First published in *Abolishing Capitalist Totality: What Is to Be Done under Real Subsumption?*, edited by Anthony Iles and Mattin (Minor Compositions, 2026)

Marxists have long been fond of rambling about ‘totality’, from the precedent of Hegel’s ‘Das Wahre ist das Ganze’,¹ through Lukács, Korsch, Adorno, Althusser. But if the truth is the whole, or the whole is the false, what actually is the whole, aside from its truth value? Is there only one whole, and if not, which one are we talking about? Though Marx was apt to make certain gestures towards a kind of holism, the elevation of totality within Marxism was a Lukácsian innovation. In Capital, Marx’s use of the term tends to be either contextually specific and non-technical—‘since the existence of commodities as values is purely social, this social existence can be expressed by the totality of their social relations alone’; ‘the capitalist class in department I comprises the totality of capitalists who produce means of production’—or implicitly defined in a technical sense as the unity of a particular set of mutually-related moments: ‘industrial capital, considered as a self-moving totality’.² In neither case do we have the sort of ‘saturated’ or all-encompassing totality that necessarily raises the ontological question of the nature of our inclusion within that whole, or of how to get ‘out’ of it, as in certain uses of the concept of ‘real subsumption’. Yet revolutionary theory seems constantly to pivot upon this problematic of inclusion and exit, and it displays a recurrent—and perhaps rather morbid—fixation on the idea that some encapsulation has, in some sense, now been completed: Adorno, Debord, Camatte, Théorie Communiste, Negri, Jameson etc. And when we think in such terms, the question of revolution easily shades into a somewhat ‘theological’ problematic of immanence and transcendence, laden with paradoxes: how does something immanent to the mode of production overcome that mode without perpetuating it?

Totality means many things in the history of Marxism. As a work of intellectual history rather than first-order theory, Martin Jay’s study of its uses in Western Marxism holds back from addressing directly what it most strikingly demonstrates: the comedy of more than half a century of intellectual exertions around a concept that in the last analysis seems to mean everything and nothing. Even in this relatively tight cluster of theorists and texts alone, totality identifies the unity of subject and object; of a lost Golden Age; of theory and practice; of a post-capitalist plenitude to come; of the working class; of history; of capitalist society; of theoretical problematics; of the state; of man with nature, and so on.³ Within a broader history of holism in Western thought, we find it also referring to artworks, the Greek polis, mankind, ‘the Spirit of the times’, community, the Parmenidean One, the cosmos, life and so on. If it was the perennial will-o’-the-wisp of post-Lukácsian Marxism—whether to be chased or fled from—in the post-Marxist epoch it still persisted through its endless exorcism, whether or not linked in ideological fashion to ‘totalitarianism’. Here we will try neither to enshrine nor vilify what is surely so basic a category as to be unavoidable (are we to avoid conceptualising the whole of anything?), but to become reflectively aware of the conceptual artefacts and effects that can come with it. What are we invoking when we invoke some ‘totality’?

The structured unity of everything

For those less philosophically inclined and more influenced by Gramscian and Leninist lines of thought—perhaps with an Althusserian admixture—totality is typically a matter of an injunction to think through all the organic complexities of a given society (state, classes, culture, institutions and so on), implicitly with a certain strategic orientation perceived as lacking in the narrow focus of ‘economistic’ Marxisms. Appealingly for intellectuals, this leaves room for such things as ‘culture’ on the battlefield of the struggle for socialism or communism. A holistic attitude to the complexities of social formations may be a more useful orientation in concrete thinking than any idea that everything must be traced back somehow to the ‘class struggle’ or some other privileged moment. But does this amount to anything more than a generic injunction to contextualise? What is the implied theory of the whole itself in such cases? What constitutes its unity? Is it something determinate, or just everything?

The polemical impetus of the Althusserian project finds its justification in modes of thinking that one still encounters fairly frequently. Althusser had a point in insisting, within the context of Western Marxism, on complexity and an opposition to mystical reductions; in pitting the ‘determinate unitary structure’ or the ‘structured unity of the complex whole’ against ‘empty unity’ and the ‘spiritual’ or ‘expressive’ totality.⁴ His enduring merit is probably to have signposted a way beyond the circumvolutions of philosophical Western Marxism, towards a theoretically self-conscious appreciation of historical and political complexity. But the positive accomplishments of his project, and the conclusions drawn from it, have always seemed more questionable.

