4

Late Roman Society And Economy

Understanding the late Roman economy presents a particular challenge. Certain topics, such as slavery, taxation and the so-called ‘colonate’, which used to occupy a special place in the secondary literature, have undergone re-evaluation, while the vast amount of new archaeological and other evidence that is becoming available means that the whole subject has been transformed. The ‘cultural’ model of late antiquity based on the work of Peter Brown and others in the past generation has been accused of ignoring or at least underplaying important issues in economic and administrative history.¹ At the same time the size and nature of the late Roman economy is still disputed; current questions include the impact of state taxation and the exaction of the grain from Egypt and North Africa, the role of the super-rich with their large estates, the level of integration, and the degree to which there was a market economy. In the later part of the period, geopolitical factors pose in acute form the question of whether and for how long this was still a Mediterranean world at all.

Consideration of the late Roman economy (which involves also the consideration of groups such as landowners, tenants and slaves) is closely tied to historiographical models of decline and collapse. While as we have seen (Chapter 2), current revisionist approaches emphasize violence as a main factor in bringing about the end of the Roman empire in the west, internal factors have also been adduced by many historians anxious to explain Rome’s fall.² A dark view of the later Roman empire is to be found, for instance, in Ramsay MacMullen’s Corruption and the Decline of Rome.³ According to these approaches, which were foreshadowed in M.I. Rostovtzeff’s Social and Economic History of the Roman Empire,⁴ the ancient world came to an end because of its own internal problems, among which were over-taxation and a decline in the capacity and willingness of elites to maintain urban life. Though for very different reasons, this was also the classic Marxist view, for which see especially Perry Anderson’s Passages from Antiquity to Feudalism⁵ and G.E.M. de Ste Croix’s The Class Struggle in the Ancient Greek World: from the Archaic Age to the Arab Conquests.⁶

This negative picture of the later Roman empire (sometimes referred to disparagingly as ‘the dominate’, in order to convey the idea of autocratic rule supported by an unwieldy bureaucracy) has been held by many in the past to apply to the period from Diocletian onwards. However, much of the evidence adduced is impressionistic. Complaints about tax collectors or soldiers billeted in towns are indeed common, but similar examples can be found in almost any society, and need to be read with caution. Whether things had really changed significantly for the worse is hard to establish, and plenty of evidence from the Principate suggests that the condition of the peasant then was hardly any better. Individual instances of peasants taking evasive action at the tax collector’s approach do not necessarily add up to a general picture of flight and collapse. Also fundamental to this view is the legal evidence, especially the often repeated laws in the Theodosian Code, by which successive emperors legislated to keep decurions in place in their towns and coloni on the estates in which they are registered. The picture of oppression and authoritarianism which these laws seem to suggest was endorsed in the past by scholars who have represented the later Roman empire as virtually collapsing under its own weight, and described it in terms such as ‘totalitarian’ and ‘repressive’.⁷ But as A.H.M. Jones recognized, when laws are constantly repeated, they must be presumed to be ineffective; moreover, laws need enforcement. Where the necessary apparatus for the latter is lacking, as it largely was in the Roman empire, it may be comforting for those in authority to repeat the law itself, but it does not necessarily follow that it was actually carried out in practice. Jones also recognized that social mobility was possible in the later Roman empire – probably even more so than under the Principate. But a wholesale re-evaluation in the last generation of the nature of the legal evidence, in particular the construction and evolution of the Theodosian and Justinianic Codes, has now led to a very different understanding of the actual significance of this legislation and of the working of law in late antique society (Introduction). Similarly, recent work on the papyrological evidence from Egypt, especially that relating to great landowning families, lies behind a new and more commercial view of the economy in the sixth century.⁸ Taking the literary sources at face value can lead to equally overstated conclusions. Many previous discussions have taken literally the apparent statement by the Christian Lactantius, a biased and hostile source,⁹ that Diocletian quadrupled the size of the army, and since the maintenance of the army was the single largest call on the state revenues, have then used this as the basis for a highly negative view of the economy in general. Similarly, despite Diocletian’s measures to ensure better collection of revenue and an elaborate system of taxation in kind, it is far from certain that the level of taxation itself increased.¹⁰ The few general statements that we have in contemporary sources on such matters as taxation tend to come from writers as biased and unsubtle as the pagan Zosimus or the fifth-century Christian moralist Salvian, and must be treated with considerable caution. Finally, in considering these methodological issues, it is also necessary to balance contrasts between east and west; if the structure of the state in the late fourth and fifth centuries was really as top-heavy and as liable to collapse from its own internal contradictions, why did the eastern empire resist fragmentation and maintain prosperity for a significantly longer period?¹¹

