The history of capitalist society is the history of the reproduction of the capitalist class relation. It is that of the reproduction of capital as capital, and — its necessary concomitant — of the working class as working class. If we assume the reproduction of this relation is not inevitable, what is the possibility of its non-reproduction?
For a brief moment the recent crisis perhaps seemed to present us with a glimpse of such non-reproduction: the phenomenon of bank runs returned to the capitalist core, a wave of fuel and food price riots swept numerous countries, stock markets slid and corporations filed for bankruptcy, the Icelandic economy collapsed, the world as a whole entered a crisis widely announced as the worst since the Great Depression, Greece was lit up with insurrection, and forms of class struggle that have not been seen for decades reappeared in the UK. For a few months empty words were thrown around about a return of Marx and mainstream economists became catastrophists, before talk of “green shoots” returned and the usual idea began to set in that this crisis was, at most, a particularly severe glitch in the normal functioning of the capitalist economy, caused by some arbitrary, non-systemic factor. In such a situation, rather than a posing of the possible non-reproduction of the capitalist class relation, it is perhaps more plausible to interpret crisis as an aspect of the self-regulation of the capitalist world economy; at most a particularly extreme “shake-out” of some excesses or irrationalities in an otherwise healthy, fully functional system.
But there is no healthy equilibrium state, no “normal”, fully functional condition at the core of capitalist society. Crisis is the modus vivendi of the capitalist class relation, the life-process of this contradiction. Insofar as the accumulation of capital is always a fraught, problematic process; insofar as, even in its victories over the proletariat, capital still approaches impasses of over-accumulation; insofar as the dance of the capitalist class relation cannot take place without both of its reluctant partners, crisis is always here. In the capitalist mode of production labour is the source of value, yet with the progress of accumulation necessary labour is a tendentially diminishing magnitude. Crisis is always with us because, for capital, labour is a problem.
Yet crisis is also a discrete event. The spectacular catastrophism that reigned in global stock markets around the fall of Lehman Brothers, the waves of mortgage foreclosures sweeping the US, the looming bankruptcy of entire states, the vast bailouts and forecasts of depression, the hailing of an end of the “neoliberal” era and the appearance — no matter how illusory — of ideas of a return to Keynes: all of these are the very real signs of a particular crisis in the capitalist class relation. The particular crisis betrays the general contradiction of this relation, as if suddenly the lid had been blown off of the machine, and all the crunching gears exposed. Like all crises, this represents the deeper shifting structure of the class relation: where an aspect of the reproduction of the relation runs up against its limits, a moment of systemic openness and a fleeting glimpse of the possibility of rupture appears. Then, where one gear had slipped from the flywheel, through some chaotic mechanics another re-engages at a now-altered momentum. The contradictory reproduction of the capitalist class relation continues for now, with some modifications; Chance the gardener’s “green shoots” announce the end of winter, and crisis is naturalised once again not as chronic or permanent condition, but as the eternal recurrence of a natural cycle.
What is the character of the reproduction of the class relation now, and how is it transforming itself? What intimations can we find in this of the possibility of its non-reproduction? What — that is to say — is the possibility now of a complete rupture with this self-reproduction? These are the questions to which a revolutionary theory must address itself. It is in the changing modalities of this reproduction that we can grasp the real history of capitalist society as something more than a contingent assemblage of facts, narratives or concepts, strategic victories, defeats or recuperations, because it is in its self-reproduction that the capitalist class relation constructs itself as a totality. For the same reason, it is in these modalities that we must look for the possibilities of an immanent destruction of that totality.
[T]he result of the capitalist process of production is not just commodities and surplus value; it is the reproduction of this relation itself [...] Capital and wage labour only express two factors of the same relation.1
If there is a defining characteristic of capital which singles it out from a mere sum of money, or some unspecified mass of materials with which one might make money, it is that it expands: it is money which becomes more money, value that self-valorises. In order to persist as capital, capital must perpetually increase its quantity. In this sense, it has a clearly “teleological” character: it has a clear goal — its own expansion — and it pursues this goal relentlessly. Since, on the systemic level, such expansion clearly cannot be maintained through the mere reallocation of value from one capital to another, in order for valorisation to take place there must be some possibility of producing new value. This possibility is labour-power.
