Viktor Deni ‘Either we kill capital or we die under the heel of capital’ (1919)

SUPPLEMENT: Imperialism and the state (Part II)

The capitalist state is a form of joint-stock operation. It is, in origin, a joint venture of the bourgeois revolutionaries

(A PDF of this supplement can be found HERE)

In the first part of this series I argued that the theory of imperialism should be disentangled from issues both of ‘monopoly’ and of capitalist decline; and that it should be used to explain, first, inequality between states and their populations, second, the phenomenon of the rise and fall of hegemon or reserve-currency states and the related recurrence of great-power war; and, third, the incorporation of the workers’ movement behind forms of social democracy (and US and other ‘liberalism’) in the high-ranked countries, and behind forms of radical nationalism in the low-ranked countries.

I went on to argue that the hypothesis would require analysing the state separately from the political economy of capitalism in order to be able to grasp the relation of the two forms. And I began this analysis with consideration of the state as such (as a form common from antiquity to the present) and of the necessary engagement of the state in the emergence of capitalism, through, successively, the role of the state in the emergence of commodity exchange; of money in general; of credit money; and of financial markets. The next step is that capitalist states are subordinated to capital - and indeed to particular capitals.


At this point it is necessary to return to some points made in the first article: class orders shape structural forms of the state, and they do so because rising classes remake states in their own interests.

We have to begin with the urban-slaveowner state of classical antiquity - not because such states are the first states (they are not), but because the form of the capitalist state arises out of the negation of the late feudal (absolutist) state, and the form of the feudal state arises out of the negation of the late antique state; so that it is impossible to grasp in theory the specificity of the capitalist state (as opposed to the characteristics of state-ness in general, which it shares with entities from ancient Mesopotamia and Pharaonic Egypt, etc onwards) without grasping it as a historical outcome.

The antique urban-slaveowner state is characterised, as far as its boundary conditions with its ‘outside’ are concerned, by the oppositions, citizen/foreigner and civilised/barbarian. The first of these oppositions maps onto the opposition between those who may not be and those who may be enslaved: the citizens may not be enslaved within the city.1 The second maps more immediately onto the opposition free(=owner)/slave: ‘barbarians’ become the subject of ideologies of their natural unfreedom.2

Internally, the structure of the state is also given by the dichotomy, owner/slave, in the form of the state as an owned entity. In the city-state periods of Greece and Rome, the state is the collective property of the citizens, whose character as owners of the state is variously expressed: for Rome, res publica Romana (the ‘Roman public thing’: ie, the state), meum ex iure Quiritium (mine by Roman-citizens’ right), etc.3 The low-level functionaries are state slaves.4 The alternative, into which large-scale regimes of this type tend to evolve, is a state owned by a single owner: imperator dominus mundi (‘the emperor is the owner of the world’).5

In late antiquity, regimes of this type decay internally by the transition from the particular gods of the city (with parallel gods of the women, gods of the slaves, etc) and directly divine emperor, into the creation of universalistic religions (Christianity, Zoroastrianism, Buddhism), which set the divine providence at a level above the state, thus freeing the emperor from immediate religious responsibility for earthquakes, crop failures, epidemics, etc. But such religions conversely tend to claim the souls of the slaves as well as the free, and the barbarians as well as the civilised, thus undermining the internal/external boundaries.

They decay externally because the pressure of imperial slave-taking at the borders generates state formation on the other side of the borders - characteristically ‘national’, because it is of groups not large enough to assert imperial claims - with the result that we get a turn to the defensive; an increase in the global cost of imported slaves; and external pressures, which increase the costs and reduce the rewards of participation in the common state regime. The external states or proto-states may merely set boundary conditions on the empire (Japan, Korea and Vietnam in relation to China; Sasanian Iran in relation to Rome); or ‘barbarian’ military groups employed on the frontier may, as happened in the western Roman empire, set up nation-states within the borders of the late antique empire (Goths, etc). The long-term outcome can be (not must be) feudalism.

In feudalism, in contrast, the boundary principle of the state is that it is the state of a nation (Kingdom of the English, etc) within a larger religio-cultural entity (Christendom, etc). The vertical principle of organisation of the society is allowed to coexist with membership in the society: villeins are subjects of kings in a way in which slaves are never citizens. The ‘unenslavable’ extend beyond the boundaries of the state to the subjects of other states, which form part of the same cultural-religious entity.

Internally, the feudal relation of lord and man, with bilateral duties, is replicated in the form of the state. The state is imagined as a sort of super-manor, with royal demesne coupled with seigneury claims over land not immediately owned by the king or managed by the king’s direct servants.

The regime of the particular state runs in parallel and in contradiction with the fully-public clerisy caste of universalistic religion - in western Europe, the Catholic church; in Japan (but extending beyond it), Buddhist monasticism.

States of this type are a problem for emergent capitalism for a number of reasons. The first and fundamental one is that the state is necessarily committed to the vested rights of the lords, the clerisy, the peasantry and the artisan guild corporations; and these are obstacles to immediate capitalist interests. This commitment is perfectly comprehensible. There is no reason (without the artificial pseudo-Romanising reasoning about dominium and iura in re aliena of 19th century lawyers in the interest of capitalist development6) to regard these rights as anything other than ordinary property rights. But the state itself necessarily self-identifies in terms of property rights (the king’s demesne and seigneury, which are replicated at every level of the feudal hierarchy down to the manor).

Secondly, and equally fundamentally, a state of this type is not a reliable debtor. The 1672 Stop of the Exchequer is a late instance of a phenomenon which goes back to - at the latest - Edward I’s 1294 seizure of his bankers’ assets and Philip the Fair’s 1307 assault on the Templars.7 Medieval law provides for the payment of debts, subject to the legal arcana of the proof rules and of the canon law rule against usury. But debts are not understood as property in the same sense as land (owned by lords and peasants) or jurisdiction (owned by lords, clerics and artisans). Moreover, subjection of the state to this, or any, law is not an organising principle of the state, which rests, in contrast, on a mixture of ‘affective’ personal allegiance, national identification and religious ideologies of monarchical authority.

