revolutionary reflections | Notes on Capital and the State: part 1

Hendrick van Cleve’s Tower of Babel

The relationship between the state and capital is a major problem in Marxist theory. Colin Barker, a longstanding member of the International Socialists (IS) and subsequently the Socialist Workers Party and rs21, has returned to this question several times over the course of his career. In this piece from 1978, which was originally published in a heavily edited form as ‘The State as Capital’ in International Socialism 2:1, Barker addressed the problem in the context of two of the ‘theoretical innovations’ of the IS: State Capitalism and the Permanent Arms Economy.

Whilst these theories had provided both an explanation of contemporary capitalism and a guide to action for the IS during the previous decade or more, by the late 1970s one of the innovators, Michael Kidron, had denigrated them as mere ‘insights’ and Barker claimed to sense ‘unease’ in leading IS member Chris Harman at a reliance on concepts which had suffered from a lack of development.

In this first installment, Barker lays out what he terms the ‘general tradition’: the framework inherited primarily from Marx and Engels with which Marxists attempt to understand the state’s role. Characteristically Barker then proceeds to put this framework to the test. 

In stating that ‘it is not from writings like The 18th Brumaire of Louis Bonaparte that we should begin, but from Capital and the Grundrisse’ Barker identifies the methodology by which one might seek to build a theory of the state rooted in the layered categories of Marx’s political economy. He employs this method in his later work ‘Value, Force, States and other problems’, which Revolutionary Reflections plans to make available later this year.

It is worth noting that in addition to criticising the work of the well-known contemporary thinkers from the left on the question of the state, Ralph Miliband and Nicos Poulantzas, Barker relates to the work of the ‘German Marxists’ – among them Elmar Altvater, Wolfgang Müller and Christel Neusüss – who today are far less read.

This installment ends as Barker begins an investigation into ‘productive’ and ‘unproductive’ labour – another thorny issue in Marxist theory – in the context of the state. This investigation is taken forward in the next installment, and forms a fascinating backdrop to the questions posed to us by Social Reproduction Theory in the present. 


In recent years, there has been a welcome revival in the Marxist critique of political economy. A good deal of this revival has been concerned with the rediscovery of Marx’s concepts, and much less with their use for comprehending the contemporary capitalist world. If there has been a field where development has been most slight, however, it has been the critique of the capitalist state and its place within the reproduction of capitalism as a whole. Partly, this has reflected the very form that the revival of ‘Marxist economics’ has taken: what Marx did write has been subjected to close (and often useful) scrutiny, while what he left unfinished has been little developed.1 Marx intended to continue his major work with an account of the state and of the world market, though to my knowledge nothing has been preserved even by way of outline of this projected continuation.2

The problem I want to try to discuss here is that of the place of the capitalist state in the capitalist system. Such a project in clearly too large for me to cover it adequately, and all I hope to do is to indicate some problems which seem to require development. Amongst other things, the Socialist Workers Party (SWP) has been distinguished on the left for two propositions about contemporary capitalism, which we seem to have been less and less concerned to develop or defend theoretically in the recent past. The first of those concerns the nature of the regimes in Russia and Eastern Europe, and in China etc.: we have unambiguously declared that they are capitalist regimes, in which production is dominated by the competitive accumulation of capital. I remain convinced by the main structure of that argument, originally developed by Tony Cliff as far back as 1948. In his work on state capitalism, Cliff outlined the relation between the Russian economy and the law of value, as it operates on the world scale, in an argument which to this day has never been adequately grasped by our theoretical opponents, who have systematically (and perhaps wilfully) misunderstood it.3 Since Cliff’s path-breaking work, however, we have not progressed very far with the theoretical development of our own ideas.4 And I’m going to continue that tradition of benign neglect on the whole in this contribution.

Secondly, we have been identified with the theory of the ‘permanent arms economy’, which theory I think we have left scandalously underdeveloped and unclear. In the contributions to recent discussion by both Mike Kidron and Chris Harman in International Socialism issue 100 I detect clear signs of unease about this theory, and I want to suggest that this unease is very proper. Not because massive arms spending in the post-1945 world has not had very significant effects on the form and length of the long boom, but because we placed too much weight on that one factor, and did not properly theorise its relation to other very important developments. In a sense, we either theorised badly, in a more or less Keynesian mode, or we did not ask exactly what it was we were trying to explain with the ‘arms economy’ analysis. The ‘arms economy’ theory demanded that we consider a number of general theoretical problems, about capitalist crisis, about productive and unproductive labour, about the role of the capitalist nation-state in capitalist reproduction, about the place of ‘luxury’ production, which we either ignored or actually got wrong. I’ll return to some of these questions below.5

The general problem of the relation between the capitalist state and capitalist production is one which, of course, has grown in significance in the course of the 20th century, not only through the emergence of fully-blown state capitals in ‘communist’ regimes, but also in western capitalism. The British state, for example, is the biggest single purchaser, the biggest single employer, the biggest single producer of surplus value operating within the British market sphere; it plays a massive role in the distribution of constant and variable capital and of surplus value between competing uses. Without an understanding of the place of the state in contemporary capitalist reproduction, modern capitalism must be a complete mystery. Misunderstandings and mystifications concerning the state and its relation to capital underpin the revival of reformist theory, both in the left of the Labour Party and the Communist Party and in the trade union movement. Combating the 57 varieties of reformism requires of us a much more theoretically acute awareness of the processes and social relations underpinning the present world crisis than we have produced up to now. We need to revive the active critique of political economy, in order to assess both what is valid in the corpus of the writings of Marx and the revolutionaries who followed him, and where in fact we need both to develop and correct their insights. If we do this, I think it fairly rapidly becomes apparent how deficient we are theoretically, particularly when we look at the question of the state.

