Origins of fortress west

What is the connection between the development of capitalism and restrictions on the movement of labour? In the first of two articles Mike Macnair traces such restrictions back to feudalism

The globalised world economy features two linked juridical phenomena - on the one hand, the increased legal freedom of movement of capital created by financial market liberalisation and the widespread abandonment of exchange controls; and, on the other, the tightening of forms of legal controls on the international movement of workers, particularly through immigration law. In this article I will approach the question firstly as a historical problem and secondly as a problem of Marxist theory. In my concluding article I will examine the question as a strategic problem for the workers’ movement.

The historical problem

The two sides of this question - freedom of movement of capital and controls on the movement of workers - have many aspects.

Capital, for example, by its nature is both free to move and not free to move, and this freedom and unfreedom of capital, which is a contradiction within capital as a social form, also finds expression as a contradiction between particular capitals. I will return to this question in addressing the theoretical problem.

As for the question of migrant labour, it is three-sided. First, there is very substantial actual migration of labour. For example, about half a million eastern Europeans have migrated to the UK since 2004, adding slightly less than 1% to the UK population. In Saudi Arabia there are currently estimated to be around five million migrant workers, as compared to a ‘native’ population of about 12 million.

Second, immigration is - at least in the imperialist countries - a political hot-button issue routinely exploited by parties of the right and far right. One might imagine that ‘nativist’ politics would have no purchase in countries like the US and Australia whose population is almost entirely composed of relatively recent generations of immigrants, but this is far from being the case. Each generation of settlers thinks of itself as ‘native’ in relation to the next. If anything, racist discourses about migration are more endemic and more violent in these countries than in European countries.

Third, there has been an unevenly progressing increase in border controls, controls on movement and controls on migrants obtaining citizenship or the legal right to remain in the destination country. A partial exception is provided by the development of the European Union, and in particular by its expansion eastwards after the fall of the Soviet regime and its eastern European satellites. It is a partial exception because the expansion has made the EU no longer simply a coalition of imperialist countries, but an entity which includes both imperialist countries and countries which are in transition to semi-colonial status. On the other hand, with legal freedom of movement of labour within the EU has come an increasing attempt - via the 1985 Schengen agreement - to tighten controls on entry to the EU from outside.

It is important to state that the conventional order in which I have presented these three sides is arbitrary. The conventional order contains within it an implied causal narrative: there is a lot of migration; hence (as it were naturally) there is a nativist reaction; hence governments appease this reaction by introducing immigration controls. But the narrative could be presented differently.

Legal controls on migrant labour facilitate intensified exploitation of migrant workers; therefore capitalists seek migrant workers in order to obtain higher profits, leading to higher migration, which reduces average wages; therefore there is a nativist reaction. Or it could be: the parties of the right promote nativism, for reasons unconnected with any real pressure of migrant labour on average wages, etc.; hence legal controls on migration are introduced; hence intensified exploitation is possible, leading capitalists to seek migrant workers, leading to increased migration. In reality, in other words, what is involved is three aspects of a single totality, and any attempt to disentangle fully the causal relations involved produces a snake eating its own tail.

Why have I described this as a historical problem? The answer is that it is widely believed that this is a new phenomenon. Maybe it is to be ‘new’ in the sense of being a part of Ernest Mandel’s ‘late capitalism’, with the black Commonwealth immigration to Britain from the 1950s, and at the same period the analogous north African migration to France, and German exploitation of Gastarbeiter. Or maybe it is to be ‘new’ in the stronger sense of being an aspect of US-led world capital’s turn to financial globalisation after the crisis of the 1970s.

The historical evidence suggests that this is not the case. The promotion of freedom of movement of capital, and unfreedom of movement of labour, and the political tensions associated with it, go back to the beginnings of modern capitalist politics with the English revolutions of the 17th century. Grasping the phenomenon theoretically will therefore require grasping it as an aspect of the inner logic of capital in general, not as a peculiar feature of the particular forms of late 20th and early 21st century capitalism.

The mobile and rootless character of capital was already identified as a political problem in the rhetoric of the reign of queen Anne (1702-14), contrasting the ‘landed interest’ and the ‘moneyed interest’. This rhetoric had antecedents in later medieval nativist rhetoric against Jews and foreign merchants, but the specific contrast of land and money in terms of the mobility of money-producing lower commitment to the nation-state was new, and reflected the emergence of the London financial markets in the 1690s and their integration with those of Amsterdam. Political rhetoric of this type had significant influence, and continued to be used - mainly by Tories - episodically into the 19th century.