Thirdly, the nationally limited character of the state is an obstacle to the international interests of ‘merchant’/shipping capital (the core of early capital), which operates across the whole sphere of Catholic Christendom and into the interfaces of this sphere with Orthodoxy and with the dar-al-Islam. This relates back to points made in the first article. Given that capital from the outset operates internationally, it is impelled to seek a state which will operate on the scale of its own operations.

This dynamic is double-edged. The bourgeoisie’s aspiration to its own state finds expression in the late 11th to 12th centuries in city autonomy, but then, as this proves to be an insufficient scale, many of the Italian city-states are torn apart by the rival universalist Guelph (papalist) and Ghibelline (imperialist) factions.8 In the 16th century, the opposition of capitalist interests to clerical ones finds expression in an international Protestant movement; the tensions which eventually destroy the Tudor-Stuart absolutist regime are primarily driven by the problem of England’s role in relation to this international movement.9

At this point we move from the logic of the bourgeoisie’s aspiration to its own state, to the relation of this aspiration to the concrete historical (European) feudalism, out of which capitalism emerged.10 Because the feudal social order is divided into nation-states ruled over by rival feudal dynasties, even though capital cannot achieve the universal state to which it aspires, it can break through at the level of the individual feudal nation-state or dynastic agglomeration; and by doing so it creates quasi-nation-states, which act on the global level: first, the Dutch republic; then, and more decisively, Great Britain.11


All modern states are directly or indirectly modelled on the British state created in two stages, in the revolutions of 1637-60 and 1688-89.

Outside - perhaps - France, the reason for this is not directly that rising capitalist classes copied the example of 1640 and 1688. It is that the capitalist creation of the British state out of the kingdoms of England and Scotland, and the revolutionary transformation of the British economic and class order which this set free, created a state which was militarily superior both to its immediate European rivals and to the extra-European states and societies with which it came into conflict. This superiority forced other states, if they were to remain independent of Britain’s global imperial aspirations, to attempt to remake themselves, at least partially, as British-style - ie, capitalist - states. In several cases this involved late-feudal states attempting to create capitalism, or social forces that were at best proto-capitalist, revolutionising these states in order to create capitalist states where (national) capital did not yet properly exist.

This history - as the British form of negation of the English and Scots feudal nation-states - shapes the resulting principles of the state forms.

First, the bourgeois state is no longer a national state, but a constitutional state. Its members are citizens defined by its law, ‘nationals’ only by analogy. It is no longer a member of ‘Christendom’ or any analogous religio-cultural order, but of an imagined and created order of international law, which extends beyond religio-cultural borders, including - for example - the conceptualisation of the Ottoman regime by the international lawyers, starting with Grotius at the latest, as merely one among the various states.12

Connected to the underlying international character of capitalism, from the moment at which it becomes a capitalist state, the capitalist state necessarily aspires to an international reach.13 This is already visible in the small-scale Mediterranean empires of Venice and Genoa, in the Dutch East and West India Companies (etc) and in Cromwell’s war with Spain. It is fully flourishing in the global character of the Franco-British wars through the 18th century, which end in the unambiguous global ascendancy of Britain after 1815.

Secondly, the capitalist state is so constructed as to be a perpetual debtor and dependent on debt for its routine administration. Since the Dutch Republic, central banks and organised financial markets have been indispensable elements of capitalist state finance and hence of the capitalist state as such.14 Where its own financial markets are weak, the state is forced to borrow on the financial markets of a stronger state, creating an outflow of surplus from the economy of the borrower territory to that of the lender.

Third, I have already said that the capitalist state is a ‘constitutional’ state, as opposed to a nation-state. The other aspect of this character is that it is a rule-of-law state. State takings are limited to those authorised by law, while the internal structure of the state regime itself is given by legal rules and the loyalty of state officials is - even where it is expressed in the form of allegiance to a hereditary monarch, as in the UK - to the constitutional order as a rule-of-law order.

The sanctity of property is at the centre of the rule-of-law idea. But its reverse side is the sovereignty of the legislator or legislative assembly. A legislative assembly represents the citizens in proportion to their property holdings: in ‘classical’ capitalist regimes this is approximated by property qualifications on the franchise, and sale and purchase of votes, so that the state is correctly described, as Edmund Burke did, as a joint stock company (in modern terms, a business corporation).15 Broad suffrage is a concession to the petty bourgeoisie, universal suffrage is a concession to the proletariat; the joint-stock effect of representation in proportion to property holdings now has to be reconstructed through the duopoly of ‘in’ and ‘out’ parties of corrupt professional politicians and the (connected) role of advertising and media.

A single-person legislator, in bourgeois dictatorial regimes, is, in effect, an auctioneer of laws, state contracts, etc, to capitalist bidders (in the form of corrupt payments): this is less satisfactory for the local capitalist class as a group, but may be the preferred form of colonial or semi-colonial regime for an imperialist power - or the only option for capitalists in conditions where an offensive of the proletariat begins to threaten loss of capitalist control of legislative assemblies.

The overall result is that, in both legislative-assembly regimes and dictatorships, legal/constitutional protection from takings of property is limited by being subject to takings through the state by large property holders/bribe payers. The ‘nation’-state is in either case perceived as a firm - in market competition, as well as in military competition, with other states in the international state system.

Particular capitals

The logic of what I have already said about a capitalist state as an institutionalised debtor is that the particular state is inherently dependent on its larger particular creditors. (The converse is, of course, also true: the scale of the state’s debts to its creditors makes the creditors at least partially dependent on their state debtor.) The clearest appearance of state dependence on creditors is in moments where the creditors demand the abandonment of particular policies - eg, in International Monetary Fund ‘restructuring programmes’, and in historical moments where creditors based outside the state territory, or a state representing them, actually take over state finances (as in the later 19th century history of Ottoman Turkey and of Egypt).16

Equally, what has already been said about the role of the legislator - individual or representative - implies that the state will be at least partially subordinated to the major particular payers of bribes (whether these are paid to elected legislators or to the single-person legislator in a dictatorial regime).