The General Tradition

The starting point for any adequate theory of the state must be the general discussions of the state in the writings of Marx and Engels. At the risk of considerable over-simplification, what they said is reducible to a small number of general propositions:

  1. The state is a product of class society, of a condition in which society has become entangled in irreconcilable contradictions. If any kind of orderly life is to continue in a society rent with internal divisions, the members of that society must lose their power to control the communal affairs of their society to specialised agencies and persons. The state is thus not an everlasting and inevitable product of human social relations; rather, it is a product of particular forms of society, whose manner of formation divides their members against each other.
  2. The very existence of the state, in this general sense, is therefore an indication of alienated social relations. The interdependence between members of society which is part of the definition and constitution of human beings is expressed in antagonistic forms. Certain forms of social relations, organized around antagonistic divisions, make it impossible for all members of society to run their own society communally.
  3. The existence of the state involves a particular form of the division of labour, through which some members of society specialise in directing the rest of society, The state is thus separated from the rest of society.
  4. What characterises the state above all is its specialisation in the use of force. The means of coercion are no longer the direct property of the whole of society, but are specialised into the hands of the members of the state, Thus the state is “… a public power which no longer directly coincides with the population organising itself as an armed force. This special, public power is necessary because a self-acting armed organisation of the population has become impossible since the split into classes… This public power exists in every state; it consists not merely of armed men but also of material adjuncts, prisons, and institutions of coercion of all kinds, of which gentile society knew nothing.” 6
  5. The state apparatus, the product of class divisions in society, is anything but neutral as between the classes. It is a crucial element in class rule.
  6. The fullest development of the state, whether in terms of the relative growth in its personnel, the relative separation of its organisation and principles from those of class society, or by any other measure, occurs with the development of capitalism, the most completely alienated form of class society.
  7. The destruction of capitalism and the construction of communism requires the smashing up of the existing state apparatus, its replacement by the armed power of the working class, and the beginning of the process of the withering away of the state altogether. A fully developed communist society, its production organised according to the famous principle ‘From each according to his ability, to each according to his needs’ will have no need of a state organisation. It will be a self-governing community, for whose members public and private interests will coincide.

At this very high level of generality, we may regard these propositions as forming the core of the Marxist theory of the state and its relation to social development. The state, its size, its form, the functions it performs, are clearly seen as a product of social relations; the form that the contradictions of class societies take shapes the form taken by the state organisation. The state is an apparatus of violence, first and foremost, separated from the control of all or most of the rest of society, which enforces onto society a ‘communal interest’ or ‘national interest’, which is not the direct product of that society’s members and which maintains the basis of the class divisions. At this high level of generality, these core propositions about the state – and thus about the need to overthrow the state and destroy it as a condition of the realisation of human emancipation through communism – form part of the ABC of Marxism both as a theory of human history and as a theory of human emancipation.

Some difficulties and unsatisfactory answers

Where problems begin to arise is when we move from this high level of generality to much more specific discussion of the precise relation between state and society within capitalism. It is this area, above all, which Marx himself left very underdeveloped, and which subsequent Marxist discussion has, until quite recently, not developed very much further. What we have inherited from Marx are some brilliant historical accounts of particular crises in the development of the capitalist state, notably in 19th century France, and a superb chapter in Capital volume I on the introduction of legislation in Britain to control the length of the working day. Otherwise, there are only scattered remarks, like those of Engels in Anti-Dühring on the general theoretical possibility of state capitalism, and very general observations by the young Marx on the alienated character of politics in capitalism.7 All of these are full of profoundly interesting insights, but hardly constitute a developed theoretical framework.

Indeed, Marx’s most famous summary statement on the question of the capitalist state is also one of the most misleading:

Each step in the development of the bourgeoisie was accompanied by a corresponding political advance of that class… The bourgeoisie has at last, since the establishment of modern industry and of the world market, conquered for itself, in the modern representative state, exclusive political sway. The executive of the modern state is but a committee for managing the common affairs of the whole bourgeoisie. 8

It is a statement with all the brilliant flair of Marx’s best aphorisms. Its very fame, however, has led to its being taken uncritically as the last word on the subject, a last word to which uncritical Marxists have then attempted to find unattainable proofs. Two problems are posed immediately by it.

First, taken literally, the statement is plain wrong. ‘The executive of the modern state is but a committee for managing the common affairs of the whole bourgeoisie.’ The whole bourgeoisie? If we quite ignore any subtleties about divisions within the bourgeoisie of this or that country, the simple fact is that the bourgeoisie is a world class, an international class of exploiters and parasites that runs a world system of production. Capitalism, from its beginning, presupposed a world market. But the state form within the capitalist mode of production is decidedly not a world state, but a system of nation states whose relations with each other are antagonistic. The whole bourgeoisie does not have a state, cannot have a state. No single, universal centre of coercion has emerged out of the contradictions of the world society that capital has forged over the past few centuries. When we talk of the capitalist state, we always mean the nation state, or rather the nation-state system.9 And no nation state, clearly, can be seen as a kind of committee for managing the common affairs of the ‘whole’ bourgeoisie.

Second, Marx’s statement has been taken – by both would-be Marxists and anti-Marxists alike – to mean that the state is the direct property of the capitalists. A good deal of ink has been wasted in the attempt to defend this proposition.10 We ourselves have been anything but immune to this idea that the capitalist class directly controls the state and runs it in its own interest. Certainly a mass of ‘radical sociology’ has been devoted to teasing out the supposed ways in which major capitalist interests are inter-linked with the leading personnel of the state through common social origins, common club memberships, common educational backgrounds, common general prejudices, directorships, lobbies, advisory committees, etc. The crucial theoretical problem with all this (admittedly often fascinating) material is that it forgets what sort of a class the capitalist class is, and imputes common interests too easily and uncritically to the bourgeoisie.11

