A person’s class is established by nothing but his objective place in the network of ownership relations’. Here we have a paradigmatic expression of social Newtonianism, wherein class is taken to be an external (or ‘objective’) structural force that occupies (determines) an individual’s social being. Among other things, class becomes here a theoretical construct that the outside observer (what Hegel calls ‘Observing Reason’) can apply to unsuspecting human agents. Concerned to exclude any sense of how class relations are lived by individuals, Cohen adds of such an agent, ‘His consciousness, culture, and politics do not enter the definition of his class position’ (Cohen: 73). Observe here how Cohen has shifted from the question of theorising class to that of merely defining it. The problem of grasping the complex social processes through which class happens in a given society is reduced to a purely taxonomic operation. In the Preface to his Phenomenology, Hegel had vigorously debunked the conflation of the problem of truth with mere definition. The serious work of comprehending a living phenomenon, he urges, has nothing in common with the trivial procedures of slapping labels on things. The analytical or definitional approach simply pursues the ‘method of labelling … and pigeon-holing everything’, which results in ‘the living essence of the matter’ being ‘stripped away or boxed up dead’ (Hegel 1977: 31). So it is with the putatively ‘structural definition of class’. In place of the ‘living essence’ of social class, we get the ‘monochromatic’ schematism of essence-categories and their ‘lifeless determinations’. Dynamic relations of becoming are reified in an effort to generate a static taxonomy that captures nothing of the rich and diverse life-processes of social class.

Invariably, proponents of the structural definition point to Marx’s distinction between class in-itself and class for-itself, as broached in The Poverty of Philosophy. Here, however, their abstract definitionalism confronts two problems. First, dialectical theorisation always begins with an initial starting point that is inevitably abstract and one-sided. But it then insists upon probing its own initial claims, exploring their partiality and the contradictions to which they give rise, so as to supplant them with more robust, complex and concrete conceptualisations. So, if a commentator pronounces that ‘class is a relation to the means of production’, the claim is not entirely wrong, though it is embarrassingly incomplete. What dialectical criticism objects to is the freezing of the initial proposition into a general truth claim. In order to avoid the empty dogmatism of such a claim, critical pressure must be brought to bear upon it – in order to disclose, for instance, that any relation to the means of production is simultaneously a social relation among human agents; that these agents are actors within a complex whole with specific features and dynamics; that in coming to an awareness of such systemic relations these actors become class agents in the fullest sense; and so on. This is why Hegel insists against analytic-propositional claims to truth that ‘the truth is the whole of the movement’ of thought as it develops ever-richer concepts that overcome the partiality of earlier claims – and therefore cannot be frozen into a single proposition. Engels had exactly this in mind when he cautioned readers of Capital that they should not search ‘in Marx for fixed and finished definitions, applicable once and for all’. Instead, ‘where things and their mutual relations are conceived not as fixed but changing, their thought images, the concepts, are likewise subject to change and transformation’ (Engels: 103). Whatever Cohen’s ‘structural definition of class’ (and similar efforts) may be, therefore, it is not dialectical theory – and it tells us remarkably little about social class.

Let us look, secondly, at what Marx actually says in The Poverty of Philosophy. Before getting to the famous passage about class in-itself and class for-itself, Marx discusses the self-constitution of the proletariat as an historical process. ‘So long as the proletariat is not yet sufficiently developed to constitute itself as a class’, he writes, revolutionary theory tends to assume a utopian (or abstract) character (Marx 1963: 125). Note the intriguing dialectical structure of this claim. On the one hand, Marx is discussing a social class – the proletariat – as if it exists, while simultaneously arguing that it has not yet managed to ‘constitute itself’ as a class. In short, one form of class exists at the same moment that another does not, at least not yet. Put differently, class exists initially in a most precarious form (merely for the observer) and thus has yet to attain the inner development by which it attains its truth, as a living and conscious form of becoming, one in which actual proletarians constitute themselves as conscious class agents. In such circumstances, the social being of class is deficient and its ‘truth’ is incomplete. Such a claim is, of course, absurd in the analytical categories of essence-thought. But it makes perfectly good dialectical sense – especially, as we shall see, for a critical theory whose inner core is freedom. Indeed, this is the spirit in which Hegel declares, in opposition to the standpoint of the Understanding, that ‘something thought of … ceases to be something alien, only when the self has produced it’ (Hegel 1977: 417). And Marx is writing in this spirit when, recognising the bare fact of the objective existence of the proletariat, he nevertheless urges that it is yet to ‘constitute itself’ as a class.