Both these points are, however, secondary to a more fundamental issue. This is that the capitalist reorganisation of warfare (the ‘military revolution’ of the early modern period)17 makes the state as a military actor dependent both on its suppliers of military matériel and on the overall balance of the economic structure within its territory.

Prior to capitalism, arms production was primarily at artisan scale and involved skills widely dispersed (blacksmiths, bowyers and fletchers ...). While powerful states could and did use concentrated state-controlled armouries or arsenals, these were not essential to military production and operations. Monopolies of military production and its technique were therefore of very limited significance in relation to state strength. The military revolution round gunpowder, which coincided and was connected with the early bourgeois revolutions, radically changed this. The manufacture of artillery,
(re‑)fortification to resist artillery, and the building of ships carrying heavy guns, are all necessarily industrial operations involving an extended technical division of labour within a single productive operation, proletarianisation of the workforce, and major capital investment. Partial capitalist development (even if under feudalist political regimes) is therefore a precondition for the full implementation of gunpowder warfare.

The first point is therefore obvious. If Ruritania’s weapons, ammunition and so on are entirely manufactured by suppliers in the territory of Westlantica, then Westlantica can cut off their supply, and will therefore have an effective veto over any military action by Ruritania which extends beyond a short period of time (identifiable by reference to the scale of munitions stockpiles held by Ruritania). It follows that the degree of practical autonomy of a state, when push comes to shove, is dependent on its arms production on its own territory - whether this is state production (which requires the state to mobilise capital and free labour) or production by locally autonomous capitalist firms. The consequence is that the state has a direct military interest qua state that independent arms production in its own territory should continue, irrespective of the profitability or otherwise of the capitalist firms that may be involved. The same is true of the transportation infrastructure, which has direct military applications.

The interest extends beyond immediate arms production. Thus, for example, in early modern Europe a strong merchant fleet and civil shipbuilding could be converted readily in wartime to naval vessels and naval shipbuilding; in modern times, heavy engineering, the motor industry and civil aviation can be readily retooled for war production, as 1939-45 demonstrated.

The second point is slightly less obvious, but has the same underlying military ground: that supplies can be cut off. Cutting off supplies of military materiel has a direct effect on ability to wage war. Cutting off other supplies is a very ancient military technique: that of besieging cities, which - by virtue of their location in the division of labour as non-agricultural producers - are vulnerable to being starved, even if their fortifications preclude direct assault.

Capitalism from the outset involves international trade connections and an international division of labour,18 and the long-term tendency is for these international connections and division of labour to intensify. Hence there is no longer a question only of cutting off food supplies, but also of cutting off raw material supplies and trade outlets. By dislocating the economy of the territory of the target state, that state’s ability to wage war is indirectly affected. The consequence of this and of the new naval technology - part of the military revolution - is that the technique of siege can be transferred from the scale of besieging individual cities to that of besieging (blockading) whole states. Attempted by the French and the British against each other in the wars of the 1790s-1800s, the blockade technique was a fundamental element of the global strategy of World Wars I and II (the Entente blockade of the Central Powers and the attempt to break it at Jutland; the counter-attempt of unrestricted submarine warfare; the Allied blockade of the Axis powers and the counter-attempt of the Battle of the Atlantic). Since the failure of direct military intervention to preserve the semicolonial status quo in Vietnam, siege/blockade, under the euphemistic name ‘sanctions’, has become a routine form of the military action of US imperialism and its immediate allies against states which in one way or another step out of line.

How are states to respond to these risks? The first answer, which is really only available to a state with immediate access to the sea and which is or immediately aspires to be a world hegemon state, is to maintain such a powerful navy that the effective use of blockade against it is excluded; and to intervene through diplomatic manoeuvres and, if necessary, military operations to prevent the creation of any land-based power or coalition capable of blocking sea trade altogether. This was in a sense already the policy of the Netherlands in the 17th century ‘Golden Age’ of the Dutch Republic; it was the general orientation of British state policy in the 18th and 19th centuries and down to 1914; and it was picked up by the US state from the 1940s, although the underlying strategic line was submerged in the particular form of the ‘cold war’ between 1948 and 1991.19

The limitation that the state has to have immediate access to the sea is obvious. It is hypothetically imaginable that air bulk-cargo transport could develop sufficiently to allow an air-based analogue available to landlocked states, but air transport technology is nowhere near this point today. The limitation that the policy is really only available to actual or immediate-aspirant world hegemon states arises because of the sheer scale of the resources involved: in order to preclude blockade, the state’s naval resources have to outweigh any likely combination of navies against it. To aspire to this is precisely to aspire to be the world top-dog state, and to achieve it is - whether deliberately or not - to make the state’s currency become the global reserve currency, for reasons which have already appeared (money is state money) or will appear below.

The second and more limited answer, which can be and is pursued by almost all states, is to have a sufficient degree of diversification and balance of the economy of the state’s territory to be able to withstand a partial blockade and to continue military operations, at least for a sufficient period to allow diplomacy and/or military counterattack to produce an acceptable outcome.

The internal logic of this policy is to extend the state interest qua state - ie, simply as a military actor - in ‘national industries’ well beyond those with immediate military applications. It is thus towards forms of protectionism and subsidy. The most transparent example is agriculture, which is subsidised and protected in pretty much all the stronger capitalist states. But a variety of forms of state support have been characteristic at one time or another of all industries which could be considered to have any degree of significance for the economic, and therefore military, autonomy of capitalist nation-states. The idea of a period of capitalism without state-backed industries having existed at some point in the capitalist past is an illusion caused by taking as good coin the claims of the Manchester liberals and similar ideologues of free-market capitalism, and by focusing on spheres of production which are as yet insufficiently significant to attract state attention.

The full and unqualified pursuit of this logic of nation-state economic autarky to back political independence is irrational. On the one hand, it produced the rival colonial empire protection systems of the first half of the 20th century, which attempted to obtain direct territorial control of raw materials sources and of markets for industry, leading to an inherent dynamic towards great-power war. On the other, it produced the national-autarkic ‘development planning’ of Stalinism and a wide range of ‘third world’ sub-Stalinist nationalist regimes: multiple duplicate heavy industry complexes, and so on.