The capitalist class is, foremost, the class which drives the competitive accumulation process of capital. It is a class which, as against the working class, is united by its demand for surplus-value. But it is also, at the same time, a class which is internally divided by competition. Exploitation and competition are, indissolubly, the two defining features of the  social relations that define the bourgeoisie. In Marx’s phrase, the bourgeoisie are ‘hostile brothers’. They share a mutual interest in extracting the maximum total quantity of surplus-value out of productive labour; they are equally engaged in permanent internecine strife with each other over the distribution of that surplus-value. The very structural antagonisms that define the internal relations within the capitalist class make it exceptionally difficult to conceive how, with all the common club-memberships, old school tie connections, intermarriages and multiple directorships in the world, the capitalists of a particular country are capable of forming a common political will sufficient to enable them to dictate the policies of a state. For the state to be a direct instrument of a class, that class must be capable of forming a cohesive view of its interests and a common will sufficiently strongly to wield such an instrument. If any class in history must, precisely because of the mode of production it supervises, find that requirement difficult, it is the capitalist class.

Three consequences follow: first, that in practice the state must have a degree of ‘autonomy’ from the capitalist class; second, that while we must by no means deduce from this that the state is ‘neutral’ in the class struggle, its class character, its specifically capitalist character, must be demonstrated on other grounds than those commonly used by ‘radical’ sociologists and political scientists; third, that the very ‘autonomy’ of the state from the immediate control of the capitalist class is a source of reformist illusions in the possibility of the working class ‘using’ the capitalist state machinery to achieve the destruction of capitalism.

Actually, there is a limiting case to these arguments. In those cases, like the Soviet Union, where the capitalist class and the personnel of the state are effectively coterminous, where capital is united in the hands of the nation-state to form a single national state capital, where therefore the distinction between ‘political’ and ‘economic’ decisions is at a minimum in practical terms, it may be possible to use the arguments misused in relation to western capitalism to prove the capitalist character of the state. In such a circumstance, the means of production and the means of violence are all in the same hands. The police are the company guards, the welfare system is the company personnel department, the corporate plan and the state plan coincide, etc. In such a case, it would seem, we might reasonably say that the state directly pursues the interests of national capital, being run precisely by those who direct the production system.

Yet, even in this circumstance, and although the processes are concealed from the popular gaze behind a facade of unity at the summit of the state system, a political struggle continues within such a system. What is at stake in these struggles is not merely personal power, as bourgeois commentators imagine. While the Soviet bureaucrat as much as the western corporation director is compelled to pursue the logic of capitalist accumulation, to pursue the ‘interests of capital’, in practice the anarchic character of the capitalist system makes it impossible in advance to know how, concretely, those interests should be pursued. Each ‘solution’ to the problem of accumulation in the context of Russian class society implies a host of problems, each resolution reveals new contradictions. An adequate military response to the West and to China implies a drain on the surplus-value available for productive investment; improving the productivity of labour requires an improvement in the production of consumer goods, including meat and dairy products; expanded production of consumer goods implies the diversion of investment resources away from productive investment in the means of production or military expenditure; better plan fulfilment requires greater scope for local initiative and decision-making, but that could undermine the bureaucratic plan and its targets; and so forth. The problem of finding ways to overcome the tendency to bureaucratic stagnation in production breeds political and economic conflicts, in which the dual need on the one hand to extract additional surplus-value from the productive workers and to preserve it for investment purposes and on the other hand the need to keep the exploited classes sufficiently loyal to the regime, only adds further dimensions to the problem. Even in the Soviet case, the simple examination of the ‘interests’ of the participants does not reveal the inner workings of the system, nor the relations between the state and the production system.

The problem of interests

The theoretical problem is that we commonly misuse the term ‘interests’. We refer to the ‘interests of capital’, and to the ‘interests’ of particular sections of capital, as in phrases like ‘the interests of the monopolies’ etc. It is not uncommon for us to explain political developments in terms of the way they express particular capitalist ‘interests’. ‘Interests’ is a popular term, but some clarity is required if we are to employ it most usefully. If we look at the way Marx interpreted capitalism, and even property more generally, we find that his approach does not marry easily with our popular, everyday usages.

Marx discusses the interests of capital in a particular way. At the most general level, the term is only another way of expressing the very form of existence of capital, as a social relation. The interests of capital are the expansion of value, accumulation for the sake of accumulation. The term in this usage refers to the form of a particular social relation, a process, which dominates those who are participants in it. When Marx, in Capital volume I, discusses the ‘valorisation’ process, through which capital enlarges itself, the ‘subject’ in his account is not the capitalist, it is ‘value’ itself. It is not to the possessor of money-capital that Marx attributes ‘a driving and dominating force’, a ‘determining purpose’, etc. It is value, a social relationship, to which Marx attributes these drives and purposes.12 Similarly, in the Critique of Hegel’s Philosophy of Right, Marx suggests that property, being a social relationship, possesses people, rather than them possessing property; thus he suggests that, under the system of primogeniture, it is not that the landlord’s son inherits the land so much as that the land inherits the landlord’s son.13 Capital, in Marx’s account, is a social relation that possesses people, much as in my Sunday School days a legion of devils shifted into a herd of swine from some poor  lunatic they’d previously inhabited.

In Marx’s account, the capitalist is the ‘conscious bearer’ of this movement. He is a capitalist, insofar as ‘the appropriation the wealth in the abstract is the sole driving force behind his operations’. This driving force becomes the ‘interest’ of the capitalist, that is, it becomes his aim and motive. This means no more than that the ‘interests’ of a particular historically given social role (here, capitalist) amount to the general content of that role. If we say capitalist we give an account of a particular pattern of action, of social relations, characterised by permanent effort after the ‘constant augmentation of value’. At this highly abstract level of analysis, there’s no difficulty in our attributing ‘interests’, for they are simply the general motives to action ascribed analytically to a general role, and comprehensible in terms of the social relations from which they spring.