No matter how many times one arrays words like ‘complex’ and ‘structured’ against others like ‘empty’, this does not in itself get one closer to anything determinate. Hegel too polemicised against invocations of a vacuous unity, in such forms as the infamous ‘night in which all cows are black’ of the Schellingian Absolute. And Althusser’s characterisation of a specifically Marxian kind of abstraction—in contradistinction to the Hegelian—as never primordial but always the outcome of prior determinations, seems unwittingly to identify precisely the aspect of Hegel that the mature Marx most self-consciously drew inspiration from.⁵ As both Hegel and Marx were well aware, concreteness is something that must be achieved laboriously, through the systematic elaboration of determinations; if it is conjured in advance of this then it can be no more than an image of determinacy—at most a ‘description’ rather than a ‘theory’, to put it in Althusser’s terms: Hegel, Marx and Althusser are all in essential agreement on this.⁶ The main question is how to proceed in that elaboration: is this an inductive or deductive process? A matter of analysis or of simple discovery? What is it that drives one forward? The questions of a socially and historically-conditioned ‘problematic’? The carrot of some positive ‘dialectic’, or the stick of simple conceptual inadequacy?

Althusser’s project of differentiating and theorising the different ‘levels’ and ‘relatively autonomous’ moments of the ‘complex structure’ of capitalist society assumes implicitly that this society can itself be identified unproblematically prior to its analysis into ideology, economy, state etc. As with most theorisations of ‘society’, that is to say, it seems to be premised on the idea of a coherent, fundamentally self-standing whole, which is surely nothing other than the image of the modern nation-state, with its unifications of internal market, territory, governing institutions, language, ideology and so on. No matter how complex, if the word ‘totality’ is to mean anything in this usage, there must at minimum be some such ultimate coextensivity to the various levels of this society. But one need not agree with Michael Mann’s whole approach to think he had a point in declaring that:

Societies are not unitary. They are not social systems (closed or open); they are not totalities. We can never find a single bounded society in geographical or social space. Because there is no system, no totality, there cannot be ‘sub-systems’, ‘dimensions’, or ‘levels’ of such a totality. Because there is no whole, social relations cannot be reduced ‘ultimately’, ‘in the last instance’, to some systemic property of it...⁷

Can one identify a single concrete society where each level or dimension—ideology, economy, state and so on—is coextensive with the others? Markets and firms straddle borders; states manage economies partially and imperfectly, through whatever limited tools are at their disposal—currency, taxation, legal regulation; ideologies and cultures both condense in specific, sub-state localities and spread trans-nationally; economic structure too is differentiated at sub-state levels. In the period of high Keynesianism and institutionalised collective bargaining, before the Nixon shock and European economic integration, it was easier to think in such terms; there really was a certain regulative unity of capital and state within individual nation-states. But even then, the globe was not divided into essentially autarkic worlds: any analysis of individual societies in abstraction from global markets and imperial structures, decaying or triumphant, was a distorted one. What, then, was the whole, and what was its theory? Was Althusser’s ‘structured unity’ ultimately that of the entire planet? What is the theory of the entire planet? Can there even be one? While we may analyse, contextualise and describe particular societies, doing so does not seem to get us closer to specifying the unity of the ‘determinate unitary structure’. This totality is not yet an object of theory.

Mann was reacting to the Althusserianism of the 1970s, and in some ways his polemical impetus aligned him with contemporaneous poststructuralist assaults on totality. But like any unsystematised series, the neo-Weberian enumeration of ‘sources’ that he offers as an alternative (politics, ideology, military, economy) invites indefinite expansion. Where, for example, does the power of the controllers of technical infrastructure or of standards-setting bodies come from? None of Mann’s categories seem right; perhaps there are five sources—perhaps more are possible. Or, if there really are only four, why this number and not some other? How do we account specifically for the four? That such impertinent questions continue to niggle suggests perhaps that the unsystematised series is not ultimately an alternative to totality-thinking, but the product of an incompletely theorised analysis. As Hegel recognised, conceptual thought drives to unite the separate. Every enumeration bears implicitly the question of the unity of the list itself.

