Chris Harman


Change at the first millennium

(Spring 1994)

From International Socialism 2:62, Spring 1994.
Copyright © International Socialism.
Copied with thanks from the International Socialist Archive at
Marked up by Einde O’Callaghan for the Marxists’ Internet Archive.

The Transformation of the Year One Thousand
Guy Bois
Manchester University Press, 1992 £9.95

There has been a controversy for 50 years and more among Marxists over the transition from feudalism to capitalism. But there has been much less debate over the earlier transition from the slave societies of antiquity to capitalism. Yet in some ways this is a more challenging topic for Marxism than the later transition.

In his classic formulation in the Preface to the Contribution to the Critique of Political Economy Marx asserts that one mode of production replaces another only when it proves itself more capable of developing the forces of production. But feudalism has usually been seen as more backward than antiquity. In its classic form it lacked the cities, the roads, the harbours, the wide literate strata of antiquity, earning it the epithet ‘the dark ages’. What is more, it has usually been seen as a form of society in which the forces of production continued to stagnate. Thus in the most influential recent account of the transition from feudalism to capitalism, that of the American Marxist Robert Brenner, the forces of production play no role at all, for he claims that they did not develop during the feudal period which was a time of ‘economic stagnation and involution’ since feudalism ‘imposed upon the members of the major social classes strategies for reproducing themselves which, when applied on an economy wide basis, were incompatible with the requirements of growth’. [1]

Such a view sees feudalism simply as the barbarous residue which remained after the Roman Empire fell apart, a dead period between one burst of civilisation and another, a thousand years later.

Yet while many Marxists have talked in these terms, a variety of non-Marxist historians have been painting a very different view of feudalism. Duby, Le Goff, Thrupp, White, Crombie and Gimpel have all shown that there was a great burst of economic development at the height of feudalism, between the 10th and the 14th centuries. A few years back the French Marxist Guy Bois integrated some of this material with his own research in parts of Normandy to explain the great crisis which beset the feudal system in the 14th century and the beginnings of the incursion into the power of the feudal lords of a new stratum of tradesmen, merchants and officials associated with ‘bourgeois property’. [2] Now research on the village of Lournand in central France has led him, in this book, to challenge much of the accepted view about the transition from antiquity to feudalism.

The conventional view goes something like this: from the 4th century onwards the Roman Empire was increasingly beset by internal crises. The slave mode of production discouraged innovation and any rapid rise in productivity at a time when the costs of maintaining the political superstructure of the empire were growing ever greater. This opened the empire up to attacks from the various peoples around its borders – the German tribes, the Goths and Visigoths, the Huns and so forth – which further increased the cost of maintaining the superstructure. Eventually invaders not only cut off outlying regions, but penetrated to the heart of the empire, sacking Rome and splitting the empire in two. The old trade routes were broken apart, the towns declined and the individual members of the old imperial aristocracy were forced increasingly to fend for themselves. They did so by working their estates, not with slaves, but with a new class of dependent peasants – some former slaves, some former free peasants forced into reliance on them by fear of the marauding invaders. And many of the chiefs of the invading armies emulated them, settling their tribal followers as the dependent peasants (serfs) on the lands they had conquered. In this way, a new way of organising production took root in the 6th and 7th centuries and persisted in most of Europe for more than 1,000 years, despite the attempt by Charlemagne to recreate the empire at the beginning of the 9th century and the enormous crises in the 14th and 17th centuries.

Guy Bois claims the empirical facts of 10th century Lournand contradict this view. His studies found that far from slavery dying out in the 5th and 6th century it still remained significant four centuries later. The big monastic estates were worked by dependent peasants – serfs. But these accounted for a relatively small proportion of the land and of output. Most of the land was in small to medium sized holdings farmed either by free peasants labouring on their own behalf or by slave families labouring for their owners. And among the toiling classes, the distinction between free peasant and slave was the vital one – since it was the slaves that provided most of the surplus for the exploiting class.

It was because the base of society still rested so much on slavery that the Frankish rulers of the Merovingian and Carolingian dynasties could attempt to re-establish the Roman Empire. And so, Bois argues, it would be better to describe society until the late 10th century as the last stage of antiquity rather than the first stage of feudalism.

But in this society the elements which were to lead to the full transition to feudalism were already growing up. Those who controlled the surplus produced by slaves had a great advantage over the free peasantry. They were able to use the surplus both to invest in more advanced forms of production on their own land and to establish new means of exploiting the peasantry, like the building of water mills. Bois insists they were not, as so many previous Marxist accounts have claimed, simply involved in ‘extra-economic exploitation’: ‘the master was by no means only a warrior or a priest devoting the majority of his time to activities of a public character, he was also an active economic agent, and ... the small domain was the scene of important technological advance’. [3] This association with the advancing sector of production put them in a strategic position once the attempt to establish a new empire fell apart. The physical resources at the disposal of the slave owners enabled them to turn themselves into a new class of ‘lords’ with effective political power in each particular locality – providing they, in turn, warded off potential external threats by putting themselves under the protection of more powerful lords. The peasantry was forced to look to them for security in an increasingly unstable world: they alone controlled a surplus that could see the individual peasant family through a difficult period if its crops failed. They alone could deploy the military force capable of offering protection against the danger of armed bands pillaging the family holding. So, Bois shows, the last decades of the 10th century saw a widespread tendency for formerly ‘free’ peasants to accept a position of dependency on – and exploitation by – the local lords.