The third option is to limit or abandon the goal of state autonomy: ie, accept at least partial political subordination to some other capitalist state. The subordination need not be complete (as in the colonial empires). It can, indeed, in the right circumstances have a very light touch: thus, for example, Switzerland and Sweden were both formally neutral through the two world wars, but this ‘neutrality’ was in substance German-friendly; or, the UK is a junior partner to the US in today’s world, as the Netherlands was to Britain in the 18th century. Space exists in any capitalist world order for a range of degrees of subordination to the world-hegemon state.

There are two factors affecting the degree of subordination. The first is the degree of antecedent development of the subordinate state before subordination, affecting its options and hence the costs of imposing fully colonial subordination. Any substantial fully-capitalist state - or state with highly developed feudalism on the verge of capitalism - has the possibility of turning itself into an aspirant world hegemon: witness Germany and Japan from the late 19th century to the mid 20th. It is therefore only in extremis that a capitalist state will actually attempt to treat as a full colony another existing capitalist state (as in occupied Europe in 1941-45). Where, however, the level of local development of the forces and relations of production before subordination is substantially lower, subordination is likely to produce both more onerous political subordination and a greater extent of adaptation of the economy of the subordinate country to the needs of the particular capitals of the superior power.

The second factor is the geopolitical needs of the superior state. In the geopolitical competition of the late 19th to mid 20th century, the smaller western European states were generally more geopolitically valuable to the great-power rivals as slightly subordinate allies or friendly neutrals than as real colonies or semi-colonies. The same was true on a more extensive scale of US policy in the cold war: contrary to Trotsky’s tentative hypotheticals,20 the US did not colonise western Europe or Japan, but promoted capitalist revival in these countries; and it did so also in a number of former semi-colonies, which were ‘front-line states’ vis-à-vis the USSR or China. In the cases of South Korea and Taiwan at least, this has meant very substantial capitalist development.

A subordinated state which is not a full colony can substantially reduce its military expenditures, but it cannot wholly dispense with concerns for military autonomy (the second option above), since the superior state will have multiple client states and cannot be wholly relied upon to regulate disputes between them. Even a state which accepts a subordinate position relative to a larger power will therefore also have concerns for domestic military production capability and for the defence of local capitals in other sectors to protect itself in case of blockade; although these concerns may be very much subordinated in the case of the geographically smallest states, whose role in international affairs is mainly interstitial (eg, Luxemburg).

The dynamics discussed here began as an impact of (proto-)capitalist military production on a predominantly late-feudal European state order. And the examples given are drawn from capitalist geopolitics in periods containing powerful late-feudal absolutist and other pre-capitalist state actors (17th-19th centuries), vast European colonial empires (mid-19th to mid-20th centuries), and the so-called ‘socialist bloc’ bureaucratic regimes (mid-later 20th century). Nonetheless, the argument of these sections is an abstract one: it discusses the inner logic of the military needs of capitalist states as - necessarily - military actors. It assumes capitalism only as industrialising the conditions of military production. And it therefore assumes only a world which is, like today’s, simply one of multiple capitalist quasi-nation-states. As the dynamics of capitalism were more transparent in 19th century England than elsewhere,21 so the dynamics of the inter-state competition of capitalist states, which were already present in the early modern period, have become more transparent, as capitalism has spread across the globe.

And these dynamics clearly involve the dependence of particular states, qua states as military actors, on the capitals in their territory - even if partially subordinated capitalist states are also dependent on the capitals of the superior states to which they are subordinated. There is a state interest in defending (some) ‘national capitals’ and not merely an interest of the particular capitals in mobilising the state in their defence.

Capitalist dynamics

So far, I have formulated grounds for supposing that the state is a form of the material division of labour, independent of class ordering and parallel to it, but necessary and subordinated to forms of class ordering; that having a territorial state (whether capitalist or that of a pre-existing class) is necessary to a capitalist economy; that capitalist states are, as they have emerged from the historical negation of feudalism, structurally dependent on capital; and that particular capitalist states are also materially dependent on particular capitals operating within their territory. The question posed is: what are the implications this analysis in understanding capitalist political-economic dynamics within the framework of the general claims of Marxism?

I will address three issues. The first, in this article, is the interrelation of tax and state borrowings with capitalist credit in general and the relevance of the average rate of profit used by Marx in his analyses of rent, etc in Capital volume 3. This may appear to be a long diversion (as well as stepping into a minefield of theoretical debates in Marxist political economy), but it is a necessary preliminary to the remaining issues.

The second, in the next article, is the relation of the state to the fundamental tendencies in capitalism identified by Marx: first, towards polarisation between an increasingly small class of capital owners and an increasingly large class of proletarians; and, second, towards cyclical crisis - periodic dislocation of the economy and impoverishment in the midst of and caused by material plenty. The second of these aspects will unavoidably involve another long ‘diversion’ in a minefield of theoretical debates.

The third issue is the effects of state interventions on capitalist dynamics: generally, I argue, to slow them down; but also (as Nikolai Bukharin argued, but more generally than Bukharin’s argument) to promote the relations of geopolitical competition between states already discussed.

In Capital volume 3, chapters 9 and 10, Marx (as edited by Engels) argues that the rate of profit tends to be equalised, because capital will naturally tend to move from low-profit activities (firms and sectors) to high-profit ones. To the extent that this tendency to equalisation takes effect, the result is that labour values are modified (‘transformed’) in the direction of ‘prices of production’, including as an element the average return on capital. Taking the simplifying assumption that the tendency is completed - ie. an equalised rate of profit is formed - allows a fairly simple mathematical treatment of redistribution of value between capitals through the price mechanism.