As I have just said, nationalist/nativist rhetoric against Jews and foreign merchants - and against the employment of foreigners by kings - goes back to the middle ages. Incidentally, to say this is pretty unorthodox Marxism, since the orthodox view - reaffirmed in ‘postmodern’ form in Benedict Anderson’s Imagined communities - is that nationalism as such is a modern, capitalist, phenomenon. However, to claim that nationalism is modern involves leaving without a theoretical home the Barons’ 1236 Nolumus leges Angliae mutari (‘We do not wish the laws of England to be changed’), the agitation against King Henry III’s Poitevin French advisers at the same period, any number of university riots between the nationes of English, French, German, etc students at medieval universities, and urban riots against Lombards and Hanse merchants, which continued episodically down to the 16th century. These are phenomena of feudal society and express its internal class contradictions. They have a late flowering in the catholic clerisy’s promotion of anti-semitism as a last defence against capitalism in late 19th and early 20th century Europe.

In 1680s England, however, we find the first instance of a new form of nativist rhetoric. Directed in the first place by supporters of the emerging Tory Party against Huguenot (French protestant) refugees settling in England, the new rhetoric blamed the Huguenots for taking English jobs and not assimilating to the Church of England. The Tory Party and its literary and artistic hangers-on were to play this card over and over again between then and now.

A Tory campaign in the 1750s against proposals to facilitate ‘naturalisation’ - ie, the acquisition of citizenship by migrants - mixed the new rhetoric with the old by focussing on the possibility of Jews being naturalised. In the 1790s French émigrés as well as feared subversive French republicans were targeted by the first Aliens Act (1793). In the 19th century the focus was on Irish migrant labour, in a campaign of defamation which gave us the word ‘hooligan’.

From the beginning of the 20th century up to World War II, the focus was again on Jews. But though conventional catholic Jew-as-capitalist rhetoric can be found among the ideological products, the campaigning edge behind the Aliens Act 1905 was east European Jews ‘taking British jobs’ and refusing to assimilate - and the fear of foreign ‘anarchism’, which transmuted in the 1920s into fear of foreign ‘Bolshevism’. From the 1950s it moved onto ‘niggers’ and in the 1970s onto ‘Pakis’. The ‘war on terror’ has its own peculiar logic of victimisation; but we seem today to be at a moment at which the core of nativist anti-immigrant rhetoric is shifting onto eastern Europeans.

American history, of course, lacks classic Toryism: Tories as such were driven out in the American Revolution. But the history of waves of migration into the US is also a history of nativist rhetoric against migrants: from Irish immigrants and the nativist ‘Know-Nothings’ of the 1840s and 1850s, to today’s US efforts to reinforce its Mexican border against Latino migration.

The earlier instances of the use of new-style nativist rhetoric about jobs, etc targeted relatively small numbers of migrants in a period when the proletariat as a class was in process of formation. Linebaugh and Rediker’s The many-headed Hydra illustrates the extent to which one of the central elements in the formation of the new class - seafarers and port workers - was from the outset migrant and international in its composition.10 

The rhetoric relates in an oblique way to a real phenomenon of labour mobility from the beginning of capitalism. Other examples of large-scale labour migration, with accompanying nativisms and legal controls, can also be found from the high period of imperialism - for example, in the importation of Indian labour into South and East Africa and Malaya and of Chinese labour into the western US.

It is so familiar to historians as to be often ignored, but probably less familiar to Marxists, that throughout the period of the rise of industrial capitalism in Britain, the state maintained legal controls on the internal movement of workers within England and Wales. Under the Poor Laws, parishes - the basic unit of the Anglican church and also to some extent of secular governance - were obliged to provide relief to the poor (principally the unemployed, but also the sick and elderly). Under the Settlement Act 1662 and following acts, unemployed workers could be deported to their parish of origin. The acts gave rise to a formidably large body of case law, and were not formally abolished until 1948, though the settlement rules were reformed in the mid-19th century and not much used in practice after this.

Under the Poor Law regime, moreover, poor children could be compulsorily apprenticed to an employer, which made it illegal for them to leave that employer for the seven years’ duration of the apprenticeship. This became a method used by several early mill-owners to supply themselves with effectively unfree labour.

But over this period, if there were efforts to control the mobility of labour, capital’s mobility was increased: witness the financial markets, witness the dramatic improvements in transportation, and witness the late 18th century textile mills of the valleys of the Pennines, and other manufacturing establishments elsewhere, created outside corporate boroughs in order to escape their regulatory powers.11 

This history points to a simple conclusion, suggested above. The whole complex phenomenon - increasingly free movement of capital, legal constraints on the movement of labour, actual large-scale employment of migrant labour in spite of these constraints and nativist rhetoric against migrant workers - is nothing new. It is an endemic feature of capitalism - more or less visible from one time, place or economic sector to another, to be sure, but nonetheless a body of dynamics of capitalism as such, not of imperialism as the ‘highest stage’, of ‘late capitalism’ or of ‘globalisation’.