Bois argues that serfdom was established in this way, not in the 5th or 6th, but at the end of the 10th century. He further argues this was a veritable revolution, with a more dynamic mode of production, feudalism, replacing a less dynamic one, antiquity. And finally, he claims the revolution in production was accompanied by the establishment of a new political framework and a whole new set of ideologies to justify it.

Much of Bois’ case is very powerful. He brings out the enormous economic advance which took place from the 10th century onwards – the adoption of new and more productive agricultural techniques, the building of literally thousands of water mills, the revival of trade and the towns. He shows the rise of feudalism to have produced a veritable advance in the forces of production for two or three centuries so that Europe, which had lagged behind most of the rest of the world, began to draw ahead of it.

Nevertheless, his picture does not quite fit together. On his reckoning it was the slave holdings which produced advance before the year 1000. Yet the outcome was feudalism, not some rebirth of antiquity. The 11th century was the century of serfdom, not slavery.

Here I think he makes the mistake of confusing a legal category with a class category. In antiquity a slave was someone whose position was not defined by his or her relation to the productive process, but by their relation to their owner. And so although most slaves may have been unpaid labourers on the country estates of the aristocracy or in the mines, slaves could also be estate supervisors who shared some of the privileges of their owners, businessmen allowed to work on their own account in return for giving a cut to their owners or even powerfully placed figures in the imperial bureaucracy. M.I. Finley has gone so far as to describe the slaves in the towns of the late empire as a ‘parasitical element’ – ’domestics and administrators’ living, like their masters, off the surplus created by slaves in the countryside and ‘free’ labour in the cities. [4] In other words, the legal category ‘slave’ could include within it people belonging not only to the main exploited class but also to oppressed but relatively privileged intermediate classes.

Bois’ class of ‘slaves’ in 10th century central France were families who worked farms for other people, but without immediate supervision. They were certainly exploited by their owners, who grabbed a large share of their output. But the way production was organised, and therefore the way in which they were exploited, was not the same as that during the classical period of antiquity. They were, if you like, ‘peasant slaves’ rather than ‘slave gang slaves’, and this had its effect on the dynamics of both production and exploitation.

A peasant household has an interest in advancing the forces of production, while a slave gang does not, for it can hope that some of the rise in productivity will accrue to itself. And so it will show more initiative and more care than a slave gang ever would. That is why, historically, small to medium size family peasant farms have been more productive than large gang labour farms, particularly when what is involved is mixed farming rather than mainly cereal production. [5]

It is this which explains why the 10th century could see farming based ‘peasant slaves’, like that based on the dependent non-slaves of the large monastic holdings, succeed where the old Roman estates had failed. A transformation in the mode of production had already begun to take place.

The transformation was not instantaneous. It proceeded at different speeds in different places and some social institutions were affected by it later than others. Elements of the old mode of production existed alongside the new. And what Marxists often call the ‘political and ideological superstructures’ lagged behind the economic ‘base’, although they too had begun to change centuries before. ‘The transformation of the year one thousand’ was the final adjustment of the superstructure to the base, the final setting up of political and ideological ‘systems that were in harmony with the new forms of production and exploitation’. But this is not the same as saying, as Bois does, that this was when the transition from antiquity to feudalism took place.

It is a mistake to see modes of production, as the Althusserians used to, as self contained ‘structures’, cut off completely from each other. Change is always taking place within each mode of production. Developing forces of production come up against the limits imposed by particular relations of production until society is thrown into immense crises, which are only resolved when new forms of production and exploitation grow up alongside the old. There is necessarily a longer or shorter gestation period, during which the new mode of production grows at the expense of the old without replacing it entirely. The coexistence between the two is never entirely peaceful and there are necessarily climactic moments when the representatives of the new mode clash physically with those trying to preserve the old. But there is rarely just one point of transition. So it was in the long shift from feudalism to capitalism. So too, Bois’ empirical evidence suggests, it was with the transition from antiquity to feudalism.

However, the importance with Bois’ book does not lie in the argument over labels. It lies in his success of showing how feudalism was a more dynamic economic system than the one it replaced, how it did advance the forces of production, how it did lead to increased output in the countryside and to a regrowth of the towns and trade after half a millennium of decline – and how it laid the basis for the development of an even newer mode of production, capitalism, that offered the possibility of further advance once feudalism itself exhausted its possibilities and entered into crises in the 14th and 17th centuries.

Bois’ book builds on the work of non-Marxist historians in such a way as to prove the value of the Marxist approach. It is a very welcome addition to the materialist study of the past.


1. The Agrarian Roots of European Capitalism, T. Ashton (ed.), The Brenner Debate (Cambridge 1993), p.213.

2. G. Bois, The Crisis of Feudalism (Cambridge 1984), p.153.

3. G. Bois, The Transformation of the Year One Thousand (Manchester 1992), p.124.

4. M.I. Finley, Ancient Slavery and Modern Ideology (London 1992), p.148.

5. For an elaboration of this argument, see T. Cliff, Marxism and the Collectivisation of Agriculture, International Socialism (Winter 1964-65, old series). This does not mean that the use of large scale gang labour agriculture is necessarily irrational for the ruling class. It can give it a much greater control over the immediate production process than peasant labour, and therefore enable them to raise the level of exploitation enormously – as happened under Stalinism in Russia.

Last updated on 4.3.2012