3
INTRODUCTION
Why should readers of Historical Materialism consider reading a book by a specialist in early Italian history, containing 831 pages of text and dealing with Europe and the Mediterranean world between the fifth and ninth centuries AD? Framing the Early Middle Ages was awarded the Deutscher Memorial Prize for 2006, which suggests that it may interest a wider audience than the fellow medievalists Chris Wickham addresses in his introduction. There, “you the reader” is assumed to belong to a group of “experts” who “often . . . know far more than I about a given set of materials.”1 In the case of this reviewer, Wickham need have no such concerns, since my area of expertise lies in a historical period that opens nearly nine hundred years after his closes, and with a country (Scotland) that he specifically excludes from discussion.2 My purpose here will therefore not be to dispute with Wickham over, for example, his explanation for why there are greater similarities between Syro-Palestinian and Italian ceramics than between either of these and ceramics of Egyptian origin.3 Instead, I approach the book in the same way as most other nonspecialist readers of this journal: as a Marxist interested in what a fellow Marxist has to say about a crucial but deeply obscure turning point in human history, and in what implications his work has for Marxist theory. As we shall see, his work is full of interest in both respects.
WICKHAM AND THE DEBATE ON THE FIRST TRANSITION
There have been recurrent debates over the transition from feudalism to capitalism. There have even been extended discussions over the transition from capitalism to socialism—an event that, as we are all too painfully aware, has not yet successfully taken place. As a result we have some idea of the relationship between economic transition and social revolution in both cases. By contrast, the emergence of feudalism has been relatively neglected by all the major intellectual traditions that seek to explain long-term socioeconomic development, including Marxism.4 In the first volume of Michael Mann’s The Sources of Social Power, for example, his conclusions concerning the decline and fall of the Roman Empire are followed by an extended discussion of Christianity and rival religions, before beginning a survey of twelfth-century Europe, considered solely insofar as it provides the setting for capitalist development.5 Mann can at least argue that, as a non-Marxist, he does not find the concept of feudalism useful, but even a work as firmly situated within the classical Marxist tradition as Chris Harman’s A People’s History of the World deals with the subject in a summary fashion that is noticeably different from the later treatment of the transition to capitalism.6
In one sense this is unsurprising, since the Marxist classics are relatively silent on the subject. The most famous discussion, by Engels in The Origins of the Family, Private Property and the State (1884), summarizes over four hundred years of history in around twelve pages. For Engels, the pressures caused by imperial taxation had already set in motion the economic crisis of the empire, as a result of which the declining profitability of slavery, in both the great estates and artisanal workshops, led to landlords settling former slaves as hereditary tenants.7 There is nothing uniquely Marxist about this explanation, except perhaps the stress Engels places on the Germanic invasions in embedding the “barbarian” gentile constitution that supposedly gave peasant society a community structure and an institutional means of emancipation from servitude. Important essays by Max Weber (1896) and Marc Bloch (written between the world wars, but published posthumously in 1947) also privileged the changing position of the slaves, although with different emphases. For Weber, the decisive point was when the territorial limits of the empire were reached, leading to difficulties in acquiring new slaves with which to replace the existing workforce, since actual reproduction—breeding slaves rather than capturing or buying them—would have required massive levels of investment that landlords were unwilling to make.8 Bloch is in some ways closer to Engels, but adds an additional component in claiming that the new class of serfs arose not only from a loosening of the conditions of absolute servitude hitherto imposed on the slaves, but from a tightening of the relative liberty previously enjoyed by free peasants.9 None of these contributions referred to revolution as such. Those that did tended to be non-Marxist and focused on a much later period. Richard Southern famously wrote of the period between 970 and 1215: “The slow emergence of a knightly aristocracy which set the social tone of Europe for hundreds of years contains no dramatic events or clearly decisive moments such as those which have marked the course of the other great social revolutions.” It was the almost imperceptible quality of the transformation that led him to describe it as the “silent revolution of these centuries.”10
Serious Marxist discussion of the subject took place over a relatively short period toward the end of the last century, culminating in a series of exchanges in Past and Present across 1996–97. Since Wickham made several important contributions to that discussion, it may be worth recapitulating the key positions, including his own, to contextualize his latest book. Two works, Perry Anderson’s Passages from Antiquity to Feudalism (1974) and Guy Bois’s The Transformation of the Year One Thousand (1989), conveniently set out the main opposing explanations and timescales for the emergence of feudalism.
For Anderson, there is a period of socioeconomic transition that begins with the barbarian settlement within the Roman Empire in the West, but is concluded only several centuries after its collapse: “The catastrophic collision of two mutually dissolving anterior modes of production—primitive and ancient—eventually produced the feudal order which spread throughout medieval Europe.”11 Given that the preexisting modes were embedded in social formations occupying geographically separate areas of Europe, feudalism was initially marked by spatial unevenness:
In effect, the core region of European feudalism was that in which a “balanced synthesis” of Roman and Germanic elements occurred: essentially, Northern France and zones contiguous to it, the homeland of the Carolingian Empire. To the South of this area, in Provence, Italy or Spain, the dissolution and recombination of barbarian and ancient modes of production occurred under the dominant legacy of Antiquity. To the North and East of it, in Germany, Scandinavia and England, where Roman rule had never reached or had taken only shallow root, there was conversely a slow transition towards feudalism, under the indigenous dominance of the barbarian heritage.12
The prolonged period during which fusion took place meant that the preexisting modes were not transformed immediately, but for Anderson there is no suggestion that they continued to exist anywhere as dominant after the sixth century, although examples could of course be found of free peasant communities on the one hand, and of slaves on the other.