It is the contribution of archaeology above all which has led historians to question the older view, and it is worth noting that A.H.M. Jones published his great work, The Later Roman Empire, in 1964, well before the current interest in late Roman and early medieval archaeology. Although Jones himself was very aware of the importance of archaeology and had travelled very widely round the provinces of the Roman empire, his work relies mainly on documentary evidence, and such a book would look very different today. The change is also a matter of new ways of looking at the subject. While the older assumptions of decline are still very much with us, many historians have been influenced by different approaches, especially comparative ones. Perhaps most interestingly, the debate about the ancient economy which has been going on since the publication of M.I. Finley’s classic book, The Ancient Economy, in 1973,¹² has extended to the later empire, thus to some extent bypassing the supposed great divide that came with the third century and the reforms of Diocletian. The older model depends on a clear periodization based on the assumption of a massive tightening up of government control and consequent increase of government expenditure, generally attributed to Diocletian. Most books still make ‘late antiquity’ begin with the reign of Diocletian, and that is under-standable, given the instability of the preceding period and the administrative innovations brought in from 284 onwards.¹³ But if many, even if not all, of Diocletian’s reforms were revisionist rather than fundamentally new, this in turn should cause us to pay more attention to the underlying economic structures and questions which hold good throughout the long history of the empire. Current research on the Roman economy focuses on such issues as technology and production, quantification, movement of goods and indicators of economic growth or decline rather than on competing overall models of economic primitivism or economic rationalism.¹⁴ Intense efforts are currently being made to quantify the Roman economy.¹⁵ But tracing the actual nature and level of market exchange remains difficult at all periods in the Roman empire, including late antiquity, not least because of the unevenness of the available source material.¹⁶

East and West

There are nevertheless certain obvious issues which affect the later period specifically, including that of the increasing divide between east and west. The basic administrative, economic and military structures of the Roman state established in the early fourth century were still in place in the eastern empire at least up to the reign of Justinian, and often beyond. We must therefore look for special factors, such as those described in Chapter 2, to explain why the west should have been different.

The late Roman tax system was designed to cope with a situation in which continual debasement of the coinage had led to near collapse, and revenues had to be collected and payments made to the troops in kind; a regular census and the five-year indiction aimed at ensuring reliable collection of tax revenue for the state, and the scheme also involved elaborate matching of need and supply. The main item of expenditure as before was on the army, who were now paid in kind as well as money. Certain obvious consequences followed: army units (themselves far more varied in type and organization than previously) now tended, for instance, to be stationed near to the sources of supply, and thus in or near towns, instead of on the frontiers. While by the end of the fourth century more payments were made in cash, the central role of the state in collecting and distributing the annona (the army supplies) remained an important feature of the economy, in terms of both organization and stimulus to production; accordingly the eventual cessation of this state function was a major factor leading to economic fragmentation, as was the end of the grain requisitions for the cities of Rome and Constantinople. Basically the same system was in force in the east as in the west, but apparently with more success. A number of factors contributed to this. The east, for instance, had been urbanized earlier and more successfully than the west and, despite the ceaseless complaints of municipal councils and their spokesmen, most of these cities continued in existence or even flourished into the later sixth century and indeed beyond (Chapter 7). We hear a great deal about their problems, not least because our written sources tend to come from just this kind of milieu; thus Ammianus Marcellinus, Libanius, Julian and later Procopius all took up the cause of the cities versus the central government. But many of their complaints had an ideological basis; in practice, the fifth and early sixth centuries seem to have been a time of prosperity for many eastern regions, especially parts of Syria and Palestine (Chapters 7). Another obvious difference between east and west in economic terms relates to the constant and in the end more serious military action in the west in the fifth century; not only was the economic base itself weaker than in the east, but the demands on it were greater. As we have seen, the western government had great difficulty in maintaining military forces adequate for their task. A deeper and more structural difference also lay in the growth of an immensely rich and powerful class of senatorial landowners in the west during the fourth century, whereas wealth in the east was by comparison more evenly spread, at least until the growth of the large estates of which we know in Egypt in the sixth century (owned by families such as the Apiones with property in Constantinople and land elsewhere as well).¹⁷ The combination of a weak government and wealthy and powerful landowners was crucial in determining the shape of the western economy from the late fourth century onwards.