What we cannot do, however, is make an easy analytical shift from the abstract to the concrete, without any mediations. If we want to understand the behaviour of actual capitalists in a particular situation, we can’t just say they are ‘pursuing their interests’. Obviously, if abstract analysis is of any value, it should assist us in attributing interests to particular people in the concrete situation. But it cannot tell us how those interests are to be pursued, how the aims and purposes implied by those ‘interests’ are to be realised.

The history of capitalism suggests, in fact, that the interests of capital can be realised in a wide variety of ways and situations. A rise in workers’ wages can be explained as being in the ‘interests’ of capital, as can a fall in wages. We cannot, by some simple process of reduction, explain the concrete actions of capitalists by simple reference to the ‘interests of capital’. One essential element that mediates the transition from one level of analysis to another is the conscious perception of the individual capitalist – the consciousness of the ‘conscious bearer’, in fact his own perception of his interests and how they may be best attained. That means, quite simply and straightforwardly, that we have to recognise a degree of ‘openness’ in the actual processes whereby the particular interests of particular capitals are both identified and pursued. Indeed, the content of capitalist ‘politics’ both at the level of the single capital and its directing body (the boardroom row, the struggle of managerial cliques and groupings, etc.) and at the level of the nation-state and of groupings of capitals is precisely concerned with the identification of the ‘best means’ to pursue the interests of capital(s). And, importantly, it is not only the capitalists themselves who participate in this process of identifying and pursuing the ‘best way’: so too do all participants in the class struggle, implicitly and explicitly, covertly and very overtly.14

The very pursuit of the ‘interests of capital’ by capitalists is an activity fraught with pains and difficulties. For one thing, given the practical existence of capital as ‘many capitals’, divided from each other by competition, there are no guarantees of ‘success’ in the activity. Looking back at their own past history, the capitalists and their spokespeople are prone to speak of their mistakes as well as of their successes.15 There are no certainties in the practical world of capital accumulation. But more than this: the very pursuit of the ‘interests of capital’ by competing capitalists itself produces the very ‘laws of motion’ of capital which Marx discussed in Capital. All the tendencies to crisis within the capital relation emerge, not as something exogenous to the activity of the many capitals, but precisely as a result and a product of those activities. Capitalism, being an anarchic system, is governed essentially by its tendency to crisis, to the forcible reassertion of principles of proportionality, and by its long-run tendencies to stagnation. The pursuit by capitalists of their ‘interests’ as they perceive them – no matter how they pursue them – inevitably produces crises for them. The particular manner in which they pursue their interests and perceive their interests may affect the precise historical form taken by the crisis tendency, but it cannot overcome that tendency in principle.

On limits of instrumentalist and structuralist theories

An adequate critique of the political economy of capitalism must therefore be founded not only in the examination of concrete inter-relations between functioning institutions and individual subjects, but also in the analysis of the structure of constraints to which the ‘conscious bearers’ of the central social relations of capital are subject. An adequate account of capitalism, that is, must be capable of revealing the interconnections between concrete historical activities of real individuals and groups, and the underlying laws of motion of the system. Accounts of the relations between the capitalists and the state which forget the structures of capitalist social relations cannot provide an adequate guide to socialist practice; nor can those accounts which focus on nothing but the structure and forget the ‘conscious bearers’ of that structure. We have to remember, in Marx’s famous aphorism, both that ‘men make their own history’ and that they do so ‘not under circumstances of their own choosing’.

The point of the rather abstrusely expressed and inadequately argued suggestions above is that we cannot understand the place of the nation-state in the movement of capital, if we rely on the kind of analysis provided by writers like Ralph Miliband – or, indeed, his ‘opponent’ Nicos Poulantzas – who forget the specific character of the capitalist system in order to demonstrate or argue the capitalist character of the state. In Miliband’s case, the problem is one of instrumentalism in that he treats the capitalist class as a class capable of acting as a purposive subject in history, pursuing common goals, when it is precisely that which cannot be demonstrated. Poulantzas, while coming from a structuralist perspective that denies conscious subjectivity of the kind pursued by Miliband, in practice regards the problem of formulating a theory of the capitalist state as a problem for ‘political theory’. His treatment makes too sharp a structural differentiation between the area of the capitalist economy and the political sphere. He does not treat as problematic the very distinction between ‘politics’ and ‘economics’, except in the most abstract sense imaginable.16 And, in his ‘concrete’ studies, despite much surface sophistication, he falls back on notions about the ‘interests of monopoly capital’ as if these were capable of being read off from history like elephant tracks.

At the most general level, we might say that the problem with what has in practice been the dominant tradition of Marxist discussion of the state and capitalism has been that it has not begun with capitalism, but with class society in general. The alternative, then, is to begin with capitalism as a mode of production – a starting point which in terms of the literature so far produced poses very considerable analytical difficulties, which are anything but solved.

Capitalism and the State: alternative starting points

An adequate Marxist analysis of the state and its relation to capital needs to begin with a consideration of the form that class relations take in the capitalist mode of production, and not merely in ‘class society’ in general. This implies that the starting point for theoretical development should be, not Marx’s occasional political writings, but rather his full theoretical treatment of the alienated forms of social relations characteristic of capitalism. That is, it is not from writings like The 18th Brumaire of Louis Bonaparte that we should begin, but from Capital and the Grundrisse.

This implies a reading of those works as much more than contributions to ‘economics’ in a narrow sense, precisely as an historical materialist critique of political economy, a critique of the whole pattern of social relations that underpins and is expressed in capitalist society. In this perspective it becomes clear that crucial ideas in Marx like ‘the law of value’, and ‘the sale of labour-power’, are not merely ‘scientific’ concepts aimed at explaining the complexities of capitalist economic development and organisation, but are critical concepts, whose very form is intended both to reveal the character of capitalist social relations, and to reveal them as historically limited and alienated social relations.