Anderson could draw on some passing suggestions by Marx himself in the Grundrisse, where the notion of “synthesis” was first deployed, as his authority.13 The main support for this position had, however, come from Russian and Eastern European academics such as Elena Mikhailovna Shtaerman, although it was by no means universally accepted by all their colleagues.14 Anderson refuses to contemplate the existence of feudalism prior to the fall of the Roman Empire, and he is, of course, scarcely alone in taking this position. As Moses Finley once wrote: “On any account chattel slavery ceased to be dominant even in Italy by the fourth or fifth century whereas it is improper to speak of feudalism before the time of Charlemagne, leaving a ‘transition’ lasting three or four hundred years.”15 Why Finley finds it improper is not clear, but the same position was also taken by his great opponent, Geoffrey de Ste Croix. The latter was prepared to acknowledge the existence of serfdom as one of the three forms of unfree labor in the ancient world (along with chattel slavery and debt bondage), but he opposed the idea that this demonstrated the existence of feudal relations of production, describing this as a “groundless connection.” Again, the grounds of his objection are not entirely clear, other than this would involve the discovery of feudalism across the Greek world prior to the Hellenistic period, although he recognizes that there are “closely related (though not identical) forms in Graeco-Roman antiquity and in the middle ages.”16 Ste Croix’s unwillingness to recognize the existence of feudalism may signal his adherence to a Social-Democratic or Stalinist notion of successive stages of social development. In the case of Anderson the reason is different. He is committed to the view that capitalism emerged as an indigenous system only in Western Europe. Although he sees feudalism as having a slightly wider territorial extent (it also includes Japan), the conditions for the emergence of capitalism are only present in Western Europe because the genesis of feudalism there took a peculiarly “synthetic” form, allowing what Anderson sees as the distinctive element—the cultural and juridical heritage of classical antiquity—to be transmitted into the new system.17 As this suggests, Anderson’s definition of feudalism is based on its superstructural characteristics—a necessity, in his view, since all precapitalist class societies other than slavery are based on the exploitation of a peasantry by landlords.18 Feudalism therefore cannot have existed during the lifetime of the Roman Empire, as these characteristics were absent. It is of course quite possible to explain the priority of capitalism in Western European history without recourse to idealist speculations about the heritage of classical antiquity. The key point in the context of this discussion, however, is Anderson’s chronology: the end of the empire in the West during the fifth century sets in train a process that led to the emergence of feudalism.
Bois would agree that feudalism did not predate the end of the Roman Empire, but in every other respect his account is the opposite of Anderson’s. Far from slavery beginning a long transformation virtually from the moment the social organization of the barbarian tribes began to interpenetrate with that of the Romans, Bois claims that it remained the dominant mode until the tenth century, notably in the areas where Charlemagne had attempted to preserve the political form of the Western Empire. Accordingly, Bois emphasizes not the process of transition but a moment of revolution around 1000, which he describes as a “European phenomenon.”19 Drawing on events in the village of Lournard in Cluny to support his thesis, he describes a situation of “dual power” between the monks of the monastery of Cluny, bearers of the new feudal order, and the existing masters, the Carolingian defenders of slavery:
The driving force behind this movement was a faction within the aristocracy, or, to be more precise, within the high aristocracy in its monastic dimension. This was done almost despite itself. The sole concern of the first Cluniacs was to assure their independence with regard to the lay powers and to reform monasticism. However, this concern led them to develop close ties with the peasantry. There was thus an identity of interest (the peasantry feeling themselves threatened by the local grandees) and even an ideological rapprochement, to the extent that monastic spirituality coincided with the moral needs of the peasantry. From this moment on the old order was threatened. As often happens in such cases, the signal for hostilities was given by the champions of the past, by that local aristocracy, warrior and slave-owning, which formed the social base of the Carolingian system, but which saw its position being eroded. By unleashing violence, it plunged society into anarchy, thus compelling the monks to assume responsibilities in the social sphere and define a new order: the first draft of feudal society.20
Bois was the last in series of French historians, beginning with Georges Duby, who had introduced the notion of a feudal revolution by way of an analogy with the bourgeois revolution.21 (Although by the time Bois’s book appeared in France, Duby had rejected both the term and the notion.22) There are two main objections to Bois’s account of the process. The first is empirical. His material is too narrowly based on one small area of France and cannot be generalized across the whole of Europe: slaves existed in estates east of the Rhine where Roman influence was minimal, and labor services were innovations in Italy during the eighth and ninth centuries, not a legacy from antiquity.23 The second is theoretical. His definition of a slave is too fixated on the legal category and not enough on the actual relationship of the direct producers so categorized to the means of production. In other words, many of these slaves were in fact nearer to the free peasants—notably in their interest in raising output—than the slaves who labored in the fields, mines, or households of antiquity.24 Nevertheless, a diluted version of the “feudal revolution” thesis has now been mainstreamed, shorn of the Marxism framework to which Bois at least had subscribed, to the extent that the term “revolution” can be used to describe changes around the first millennium without specifying the transition to any particular mode of production, as in Robert Moore’s The First European Revolution, c. 970–1215 (2000).25
Where did Wickham stand in this debate? In two important articles, “The Other Transition” (1984) and “The Uniqueness of the East” (1985), he established his own distinct position. Against Anderson, he claimed that feudalism already existed in 300 AD, so that it could not have been the result of a synthesis of German barbarism and Roman slavery. Indeed, as Wickham points out, “in so far as the German invaders had such things as a landed aristocracy, these largely resulted from Roman influence.”26 Against the French tradition of “feudal revolution,” which was shortly to culminate in Bois’s work, he claimed that this preexisting feudal mode of production had become the dominant mode by 700 AD, by which time “the balance shifted” from the hitherto dominant ancient mode.27 The transition from slave to serf, through the mechanism of labor service, was, he claimed, “marginal” to the transition—indeed he sees the peasantry as major beneficiaries of the entire process. He does note that increased surplus extraction from peasants occurred during the ninth and especially tenth centuries, but this is characteristic of the end of the first phase of feudal development, not the transition to feudalism itself. Whatever happened around 1000 AD could scarcely have been a revolution, then, since the fundamental change had already been completed 300 years earlier.