Grooming as totalisation

For Durkheim, ‘the concept of totality is but the concept of society in abstract form. It is the whole that contains all things, the supreme class that contains all classes’.⁸ But what if there is no concrete concept of society? What if, on the contrary, the concept of society itself is an abstract, indeterminate totality? After all, it is hard to see how one would go about elaborating theoretically the whole of ‘society’. What level of abstraction would be the right one? Should the theory map all the concrete institutions, customs, cultures, groups, individuals, buildings, classes, territories, kinship structures, transactions, laws and so on that would typically be seen as constituting a society? What sort of theory would that be? Where should it stop? Or, if it is to involve something more abstract, what specifies that abstraction? In itself, the concept of society is distinctly lacking in determinate content. While sociological analyses might imbue it with some, to what extent will those analyses be generalisable as a matter of the actual concept of society? Durkheim’s pronouncement on totality, that is to say, seems to have things precisely inverted: concepts like ‘society’ are examples of the most abstract, indeterminate kind of totality, for how could a whole that contains all things be concrete at the level of theory? In our imaginations we may conjure up a picture of concreteness—some mass of details, like a ‘Where’s Wally?’ picture perhaps—but that is a long way from actual determinacy.

Social wholes in general—at least those beyond a certain scale—seem to be characterised by such elusiveness. What does a nation or civilisation contain, concretely? What about ‘the people’? Perhaps this has something to do with the famous ‘Dunbar number’: an apparent limit to our cognitive capacities in terms of thinking about groups larger than roughly 150 members. For Robin Dunbar, the social bonds of other apes are maintained by direct grooming processes, which set determinate limits on group size, since the number of social connections grows exponentially as members are added, and there is only enough time in the day—and neocortices are only ever big enough—to cultivate and map immediate relations with so many individuals.⁹ Radical imaginings of a purely immediate kind of ‘society’ reduced to direct inter-individual relations—or perhaps of the mythically integral Homeric Greece in the visions of a Winckelmann or the young Lukács—might be said to conceive sociality on the model of the non-linguistic primate group, where it is always in principle within the capacity of the individual to track the concrete relations between all others. This renders a very distinct kind of social totality, where the social group is always potentially graspable in all its determinations, without abstraction.

Dunbar speculates that the origins of human language, with its capacity for abstraction, may lie in the evolutionary pressure towards the formation of larger groups. He sees evidence for this in all sorts of places: the typical organisational structures of differently sized firms; human neocortex scale relative to other apes; the social-regulative practices of Hutterite communes, and so on. Once group size surpasses this level, new phenomena come into play, such as increasingly hierarchical structures, and a displacement of direct personal loyalties by ‘more nebulous and less inspiring concepts such as “the regiment” or “the Queen”’.¹⁰ It is, we might say, as if the totality at play in non-human social groupings were constantly in the active process of totalisation—always deliberately cultivating and remaking itself in each of its links, and thus always potentially present to itself in all its determinations—but that what some archaeologists and anthropologists term the ‘human revolution’ introduced an abstractive rupture into the continuity of animal social forms, after which the social totality would need to be thought through abstract concepts such as the king’s second body, the nation-state, God’s chosen people, the corporation or cosmopolitan world society.

The indeterminate totalisation of the social whole would then be not simply a failure of thought, but an abstractive function that enables that whole to exist in the first place. Social groups unify themselves under the sign of one or another concept that can stand in for the whole, yet without specifying it concretely. The nation comes into being in referring to itself as such; it performatively establishes a social actuality for itself, embodying itself in institutions, cultures and so on; yet it remains in the nature of such concepts to resist pinning down definitively.¹¹ This projected unity of the indeterminate is thus perhaps a structural facet of social thinking, and indeed of ‘society’ itself, beyond the smallest scales—a sort of placeholder, a zero or an ∞, always in lieu of substantiation, but always essential to the very possibility of the large-scale social entity. If this is the case, then we perhaps have some explanation for the elusiveness of the totality. Real, but indeterminately so, the social whole gives to our thought a basic compulsion to totalise with similar indeterminacy. But such totalisations necessarily remain pre-theoretical: if the work of theory is to illuminate and specify something previously implicit or unclear, we might reasonably doubt that any amount of theorising will ever elaborate the concept of something like ‘society’. What we might shed light on instead, is the problem of this indeterminacy itself.