In this perspective, the character of the state and its relation to the processes of capital accumulation becomes a matter of great significance for the critique of political economy. What we have to do, in this light, is to tease out the forms in which the state is itself the expression of the same fundamental alienated social relations that are expressed in the notions of value and surplus-value, accumulation and competition, and examine the limits of the possibility of ‘state intervention’ in overcoming the contradictions of capitalist social relations.

It is characteristic of reformist thought that it looks to the state to do just that, to overcome the contradictions of capital. For reformism in its various guises, the state stands over and outside capitalism, as a source of power which is capable under favourable circumstances of being captured by socialist principles and employed to destroy capitalism. Nothing, of course, could be further from Marx’s whole analytical conception than this, as the most cursory examination of his early writings reveals. There he unambiguously treats the modern state – in particular the modern democratic state – as the expression of the most complete form of alienation. While, however, it is clear from the writings of his maturity (e.g. The Civil War in France, and the Critique of the Gotha Programme) that he never abandoned this position, it is unfortunately the case that he never concretely worked out the links between the fundamental forms of social relations he analysed in the three volumes of Capital and the forms in which the capitalist state appears. This was to have been achieved in a projected volume continuing the analysis in Capital, which was never even drafted in outline.

Work along these lines has been taken up by a number of recent Marxist theorists, especially in Germany (cf. Holloway and Picciotto’s recent collection), who have attempted to derive a theory of the state from an analysis of capital. I cannot cover all the ground or all the ideas they develop, or offer an adequate critical assessment of their work here. What I will do is to take some ideas from their work, and show how they make – I think universally – a mistake whose significance increases with the development of ‘late capitalism’.

On the state-derivation argument

The starting point for the German theorists, analytically, is a question posed in the 1920s by the great Bolshevik legal theorist, Pashukanis.17

Engels, Pashukanis suggests, had theorised the state as a necessary product of class society and class conflict. But in so doing he had failed to ask why the state takes on the specific form it does, and why it performs the specific functions it does, within capitalist society.

Behind all these controversies one fundamental problem lies concealed: why does the dominance of a class not continue to be that which it is, that is to say, the subordination in fact of one part of the population to another part? Why does it take on the form of official state domination? Or, which is the same thing, why is not the mechanism of state constraint created as the private mechanism of the dominant class? Why is it dissociated from the dominant class taking the form of an impersonal mechanism of public authority isolated from society? (Pashukanis, ‘The general theory’, p. 185)

To this question, several kinds of answers have been given, which we can quickly outline, not in opposition to each other, but rather as if they were all additions to a rounded account. First, the fact that capital can exist only as many capitals in competition with each other means that the reproduction of the whole society is a problem. In a society constituted as a market, in which production is carried out in isolated units related to each other antagonistically through exchange, some institution outside the production units must enforce order on them, guarantee the rights which each must recognise if social production as a whole is to continue. The maintenance of contractual relations requires the development of law, and of an institution of coercion to enforce contracts and rights. Put another way, since the interests of each isolated individual in commodity production include no conception of a general interest of society, the state is required to enforce a common interest on all participants. (The state, in this perspective, is required as Thomas Hobbes’s Leviathan was required, to save humanity from the consequences of its own nature ‘red in tooth and claw’).

The fierce competition between capitals means that capitalist society cannot ensure its own continuance, its own necessities, without some other institution outside capital to enforce a common interest on capitalist society. In his analysis of the development of the Factory Acts, Marx showed that the intervention of the state became necessary to prevent 19th century British capitalists from so extending the working day that they risked destroying the very foundation of their capital accumulation, the living labour-power of the working class.18

The paradoxical result of the workers’ struggle to limit the working day was that, through the state, they enforced on capital its own interest in further development. The point is taken further in an argument (derived from a section of the Grundrisse) that capital accumulation involves the necessity of certain ‘general conditions of production’ which capital itself – being constituted of many mutually antagonistic capitals – is incapable of providing. These include: the provision of certain parts of the economic infrastructure – e.g. roads, postal communications, nationalised industries etc; the provision of a legal system through which rights and contractual responsibilities are enforced; the regulation of the conflict between workers and capitalists, and where necessary the political coercion of the workers; safeguarding the expansion of national capital on the world market (the imperialist function). Here the state is seen as providing certain necessary general conditions of production which capital itself cannot directly provide. The existence of the state as a coercive institution separate from capital, a product of the capital relation but not itself capital, is posited as a necessary aspect of the capitalist mode of production.

Capitalism, of course, is not merely competitive production of commodities; it is a mode of production based on the production and accumulation of surplus-value. It is not merely an anarchic system of production and distribution, it is a class system. The heart of capitalist social relations is constituted not simply by relations of antagonistic separation of producers, but by relations of domination, but the system of domination is hidden. On the surface of society, in the market relations between capitals and between capital and labour, it appears that everything is fair and equal; everyone gets, more or less, the value of their commodities, including the worker who sells their labour-power. In Capital, vol. I, Marx suggests that force is not a permanent necessity for capitalist exploitation to continue. Workers continue to produce surplus-value for their employers, not because there is a gun permanently at their backs, but because if they don’t they will starve:

The advance of capitalist production develops a working class which by education, training and habits looks upon the requirements of that mode of production as self-evident natural laws. The organisation of the capitalist process of production, once it is fully developed, breaks down all resistance. The constant generation of a relative surplus population keeps the law of supply and demand of labour, and therefore wages, within narrow limits which correspond to capital’s valorisation requirements. The silent compulsion of economic relations sets the seal on the domination of the capitalist over the worker. Direct extra-economic force is still of course used, but only in exceptional cases. In the ordinary run of things, the worker can be left to the ‘natural laws of production’, i.e. it is possible to rely on his dependence on capital, which springs from the conditions of production themselves, and is guaranteed in perpetuity by them. 19

Thus, the permanent exercise of force in exploitation is not necessary, and indeed may hinder production. The separation of the state – as an institution concerned with the use of coercion – from the immediate exploiters is thus a necessary aspect of capitalist exploitation.