The marginalists argued that this procedure involved a fatal logical flaw in Marx’s political economy. The first problem was that ex hypothesi the input costs of production in the transformation schema also needed to be ‘transformed’. In the first place this would lead to an infinite regress. This is a relatively trivial objection, as WP Cockshott and his co-authors have pointed out in Classical econophysics (like other unavoidable infinite regresses reflecting features of the real world, this one can be made practically tractable by the use of iterative computational procedures).22 The more fundamental objection is that, if the input costs of production need to be ‘transformed’, Marx, after constructing a systematic theory on the basis that labour is the basis of value, has smuggled back in an inherent contribution from capital; and/or that, having regard to the transformation procedure, the original claims for labour values are redundant. The marginalists go on to argue that rent on shortage goods (‘goods’ here including services) is the more fundamental category, wages being reinterpreted as just another form of rent. The large majority of ‘Marxist’ economists have responded to this objection by modifying Marx in the direction of Piero Sraffa’s neo-Ricardianism or of Keynesianism.

Advocates of the ‘temporal single-systems interpretation’ (or TSSI) - Guglielmo Carchedi, Alan Freeman, Andrew Kliman and others - argue that the logical contradiction is only apparent. If historic costs rather than current costs are used as the basis of the input costs in the transformation, the contradiction disappears (and a whole range of other problems, like polarisation on the national and international scales, and the boom-bust cycle, become more transparent).23

Cockshott and his co-authors in Classical econophysics reject the TSSI on the basis that it involves a forced reading of Marx, and accept that there is a contradiction in Marx’s argument, but argue that the transformation procedure is redundant. Their reasoning is that profit rates do not tend to converge on equality, but to move towards a statistical equilibrium in which they are quite widely distributed - an argument originally made by Emmanuel Farjoun and Moshé Machover.24 Within this framework the empirical evidence suggests that untransformed labour values are as predictive of actual prices as the cost of production, or more so.25

What is missing in the argument that the transformation procedure is redundant is the role that the ‘general rate of profit’ plays in the larger argument of Capital volume 3 in accounting for market rents and rates of interest and for ‘merchant profit’: ie, the profit on commodity speculation, as forms of redistribution of an aggregate body of surplus value originally created in capitalist exploitation of labour.26 The connection is in some ways more transparent in Marx’s critiques of David Ricardo and Adam Smith on the issue in Theories of surplus value and in his 1868 letter to Engels which foreshadows this part of the argument of Capital volume 3.27 If labour-time is to be the basis of value, it is necessary in some way to account for the persistence in developed capitalism of forms of income which are not directly derived from the employment of free labour. The idea of a redistribution of profit according to capital advanced - if true - fulfils this function - or, more exactly, provides an account of the dependence of rent and interest on profit from the exploitation of labour in a fully capitalist economy.

There were, of course, both rent and interest in pre-capitalist societies. But neither obeyed the market laws discussed in Capital volume 3. Rent, on the one hand, was a primary form of the direct exploitation of peasant labour, and a secondary form by which broad ruling classes produced from their own ranks concentrated elites capable of governing (Roman latifundistas, medieval baronial elites). At base, it ‘grew out of the harvest’. At this level there were local, customary rents, not market rents in the modern sense, and harvest failure normally implied at least partial rent relief.28 Interest, on the other hand, was ‘absolute rent’ on money - a tribute taken by the money-holder merely in the right of ownership; in effect a tax to support the public benefits provided by the existence of the social institutions of private property and money, falling randomly on borrowers and benefiting lenders equally randomly. But the rate of interest was in no sense market-controlled, but rather so arbitrary that there could be no market rate, with the result that it was subject to legal prohibition or legal limits on interest rates.29 What changes in capitalism is the formation of market rates of rent and interest, which entail subordination of rent and interest to profit on productive activities. Indeed, if capital is to be the ruling class, this transition - the subordination of rent and interest to capitalist profit - is necessary.

If capital as such attracts the average or general rate of profit, then the same must be true (with the necessary modifications) of capital invested in land, minerals, etc (rent) or of capital lent at interest. Rent and interest are thus subordinated to capitalist profit. The problem, however, is how there comes to be an average or general rate of profit to play this role.

Marx argued that capital tends to move into more profitable activities, and therefore, in order to render the mathematical exposition tractable, we can make the simplifying assumption that it has done so - with the result that there is a uniform rate of profit. But this argument clearly will not hold water. The stage of adopting the ‘simplifying assumption’ is unacceptable. The tendency of capital to move into more profitable activities, considered as a law of tendency, is a long-run one. Analogously, it is a biological fact that eventually all organisms will die. To make the ‘simplifying assumption’ that all organisms are dead would leave biology with nothing of interest to say. Or in thermodynamics entropy tends to increase, but to assume for the sake of simplification that entropy was already at maximum would destroy any use of the theory. The tendency of capitals to seek opportunities for higher profit is a fundamental motor of capitalist dynamics. To assume that capitalism had attained a state of a uniform general rate of profit would therefore negate its character as capitalism; what would be left would no longer be capitalism, but a mere generalised rentier system.

Secondly, assuming that capital tends to seek higher-profit activities, this tendency is not powerful enough - or is sufficiently affected by powerful obstacles and counter-tendencies - that there is not a general tendency for rates of profit to converge on the average rate or formation of a real general rate of profit through simple competition within any period of time which is relevant to real-world analysis.

I stress general tendency and general rate. Suppose groups of firms which are in immediate competition with one another selling inter-substitutable commodities (eg, cars, but also cars and motorbikes; pens, but also fountain pens, biros and rollerballs). We can call such groups ‘sectors’ for the sake of a shorthand expression, but subject to the clear understanding that the definition above is different to that of a ‘sector’ in aggregate economic statistics. There is a powerful tendency within such a sector to the convergence of rates, not only of any hypothetical ‘transformed’ profit, but also of surplus value. The reason is that within such a sector it is immediately transparent to the competing firms that the price of goods in the sector follows the average production costs, and hence that one firm within the sector’s productivity gain, or use-value (product feature) gain, is another such firm’s productivity, or use-value, relative loss. Hence the firms within the sector are immediately driven to copy the most successful material productive technique in the sector and, as a result, converge on a common organic composition of capital and rate of surplus value.

This tendency also drives centralisation of capitals within the sector: ie, a tendency to monopoly/oligopoly and cartels. This in turn allows a monopolised sector, if its output is necessary to other economic activities, to charge above-value prices, leading to a redistribution of surplus value in favour of the monopolised sector. Between sectors, however, there is no such transparency to drive convergence; and, indeed, the material-practical applicability of fixed capital may vary dramatically from sector to sector.