Revolution against everything

This matters for revolutionary thinking because it has a bearing on the question of what is actually to be overcome. What is the scope of revolution? If it is a whole that contains all things, then we must wonder where to begin; anywhere might be as good as anywhere else. If the object of revolution is to be a hazily defined ‘society’, this is liable to leave the notion of revolution itself hopelessly indeterminate. As long as we are operating at this level, revolution seems to remain a matter of what Hegel termed ‘picture thinking’ (Vorstellung). It may be brought to mind as a fantasised event, such as a global riot, insurrection or the formation of an organisation—depending on your identity and affiliations. Such speculative imaginings may at least perform a certain role in making present to consciousness the question of radical social change, but they betray a poverty of actual content, as indeterminate as the world that they are arrayed against. In order to move from picture to theory, it is necessary to address this problem of scope, and to question the compulsion to totalise indeterminately.

Not all totalities are indeterminate. The capitalist mode of production is what we term a determinate totality, and it is such a totality that Marx strove to elaborate in the volumes of Capital. It is determinate because it is possible to specify in detail the mediations that constitute it—labour, capital, money, means of production and so on—and to show how these relate to each other in such a way as to constitute an integral whole. Capitalist society by contrast, with all its people, cultures, institutions, practices, beliefs and so on, is an indeterminate totality: no theoretical analysis could ever enumerate its mediations as an internally self-related system. In Marxist theory, things like ‘society’ tend to be closely identified with concepts derived from the critique of political economy such as ‘the capitalist mode of production’, or more loosely, ‘capitalism’. There is often an implicit telescoping of terms, such that ‘society’—with all its indeterminacy—is identified with ‘capitalism’, which is in turn identified with ‘the capitalist mode of production’, or simply ‘capital’.

This telescoping identification of concepts can lead to certain illusions of explanation, such that one seems to promise to account for the other—as if all the moments of capital’s systematic self-positing could be enough to explain whatever ‘society’ is. Marx constructed lucid explanations of what capital is, how individual capitals relate to each other, how they form something more general, how capital relates to workers, why the classical political economists were systematically blind to certain things about all this, and so on. This does not however mean that he had a general theory of whatever it might be that ‘society’ denotes, or even ‘capitalism’ (if we take this to be anything more than a contraction of ‘the capitalist mode of production’). The theory of capital does not in itself account for the projected unity of the social world; although no doubt in many ways related, these are quite distinct. If we cannot construct a theory of ‘society’, or even of ‘capitalism’, in the way that we can construct a theory of capital, nor can a theory of the one stand in as a theory of the other.

Capital as society

There are, however, some concepts that seem to provide justification for the identification of the mode of production in a strict sense with other aspects of the social world. Subsumption and reproduction are key examples: the first, in summary, because it seems to identify a way in which capital incorporates within itself a world beyond it; the second because if society reproduces itself through the mode of production, it might seem to follow that everything that occurs within ‘society’, and which can be characterised as reproductive in some sense, is part of the mode of production.¹² Let’s consider these concepts in more detail.

In a fairly loose sense, ‘subsumption’ has entered the lexicon of critical social theory: this or that aspect of society, reason, culture, nature may be said to have been at some point ‘subsumed’ by ‘capitalist logic’ or suchlike.¹³ This non-technical usage names a certain basic intuition about how capital profoundly shapes the world in general, and there is of course much truth in this. But there is also a stronger, more ostensibly technical sense in which it has entered some conceptual toolkits. As we noted back in Endnotes 2, since the 1970s or so, when many people first discovered the ‘missing sixth chapter’ of Capital, some—particularly on the French ultra-left—have perceived in Marx’s distinction of formal and real subsumption a periodisation of the history of capitalist society.¹⁴ Crucial to this is the way in which subsumption seems to provide a bridge between the capitalist production process in the narrow sense and more expansive notions such as society or humanity. It can thus seem to perform a theoretical role reminiscent of ‘reification’ in the Lukács of History and Class Consciousness.

For Jacques Camatte, an ‘anthropomorphosis of capital’ had occurred, such that capital became directly identified with human community as such:

The fact that capital is erected as a material community allows capital to avoid losing itself in all the various processes. This community flourishes on the dead and the reified. But, with the movement of anthropomorphosis, capital becoming man, its community poses as Gemeinwesen. Thus men are snared by the being they themselves produced.