The way in which the state concretely performs its functions of preserving the conditions of capitalist accumulation is not something that can be derived in the abstract from the categories of Capital. Given that the very processes analysed in Capital are processes of class struggle, the actual content of state activities and policies is determined by exactly those processes of struggle between and within the classes. Just as we argued earlier, in relation to the ‘interests’ of single capitals, that there is no objective correlation between the ‘interests’ of capital in general and the actual perceptions and policies of actual capitalists, so too with the state’s functions. How, concretely, states have perceived and pursued the ‘interests’ of  capitalist society, of the maintenance of the conditions of capital accumulation, is a matter which is determined only by the permanent struggle of capitals and of workers to influence and command the state to their own perceptions of their interests.

At the same time, there are objective constraints on the possible strategies and policies of the state, deriving from its relation to the process of capital accumulation. The financing of state activity, the maintenance of its apparatuses of domination, the provision of the means for the state to supply the ‘general conditions of production’ all depend on the maintenance of surplus-value production. For the state is funded out of deductions from total social revenue, through taxation and other means. The state is objectively constrained to be a capitalist state by its dependence for its own existence on a supply of surplus-value to fund its manifold operations.

Further difficulties

There is a good deal more to the contribution made by the German Marxist theorists to the understanding of the relation of capital to the state than appears above, and their work – with its many controversies and insights – deserves serious study. At the same time, in a few respects all the German theorists appear to agree on a particular approach to the question of the state which I wish to question. I can sum up the problem by suggesting that they all tend to see the state in two unsatisfactory ways. Firstly, they all treat the state as a necessarily unproductive sphere of activity, as being outside the concept of ‘capital’ itself, as not-capital though necessary for capital; and secondly, they all tend to treat ‘the state’ in the singular.

The two problems, I suggest, are related. The implications of their theorising of the state are such as to rule ‘state capital’ out of court as a theoretical impossibility, a ‘non-thing’ – a position which must tend, I suggest, to make it more and more difficult for them to apply the concepts they have developed to the analysis of contemporary capitalism, in both the East and the West (not to mention the South).

As to the first point, Müller and Neusüss treat the state as ‘a particular social institution which confronts productive society’ (Holloway and Picciotto, State and Capital, p. 38). Elmar Altvater writes:

… capital cannot itself produce through the actions of many individual capitals the inherent social nature of its existence; it requires at its base a special institution which is not subject to its limitations as capital, one whose transactions are not determined by the necessity of producing surplus value, one which is in this sense a special institution alongside and outside bourgeois society. (Holloway and Picciotto, State and Capital, p. 41, my emphasis, C. B.) 20

And Hirsch writes that,

The function of the bourgeois state can never be more than the creation of the “external” conditions for the social reproduction process which regulates itself on the basis of the law of value. The social process of production and reproduction cannot be the direct object of state activity… (Holloway and Picciotto, State and Capital, p. 64, my emphasis, C. B.)

And before them all, the Bolshevik theorist of law Pashukanis had declared,

The principle of competition – which is ‘dominant in the bourgeois-capitalist world’ as we have already noted supra – provides no possibility of associating political authority with an individual enterprise – after the analogy of the association of such authority with great holding of land under feudalism. (Pashukanis, ‘The general theory’, p. 186)

Thus a rigid conceptual distinction is maintained between ‘state’ and ‘capital’. The German theorists, and Pashukanis, are not, of course, alone in this view, as the opponents of the theory of state capitalism have shown (see 3). Nor are they alone in their tendency to theorize more or less generally about ‘the state’, especially when conducting relatively abstract discussion. The ‘external’ concerns of states tend to be tagged on as an afterthought21, not taken into the general analysis of the form and functions of the capitalist state.22

Contrary to the position implicitly or explicitly assumed in much of the contemporary discussion, I suggest that the state can itself be a ‘capitalist’, in the sense that the state institutions can directly become the ‘conscious bearer’ of the capital-relation, that the state can have as its direct object of activity the social process of capital production and reproduction within a national sphere, that ‘political authority’ can be associated with ‘an individual enterprise’ even while the total activity of the state cannot be thus defined as directly capitalist. In particular, it seems to me to fly in the face of current actual developments in the world economy to deny that the state can be an active capitalist, that is, a capitalist under whose direct dominion surplus-value is produced. In other words, the state can be funded not only through its taxation of revenue from productive capitals falling within its ‘political sphere’ but can itself also directly exploit productive labour.

In part this position assumes that what Marx termed ‘the general conditions of production’, and which Marx discussed very briefly in the Grundrisse, are not produced by this or that nation-state for ‘capital in general’, as much of the discussion tends to assume. Rather, they are produced by this or that nation-state for ‘its’ capital, or capitals, not to facilitate capitalist production in general, but to facilitate national capitalist production conducted in competition with capitalist production under the aegis of rival nation-states. This also therefore assumes that the production by states of the ‘general conditions of production’ includes state activities which are both ‘productive’ and unproductive, in capitalist terms.23 (I also assume, though this point is not much further developed here, that the production of the ‘general conditions of production’ is not an activity necessarily restricted to states, but is also performed by other capitals and by such forms as the family, under conditions which require to be analysed in their own right.24)

Such a position requires that we have some clarity about the terms ‘productive labour’ and ‘unproductive labour’, which is taken up in the second part.



1 In one area, this is an unfair assessment. The extensive debate over the last few years concerning the place of ‘domestic labour’ in the Marxist system – on which see the bibliography [included in part 3] – has, if unevenly, contributed considerably to our understanding of wage labour and the reproduction of labour power, and of the family form and the situation of women in capitalist society.