We are now in a position to make sense of the famous formulation that arrives later in the text: ‘This class is thus already a class as against capital, but not yet for itself’. Here again we encounter the idea of different modalities of class, one precariously incomplete that requires its sublation by a fuller and more complete form. It thus comes as no surprise when Marx informs us that in the course of the struggle the working class ‘constitutes itself as a class for itself’ (Marx 1963: 173). Class-being without class-consciousness is an inherently partial and inadequate mode of existence of class, meant to be superseded by a more complete and truthful form. Indeed, for the working class to fail to attain this truth (its class being for-itself) is for it to fail to realise its actual (revolutionary) possibilities. Critical-dialectical theory pursues an account of the living circumstances in which working classes are moving toward or away from their ‘truth’ as revolutionary social movements. Seen in this light, static, structural accounts are trivial at best: ‘The dynamic of class relations is not a principle of sectional classification’ (Bensaid 2002: 178).

It is in this light that we might interpret E. P. Thompson’s famous assertion that ‘class is not ‘a “structure” nor even a “category”, but … something which in fact happens (and can be shown to have happened) in human relationships’ (Thompson 1963: 9, 8). Thompson here takes his distance from the essence-category approach, which posits class as a structure that inhabits people’s social locations, or as a conceptual category that the social analyst can externally apply to them. Again, this is not to claim that such approaches capture nothing real; it is to insist, however, that they are deeply impoverished, incapable of comprehending class as a complex and dynamic social process whose highest form is self-constitution – the only understanding of class congruent with the mission of the historian (or the serious political activist) to understand its lived reality in historical time. In rightly telling us that the treatment of class ‘as a thing’ was foreign to Marx’s meaning, Thompson gestures toward the dialectical approach we have delineated. This is not to suggest, however, that Thompson’s approach is without its shortcomings. After all, feminist and anti-racist historians have demonstrated the ways in which his historical narrative is insufficiently attentive to the specific experiences of black and women workers in the making of the working class in England, and of the distortions this introduces into his reconstruction of working-class self-making (e.g. Clark; Linebaugh). But in so doing, they have extended and deepened Thompson’s injunction to grasp class as ‘a historical phenomenon’, rather than retreating into a sterile structuralism of class.

These efforts to deepen our grasp of how diverse social actors comprise a working class return us to the debate over personal-identity and class as well as to the Hegelian-Marxist problematic of concrete universality. In these works we encounter classes not as abstract universals but as ‘multidimensional formations constituted in time’ (Camfield 2004: 421-2). Moreover, processes of working-class formation are seen to depend heavily upon ‘consciousness, culture, and politics’ – the very terms Cohen wants to banish from his ‘structural approach’ to class. As one feminist labour historian notes, the practices of radicals in helping to organise people ‘in new ways’ is frequently decisive in creating ‘a unified class consciousness out of divided communities’ (Clark: 4). And such practices are part of historically specific cultures of resistance that frame the ways in which class is experienced and understood. We can even go further and assert that the most dynamic working-class movements frequently emerge where radical organising names these divisions – of race, gender, nationality, sexuality, ‘skill’, and more – and frontally challenges the oppressions associated with them. To take just one recent set of examples, we have witnessed powerful projects of subaltern mobilisation in much of Bolivia and in Oaxaca, Mexico, where demands, forms of organising, and social visions have been articulated in simultaneously indigenous and working-class idioms (McNally 2013). Rather than trying to subsume indigenous identities under the rubric of class, we observe here forms of working-class identity and organising constituted in and through rich traditions of indigenous resistance, generating a dialectical configuration that Jeffery Webber has called ‘combined oppositional consciousness’ (Webber 2011). It is precisely the diversity of these working-class movements that gives them their strength and resilience. ‘The concrete is concrete’, wrote Marx ‘because it is the concentration of many determinations, hence unity of the diverse’ (Marx 1973: 101) – and this applies with special force to working-class movements. There is no such thing as a vibrant, contestational working-class movement which does not draw together the grievances and oppositional practices of particular oppressed groups into a dynamic totality that expresses (rather than suppresses) its discrete parts. A truly comprehensive working-class movement requires a self-activating ‘unity of the diverse’ in which distinct parts of the dominated class, with their specific experiences of oppression, find avenues of self-expression and self-organisation within the wider class movement. Only in this way can the working class ‘constitute itself’ as a class, i.e. as a concretely universal social power. And only in such ways can socially differentiated groups of workers come to see how their distinct experiences of oppression are in fact internally related, discrete but interconnected parts of a totalising system.