Wickham began by identifying a contradiction within the Roman ruling class, which was heightened from the beginning of the fifth century. The acquisition of land made individual members liable for tax, which they tried with increasing success to evade, this reducing the resources available to them collectively as state managers. The main recipient of state funding was the army, engaged in increasingly futile attempts to repel the Germanic invasions—attempts whose lack of success provided an even greater incentive to tax evasion. Meanwhile the German invaders began to appear an attractive alternative to supporting a declining but acquisitive state apparatus. As Wickham stresses, however, “tax-evading aristocrats,” although important, were not the only social actors involved in achieving the transition.28 Peasants played a far greater role, but (despite several important risings from early in the fifth century) not principally as participants in open class struggle. Instead, they hastened the internal disintegration of the empire by placing themselves under the protection of landowners, effectively renouncing their independence on the assumption that not only would their new status as tenants not carry tax liabilities, but their new lords would be capable of avoiding such responsibilities themselves, and consequently would not pass them on. In effect, both the landowners and the peasants had reasons to choose what would later become known as feudal social relations.
These pressures also applied in the East, but the outcomes were different, mainly because in the West the crisis of taxation coincided with an additional factor: the Germanic invasions. The triumph of the barbarians did not immediately lead to total transformation: “The new Germanic states were not yet feudal.”29 Taxation continued, but without the need for a centralized army—since the new states raised armies from their own landowners and retainers—the main purpose for raising taxation no longer existed. Taxation became increasingly fragmented: inessential for supporting monarchs, whose wealth derived from their own estates, it became principally used for securing support through gifts or bribes. Previously, members of the ruling class had sought to acquire land in order to gain access to control of the state apparatus, but now it became an end in itself: “Private landowning was henceforth no longer the means to the obtaining of power; it was itself power.”30 The scene was by no means uniform: some areas, such as the British Isles, reverted to preclass agrarian societies; in others, such as the German states, preclass societies coexisted with feudal relations in a subordinate position; but in terms of the state, “all were feudal, for they were based on the politics and economics of landowning, expressed in different ways.”31
Why then the difference with the East? Wickham originally argued that, in addition to the slave mode, the Roman Empire had at different times also involved the feudal and the tributary modes, based respectively on rent and tax. He originally argued that these two modes emerged as dominant from the fifth century, effectively maintaining different aspects of the later Roman Empire: feudalism in the West and the tributary mode in Byzantium. And while Wickham was clear that the tributary mode was not simply a relabeling of the “Asiatic” mode, which he rightly dismissed, he also emphasized that it did exist in other regions, above all in the Chinese Empire.32 The distinction between feudal and tributary modes drew far more response than his account of the transition. In particular, Halil Berktay and John Haldon pointed out that, in terms of the central exploitative relationship with the peasantry, there was no difference between these; the difference lay in the extent and nature of state power, but Marxists do not distinguish between modes on superstructural grounds—that would be to fall into precisely the error for which all contributors to the debate criticized Anderson.33 Wickham accepted this criticism, as he pointed out upon the re-publication of his early essays in Land and Power:
The basic economic division inside class societies thus becomes simply that between societies based on taking surpluses from peasants (or, for that matter, household-based artisans) and those based on withholding surplus from wage laborers. . . . It does not mean that the Chinese or Roman empires, the Frankish kingdoms, and the feudal world of the eleventh century were exactly the same, for an essential structural difference remains between the first two, tax-raising state systems (with aristocracies subject to them), and the second two, polities dominated by aristocratic rent-taking and Marc Bloch’s politics of land.34
Like the positions to which he was opposed, Wickham could find support for his alternatives in respect of both chronology and modes of production in Marx’s own writings, specifically in the Grundrisse, that most ambiguous of his major works. As Hobsbawm wrote in an important early commentary: “Feudalism appears to be an alternative evolution out of primitive communalism, under conditions in which no cities develop, because the density of population over a large region is low.”35 Similarly, although Wickham derived his use of the tributary mode from Amin, the concept, if not the actual term, can also be found in the pages of Marx’s notebooks: “In the case of the slave relationship, the serf relationship, and the relationship of tribute (where the primitive community is under consideration), it is the slave-owner, the feudal lord or the state receiving tribute that is the owner of the product and therefore its seller.”36 Wickham therefore had at least as much reason to claim a relationship to the Marxist classics as his opponents.