Since workers do not necessarily need to spend the entirety of the working day producing enough to be able to reproduce themselves as workers for the next day, a surplus can exist between the amount of labour actually performed by workers, and the social average of labour that is expended in producing the goods with which these workers reproduce themselves. A distinction between labour and labour-power thus arises, and it is reasonable to say that the entire edifice of capitalist society is erected on the basis of this distinction.
Whilst of course workers must be compelled to work this surplus, this compulsion is a systemic one. What, for the worker, is merely the number of working hours necessary to earn the wage requisite for reproducing her life at a given level, is for capital both an outlay in wages and the possibility of profit beyond the mere value of these wages. Whilst the position of the worker with regards to property means that her formal freedom is at the same time coupled with systemic coercion, both parties in this arrangement remain consenting “bourgeois subjects,” freely taking themselves to market. This meeting on the labour-market between capital and labour has — of course — certain inherent frictions, and like all good traders, both parties will always be looking for ways to obtain more for less. Workers drag themselves reluctantly to work, steal back as much time as possible, and sometimes strike for higher wages, whilst capital imposes the working day as rigorously as possible and will always be searching to expand the surplus portion of the labour that takes place in its production process.
This day-to-day meeting of capital and labour is not merely a contingent fact. If it were, then the persistence over time of capitalist society would be nothing short of miraculous. It is not a fact because it is a process in which we are all ceaselessly involved, and it is not contingent because — in its repetition — we can trace a certain systematicity to the way in which this meeting comes about.2 Workers do not merely happen to meet capital on the labour market with only their labour-power to sell, and capital does not merely happen to confront these workers as amassed means of production, possessed as private property. Rather, workers as sellers of labour-power and capital as amassed means of production are both produced as such by a determinate process. This process is the process of production itself: as well as producing value and distinct use-values, the production process at the same time is the process of production of the capitalist class relation.
If we consider not the start of the production process but its result, the successful capitalist has appropriated surplus-value from the workers, realised it in exchange, and can now employ this value in the next cycle of the production process; whereas the worker, being paid for her labour-power only, leaves the production process only with a wage to cover the cost of her reproduction for the next cycle of production. Both parties thus return, at the end of the process of production, to the structural locations from which they entered it. The worker has little choice but to sell her labour-power again, since she has not amassed anything of her own in the course of the production process, and the capitalist is impelled by the expansive logic of capital to employ her once more. Once the capitalist process of production has begun, its continuity is — at least in this sense — automatic. There is a necessity to the continuing reproduction of the capitalist class relation which follows from the character of the capitalist process of production itself.3 Since the process of production is nothing but this class relation in actu, we may say that the reproduction of the capitalist class relation follows necessarily from the character of this relation itself.
The self-founding of the capitalist class relation is also that of the totality of capitalist social relations. With this process of self-reproduction, it is not only workers and capital that are reproduced, but also the state and all its organs, the family structure and the system of gender relations, the constitution of the individual as a subject with a specific internality opposed to the world of production and so on. It is only through the repetition of their reproduction — pivoting upon that of the capitalist class relation — that these many moments come to bear any systematicity, and thus to constitute a totality.