We are delineating here the dialectical concept of totality, understood as a process of totalisation that unifies (without suppressing) the partial totalities constitutive of it. The social totality is thus grasped as existing ‘in and through those manifold mediations through which the specific complexes – i.e. “partial totalities” – are linked to each other in a constantly shifting and changing, dynamic overall complex’ (Meszaros: 63). In contrast, the essence-approach hypostasises the whole by ‘favoring it over the parts’ – and in so doing posits a ‘false totality’ (Kosik: 27). Treating class unity as something external to the individuals who comprise the class in question, it constructs an empty and lifeless abstraction bereft of the dynamic interactions among the parts that make a whole a living and organic one. In the dialectical conception of concrete universality, however, the totality lives in and through the ‘moments’ or ‘mediations’ (‘partial totalities’) that are constitutive of it.

It is highly instructive that this approach to concrete totality is at the heart of the compelling project of a feminist and anti-racist Marxism that has been advanced by Himani Bannerji. While not using the terminology of internal relations, Bannerji hews to its protocols, particularly in her call for theory that shows how ‘social relations and forms come into being in and through each other’. Arguing for a non-reductive historical materialism, she urges that such a theorisation

must challenge binary or oppositional relations of concepts such as general and particular, subject and object, and display a meditational, integrative, formative or constitutive relation between them which negates such polarization. This could be done by further developing Marx’s concept of mediation … The sole purpose of this concept is to capture the dynamic, showing how social relations and forms come into being in and through each other, to show how a mode of production is an historically and socially concrete formation. (Bannerji 1995: 82-3)

Note that Bannerji’s approach also moves us beyond the aporias of intersectionality theory, not the least of which is its tendency to imagine different forms of social oppression as discrete and autonomous social relations which are then brought into contact by means of ‘intersection’ with each other – rather than as ‘social relations and forms [that] come into being in and through each other’. In Bannerji’s mediational perspective, there is no social relation of, say, race, that is not internally related to sexuality, gender and class, and therefore constituted in and through these relations. To be sure, these different social forms can be analytically distinguished, just as they are distinguished in experience; but this should not entail the error of imagining that they actually exist as discrete ‘things’, which then enter into external contact with each other. Yet precisely this Newtonian assumption haunts much of intersectionality theory with its tendency to imagine distinct ‘forces’ whose intersection must then be accounted for. The mediational approach of co-constitution, however, grasps particular social relations as always-already ‘moments’ of ‘an historically and socially concrete formation’, a concrete ‘unity of the diverse’ that is understood organically, rather than atomically. Turning more recently to the co-constitution of race and class, Bannerji has argued, ‘“Race” therefore is a collection of discourses of colonialism and slavery but firmly rooted in capitalism in its different aspects through time. As it stands, “race” cannot be disarticulated from “class” any more than milk can be separated from coffee once they are mixed, or the body divorced from consciousness in a living person’. Moreover, she continues, ‘this integrity of “race” and class cannot be independent of the fundamental social organization of gender’ (Bannerji 2005: 149).

Bannerji’s approach represents a fundamental challenge to a ‘fanaticism’ of class based on ‘the refusal to admit particular differences’. At the same time, however, it also challenges ‘the refined particularism and individualism of the politics of “difference”’, which ‘lack the potential for a revolutionary politics’. Where class fanaticism refuses the rich narratives and struggles essential to genuine anti-racism, socialist-feminism and queer liberation, to take three examples, and in so doing posits class in the form of abstract universality, the politics of personal identity commit a parallel reification of the particular, refusing to acknowledge its internal relations with all the partial totalities that compose the social whole. A genuinely transformative radical politics requires that, as much as ‘a rich description of immediate experience is an indispensable point of beginning … it must expand into a complex analysis of forms of social mediation’ (Bannerji 1995: 73, 75, 4). Indeed, only this approach can generate a genuinely working-class politics capable of expressing the particular ways in which different sections of the working class live and experience the multi-faceted relations of the capitalist totality. Our theoretical and political project requires, therefore, a systematic attentiveness to the complex organisation of all the partial relations that comprise the social system in its dynamic processes of becoming. Anything less is to repudiate dialectics – as well as the emancipatory project of concrete universality.