THEMES, THEORIES, ABSENCES
After a professional career as a historian principally of the Tuscan region of Italy, Wickham has now returned to the subject of the post-Roman world as a whole. During the debate on the feudal revolution, Wickham commented on Bisson’s dating of that process to between 850 and 1100, noting that “250 years is a long time for a revolution,” and he argued that it was preferable to see the period as one of consolidation or formalization of a feudal system that had already been established: “Like the Industrial Revolution, or the varying moments of middle-class political assertion in Europe that began with the French Revolution, this major shift could be fast or slow, (relatively) peaceful or sharply violent, and the variations themselves shed light on the structural differences between one region and another.”37 In effect, Framing the Early Middle Ages is a massive depiction of the prior process of feudal emergence, empirically substantiating the picture he sketched out over twenty years ago, taking into account his changed position on the modes of production debate. This is no mere coda to the earlier debate on the transition to feudalism, but a re-engagement with the issues that has greater significance than any of the original contributions.
Starting from the dissolution of the political unity and relative economic homogeneity of the Roman Empire, he traces how the constituent regions diverged from each other as successor societies adopted particular aspects of the imperial experience. As we should expect from his previous work, Wickham is particularly interested in the nature of the fiscal regime, which he sees as simplifying and in some places disappearing altogether, with obvious implications for whether society was dominated by feudal politics of land or tributary extraction by the state. Central to the book is the fate of the two great classes: the peasantry and the aristocracy, sometimes antagonistic, sometimes cooperative. The former achieved greater autonomy and, in some areas, freedom from exploitation altogether; indeed, in some respects Wickham describes a golden age for peasants, compared to both the oppression from which they had been released and the oppression to which they would eventually be subjected. A condition of peasant freedom was the weakening power of the aristocracy, the character of which also changed, becoming more narrowly focused on its military role and abandoning the literary culture that had been just as important to the Romans. This too was only a temporary condition, before the reassertion of their dominance by the end of the period. Throughout it, however, Wickham is clear that the aristocrats are in most respects the key social actors. Large-scale production at the regional level, let alone interregional long-distance exchange, was structured by “elite consumption,” in the absence of mass peasant demand: the wealthier the aristocracy, the greater the demand. But that wealth was in turn determined by two of the elements Wickham sees as constitutive of the transitional economy: the extent, reach, and effectiveness of the tax system, and the level of exploitation of the peasantry by the lords. To these must be added two more elements, one contingent and the other deeply structural: the retarding effect of war and the extent to which a region continued to be integrated into the post-imperial Mediterranean world system.
The scope of this survey, and the command with which Wickham presents it to the reader, means that Framing will inevitably and rightly be compared to the other great Deutscher Prize–winning work of premodern history, Ste Croix’s The Class Struggle in the Ancient Greek World. That work famously begins with a lengthy consideration of the theoretical concepts that Ste Croix then uses to structure his argument.38 Wickham does not adopt this strategy. Although there is of course a theoretical apparatus at work behind the scenes, Wickham draws back the curtain to reveal it only in tantalizingly short passages, usually where some sort of definition is required. Typically, this was also the approach taken by the earlier generation of British Marxist historians like Rodney Hilton—to pick one specifically identified as an influence by Wickham—at least in their substantive works.39 This approach has much to recommend it, especially when compared to the endless theoretical preliminaries that were typical of the work of, for example, Barry Hindess and Paul Hirst, at the time when Wickham began to publish. But here the very richness of empirical detail means that underlying theoretical positions have sometimes to be inferred. This lack of explicit discussion creates a barrier for the reader, not to checking Wickham’s conformity to some Marxist orthodoxy or other, but to assessing how his assumptions have shaped his use of the material. On one methodological issue, however, Wickham is explicit: his rejection of teleological explanations of capitalist development, particularly, “the metanarrative of medieval economic history which seeks to explain the secular economic triumph of north-west Europe.” Instead, he emphasizes “the variegated patterns of social development” and argues that “social change is overwhelmingly the result of internal factors, not external influences.”40 By external influences Wickham seems to mean the view (which he identifies with Henri Pirenne, although it can be traced back to Adam Smith) that feudalism arose as a result of outside pressures.41 Wickham argues that the roots of feudalism have rather to be discerned in the internal development of the regions he discusses.
These regions extend from the Irish edge of Western Europe to the present-day Middle East. The geographical compass of the book is set by the subject, the after-world of the united Roman Empire, and Wickham does full justice to the range of different societies this involves, focusing as much on Egypt and Syria as on Denmark and England. The book is not completely exhaustive: Scotland and (for the most part) Saxony are excluded without greatly affecting his argument, since similar societies to these are included, although the exclusion of the Balkans perhaps passes up an opportunity to compare tributary formations in Europe with those in North Africa and Asia. But these are minor issues. Far more important than absolute comprehensiveness is the fact that the parallels and similarities he draws to our attention help undermine another form of teleology. In this case, it is not the Western origins of capitalism so much as the broader narrative that distinguishes Europe, or sometimes simply “the West,” from the rest of the world on the basis of the “Judeo-Christian” heritage, or similar inventions, which supposedly date back to this period.