It is a trivial truth that the social structures which constitute this totality cannot persist without the founding of society in production. Taken in only its immediate material aspect production presents itself as a quasi-natural basis for the reproduction of “society”. Yet in the capitalist mode of production it is value — not the general production of human life through any “human metabolism with nature” — that is the direct object of production, and it is first and foremost not “society”, but the capitalist class relation that is reproduced. “Society” as such — or the social formation — is the appearance in the abstract of the totality of relations that are reproduced through the self-reproduction of the capitalist class relation. A theory which sets out from the self-reproduction of the social totality in the abstract can only express the existence of this totality tautologically: the persistence of the parts is functionally necessary for the persistence of the whole, and the persistence of the whole is nothing but the persistence of these functional parts. The Althusserian notion of “structural causality” takes this tautology for a metaphysical principle — a mistake inseparable from the functionalist tendency within Althusserian Marxism.4
But an assertion of the contingency or open-endedness of class struggle, or a “Copernican turn” to the working class as subject of such struggle, is not an adequate alternative to a functionalism or naturalism of social reproduction. In its systematic self-reproduction the class relation is specifically not a contingent affair, and as the concomitant pole to capital in a relation of mutual reproduction, the working class as such cannot be the focus of revolutionary theory. The totality, of course, has many levels of concreteness, and is cut through with complex and contingent factors that cannot all be adequately accounted for through some simple liturgy of class relations. But as the locus of capitalist production, as the point from which it sets out, and to which it always returns, as the moment of the self-founding of the mode of production, the reproduction of the capitalist class relation has a centrality for any theory of revolution.
[F]or any era, to be present means having horizons. To pass is to lose those horizons.5
To pose the question of revolution is to put at stake the continuing existence of this capitalist class relation itself. Revolution cannot be the mere expropriation of capital, the seizing of the means of production by or on behalf of the working class. It must be the direct destruction of the self-reproducing relation in which workers as workers — and capital as self-valorising value — are and come to be. The revolution will be communist, or it will not be. We call the revolution thus conceived “communisation.”
The immanent self-perpetuation of the capitalist class relation presents itself as an eternalization: in its self-founding the class relation appears infinite, without a beyond. Since this relation projects itself onto an infinite future, revolutionary theory necessarily concerns itself with rupture, with an interruption in the very temporality of the relation. But self-reproduction is not a simple tendency towards equilibrium, or the dynamic preservation of an essentially static state. To posit the self-reproduction of this relation is not to take a starting point which can only ever demonstrate the functional closure of the system, and against which we must assert the radical open-endedness of class struggle, or a vision of revolution as radically exterior, messianic or transcendent. An organic metaphor is perhaps more appropriate than a cybernetic or mechanical one: an organism is inherently homeostatic, but it necessarily changes throughout its life span, it still must die, and its tendency towards death cannot be understood as exterior to its very living. Yet the capitalist class relation does not merely reproduce itself with a unity of function that must, like all good things, one day come to an end. Rather, as a class relation — a relation of exploitation — it is inherently antagonistic. Insofar as each has a directionality to its assertion against the other, the logical culmination of which would be final victory, both poles in the relation can project themselves as its ultimate truth, its final victor. Both capital and the proletariat can legitimately lay claim to being the essence at the heart of capitalist society, but such claims will always be contradictory, since neither pole in this relation is anything without the other.
Since each pole of this relation can claim contradictorily to be its truth, and since it is a dynamic relation with a directionality at its heart ensuing from the future-orientedness of capital’s valorisation process, the class relation always bears within it an immanent temporal horizon. It does not simply eternalize itself as a monolithic, closed totality. Rather, as a relation of struggle it carries as its own horizon a vision of the future as projected resolution to this antagonism. The final victory of the working class, the permanent establishment of liberal capitalism, looming barbarism, or ecological apocalypse: the class struggle always has a singular horizon, and depending on the dynamic of the class relation at any given moment, this horizon has a variant quality. Within this horizon, a supersession appears which may be more or less contradictory. If the overcoming of the capitalist class relation on the basis of the simple victory of one or the other of its poles is impossible — for each pole is nothing without the other — then, insofar as the affirmation of the working class as working class was their content, the revolutions of the 20th Century can be said to have posed an impossible overcoming of the capitalist class relation. In contrast, the revolution as communisation appears only in the struggle which carries the direct non-reproduction of the class relation in its immanent horizon.