In the process, for Camatte, capital had integrated the proletariat by ‘capitalizing’ it, and ‘everything’ had ‘become capital’ in a kind of total subsumption.¹⁵ This had practical implications: the Bordigist ‘theory of the proletariat’; the problematic of class consciousness; the revolutionary organisation; the orthodox faith in the progressive development of the productive forces—all were to be abandoned in favour of an orientation to a human species viewed as domesticated and decadent, and hope could only lie in something external to this monolithic totality. Yet since capital was identified with human community as such, ‘having absorbed all the old contradictions’, it became hard to see what the positive content of such an outside could be; were we to be left merely positing an empty transcendence?¹⁶

There are echoes of these formulations in Théorie Communiste’s work, albeit shorn of the humanist mysticism. Rather than humanity, here we find, in a particular historical moment, capital ‘becoming’ capitalist society through a process of subsumption:

It is not possible to understand the real subsumption of labour under capital without considering that what occurs in the labour process only resolves itself outside of it. Capital, as society […], is a perpetual work of the formation of its inherent contradictions at the level of its reproduction which undergoes phases of profound mutations. It is possible to go so far as to say that the real subsumption of labour under capital is defined as capital becoming capitalist society, i.e. presupposing itself in its evolution and in the creation of its organs.¹⁷

The justification for this identification seems to be that the main prerequisite of the capitalist production process—the separation of capital and labour—is also its major outcome. At a general social level, this means the reproduction of the fundamental classes of capitalist society, and the perpetuation of capitalist relations of production. And once we are talking at this level of social reproduction it becomes tempting to slide from the narrow gyrations of capital’s own circuit to the reproductive functions of all sorts of aspects of ‘society’:

As far as real subsumption is concerned, the criterion for its dominance has to be sought out in the modalities of reproduction of labour-power (social and political modalities): social welfare systems, the invention of the category of the unemployed, the importance of trade unionism, etc. All this naturally accompanies the transformations in the labour process: the decline of handicrafts and domestic industry caused by the first phase of large-scale industry. In order for there to be real subsumption, […] modalities of reproduction of labour-power must be created which are adequate to the transformations accomplished in the labour process.¹⁸

Since the reproduction of capitalist relations of production can be conceived as the reproduction of capitalist society, an elision can be performed whereby the whole of society becomes little more than the capitalist production process writ large. And since that process has a peculiar developmental dynamic, it becomes tempting to read this into the history of capitalism in general. Thus Marx’s logical schematisation of the phases in capital’s relationship to its production process—moving from formal to real subsumption—becomes a historical periodisation of capitalist society as a whole.

The theoretico-historical apparatus that Théorie Communiste have constructed on this basis is in many ways a sophisticated and compelling one. And the language of subsumption seems to chime with a general intuition that some fundamental incorporation has taken place. But this is a striking case of that telescoping pattern of identification that we encountered earlier, where the logical totality of the moments of capital’s circuit comes to account somehow for the indeterminate totality of society, for the imagined unity of the social world. The telescoping movement provides an illusion of systematicity as well as of explanation: if we can elaborate in some theoretical detail what it is for capital to subsume the moments of the production process, and if this subsumption gives us a ticket to the level of ‘society’ as such, then volume 1 of Capital (read through its ‘missing’ sixth chapter) becomes a total theory of society and everything in it, and the problem of revolution becomes the ontological one of inclusion within a capital-saturated totality, a problem of immanence and transcendence, while we circle infinitely around a seductively enigmatic paradox: ‘How can the proletariat, acting strictly as a class of this mode of production, in its contradiction with capital within the capitalist mode of production, abolish classes, and therefore itself, that is to say: produce communism?’.¹⁹

How we become capital

Let’s revisit this question of subsumption once more, to clarify what it actually involves. If, through ‘subsumption’, capital does not merely shape the world but actually seems in some sense to incorporate it within itself, what is the nature of this incorporation?

Means of production are subsumed by capital in an ontological sense when it takes hold of them and makes them serve the ends of its self-valorisation. The employment of workers is part of this process, and in this case the wage is the key mediation. As part of capital’s circuit, that wage posits the worker’s labour as its content. Here our labour appears as ‘part of’ the capitalist mode of production not just in the manner of this or that incidental aspect of the forces and relations of production of a given capitalist society, but as a systemic aspect of capital’s own production process—indeed, as part of capital itself. In the wage, an abstract monetary magnitude ranges our labouring capacity under itself, such that certain of our concrete exertions appear as a mere facet of capital’s own self-aggrandising project. The peculiarity of such arrangements led Marx to conceptualise them with a word borrowed from logic and ontology: the technical term for some ranging of particulars under a universal—such as that of animals under their species—is subsumption.²⁰

This is different from mere ‘domination’, which is a matter of being ruled over by the master of the house—the domus. In a relation of domination the two poles remain relatively external to each other, unimplicated in one another’s being. But when some particular is subsumed under a universal, the relation is between two inextricable logical facets of some self-same being: a whale is both this whale, and a whale. The universal presents itself as the essential truth of the particular, the particular as a mere instance of the universal. Monetary exchange is peculiar in realising something like this not just as a mental operation, but as something practically real and socially objective. In every act of exchange, a monetary quantity presents itself as the true measure of some thing, in all its qualitative irreducibility; a mass of particulars is subsumed.