For example, the societies that showed the most signs of agricultural intensification during the period lay not only at opposite extremes of the tax/rent continuum but also at opposite extremes of the territorial limits of the empire: Francia on the one hand, and Egypt and the Levant on the other. These two regions, respectively involving “a rich aristocracy in the Carolingian world, a powerful state in that of the Umayyads and the Abbasids” were “the regions with the most potential for exchange, and thus the most stimulus for agricultural intensification.”42 But the same types of parallel are also apparent in less complex forms of society, where the Roman state collapsed: Mauritania, in the Berber lands of North Africa, had a pattern of “social development” that Wickham claims “resembles Britain,” although “its closest British analogues would be with more traditional highland Wales” rather than lowland England. Mauritania retained its own political traditions under the empire, while Britain wholeheartedly embraced those of Rome; yet the results were similar, not least in terms of social simplification and economic retrogression.43 This would suggest that, whatever the origins of the differences between West and East, they are clearly not intrinsic to the societies involved and have to be traced instead in subsequent historical developments.
Wickham’s account would also suggest that religious differences between Christianity and Islam are less important in determining the regional character than the material conditions upon which he focuses, but this has to be inferred since ideological issues are nowhere discussed, except briefly in relation to aristocratic hegemony. To be fair, Wickham makes clear from the outset that, because of the already great length of the book, his focus will be on the social and the economic. Take as a comparison Fernand Braudel’sThe Mediterranean and the Mediterranean World in the Age of Phillip II, in many ways a model of this kind of large-scale history. It begins with a part on the physical geography of the region, where change is “almost imperceptible,” and ends with one of the political events of the fifty-year period beginning in 1550, where change occurs in “brief, rapid, nervous fluctuations.” Between is a part dealing with “social history,” where change is “slow” but nevertheless has “perceptible rhythms.”44 Wickham’s book deals with themes similar to those in this middle-range part of Braudel’s book (“Collective Destinies and General Trends”).45 Political developments are only discussed to provide essential background to the regions under discussion, and culture is excluded completely—although not cultural artifacts; potsherds appear with great regularity, but only as traces of economic activity.46 But “politics” here has to be understood primarily as what we would now call “geopolitics”—or more simply, war—since the state, the political institution par excellence, is certainly of paramount importance to Wickham, not least because its form was “the arena that saw most change.”47 Indeed, following a brief survey of geopolitical developments, he privileges the state as the first area of discussion before analyzing the position of the two main classes, aristocrats and peasants. This is not because he sees the state as the “prime mover” in social change, “with the form of the state somehow determining every other aspect of society and the economy, in a statist version of a very traditional Marxist analysis.”48 Wickham is not proposing to substitute a superstructural determinism for one that privileges the base, but the structural focus of the book does mean that the actual moments of change—above all the moments of peasant expropriation—tend to be subsumed within discussions the main focus of which is on other aspects of the period.
STRUCTURE: SOCIAL CLASSES, MODES OF PRODUCTION, AND STATES
Wickham argues that neither slavery nor (less controversially) wage labor was of any great significance to the economy of the early Middle Ages:
Throughout our period the slave mode was only a minor survival, everywhere marginal to the basic economic structure, of the landlord peasant relationship (where there were landlords at all). . . . The marginality of the slave mode in our period is matched by the relative unimportance of wage labor, at least outside Egypt; essentially, throughout our period, agriculture on estates was above all performed by peasant, tenant cultivators.49
Slavery and unfreedom more generally had a different significance depending on the extent of peasant autonomy within the locality. Where a lord had superiority over an entire area, as in the Ile de France, it could be a status distinction, with the possibility of movement between the free and unfree. Ironically, it was where peasants were most free of lords, as in England before 700 AD, that the distinction had its greatest significance, indicating a potentially exploitative relationship within the household economy.50 Only “potentially,” Wickham argues, because the position of the unfree within the peasant household would only involve “class exploitation” in circumstances where “the members of the free family all stopped working, and simply lived of the labor of the unfree.”51 I am not sure whether this argument is sustainable. Exploitation still takes place in situations where small commodity producers supplement the labor of themselves and their families with wage labor, since wage laborers produce a surplus over and above what they receive. Why would the situation be different in the case of a peasant family supplementing their labor with that of slaves or otherwise unfree workers? If anything the surplus would be greater in the second case. This does not affect Wickham’s argument about the irrelevance of slavery, but it does raise a question about his treatment of the one mode of production that he sees as seriously posing an alternative to feudalism: the peasant mode.