It is only through its systematic reproduction that this relation presents itself as a unity rather than as an ad hoc arrangement, and — if by history we understand more than the impossible description of a formless flux — it is only as such a unity that it is capable of having a history. Just as the basis of the accumulation of capital is internal to the capitalist class relation, so — on the social level — are its effects. Falling profitability directly affects the ability not just of capital to reproduce itself, but also of the working class. Incessant technical reorganisation of the labour process brings radically varied patterns of experience to the lives of workers. Reorganisation of gender roles away from the single wage family through the increasing employment of women brings a different shape to the family and the experience of “personal life” outside of the production process. The expansion of the credit system enables capital to move globally with an increasing fluidity that alters the roles of states in the world system, and undermines national-level bargaining on the part of the working class. The tendency of labour-saving innovations to expel workers from the production process and generate a surplus population, where this population is able potentially to join the labour market, puts a downward pressure on wages and job security, and where it cannot join the labour market, vast slums are thrown up to house a human surplus whose reproduction is increasingly precarious and contingent. All of these tendencies are immanent to the capitalist class relation. The history of the development of the capitalist mode of production is that of the unfolding, within the capitalist class relation, of these tendencies, and thus the internal alteration of the quality of this relation itself.
The horizon of supersession which the class relation carries within it has a variant quality: its character at any given moment is inextricable from the historical modification of the class relation. What is invariant is that there is such a horizon at all. The changing character of this horizon is the primary basis and object for revolutionary theory. In posing the question of the revolutionary overcoming of the capitalist class relation, we traverse the theoretical terrain of this horizon as it presents itself now, to us. This is a stratified terrain with its own geology of sediments, irruptions, and fault lines. We trace the line of this horizon as it exists — approaching as close as possible to the conceptualisation of our exit from this landscape — and as it once was, differentiating the landscape which faces us from those of the past. Communist theory is the theory of the immanent horizon of the class struggle. In tracing this horizon, and in conceptualising its passing-over, we render the class struggle in its historicity a determinate object of theory and take it up in its finitude. In putting the class relation itself at stake through positing its ultimate supersession we can view this relation for what it is. We can grasp its truth not through the projection of a spurious neutrality, but through the opposite: through assuming the partisan standpoint of its overcoming, an overcoming that exists not merely in “theory” but in the immanent dynamic of the class relation itself.
To the degree that labour time — the mere quantity of labour — is posited by capital as the sole determinant [of value], to that degree does direct labour and its quantity disappear as the determinant principle of production — of the creation of use values — and is reduced both quantitatively, to a smaller proportion, and qualitatively, as an, of course, indispensable but subordinate moment […] Capital thus works towards its own dissolution as the form dominating production.6
If the capitalist class relation is a contradictory one in which reproduction is never a simple matter of the preservation of a stable state, this is because — as we indicated above — labour is a problem for capital. As the sole source of surplus value, surplus labour is always something which capital requires more of in its constant drive to accumulate. In increasing the productivity of labour, capital benefits by increasing the ratio of surplus to necessary labour, yet at the same time it thereby diminishes the role of labour as the “determinant principle of production.” This ultimately means that fewer workers are required to produce the same mass of commodities, and with this reduction comes a reduction in the possibilities for valorisation. From this simple contradiction we can derive some of the fundamental tendencies within the reproduction of this relation, and it is in this simple contradiction that we can see how capital “works towards its own dissolution.”
The fabled law of the tendential fall in the rate of profit expresses aspects of this simple contradiction. In its canonical formulation this law derives from the fact that in its competitive battle against other capitals, any capital will tend over time to increase the productivity of its workers through technical developments in the production process: its technical composition will tend to rise. With productivity increases it takes less labour-time to produce the same commodity, and the individual capital thus gains an advantage over other capitals, but in time these same productivity gains become generalised, wiping out the initial gain, and leading to a lower value of the commodity, since its production now requires less socially necessary labour-time. Thus even at this abstract level we can locate a first appearance of this simple contradiction, for the drive to accumulate surplus value through the production of commodities — a surplus which is constituted from surplus labour — leads to a reduction in the labour-time, and thus the scope for surplus labour, involved in the production of the very same commodities.