If the claims of money over what it buys have always had a certain ontological character, however, as long as we are focused only on the moment of exchange this inclusion appears as strictly momentary. If the monetary measure seems to subsume the qualities of the commodity, it does so only fleetingly and from a specific standpoint: once bought, this book ceases to perform as a commodity, subsisting simply as your book, until you decide to sell it on to someone else. Likewise labour-power, once bought with a wage, ceases to present itself as a commodity, dissolving itself into the flows of daily activity; the wage presents itself as the truth of the worker’s labour only when they go to market—which is to say, only as long as we are considering their labour-power itself as one side of a market-mediated transaction. Thus this ontological mode of inclusion so far signalled by the term ‘subsumption’ appears, by the same token, extremely weak. Insofar as labour is only ever posited passingly and contingently as the content of the wage, in instants of mere exchange, it barely raises the ontological question of inclusion and exit.

A greater necessity comes into play when we consider not simply monetary exchange in general, but exchange as a moment of capital’s circuit. The various transactions which make up capital’s valorisation process constitute a certain unity, which folds back upon itself to reproduce some of its own conditions: profits are reinvested in subsequent iterations of the process. These exchanges can be understood as the moments of a single process, and thus so can the accompanying dynamic of subsumption. Where we are dealing not just with financial or merchant’s capital, but with a capital that has taken hold of some production process, its various exchanges can be seen as enacting the subsumption of another unified process. It is insofar as my labour participates in this process that it is identified specifically with capital and not just with the wage that I am paid. Insofar as it inherits a kind of necessity from capital’s circuit, subsumption may seem to have deepened here, but it has also inherited the contingency of the mere moment of exchange, and is still distinctly weak. For though my labour feeds the capitalist glutton during the day, my capacity to work comes home with me in the evening, itself unsubsumed by this particular capital, and free in principle to go elsewhere: this is part of the definition of wage labour as opposed to dependent labour or slavery.

If I am in a sense ‘included’ in a particular capital during the working day, this always remains a particular capital, for the drive to accumulate has the individual firm as its irreducible nexus. No worker is employed by—and no means of production are set in motion by—‘capital in general’: it is always a particular firm that offers a wage in order to keep in motion its own production process, that aims to sell the result at a profit, and that returns again to the labour market with the resulting revenues. Strictly speaking—that is to say, understood rigorously not simply as domination, but as a logical or ontological matter—‘subsumption under capital’ occurs only within this basic circuit. Whether merely formal or fully real, subsumption under capital by definition occurs only within individual production processes, for it is only there that anything is directly identified with, or included within, capital itself. And thus, insofar as the wage permits no capital to claim me in my entirety, this ontological inclusion identified by the concept of subsumption still remains weak. There is always a world outside, and we are always part of it. In this sense subsumption under capital does not in itself even raise the ontological question of inclusion within—and exit from—the mode of production.

Infancy to dotage

Of course, capitalist production processes have major implications beyond their own narrow circuits, encompassing such things as the carbonisation of the atmosphere and the melting of the ice caps. But implications are not identifications: things do not have to become capital to be affected by it, even profoundly. Whatever its outcomes, if subsumption under capital, strictly speaking, occurs only at the level of individual capitals, it cannot directly characterise society in general, or its history, especially not at a global level.

That subsumption occurs strictly at this level also has implications for any attempt to periodise capital itself in these terms. Each capital will negotiate its relationship to each of the things that compose its production process in a particular way, accepting them as is, reorganising or transforming them. With time, these reorganisations and transformations will of course accumulate and deepen. Thus these subsumptive relationships will tend to chart a trajectory from a ‘formal’ to a ‘real’ character, and such developments will spread from one capital to another, tendentially generalising throughout the global economy. Thus the aggregate too will tend to chart a trajectory from ‘formal’ to ‘real’. But if everything tends to run in one direction, that ‘everything’ is a vast chaotic manifold that cannot itself be reduced to single stages at given points in history, as if last year the wider world was only formally subsumed while this year it really is.²¹ This is because the distinction between formal and real subsumption schematises, in the abstract, the generic temporality of the capitalist production process, not its social history. The human ageing process too is generally unilinear—one must be young first before one can be old—and we may say that the human race in aggregate is now demographically older than it was a few decades ago, but any claim that human history as a whole could be divided into such things as an ‘adolescence’ and a ‘maturity’ could only ever be a dubious metaphor; here too, there is a key distinction to be made between temporality and history.