Wickham argues that feudalism had become a universal system across Europe, North Africa, and the Middle East by the ninth century, but what does he mean by feudalism? As I noted earlier, Wickham rejected his earlier distinction between rent and taxation as the basis for distinguishing between the feudal and tributary modes of production, and he retains that position here. He has not, however, abandoned the tributary mode itself.52 Indeed, following Haldon, he writes: “it now seems to me that both [feudalism and the tributary mode] are sub-types of the same mode of production, in that both are based on agrarian surplus extracted, by force if necessary, from the peasant majority.”53 The process of exploitation is the same in each case, but the mechanism for rent or tax collection is different and, as Wickham stresses, this leads to corresponding differences in the state, above all in two respects. With the important exception of Merovingian and Carolingian Francia, “tax-based states were . . . richer and more powerful than rent-based, land-based, states.” More important even than wealth, however, was stability, which Wickham illustrates with the Byzantine example:
Even at the weakest point of the eastern empire, roughly 650–750, Byzantine political structures were more coherent than those of even the best-organized land-based states, such as Lombard Italy in the same period; tax-based structures had more staying-power, and the risk of decentralization, a feature of all land-based states, was less great. If taxation disappeared as the basis of any given state, then, no matter how much cultural, ideological, or legislative continuity there was . . . it would not prevent fundamental changes in political resources, infrastructure and practice.54
Nevertheless, both variants stand at a far greater distance from the peasant mode of production, involving “an economic and political system dominated by peasants, in a ranked society” than they do from each other.55
Societies based on the peasant mode involve “clear status differences . . . but they are not necessarily stable or heavily marked, except for the distinction, always present, between free and unfree.”56 According to Wickham there were many varieties of the peasant mode, but the essential features for him are that the productive unit is the household and that each household works land that it directly controls. Relations between households are governed by reciprocal exchange, partly to consolidate community relations, partly to acquire goods to which individual households would not otherwise have access. Since communities based on the peasant mode do not have to produce a surplus for an exploiting class, the main impulse behind production is to allow maximum leisure compatible with the satisfaction of physical needs and cultural norms; indeed, there are strong social pressures on individual households not to increase production beyond certain limits, since the output will either be given away to other, less productive neighbors or, if retained, lead to the household being ostracized by the rest of the community. Society under the peasant mode should not of course be regarded as a “primitive communism,” since, in addition to the use of unfree labor, it is inegalitarian with respect both to gender relations and to the act of giving itself, which confers high status or rank upon those who can give the most. The latter relationship is not, however, fixed, in that positions within the status group can change. During this period the peasant mode would have existed in two forms: either in a dominant “tribal” form, as in large parts of Northern Europe but also Spain and North Africa, where a relatively small external tribute might have to be paid to a local lord; or scattered like islands (Wickham writes about “leopard spots”) among territories otherwise dominated by the feudal mode of production, as in Francia and Italy.57
Wickham assembles an impressive array of evidence to demonstrate the existence of the peasant mode, and his historiographical achievement also supports an important socialist argument. The existence of an original classless society, “primitive communism,” is regularly denied by supporters of capitalism, for whom it is an enormously dangerous idea, suggesting as it does that inequality and exploitation are not, as it were, natural conditions. Wickham rejects both the term and, as we have seen, the implication that it involved complete equality in relation to this period; but if he is right, then it means that in some regions at least, the collapse of class societies in their slave and tributary forms did not lead to the “war of all against all,” but rather to a situation in which cooperation was the dominant characteristic. Although the situation is scarcely likely to be repeated should capitalism collapse, it is nevertheless an important historical contribution to the debates over human nature. But is the peasant mode effectively the same as the “Asiatic” mode, where the latter is taken to be a general term for mode dominant in transitions between classless and class societies?58 In other words, although the peasant mode seems to be the “fallback” position for peasants where precapitalist class society collapses, is it also a dynamic mode that would in time produce a new or revived form of class society? I will return to these issues below.
The absence of classes, or at any rate the absence of classes relating to each other as exploiter and exploited, also suggests the absence of a state, and Wickham accordingly argues that this was the case where the peasant mode was dominant. He defines the state as an institution combining a series of key elements: a centralized public authority apparently distinct from the public itself; “the centralization of legitimately enforceable authority (justice and the army); the specialization of governmental roles, with an official hierarchy which outlasted the people who held official position at any one time; the concept of a public power, that is, of a ruling system ideologically separable from the ruled population and from the individual rulers themselves; independent and stable resources for the rulers; and a class-based system for surplus-extraction and stratification.” On this basis he identifies three types of state: “strong,” as in the Roman, Byzantine, and Arabic empires; “weak,” as in Romano-Germanic kingdoms of southern Europe like Gaul, Italy, and Spain; and nonexistent (“pre-state”), as in the non-Roman kingdoms of northwestern Europe like Ireland, England, and Denmark—in other words, where the peasant mode was strongest. As Wickham rightly remarks, the point of a definition is its usefulness: how useful is this one?59
It is useful insofar as it helps us to remember that state formation is a lengthy process, which if captured by the historian before the end will reveal an institution that is not yet a state, but is (to use Hal Draper’s terminology) a “proto-government” exercising “proto-political” power. States take as long as classes to form, but this indicates my first difficulty with Wickham’s definition, namely that he places too much emphasis on region-wide formal attributes. If classes do exist, and Wickham accepts that lords tended to coexist with peasants even under the peasant mode, then the imposition of coercion and control is no longer exercised entirely by the community as a whole, but by a part with separate juridical powers.60 In this context, aristocrats and landowners more generally can act as a “state,” can embody state functions, at quite local levels.