This is, however, by itself not a loss for capital, since in increasing the productivity of labour it also lowers the cost of labour by cheapening the goods which workers consume. Wages can thus be relatively decreased, and the part of the working day spent producing surplus value for capital can be extended. If however we assume that, over time, such rising technical composition will lead, at the level of total social capital, to a rising value composition — a rising ratio of capital devoted to means of production (constant capital) in relation to that devoted to wages (variable capital)7 — this means that a capital of which a growing proportion is devoted to means of production must valorise itself on the basis of a diminishing proportion of variable capital. Since the working day cannot be extended indefinitely (the day has only 24 hours, and the worker must spend some of these reproducing herself as a worker), and the part of the working day devoted to necessary labour can only be reduced towards zero, the amount of surplus value which capital can extract from an individual worker has definite limits. Thus eventually capital will be unable to extract enough surplus value to continue accumulation at the same scale. If the direct reduction — through productivity increases — in the labour-time necessary for the production of a given commodity represented a first appearance of the problem of labour for capital, we see here a further appearance of the same contradiction at a more concrete level.
All of this follows quite simply from a rising value composition of capital. For the sake of this argument, rising value composition is something assumed to follow from a rising technical composition. However, various factors complicate the relation between the technical and value composition, and allay the tendency for the rate of profit to fall as a result of the direct effect of the former on the latter. In particular, it must be noted that the same rising productivity of labour which would otherwise directly increase the ratio of constant to variable capital, at the same time decreases the value of means of production, thereby at least mitigating any tendency towards such an increase. Thus it is by no means self-evident that such a tendency will manifest itself in the actual unfolding of capitalist accumulation. However, if the theory of the tendency of the rate of profit to fall helps to highlight the extent to which labour is a problem for capital, Marx’s theory of the “general law of accumulation” and of the constant generation of surplus populations, is both more revealing and more historically palpable in this respect.8
The relative decline of necessary labour appears as a relative increase of superfluous labour capacities — i.e. as the positing of surplus population.9
It is self-evident that capitalist production tends to massively increase the productivity of labour. We do not need to concern ourselves with the relation between the technical and value compositions of capital to establish this. This means quite simply that, over time, fewer workers are required to produce the same quantity of use values. There is thus a tendency within capitalist accumulation to reduce the contribution of direct labour. If this tendency is not cancelled by any opposing tendency, and is left to play itself out historically, this will mean that more and more workers will be rendered superfluous to the production process. Viewed in terms of population, capital thus tends to produce a proletarian population that is surplus to the requirements of production: a surplus population. This is another mode of appearance of the basic problem of labour for capital.
This tendency is not an absolute one, and as in the case of the falling rate of profit there are countervailing factors. Capital may find new use values in the production of which workers can be employed, and with an increasing scale of production in any given line, productivity increases need not translate directly into an absolute decline in productive employment. Though of course environmental destruction presents itself as a very real problem of capitalist accumulation, the quantity of use values that can be consumed does not have clearly defined limits. It might thus reasonably be argued that, even if capital tends over time to reduce the number of workers required to produce any given quantity of use values, it can prevent this tendency from becoming a chronic problem by moving into the production of different use values — and, concomitantly, developing new needs for such use values — or expanding production of existing goods.
Of course, a number of factors complicate this. A given population can only consume so much of a particular type of commodity, and the productivity of labour is not simply a blank slate in the production of any new use value. Productivity-enhancing techniques will very often be generalised across different lines of production, meaning that production in new lines often quickly takes on the productivity gains developed elsewhere, as well as bringing about further advances which may themselves be generalised. The ability of total social capital to overcome its own tendency to reduce the number of productively employed workers is thus dependent upon its ability to keep pace with a growing rate of social productivity gains.