An average rate of return on investments is, however, formed - in financial markets, and as an item of information which provides the basis of investment decisions in these markets, which then spread to the basis of investment decisions in commercial and agricultural land and loan markets - with weaker, but still present, consequences for housing rent and mortgage, and consumer loan, markets.

Engels in his supplement to Capital volume 3 (1895) accepted that the ‘general rate of profit’ is not satisfactorily accounted for in chapters 9 and 10 of that volume, saying that if Marx had lived he would have elaborated further. He offers his own elaboration. This is that labour values hold in petty commodity production; capital as such, in contrast, grows out of joint-stock operations in shipping, textiles and mines. These in turn develop in a late-medieval context, in which small merchant and artisan proprietors have holdings in common, controlled by narrow collectives, like the German peasant Mark and artisan guild monopolies. The joint-stock operation reflects this origin in a ‘capitalist communism’, in which the profit of the joint stock is returned in proportion to capital contributed.30 This is reflected in the character of fully-developed capitalist society: ie, that a general rate of return on capital contributed is expected. The result is a dynamic away from simple labour values and towards a general rate of profit.

The weakness of this argument is at the point of the transition from the joint-stock enterprise in late feudalism to capitalist society.31 The joint-stock enterprise is not merely an abstract distributive principle among capitalists inter se, but a concrete, organised institution - and within the economy as a whole it is a firm in competition with other firms (inter alia for investment capital). Engels’ argument does not provide analogous organisational forms, or analogous motivations, for the formation of anything resembling a generalised rate of profit across the capitalist economy as a whole.

The missing term in Engels’ argument is (as might have been expected from my earlier argument) the state. The capitalist state is a concrete organised institution, was conceived by some writers of its early period as a joint-stock, and is certainly conceived in modern times as a firm (talk of ‘British competitiveness’ and so on). However, the capitalist state is not like a joint-stock in the sense simply of paying dividends in proportion to capital invested. The connection between the capitalist state, the average rate of profit and the subordination of rent and interest to capitalist profit is more complex.

The problem confronting the revolutionaries who created the early capitalist states was of how to raise funds to wage war in the absence of the inherited rights to revenue and related powers, which were vested in princes and feudal lords and formed the core of the finance of the feudal states - and, in fact, before the new states had full control of territory from which to raise revenue. The solution was a combination of, first, taxes agreed by some form of ‘representative’ consent, which was an element in the revenue of the later medieval feudal monarchies, but not the stable core of their revenue; with, second, borrowing - initially from a politically defined group of supporters of the new regime - on the basis that future tax revenues were in the first place hypothecated (mortgaged) in favour of the payment of interest to the creditors. The system involved the creation of central banks to form a ‘central committee of the creditors’ vis-à-vis the institutions through which consent to tax was obtained, and the explicit transferability of the rights created in the state’s debts.32 The result is that the capitalist state is a form of joint-stock operation. It is in origin a joint venture of the revolutionaries. The major taxpayers and their (bribed) parliamentary representatives are analogous to, though not identical with, the equity shareholders in a corporation, with the ‘dividend’ taking the indirect form of the surplus profit of firms protected by the state over tax liabilities; and the state creditors are precisely analogous to a corporation’s bondholders.

The creation of transferable state debt instruments secured on future tax revenues has profound effects. As I indicated above, in pre-capitalist societies rent grows out of the harvest, but interest has no predictable character and is to some extent discountenanced by the laws; the legal systems protect interests in land strongly, but debt claims have a much less secure character and, indeed, a speculative one. Hence rights in or over land (including local jurisdictions and mortgages) are the only secure long-term investment from which income can be drawn to support rentier purposes: ruling class widows, orphans and elderly, and public purposes not directly run by the state (temples, churches, monasteries, local civic distributions, local public works, and so on). The creation of transferable state debt instruments secured on the future tax take transforms the situation. Investment in the public funds now offers an alternative to investment in land; a fortiori as the central bank and transferability creates an organised investment market (Amsterdam, London ...).

I said above that the state is like a joint-stock corporation. This is only a half-truth. The joint-stock predates the capitalist state and informs aspects of its form. But the state - and especially the organised investment market in the public funds - reacts back on the form of capitalist joint ventures, inter alia leading these to participate in the organised investment market by issuing transferable shares and stocks tradeable on this market.33 The more the investment markets grow, the more they also react upon rents and on rates of interest more generally.

The effect of these processes is not to lead to an actual equalisation of the rate of profit on capital advanced. Rather, the shares and debt stocks, along with government securities, are subjected by the investment markets to a constant process of upwards and downwards revaluation of capital, relative to the return on government securities. The capital advanced may be entirely lost by the individual lender or shareholder by a pure process of downwards revaluation by the investment markets, even though the physical enterprise is making a real return on the actual money originally advanced and/or there is still substantial tradable value in its material fixed capital assets; conversely, once-profitable enterprises, which are really losing money, may continue to pay dividends in order to keep up their market valuations;34 or (as happens in bubbles - most strikingly in recent decades in the 1990s dot-com bubble) the individual capital advanced to an enterprise which has never made a profit may temporarily multiply enormously on the basis of pure hope for its future.

Through these processes money comes to be seen as fructiferous and rents, interest and the return on speculative operations are subordinated to profit - the processes described by Marx in Capital volume 3 - without an actual convergence of rates of profit or more than a limited ‘transformation’ in the price mechanism. There is some redistribution through the price mechanism and a degree of ‘transformation’, but what is involved is not a direct transfer from low-capital firms to high-capital firms: it is transfers into and through the rentier investment financial markets in the form of interest charges on individual firms’ debt, rent, etc.

The fixed point, which enables all these processes, is the basic idea that government securities are equivalent to land as a rentier investment. Without the safeguard of government securities, the investment markets more generally will be destabilised and fail. It is thus through the creation of the capitalist state that the capitalist class constitutes ‘abstract capital’ and constitutes itself as the ruling class - ie, constitutes the idea that profit is the fundamental basis of ‘civilisation’.