I have noted above that the problem of concrete universality simultaneously raises the question of freedom. Theorising wage-labour in the image of capital obstructs the project of universal freedom, however. In essence-thinking, as Hegel writes, determinateness (i.e. the determinate qualities and elements of specific parts), ‘is presented only as posited by essence itself; it is not free, but is present only as connected with its unity’. If the determinate relations of class are merely posited by essence – whether this is represented by the self-appointed radical theorist, or the putatively omniscient leadership of a party – then class is not the product of the self-activating agents who comprise the group in question, which can only mean that this class ‘is not free’. Yet Marx’s political project is distinguished precisely by the way it pivots on the principle of working-class self-emancipation (Draper 1977: ch. 10). In the dialectical terminology inherited from Hegel, self-emancipation is not possible if class (essence) is posited as ‘something unaffected by, and subsisting in independence of, its concrete phenomenal embodiment’ (Hegel 1892: 210). Racialised workers, women workers, queer workers, disabled workers must all be able to determine, in and through their own activities and aspirations, the very meanings of working-class identity and struggle. Put differently, the project of working-class self-emancipation means the development of a living movement in which the parts (particularised experiences of the complex configuration of the orders of race, gender, sexuality, ability and ‘skill’ that comprises the social whole) self-consciously articulate themselves as dynamic aspects of a totalising project of class (and hence universal) emancipation. Such a politics seems consistent with Fanon’s call for a ‘new humanism’ that is at its core anti-colonial, anti-racist and anti-capitalist (Fanon 1968). And the trajectory of Marx’s own political and theoretical development seems to have involved an increasingly complex and differentiated sense of anti-capitalist politics (Anderson).

If it is true that a concept ‘ceases to be something alien, only when the self has produced it’ (Hegel 1977: 417), then this is doubly so for the concept of the working class, which only becomes living and true when a preponderance of workers have produced it – which, of course, is what Marx intends when he refers to the workers ‘constituting’ themselves as a class. And this is why it is essential that working-class movements develop a political vision that, as Bannerji puts it, grasps how differentiated ‘social relations and forms come into being in and through each other’ and, therefore, how only a dynamic and deepening ‘unity of the diverse’ can build the oppositional and counter-hegemonic power necessary to freedom. It is self-recognition – as a worker – in the experience of other workers that is constitutive of the universality of the working class as an authentically oppositional force. Precisely this betokens freedom.

As I have argued elsewhere, ‘a society of freedom … is one in which the world-building projects of humans obey not an external project but an immanent one’: a society in which human practical activity is ‘self-mediating’, and in which history, as Hegel puts it in his Phenomenology, becomes ‘a conscious self-mediating process’ (McNally 2003: 20). The political project of a society of freedom requires the building of a social-class movement whose animating principle is the self-determination of its parts as universals. Genuine freedom in the Hegelian-Marxist tradition is accomplished in the self-determination of the parts (e.g. revolutionary citizens) as identities in and through difference, i.e. as internally differentiated universals. And this requires a structure of individual and collective self-determination based on mutual recognition, and genuine reciprocity among its members. While Hegel utterly misreads the possibilities of liberty in bourgeois modernity, and misunderstands the ways in which capital is systematically antithetical to freedom, this problematic of liberation is inherited by Marx and reworked as the project of working-class self-emancipation. So much does Marx take for granted the inherent connection between concrete universality and freedom, that in his 1844 manuscripts he describes the human being who directs his activity to the whole species as ‘a universal and therefore free bring’ (Marx 1975: 327). But this can only be achieved by a theory and practice that, rather than modelling the unity of labour on the abstract and external universality of capital, recognises the working-class project of concrete universality as one grounded in the identity of unity and diversity.