Historically, this has not occurred. At the global level, the number of wage-labourers productively employed in first agriculture, and now in manufacturing too, have declined relative to world population. This is the real meaning of the “deindustrialisation” that has taken place in the last 30 years. Though it is of course easy to demonstrate that plenty of industrial production still takes place, and that this is not only in important exporter nations such as China, the share of workers actually employed in manufacture has now been declining for almost two decades at the global level.10 As we explain in the article which follows, the result has been a rise in low wage (and formally subsumed) service work, and vast slums in what used to be known as the “third world”.
If the reproduction of the capitalist mode of production takes place essentially through the double reproduction of workers as workers, and of capital as capital, each producing the other; if the two wheels of this double moulinet meet at the point of production through the mediation of the wage form; as capital tendentially renders the proletarian population superfluous to production, the integrity of the double moulinet is undermined.11Increasingly it is no longer a reciprocal and cyclical relation in which the proletariat reproduces capital, and capital reproduces the proletariat. Rather, the proletariat increasingly becomes that which is produced by capital without producing capital. As the population that is simply superfluous to capitalist production, yet one which has no autonomous mode of reproduction, the surplus population is reproduced as a side-effect of capitalist production. Since its self-reproduction is not mediated through the exchange with capital of productive labour for the wage, it does not close the circuit with capital, and its existence thus appears as contingent or inessential relative to that of capital.12 Such a consolidated surplus population represents the tendential disintegration of the double moulinet of capitalist reproduction.
The very concept of the free labourer already implies that he is a pauper: a virtual pauper. […] If the capitalist has no use for his surplus labour, he cannot perform his necessary labour; nor produce his means of subsistence. He cannot, in this case, obtain them by means of exchange. If he does obtain them, it can only be because alms accrue to him from the revenue.13
For Marx, to the extent that she has only her own labour-power to sell, and is not even guaranteed of being able to do this, the worker is a virtual pauper. For the consolidated surplus population whose reproduction has ceased to be mediated by the exchange of productive labour for the wage, this pauperisation has become actual. The labour-power that the class of “virtual paupers” must sell is itself, in the long run, that which reduces it to a class of actual paupers. The proletarianisation of the world’s population thus does not take the simple form of the conversion of all people into productive workers, for even if they become productive for capital, these same workers ultimately produce their own superfluity to the process of production.
As that part of the global population diminishes whose reproduction is mediated through the exchange of productive labour for the wage, the wage form as the key mediation in social reproduction may appear increasingly tenuous. With these shifting conditions, the horizon of the class relation, and the struggles in which this horizon presents itself, must inevitably change. In this context, the old projects of a programmatic workers’ movement become obsolete: their world was one of an expanding industrial workforce in which the wage appeared as the fundamental link in the chain of social reproduction, at the centre of the double moulinet where capital and proletariat meet, and in which a certain mutuality of wage demands — an “if you want this of me, I demand this of you” — could dominate the horizon of class struggle. But with the growth of surplus populations, this very mutuality is put into question, and the wage form is thereby decentred as a locus of contestation. Tendentially, the proletariat does not confront capital at the centre of the double moulinet, but relates to it as an increasingly external force, whilst capital runs into its own problems of valorisation.
In such conditions the simple self-management of production by the proletariat no longer presents itself on the horizon of the class relation. As production occupies a diminishing proportion of the proletarian population — a proportion which is itself rendered increasingly precarious as it potentially competes on the labour market with a growing mass of surplus workers — and as this disintegration of the reproductive circuits of capital and proletariat gathers pace, the horizon of the overcoming of this relation perhaps appears apocalyptic: capital gradually deserts a world in crisis, bequeathing it to its superfluous offspring. But the crisis of the reproduction of the capitalist class relation is not something that will simply happen to the proletariat. With its own reproduction at stake, the proletariat cannot but struggle, and it is this reproduction itself that becomes the content of its struggles. As the wage form loses its centrality in mediating social reproduction, capitalist production itself appears increasingly superfluous to the proletariat: it is that which makes us proletarians, and then abandons us here. In such circumstances the horizon appears as one of communisation; of directly taking measures to halt the movement of the value form and reproduce ourselves without capital.