Historical attempts to distinguish between eras of formal and real subsumption do have some formal resemblance to another, more legitimate, distinction between phases. At the level of the most abstract form it makes obvious sense that an ‘extensive’ moment would have to precede any ‘intensive’ one. And it is empirically true that pre-capitalist exploitation was typically extensive, while the coming into being of capitalism involved a shift to a kind of intensive growth. Prior to the emergence of agrarian capitalism—and continuing until relatively late in contexts such as the European absolutist state—exploiting classes tended to gain greater surpluses just by squeezing more out of their inferiors through extra-economic power. The peculiarity of the English state and agrarian economy immediately prior to the formation of capitalist relations of production was the way in which the configuration of these two removed that option, compelling landowners to extract a surplus from their tenants by means of the market in rents, while driving producers to increase their productivity and ‘improve’ the land.²² It is in such intensive development that we find the origins of capitalism, for it set in train an ever-expanding process of accumulation and dispossession of which both proletariat and capital were products. If capitalism finds its origins in intensive development, it then drives that development further as market competition ceaselessly compels producers to make transformations in the production process: in this sense, capitalism is real subsumption.

Any attempt to tell this history on the basis of shifts to formal then real subsumption would seem implicitly to assume both a pre-existing capital, always already capable in principle of grabbing hold of its share of the world, and a pre-existing proletariat, always already lining up at the factory gates. Accounts of so-called primitive accumulation, which imagine the origin of capitalism as something constituted more or less ‘politically’—through simple violence and a sort of extensive development which had the effect of heaping up capital on the one side and labour on the other—seem to make a similar assumption.²³ For how do we account for the separation of the political and the economic, which itself is something peculiar to capitalist society? In ancient empires rapacious violence and expropriation tended to increase such things as tax revenues and supplies of slave labour rather than laying the bases of capital accumulation.

Policing the exit

If capitalism is real subsumption, and has always had this intensive character, then it does not make sense to think of real subsumption as posing the sort of problem of inclusion and exit that we have been interrogating here—at least, no more so than the capitalist mode of production always has. Real subsumption in itself cannot characterise either an epoch or a society, and it has not, in itself, enchained every reproductive aspect of the entirety of society to the production process. But if—as we found in our look at the ontology of subsumption under capital—we always inhabit a world beyond the capitalist production process strictly speaking, then why can we not simply abandon this mode of production and do something else? What is the real nature of the problem? There are surely reasons for the popularity of such ideas; they seem to correspond to a certain intuition.

There are several possible answers to this question. Firstly, given their historical concurrence, it is easy to identify the capitalist mode of production with the modern nation-state. As the latter encompasses us within its borders, manages us through its bureaucratic institutions, regulates us through its laws, inculcates in us a specific culture and language, its relation to us can seem in a sense deeply ontological. And in its own way—though the logics are fundamentally distinct—the state may be just as subsumptive as capital: political sovereignty involves an identification of citizens and subjects with a greater whole.²⁴ Rather than the capitalist mode of production, it may be that it is the programme of national state-consolidation that helps explain the holisms of Hegel, Croce or Gramsci—intellectuals self-consciously struggling to do their universalistic, integrative bit—and thus a significant aspect of the Marxist discourse of totality.²⁵

In themselves, the fearsome encaging effects of political power are of course not something particular to the capitalist epoch. Consider Gibbon’s terrifyingly totalistic description of the imperial Roman state:

The empire of the Romans filled the world, and when the empire fell into the hands of a single person, the world became a safe and dreary prison for his enemies. The slave of Imperial despotism […] expected his fate in silent despair. To resist was fatal, and it was impossible to fly. On every side he was encompassed with a vast extent of sea and land, which he could never hope to traverse without being discovered, seized, and restored to his irritated master. Beyond the frontiers, his anxious view could discover nothing, except the ocean, inhospitable deserts, hostile tribes of barbarians, of fierce manners and unknown language, or dependent kings, who would gladly purchase the emperor’s protection by the sacrifice of an obnoxious fugitive.²⁶

Given the relative constraints on their communicative capacities, it is hard to believe that earlier forms of political power could ever have felt quite as total as those of the present, yet Gibbon’s fugitive of Rome is not all that far from the despairing Foucauldian, fixated on the repressive powers of the modern state. That is, in short, because the encompassment and repressive regulation of a population is just what states do, and have done since their Neolithic origins.