A further theoretical problem is suggested by the relationship between the fourth and fifth characteristics of a state in Wickham’s definition (independent and stable resources for the rulers; and a class-based system for surplus extraction and stratification): “It is worth distinguishing between the resources of rulers and those of the ruling class, because one can often, even though not always, draw a distinction between the two (e.g., tax versus rent).” Wickham acknowledges that, where taxation was the overwhelmingly dominant method of surplus extraction, it could be subsumed into the provision of resources for rulers, “and the ruling class were simply public employees”: “In practice, however, dominant classes have almost always been distinguishable from state institutions; they are independently wealthy, although they characteristically seek wealth as well as power from official positions in public hierarchies.” And in landed societies where rent is the dominant method of surplus extraction, the subsumption operates in reverse, with “the resources of kings . . . nearly identical with that of the ruling class as a whole.”61 The distinction between “rulers” and “ruling classes” here seems to be unnecessary to Wickham’s argument. In the analysis of contemporary capitalism, there are usually some differences of interest between those who manage the state and those who own or control capital, although these always overlap, are currently decreasing, and in any case tend to be overridden by the joint class membership of the bourgeoisie. To the extent that these differences do exist, they are a reflection of the (much exaggerated) “separation of the economic and the political under capitalism.”62 But under feudalism, or any other precapitalist mode of production, the separation does not exist. Consequently, until the emergence of the absolutist state from the late fifteenth century, the possibility of a clash of interests between what one might call the political and economic wings of the ruling class does not arise. The resistance of Roman aristocrats to being taxed by the Imperial state, to which Wickham gave a central explanatory role in his initial account of the transition, might be cited as an example that supports the rulers/ruling class distinction—indeed, it lent plausibility to the claim that the feudal and tributary modes were distinct. But precisely because it occurred at an exceptional moment of systemic breakdown, it scarcely reflects the “normal” operation of class society. It is not clear what behaviors are explained by this distinction that would not otherwise be so. Indeed, to maintain it would seem to suggest that both groups operated with potentially different “logics,” which undermines Wickham’s—in my view, correct—argument about the fundamental unity of the feudal and tributary modes.
AGENCY: SOCIAL REVOLUTION, SOCIOECONOMIC TRANSITION, AND THE CLASS STRUGGLE
Who were the agents behind the transition to feudalism, where it did not emerge directly from the end of the slave mode? Although Wickham broadly endorses what he calls the historiographical “cliché” of serfdom emerging as a combination of tightened constraints on formerly free tenants and loosened constraints on the formerly unfree, his own emphasis on the relative unimportance of slavery suggests that the former was of considerably greater importance.63 The fate of the free peasantry is a central issue; the question is the extent to which it was undermined from within, by the emergence of class divisions, or overthrown from without, by submission to an existing class of aristocrats and landowners.
Wickham points to an inconsistency in Marx’s own approach as to the “prime mover” behind changes from one mode to another, contrasting the more abstract formulations (such as those of the 1859 “Preface,” which privilege the development of the forces of production) with actual historical analysis (such as in Capital Volume 1, where he shows that changes to the relations of production take priority). For this reason Wickham supports the position taken by Robert Brenner, which follows the latter aspect of Marx’s own work.64 The problem for Wickham, as for Brenner, is what drives changes to the relations of production. Wickham repeats the oft-expressed claim that, for Brenner, it is the class struggle.65 However, as I argued in the 2004 Deutscher Memorial Prize Lecture, Brenner is actually far less interested in the class struggle than is generally supposed. Since he believes that there is no impulse for the forces of production to develop, class struggle acts instead for him as a mechanism for producing (or failing to produce) a set of “unintended consequences” that in turn lay the initial conditions for the formation of capitalist social relations of production.66 As presented by Wickham, the peasant mode before the consolidation of feudalism bears a close resemblance to the description of the peasantry under feudalism as it is presented by Brenner, prior to the emergence of capitalism—particularly in relation to its internal stability and the lack of motivation to develop technology or increase productivity beyond a certain point.67
Leaving aside the question of whether both can be right, Wickham is more open to the possibility of feudalism emerging from internal developments within the peasant mode than Brenner is of capitalism emerging from internal developments within the feudal mode. Wickham allows, for example, that a “feudal economic logic” may be set in train when peasants with high status begin to require others to provide them with goods in return for more specialized but less material services, of which he specifies military protection (although presumably religious functions would also be relevant here).68 He also notes that peasants can acquire sufficient land to be able to lease it out to tenants, thus elevating themselves into the landowning class, with the potential to ultimately join the aristocracy.69 But these are mainly presented as hypothetical cases, rather than a process that can actually be traced. Of the two examples he offers, one is from the actual Danish village of Vorbasse and the other is from his invented archetypal village of “Malling.” It is clear that evidence is lacking, and Wickham explains that he must rely on models of change because “we can so seldom see them happening in our sources.”70 Consequently, when Wickham discusses the shift to feudal relations of production, he generally does so in contexts where they are introduced by an agency from outside the peasant community, namely existing landowners and aristocrats. But it is not clear whether this was the main path to the establishment of a “feudal economic logic” or simply the most visible from this distance in time. The extent to which feudalism was a “bottom-up” in addition to a “top-down” affair remains an area that still requires further research.
There are three types of class struggle “from below” recorded by Wickham, none of which would necessarily contribute to the rise of feudalism. The first are slave revolts. Wickham gives only one example, a tantalizing reference to what he calls the “famous” Zanj slave revolt in southern Iraq during the 870s.71 This event may enjoy fame among scholars of the medieval Middle East, but it might have received greater consideration for the benefit of nonspecialists, particularly given the emphasis Wickham places on the relative unimportance of slavery during his period.