  1. In relation to Athens, see DM MacDowell The law in classical Athens New York 1978, p80. On Rome: A Watson Rome of the XII tables Princeton 1975, pp122-23.↩︎

  2. Eg, MI Finley Ancient slavery and modern ideology London 1980, pp117-99; and O Paterson Slavery and social death Cambridge MA 1982, pp58, 176-79.↩︎

  3. Res publica Romana: eg, Cicero De re publica I, 41. Meum ex iure Quiritium: the formula of claim of ownership used in the legis actio sacramentum in rem: G416 - ‘Mine by Roman citizen law’ is an alternative translation; both F de Zulueta (The institutes of Gaius, part 1, Oxford 1951, p237) and WM Gordon and OF Robinson (The institutes of Gaius London 1988) give the uninformative “Quiritary right”.↩︎

  4. PRC Weaver, ‘Social mobility in the early Roman empire: the evidence of the imperial freedmen and slaves’ Past and Present July 1967 collects together a good deal of information for the early empire, but also for the late republic.↩︎

  5. See the discussion in M Macnair, ‘Historical blind alleys’ Critique No58 (2011).↩︎

  6. Dominium in this context means absolute (or unlimited) ownership - iura in re aliena (limited) rights over property owned by another. The account of Roman property law in terms of absolute ownership and iura in re aliena constructed by the 19th century Pandectists is now regarded as radically oversimplified; cf, for example, A Rodger Owners and neighbours in Roman law (Oxford 1972), and, for a very brief wider-ranging survey, see A Borkowski and P du Plessis Textbook on Roman law Oxford 2010, pp157-59. JQ Whitman The legacy of Roman law in the German romantic era (Princeton 1990), chapters 5-6, brings out clearly the links between this reasoning and the development of capitalist agriculture, which was already hypothesised in 1949 by Karl Renner in The institutions of private law and their social functions (New Jersey 2009) chapter 2, on the basis of hints in Marx.↩︎

  7. On Edward I, see a recent summary account in AR Bell, C Brooks and T Moore, ‘The credit crunch of 1294: causes, consequences and the aftermath’: www.voxeu.org/index.php?q=node/3563. For Philip the Fair see M Barber The trial of the Templars Cambridge 1978, pp27-44 and 246-47.↩︎

  8. Cf discussion of the ideological aspect of the factions in S Ferente, ‘Guelphs! Factions, liberty and sovereignty: enquiries about the Quattrocento’ (2007) History of Political Thought No28.↩︎

  9. This is clear in both R Brenner Merchants and revolution (Cambridge 1993) and J Scott England’s troubles (Cambridge 2000).↩︎

  10. The case of Japanese feudalism is different, because, while Korea came close to feudalism, it did not achieve it, and the region/common culture remained dominated by the late antique empire in China until the entry of European capitalist states into east Asian regional politics.↩︎

  11. I say quasi-nation-states because capitalist states tend in practice to be sub-national (Flemish-speaking Belgium is part of the same nation as the Netherlands; Germany and Austria are parts of a single nation), supra-national (Britain includes England, Wales and Scotland, and at its height Ireland; France includes Brittany, Corsica and part of Euzkadi; etc) or simply constructed by colonialism (most states outside Europe).↩︎

  12. Cf C Miéville Between equal rights Leiden 2005, chapter 5, for a summary account of the history within a Marxist framework (albeit one different from the present). B Teschke The myth of 1848 (London 2003) builds a different narrative on the basis of the ‘Brenner thesis’, but at the end of the day the new narrative merely displaces the formation of international law from Westphalia (1648) to Utrecht (1713). Neither in my view takes sufficient account of earlier developments in the direction of the international law conception within the frame of the Mediterranean interstitial-capitalist city-states (cf, for instance, AP Sereni The Italian conception of international law (New Jersey 2009) part 1; and G Mattingly Renaissance diplomacy London 1973, part 2.↩︎

  13. Even if it may in practice be forced to accept a subordinate position (see below).↩︎

  14. See, for example, BG Carruthers City of capital: politics and markets in the English financial revolution Princeton NJ 1996; MC ’t Hart The making of a bourgeois state Manchester 1993.↩︎

  15. E Burke Reflections on the revolution in France: www.bartleby.com/24/3/4.html. The image was common at the period: Gregory Dart in Rousseau, Robespierre and English romanticism (Cambridge 2005) cites Sièyes as using it in the 1780s. Of course, modern media and politicians’ talk of ‘British competitiveness’, and analogous formulae in other countries, continues to assume the idea of the state as a business firm.↩︎

  16. On Turkey see M Birdal The political economy of Ottoman public debt London 2010. On Egypt there are useful narratives in DS Landes Bankers and Pashas (Cambridge, Mass 1958) and RC Mowat, ‘From liberalism to imperialism: the case of Egypt 1875-1887’ The Historical Journal No65 (1973). A review of Turkey, Greece and Egypt from a creditor standpoint, using a wide range of literature, can be found at discovery.ucl.ac.uk/id/eprint/1514428/1/Tuncer_Accepted_05092016.pdf.↩︎

  17. This has been extensively debated. See (among others) G Parker The military revolution Cambridge 1996; J Black A military revolution? Basingstoke 1991; CJ Rogers (ed) The military revolution debate Boulder 1995; BS Hall Weapons and warfare in Renaissance Europe Baltimore 1997); A Ayton and JL Price, ‘Introduction’ in The medieval military revolution London 1998. However, the debate over the proper periodisation of military change (and about what can properly be called a ‘revolution’) which dominates this literature is of completely secondary importance to my concerns, which are with the transition from pre-capitalist to capitalist forms of military production. This transition should a priori be expected to parallel the early development of capitalism: ie, appearing in gradual and limited forms alongside feudal military service in the later Middle Ages, and becoming increasingly dominant from the early modern period.↩︎

  18. See my review of Boris Kagarlitsky’s Empire of the periphery: Weekly Worker April 2 2009 (weeklyworker.co.uk/worker/763/studying-the-past-to-grasp-the-future) and April 9 2009 (weeklyworker.co.uk/worker/764/marxism-and-the-inequality-of-nations).↩︎