Nonetheless, states must have seemed more porous and provisional prior to the consolidation of the bureaucracies and infrastructures of the modern world: feudal boundaries were constantly shifting due to the vagaries of aristocratic succession; the physical walls of the polis gave way to relatively unprotected countryside which might readily change hands with some shift of fortune; the inhabitants might be moved en masse to some other location or colony; the ebbs and flows of imperial conquest could displace large populations; and throughout much of human history, nomadism was still a relatively familiar phenomenon from beyond the bounds of the agrarian state. Such things must have had a bearing on the character of the social imaginary; active membership of the citizen body, and its absence or withdrawal, could be central in policing the bounds of the polis (thus the importance of exile and ostracism in the Greek world), but could the individual ever have imagined themselves as included within some saturated, all-encompassing totality as we are prone to now?

Is, then, the real problem of inclusion and exit that of the state, rather than the mode of production? Though its precise definition may remain somewhat blurry, with all its identifiable institutional articulations this state is a more determinate totality than is society as such. And in an important sense the state also helps tether us within specific relations of production: its repressive apparatuses prevent us from simply walking out of the workplace and finding another way to feed ourselves. But this is not simply a matter of policing the exit—after all, the land that we might use for such ends is already possessed by someone else, and it is specifically the maintenance of that possession that matters. And thus we quickly find our way back to capitalism and its fundamentally agrarian basis.²⁷ Though it is hard to come by reliable statistical data (since many landowners like to hide behind layers of legal obscurity), holders of land amount to a tiny and dwindling portion of the population, and some of them are extremely powerful. Such people have of course ensured their protection, and any genuine attempt at some proletarian reappropriation of agricultural land would immediately run up against the full force of a heavily militarised surveillance state which has us hemmed into its borders. This is the real bind that we will one day need somehow to find our way out of.

Capital’s infant

The crux here is not an ontological matter of incorporation or identification. It is a problem of dependence and its modalities. Market dependence is the principal prerequisite for the establishment and maintenance of the capitalist mode of production; it is firstly market dependence that will need to be overcome if we are ever to move beyond it; and market dependence is premised upon a separation from means of subsistence. There is at least some scope for a nominal reciprocity within this basic dependence: though workers need commodities, capital too has a need for labour. But the tendency of this relationship to issue in a surplus population amounts to a shift from mutual dependency—albeit within an asymmetry of power—towards an increasingly lopsided situation, in which a large part of the human population clings on for dear life to its dysfunctional supplier of subsistence goods. Bound to the extraction of fossil fuels and a broader despoliation of nature, this supplier threatens ultimately to kill its dependents. Such profound dependence on something that threatens ultimately to kill us might be compared to a narcotic addiction, but the reproductive, nutritive nature of the bond here invokes something even more thoroughly existential: the fragility of the infant. The proletariat is the dependent child of capital, infantilised by capitalist society and insecurely attached to a parent constantly threatening to withhold their affections.

It is easy to mistake thoroughgoing dependency for perfect identification, as if the capital relation permitted some womb-like ‘oceanic feeling’. But we are not enclosed within this mode of production as in a uterus. Though the dependency may be complete, it does not engulf us. And even a baby does a great deal more than eat and shit: it is a self-developing neural system which learns spontaneously to integrate and regulate its sensory organs, limbs, interior; to locate itself in space; to form a posture oriented to the world of things and people stretching beyond its parents; to extend the neural maps of its limbs into tools, and to begin to make that world act upon itself—all alongside a complete alimentary dependency. Capital’s dependents too will have to develop a posture, an orientation, tools, if they are to stand a chance of eating after their parent’s demise, and they cannot and do not merely wait for that moment. Everywhere people strive to stand and become mobile; to set and follow their own rules, if given half a chance. They find ways with the stuff of the world to construct themselves and to reconstruct that stuff. Insofar as capitalist society inhibits or prohibits that development of autonomy, it infantilises us. But the infant must learn to feed itself.