The theoretical problem

To recognise a general dynamic in capitalism as such towards freedom of movement of capital, towards labour migration and towards attempts to control the movement of labour by law poses a theoretical problem for Marxists: or, more exactly, a group of theoretical problems. The underlying problem is that Marx’s Capital postulates a ‘pure capitalist’ economy, in which there is neither unfreedom of movement of capital, nor unfreedom of movement of labour.12 

To a considerable extent, moreover, Marx explained the nation-state in the terms of a movement towards such a ‘pure’ capitalism. In this explanation, by virtue of the very ‘purity’ of capitalist property rights, the capitalists are necessarily driven to expand beyond the horizons of the city and borough and create a national state. But this involves the freedom of movement of labour as well as that of capital: because without the freedom of movement of labour there is no national market either in labour-power or in the goods consumed by labour.13 

Hence Marxist theories of the emergence of capitalism have tended to include strong elements of the emergence of freedom of movement of both capital and labour within the emergent nation-state through the overcoming both of forms of personal servitude and of the regulatory powers of boroughs, guilds and the monopoly corporations of absolutism.

So the explanation of the nation-state in terms of the development of the national market will not enable us to derive the present phenomenon, which is both international and sub-national in character.

More fundamentally, Jairus Banaji in a 2003 article in Historical Materialism has argued that the sharp differentiation between free and unfree labour made by many Marxists is unsound and internalises the liberal ideology of freedom of contract. Rather, he argues, there is a spectrum of capitalist relations to labour which runs from, at one extreme, a characterisation of the slave plantations of the American South, etc as capitalist (found at several places in Marx’s writing) through various intermediate forms, to, at the other extreme, the conventionally ‘free’ worker who is bound to capital as such, but not to any individual capitalist.14  Our English 18th century apprentices bound for seven years and workers deportable if they become unemployed because of the law of settlement, discussed above, fall at intermediate points on this spectrum.

The difficulty Banaji poses, though he does not express it in his 2003 article, is that the precise distinguishing feature of Marx and Engels’ strategic conception of the way to get beyond capitalism rests on the difference between free labourer, serf and slave. Marx and Engels argue that because the proletariat is freed from the means of production - neither owned as a means of production, like a slave, nor owned with the land and tied to it, like a serf - the proletariat can, by emancipating itself, emancipate the whole of society. Banaji’s spectrum of unfreedom calls this argument into serious question.

If we could really have a capitalism which, as a whole and effectively, denied the freedom of movement of labour, we would be forced to abandon the whole idea that the proletariat as a class could by emancipating itself emancipate the whole of humanity.

We might still find Marx’s economics useful in the analysis of capitalist dynamics - though, frankly, because Marx’s Capital presupposes free movement of labour (as I said before), it would probably be of very limited use. But we would certainly have to abandon any idea of a connection between the workers’ movement and the general emancipation of humanity in favour of some form of utopian or ethical socialism.

Banaji is perfectly well aware that, in fact, we cannot have capitalism with a universal unfreedom of labour: he cites for this point one of his own previous articles from 1977. But it is almost a throwaway point; and in his Agrarian change in late antiquity (Oxford 2001) Banaji’s methodology of the spectrum of unfree wage-relations leads him to find proto-capitalism in late Roman and Byzantine Egypt, and to make extensive analytical use of economic categories which are in substance neo-classical. That is, like neo-classical and institutionalist economists studying other aspects of the past, he projects back capitalist categories onto the pre-capitalist world. Under this method, the logic of capital becomes a timeless truth which is merely prevented from appearing effectively until the political economists, from Adam Smith onwards, opened the way to its general acceptance.

The key to escaping from this problem is a group of understandings. First, Marx’s Capital offers an abstraction from the real development of capitalism, or a counter-factual analysis of what the consequences would be if the pure capitalism recommended by the political economists was actually created.15  Second, the project of which Capital Vol 1 and Engels’ reconstructions of volumes 2 and 3are part is very radically unfinished (as Harvey and others, most recently Lebowitz, have argued).16 

I would add: this is true in particular because (as I have argued elsewhere) the nation form of the state cannot be derived directly from the laws of motion of capital or the requirements of capitalist development, but is an inheritance from feudalism.17  Third, the laws of motion of capital which the text discusses are laws which are instantiated as probabilistic laws of tendency in the real world, not as exact laws, as several authors, notably Carchedi, have argued.18 

An associated point to this last is that made by de Ste Croix in discussing classical antiquity.19  A ‘mode of production’ or ‘class order’, as applied to a society, refers to the dominant mode by which surplus is extracted. It does not imply that all other forms of organising production are absent. In particular, petty proprietorship and petty family production are present in classical antique ‘slave’ society as well as in feudalism, and continue to be present on a large scale in capitalism.