2 On Marx’s plans for the whole work, see: Roman Rosdolsky, The Making of Marx’s ‘Capital’ (London: Pluto, 1977), ch. 2.

3 Cf. Ernest Mandel, The Formation of the Economic Thought of Karl Marx (London: NLB, 1971), p. 111. The treatment (in Marx’s Grundrisse) of the limits set on the concentration of capital is interesting as a ‘refutation in advance’ of the theory of state capitalism: ‘Capital does not exist and cannot exist except in the form of a number of capitals, and its self-determination thus appears as the interaction of these many capitals one with another’.The same ‘argument’ is used in: David Yaffe, ‘Value and price in Marx’s Capital’, Revolutionary Communist, 1 (1976), p. 42 n. 100. The generally admirable Roman Rosdolsky falls into the same trap as well (Rosdolsky, Making, p. 42, n. 117; p. 210, n. 27).

As Cliff’s whole argument about the capitalist character of the Russian economy after 1929 depended precisely on this point of Marx’s (without Cliff having the benefit of having read the Grundrisse), the objections display simple ignorance: Tony Cliff, Stalinist Russia: A Marxist Analysis (London: Pluto, 1970), ch 7.

The general point is this: a single, universal ‘capital’ is, as Marx remarked, a contradiction in terms, a ‘non-thing’: Karl Marx, Grundrisse (Harmondsworth: Penguin, 1973), p. 421. A single state capital, ‘one without alien capitals confronting it’, would similarly be a ‘non-thing’. If the capitalist state form were such as to cover the entire world, as a single state, then state capitalism would indeed be a theoretical impossibility. But the actual form that the capitalist state takes is that of a multiplicity of nation states. Just as capital exists only as many capitals, the state exists only as many states, Whether it can exist as one state is more than dubious: cf. Lenin’s critique of Kautsky’s theory of ‘ultra-imperialism’. In our world, there is the small matter of World War III to be experienced this side of any single-state formation. See: Colin Barker, ‘A note on the theory of capitalist states’, Capital and Class, 2:1 (1978), 118-26.

Here we have, perhaps, fallen down is in our slowness to develop and defend our own theory along these lines. For instance, Martin Shaw, in a generally useful article (‘The theory of the state and politics: a central paradox of Marxism’, Economy and Society, 3:4 (1974), 429-50), misses this essential partial character of the capitalist state form in his defence of state capitalist theory (a point accepted by him in correspondence).

4 Chris Harman, Bureaucracy and Revolution in Eastern Europe (London: Pluto, 1974) is an excellent book, but its merits are those of fine historical analysis. Chris is largely concerned to repeat Cliff’s arguments, rather than to develop them theoretically.

5 Again, our critics were also confused. Ernest Mandel’s discussion of the problem of productive and unproductive labour in Late Capitalism (London: NLB, 1975), ch. 12, is just a muddle, confused and confusing throughout. David Yaffe and the comrades around Revolutionary Communist initially got the question of productive and unproductive labour entirely wrong, along similar lines to Mike Kidron’s mistakes in Capitalism and Theory (on which see below). See, for example, Paul Bullock, ‘Categories of labour power for capital’, Bulletin of the Conference of Socialist Economists (Autumn 1973), 82-98 and Paul Bullock, ‘Defining productive labour for capital’, Bulletin of the Conference of Socialist Economists (Autumn 1974). To be fair to them (a difficult stance, given the intemperate character of their own contributions to debate), they have since largely corrected their earlier errors on this question: Paul Bullock and David Yaffe, ‘Inflation, the crisis and the postwar boom’, Revolutionary Communist, 3:4 (1975), 5-45.

6 F. Engels, The Origin of the Family, Private Property and the State.

7 In particular, Critique of Hegel’s Philosophy of Right; On the Jewish Question; Introduction to the Critique of Hegel’s Philosophy of Right. All to be found in: Karl Marx, Early Writings (Harmondsworth: Penguin, 1973).

8 Karl Marx and Friedrich Engels, Manifesto of the Communist Party, in: David Fernbach, (ed.), The Revolutions of 1848 (Harmondsworth: Penguin, 1973), p. 69. (My emphasis, C. B.)

9 It is odd how little theoretical reflection there is on this question in the Marxist literature. Yet a moment’s thought suggests it is vital for the comprehension of capitalist imperialism, for the long debate over the ‘national question’, etc.

10 Ralph Miliband is probably the best-known of all the proponents of this thesis, according to which the modern state is the ‘instrument’ of the capitalist class: see The State in Capitalist Society (London: Weidenfeld and Nicolson, 1969). With the possible exception of Claus Offe (‘Structural problems of the capitalist state’, German Political Studies, 1 (1974), 31-57), none of Miliband’s critics have really attacked him directly on this question, although they have made useful contributions to the debate in other ways (e.g. Poulantzas, Balbus).

11 Very often – and Miliband is an exemplar here – the capitalist class is not really the object of attention at all, but rather ‘the rich’ become the chief focus of interest. Thus attention is diverted from the relations of production to the sphere of distribution, and the specific character of the capitalist class is forgotten. The difficulty with Miliband’s propositions is that they could equally well be applied to feudal England or slave-owning Rome, with only marginal adjustments here and there. Miliband’s whole analysis really does not belong to the Marxist tradition of analysis at all: it lacks historical specificity, and lapses into a mixture of sociological elite theory and sociological functionalism. I have an overlong unfinished typescript on the problems posed by Miliband’s kind of analysis, if anyone’s interested.

12 See especially Capital, vol. I, ch. 4: ’The general formula for capital’

13 Karl Marx, Critique of Hegel’s Philosophy of Right, in: Marx, Early Writings, p. 175.

14 One of the most amazing aspects of the work of writers like Miliband is the practical absence of the class struggle from the centre of their analysis. They provide an account, not of a society riven by class contradiction and struggle, but of a society organised by and dominated by the rich and powerful. (The most effective critique of Miliband along these lines is that provided by Isaac Balbus.)