  19. For references on Dutch navalism, see GE Halkos and NC Kyriazis, ‘A naval revolution and institutional change’ European Journal of Law and Economics No19 (2005). British naval policy: P Kennedy The rise and fall of British naval mastery London 2004, chapters 3-6. US navalism: MA Palmer Origins of the maritime strategy: American naval policy in the first postwar decade (Annapolis 1988). JR Holmes and T Yoshihara Chinese naval strategy: the turn to Mahan (London 2009), though addressed to analysis of Chinese naval policy, illustrates throughout the continued assumptions of global naval dominance of US policy - the authors take it to be a threat to US security if the US Navy were to lose uncontested access to the Chinese coastline.↩︎

  20. L Trotsky Europe and America (1926): www.marxists.org/archive/trotsky/1926/02/europe.htm; also ‘On the national question’ (1938): .↩︎

  21. K Marx Capital Vol 1, preface.↩︎

  22. WP Cockshott et al, Classical econophysics Milton Park 2009.↩︎

  23. A Kliman Reclaiming Marx’s ‘Capital’ Maryland 2006.↩︎

  24. ED Farjoun and M Machover Laws of chaos London 1983.↩︎

  25. WP Cockshott et al, Classical econophysics Milton Park 2009.↩︎

  26. There is a great deal of confusion among Marxists about ‘merchant capital’ and ‘merchant profit’. Marx argues that in its pure form merchant capital is unproductive. However, though Marx does not make this clear, this ‘pure form’ is simply buying cheap and selling dear, abstracted from everything else: that is, it is commodity futures speculation. Transportation and warehousing are productive activities: ie, they increase the sum of available use values to the end users, and the fact that before capitalism they were conducted by people who were also concerned in buying cheap and selling dear does not render them unproductive, any more than the efforts of industrial capitalists to buy cheap and sell dear do so. Hence, to the extent that later medieval and early modern ‘merchant capitalists’ who were engaged in transportation and warehousing carried on their businesses by employing wage labour, they were not a class fraction distinct from manufacturers, but simply capitalists. In this frame, the decisive difference between interstitial pre-capitalist mercantile activity, and capitalist mercantile activity emerging within feudalism, is the scale - at base a technical scale of ship construction, etc - which implies the employment of wage labour both to build and to run ships, as well as bulk transportation with effects beyond luxury markets (another angle of approach to this issue is taken by J Banaji in ‘Islam, the Mediterranean and the rise of capitalism’ (2007) Historical Materialism No15). The ‘pure form’, on the other hand, when fully abstracted from transportation and warehousing, is revealed as merely a financial operation.↩︎

  27. Marx to Engels, April 30 1868 CW Vol 43, p20.↩︎

  28. For Roman law see DP Kehoe Investment, profit and tenancy Michigan 1997) pp214-34. For medieval ius commune R Zimmermann The law of obligations (Oxford 1996), pp371-74, is useful. In English medieval law there was very indirect recognition of the principle through the action of waste, held inter alia to create liability of landowners holding less than full ownership (as guardians, widows, widowers, etc) for driving villein tenants to flight by exactions, which, though in a strict sense legal, were excessive. See PR Hyams King, lords and peasants in medieval England Oxford 1980, pp27-29.↩︎

  29. For statutory regulation in Roman law, see J Andreau Banking and business in the Roman world Cambridge 1999, chapter 9. On theoretical prohibition in medieval Europe see the summary in R Zimmermann The law of obligations Oxford 1996, pp170-75. In regard to theoretical prohibition in Islamic law, J Schacht An introduction to Islamic law (Oxford 1964), though controversial in its general approach to the origins of Islamic law and very dated in relation to recent developments of Islamic law in relation to finance, provides a convenient short summary of the principle at pp145-46. For theoretical prohibition in ancient Hindu law, see W Doniger and BK Smith The laws of Manu London 1991, pp59, 62, 93, 94 and 257. Regulation of interest rates: P Olivelle (trans) Dharmasutras Oxford 1999, p99. Statutory regulation in pre-modern China: V Hansen Negotiating daily life in traditional China New Haven 1995, p43 (medieval); PCC Huang Civil justice in China Stanford 1996, pp78-79, 83-84 (19th century). Statutory regulation in early modern England: M Bacon A new abridgment of the law London 1807. Usury Vol 7, pp187-214, though relatively late, provides a convenient treatment.↩︎

  30. The phrase ‘capitalist communism’ was already used by Marx in his 1868 letter to Engels (see note 27). In reality this was by no means the universal principle of early joint-stock arrangements and it is not even now the universal principle of the distribution of profits within capitalist business corporations. High executive salaries and ‘bonuses’ are in reality a distribution of profits to traders and managers who have contributed only unquantifiable assets - trading skills and contacts - to the business. Beyond this point, corporate capital structures are commonly highly complex mixtures of long-term loan capital and different classes of equity with distinct rights.↩︎

  31. An alternative systematic critique of Engels’ argument is offered in John Weeks in Capital, exploitation and economic crisis London 2011, chapter 1, on the basis that Marx’s argument as a whole presupposes a fully capitalist society and labour values are impossible in pre-capitalist society. In the pure form of Engels’ argument this is in a certain sense clearly valid, the point being that the peasant or smallholder does not pay for all inputs, but generates some of them by labour which directly produces use-value. However, at this level of requirement of ‘purity’, Marx’s model is not true of present-day capitalism: witness, for the most obvious example, domestic labour.↩︎

  32. See Carruthers, ’t Hart, etc, cited in note 14.↩︎

  33. When I say inter alia, the alia includes the adoption of state-bureaucratic forms of internal control and accounting, as well as formal constitutions and rules of procedure borrowed from local government.↩︎

  34. Examples include the Dutch VOC (East India Company) in the mid- to later 18th century and the British Motor Corporation/British Leyland in the 1960s and 70s (see the 1975 Ryder report, cited in the obituary of Lord Stokes published in The Guardian July 22 2008: www.guardian.co.uk/business/2008/jul/22/automotive.past).↩︎