 1. E Mandel Late capitalism London 1978.

 2. See G Holmes British politics in the age of Anne London 1987, chapter 5.

 3. The language of the ‘landed interest’ and the ‘moneyed interest’ was still being used to describe British politics as late as the 1850s: eg, K Marx, ‘The elections in England - Tories and Whigs’ New York Tribune August 6 1852: www.marxists.org/archive/marx/works/1852/08/06.htm. Compare also the arguments of Virginia conservatives in the early 19th century discussed in R Bruce The rhetoric of conservatism San Marino 1982, pp76-79.

 4. B Anderson Imagined communities London 1991. In arguing against this view I should say that I am arguing against the view that is ‘orthodox’ among CPGB comrades: see, for example, Jack Conrad, ‘Nationalist myths are not Marxism’ Weekly Worker July 27 2006; Bob Davies, ‘Fact and fictions’ Weekly Worker September 14 2006.

 5. Of course, modern historians are apt to look for such evidence by virtue of nationalist predispositions, as Anderson argues. But the point is that if they look before the fall of the Roman empire, they will not find evidence of nationalism: while in the medieval period the evidence for national self-identification of states is abundant, including among lower social strata. Cf also S Reynolds Kingdoms and communities in western Europe 900-1300 Oxford 1997, chapter 8.

 6. DI Kertzer Unholy war: the Vatican’s role in the rise of modern anti-semitism London 2003; and see also my review of Kertzer: ‘The politics of purity’ Weekly Worker July 22 2004.

 7. Discussed in D Statt, ‘The City of London and the controversy over immigration’ (1990) 33 Historical Journal 45-61. Though Statt minimises the party-political aspect of the controversy by the usual means of ‘revisionist’ historians of 17th century England - ie, to attribute it to local motivations - in the sources he cites and the periods of anti-immigrant agitation its party character is apparent.

 8. According to the Oxford English Dictionary, ‘hooligan’ appeared in the 1890s and is generally accepted to be derived from the Irish name, Houlihan.

 9. On the Know-Nothings, Wikipedia has a convenient short account and bibliography: http://en.wikipedia.org/wiki/Know-Nothing_movement.

 10. P Linebaugh, M Rediker The many-headed Hydra: sailors, slaves, commoners and the hidden history of the revolutionary Atlantic London 2000.

 11. PJ Corfield argues against an earlier view (also held by 18th century authors, some of whom he cites) that urban guild regulation led to capital flight (PJ Corfield The impact of English towns 1700-1800 Oxford 1982, pp 87-93). Notwithstanding Corfield’s arguments, it is clear that capital not tied to a particular locality (eg, mining) could move its operations, and that some capitals did move to escape regulation.

 12. In the preface to the first German edition Marx says: “The physicist either observes physical phenomena where they occur in their most typical form and most free from disturbing influence or, wherever possible, he makes experiments under conditions that assure the occurrence of the phenomenon in its normality. In this work I have to examine the capitalist mode of production, and the conditions of production and exchange corresponding to that mode. Up to the present time, their classic ground is England. That is the reason why England is used as the chief illustration in the development of my theoretical ideas ... Intrinsically, it is not a question of the higher or lower degree of development of the social antagonisms that result from the natural laws of capitalist production. It is a question of these laws themselves, of these tendencies working with iron necessity towards inevitable results” (www.marxists.org.uk/archive/marx/works/1867-c1/p1.htm). The point that a ‘pure capitalist’ economy is postulated except insofar as the analysis is explicitly historical is made by most later expositors of Capital: eg, D Harvey The limits to capital London 1999, chapter 1.

 13. For the first point, see, for example, The German ideology. For the second, see the preface to Capital Vol 1: “Apart from higher motives, therefore, their own most important interests dictate to the classes that are for the nonce the ruling ones, the removal of all legally removable hindrances to the free development of the working class.”

 14. J Banaji, ‘The fictions of free labour: contract, coercion and so-called unfree labour’, (2003) 11 Historical Materialism pp69-95.

 15. John Harrison in Marxist economics for socialists (London 1978, chapter 2) argued that this was the character of the treatment of simple commodity production in the early part of Capital. But since the rest of Capital is built on this treatment (subject to the historical account of the development of the proletariat and associated matters in Capital Vol 1),it would logically follow that the whole text is to be read in this way; and this would certainly avoid some problems of interpretation.

 16. D Harvey Limits to capital introduction and passim; M Lebowitz, Beyond capital: Marx’s political economy of the working class London 2003.

 17. M Macnair, ‘Law and state as holes in Marxist theory’ (2006) 34 Critique 211-236.

 18. G Carchedi Frontiers of political economy London 1991, chapter 1 and appendix; Harvey’s Limits to capital displays the same method.

 19. GEM De Ste Croix The class struggle in the ancient Greek world New York 1981, chapter 2.