The second are tax revolts. The examples to which Wickham devotes most attention took place in Umayyad, then Abbasid Egypt between 726 and 832 AD. These were not the actions of a particular class, like slaves, but overlapping risings, first by a preexisting religious community (the Christian Coptic sect), then by Arab settlers, and finally by an alliance of the two. Wickham argues that these were provoked not by higher levels of taxation than under the Roman Empire but rather because taxation tended to be more arbitrary and, above all, “more stringently enforced, and more aggressively policed.” The reason lies in the fact that the Arab rulers did not transform the societies they occupied but established themselves as a “state class” maintained solely by taxation, “with no structural social links to taxpayers,” meaning that patronage of client groups as a channel for allowing the latter to mitigate or avoid tax was not an option.72 Peasants also rebelled against taxation, in some cases supported after the fact by local religious figures: Wickham records one episode in Eukraoi in West Galatia (modern-day Turkey), where the local bishop and “holy man,” Theodore of Sykeon, dismissed the local aristocratic administrator and tax-collector after a rising by the villagers.73 This was not, however, the main focus of peasant action.
The third are peasant revolts, but these are different from predecessors under the Roman Empire and from successors under the consolidated feudal regime. Earlier peasant revolts, above all of the Bagaudae against the Roman Empire in Gaul, were essentially directed against taxation and injustice at a time when the state was weakened, and therefore the possibility of change beneficial to the peasantry became possible. Later peasant revolts too were conducted against the state in relation to “military service, laws on status and, above all, taxation.”74 In this period, revolts have a different impetus. Wickham notes that “a detailed knowledge of peasant states of mind is largely closed to us before the fourteenth century.” He nevertheless speculates that aristocratic hegemony did function in certain areas where the peasants had to rely on aristocrats for external support, as in eighth-century Lucchesia (in modern Italy), although this did not, of course exclude “small-scale signs of disobedience”—but these are compatible with overall acceptance of ruling values. At the other end of the spectrum, as in eighth-century Paris, the aristocrats dominated through “overwhelming physical force” and did not require peasant acceptance of their rule, which they any case did not receive. Between these lies a third type of area, such as sixth-century Galatia, where neither situation prevailed: that is, where aristocrats could rely on neither ideology nor violence to secure compliance. As Wickham notes, the latter situation is where revolts are most likely to take place, but “the absence of hegemony is only one reason why peasants revolt, of course; they have to have something concrete to oppose as well.”75 In this case peasant revolts are signs of resistance to attempts by the emergent ruling class to impose serfdom. England is exceptional in its lack of peasant revolt, which seems to have two causes. First, because initially both the “rulers” and the “ruling class” (to use Wickham’s distinction) had less control over the peasantry than in any other part of Europe, while at the same time they exercised superiority over exceptionally large territories. Second, in that when the lords did move to subject or expropriate peasant communities they did so slowly and in piecemeal fashion, attacking the weakest while leaving the strongest and wealthiest untouched until the basis of possible collective resistance was eroded.76 Elsewhere, the gradual encroachments of the emergent feudal state led to what Wickham calls “frequent small-scale resistance,” which erupted into one of the great risings of the period: the Stellinga revolt in Saxony of 841–42 AD, a revolt that took the opportunity of a civil war among the local Saxon ruling class to launch a program for the return to the pre-aristocratic social order.77
Could war be treated as the functional equivalent of revolution, in the sense that revolutions from outside and above were conducted by the Cromwellian and Napoleonic armies a thousand years later? One of Wickham’s critics has certainly accused him of ignoring the impact of war—indeed, of being in thrall to an “Austro-German model of explaining great transitions in human history,” a model with “little or no theorization of war or violence.”78 The “Austro-German model” evidently includes Marxism, and Wickham himself has made a similar point about it in the past; but it is untrue.79 More to the point, in this case it is also irrelevant: war can be important as an agent of social change, but it is scarcely autonomous. It is difficult to see how war could be the source of a new form of society, since all the societies involved were based on variations of the same mode of production, and the most ideologically innovative—the Arab invaders of the later seventh century—tended to allow the continuation of existing social structures in the territories they conquered. The most significant social impact of war was in fact an inadvertent consequence of the Vandal invasions of North Africa, of which Wickham writes that “Geiseric’s conquest of Carthage in 439 is arguably the turning point in the ‘fall’ of the western empire”—the significance of the conquest being that it “broke the tax spine” connecting the Roman world economy.80 But elsewhere the impact was muted: as Wickham notes, “only Italy in the sixth century and Anatolia in the seventh saw wars that really devastated economies and societies on the regional level for more than short periods.”81
In short, class struggle clearly occurred in different forms throughout the period, but revolutionary movements for the transformation of society, or plausible surrogates, are absent—except of course in the form of the piecemeal revolution from above, imposed by the lords to either abolish the economic logic of the peasant mode or erode the autonomy of peasants where the feudal mode already existed.
CONCLUSION
Wickham has argued that the interaction between the forces and relations of production, and between them and the superstructure, may vary from one mode of production to another. Be that as it may, what he makes unmistakably clear in this work, among many other things of value to historical materialists, is that the nature of the transitions from one mode to another are certainly distinct, and that we proceed by analogy with later ones at our peril. Whatever questions Framing the Early Middle Ages still leaves unanswered—and many of those we are only able to ask because of Wickham’s achievement—we should be grateful that we now have an account of the transition to feudalism that rivals those on the transition to capitalism, which have for so long been the staples of Marxist historiography.