15 See, for example, Sir Frank MacFadzean (former chairman of Shell), Galbraith and the Planners (Glasgow: University of Strathclyde, 1968).

16 There is a brief, but useful, methodological critique of the work of both Miliband and Poulantzas in the ‘Introduction’ to John Holloway and Sol Picciotto (eds.), State and Capital: A Marxist Debate (London: Edward Arnold, 1978).

17 Evgeny Pashukanis, The General Theory of Law and Marxism (1923). This is a regrettably little-known work, which is currently available in English in a classically appalling translation by John Hazard in a volume entitled: Soviet Legal Philosophy (Cambridge, MA: Harvard University Press, 1951). Hazard’s text can only be understood if the reader can re-translate as he or she goes, for ‘labour-power’ appears as ‘worker-strength’, ‘commodities’ as ‘goods’, ‘exchange’ as ‘barter’, etc. A new translation has been announced for later in 1978 by Inklinks Books. There are useful introductions to Pashukanis in: Chris Arthur, ‘Toward a materialist theory of law’, Critique, 7 (1977), 31-46 (to which a forthcoming critical note in the same journal by Steve Redhead should be noted) and in: Eugene Kamenka and Alice Kamenka, ‘The life and after-life of a Bolshevik jurist’, Problems of Communism (January 1970), 72-8.

18 Cf. the discussion of this in: Wolfgang Müller and Christel Neusüss, ‘The illusion of state socialism’, Telos, 2 (1975), 13-90.

19 Karl Marx, Capital, vol. I (Harmondsworth: Penguin, 1976), p. 899.

20 Elmar Altvater is the most explicit in his insistence on this point, and in a way the most revealing. He writes: ‘…the state is never an actual, material, total capitalist, but rather always simply an idealised or fictitious total capitalist’: ‘Notes on state interventionism’, Kapitalistate, 1 (1973). The State, he suggests, is able to play its particular role, of supplying the deficiencies in capitalist reproduction as a whole,

…precisely because the state, as a special institution, outside and above bourgeois society, is not subservient to the necessities of surplus production, as are capital units, no matter how big. The appropriate form of the state under capitalism is therefore its special existence counterposed to capital units…

Altvater is here making explicit what is implicit in other writers. The state is necessarily ‘outside and above’ bourgeois society; ‘is not subservient to the necessities of surplus (value) production as are capital units’; ‘is counterposed to capital units’; etc. Altvater’s whole argument is construed on a reading of Engels’ Anti-Dühring, in which Engels discusses the situation where the centralisation of capital reaches the level of the state, and the state itself becomes a capitalist. Engels writes:

The modern state, no matter what its form, is essentially a capitalist machine, the state of the capitalists, the ideal personification of the total national capital. The more it proceeds to the taking over of the productive forces, the more does it actually become the national capitalist, the more citizens does it exploit. (my emphasis, C. B.)

Altvater disagrees with Engels’ formulation, but in so doing revealingly misquotes what Engels says. Where Engels refers to the state as the personification of total national capital, as national capitalist, Altvater leaves out the word ‘national’ and reads Engels as saying that the state becomes the ‘total capitalist’.

Had Engels said that, he’d have been uttering a Marxist nonsense. A total capitalist would imply a single capitalist controlling the entire capital within the system i.e., precisely the abolition of capitalism. Capital, pace Marx’s Grundrisse, can only exist as many capitals. The total unification of all capital into one set of hands would imply the direction of production from one centre, whether with or without proletarian revolution. The law of value would be abolished; competitive accumulation could no longer provide the motive of production (cf. the discussion of the abstract possibility in Leon Trotsky’s In Defence of Marxism). To repeat, had Engels said that, he’d have uttered a nonsense. But he didn’t say it. He referred only to the ‘national capital’, which is a very different thing. Nor, it’s clear, did Engels regard the state takeover of production as the abolition of the capital relation:

The workers remain wage-workers – proletarians. The capitalist relation is not done away with. It is rather brought to a head. But, brought to a head, it topples over. State ownership of the productive forces is not the solution of the conflict, but concealed within it are the technical elements that form elements of that solution.

In the 20th century, we might find Engels a bit optimistic here. But, though he doesn’t theorise the question any further (presumably he assumed his readers would understand him), his meaning seems clear enough. It is of course only comprehensible if we remember a crucial element in the capitalist system which Altvater has forgotten, namely, that capitalist society is a world society consisting of many nation states. A nation-state takeover of the means of production in one country is not, by any means, the centralisation of all capital into one set of hands, thus not the formation of an impossible ‘total capitalist’.

21 This is very clear in the article cited above by Altvater, who lists four general functions of the state, the fourth of which is ‘safeguarding the existence and expansion of total national capital on the capitalist world market’. This fourth function is the only one to which he devotes no further analysis.

22 This is a counsel of excellence, which I’m currently unable to fulfil. At least implicitly, some pointers in this direction are suggested below – though I hasten to add that they’re not theoretically integrated.

23 It would be improper of me not to draw attention to the fact that my use of the term ‘general conditions of production’ goes beyond Marx here. For Marx – in a rather abbreviated discussion – defined these ‘general conditions’ as definitely unproductive forms of labour. These conditions of production, he wrote,

…so long as their production cannot yet be accomplished by capital as such and under its conditions – are therefore paid out of part of the country’s revenue – out of the government’s treasury – and the workers do not appear as productive workers, even though they increase the productive force of capital (Marx, Grundrisse, p. 532).

24 On the production in the family of capital’s ‘general conditions of production’: Ann Foreman, Femininity as Alienation (London: Pluto, 1977); also relevant to her argument is: Jane Humphries, ‘Class struggle and the persistence of the working class family’, Cambridge Journal of Economics, 1:3 (1977), 241-58, and: Eli Zaretsky, Capitalism, the Family and Personal life (London: Pluto, 1976).



Please enter your comment!
Please enter your name here