Thus east and west were both similar and dissimilar in this period, and local factors are increasingly important from AD 395 onwards; yet many shared features remain and some similar trends can be observed, even though the rate of change may differ. The standard accounts of decline and collapse obscure these real differences. In contrast, the present lively state of archaeological investigation invites us to compare one site or area with another, and encourages the broader view; it also invites the question of how the traditional textual evidence and the increasing amount of material evidence relate to each other. This approach is inherent in Wickham’s magisterial work. By the end of the sixth century, while it is still possible in some ways to speak of a Mediterranean world (see Conclusion), the west has largely fragmented. The eastern government and its provincial and defensive structures were tested both in the Balkans and in the east in major wars against the Sasanians, and Justinian’s wars of reconquest may have overstretched the capacity of the eastern empire; while it is hard to quantify the impact of the plague which first hit the empire in 541, contemporaries viewed it as a catastrophe of major dimensions (Chapter 5). There were also structural factors which we can see reflected in the gradual metamorphosis of many cities in the eastern empire from late antique towns with public buildings and grand houses into smaller, more defensive and village-like formations, a process which had begun before the end of the sixth century (see below and Chapter 7). The Persian invasions of the early seventh century dealt another severe blow to parts of the east, and seriously challenged the ability of the state to maintain a military response. Eventual victory over the Persians was followed by the first Arab incursions into Syria in the 630s, which the eastern empire was unable to repel (see Chapter 9). Taking a long view, it is possible to argue that the east and west underwent similar processes, but at different times, the speed of change being regulated by the operation of local factors. Yet the resilience of Constantinople in maintaining its political existence is a remarkable feature of the seventh and eighth centuries, and there is ample evidence of cultural and local economic continuity through the Umayyad period.

The Organization of Labour

Large-scale slavery declined in the Roman empire but the sources make it very clear that slaves continued to exist,¹⁸ sometimes in very large numbers – for example, on the estates of senatorial landowners – and this continued in the east into the Byzantine period. As we saw in Chapter 3, when late Roman landowners became Christian, they sometimes sold their property in order to use the wealth for Christian purposes, in which case the slaves were sold too; this was the case with estates which belonged to Melania the Younger in the early fifth century. Slaves also often resulted from war, in which Romans as well as barbarians might be captured and enslaved. Slaves could also be bought, and were easily come by along the frontiers, and their existence is taken for granted in the barbarian law codes;¹⁹ the example of Caesarius of Arles in the early sixth century shows that ransoming such slaves came to be seen as one of the duties of a Christian bishop.²⁰ Legal sources demonstrate the continued existence of slaves on the land and elsewhere, and the church also soon became a major owner of slaves. We can assume that part of the labour force on the land and in many forms of production will still have been servile. However, it is less clear what this meant in practice, or how slaves related to coloni – technically free tenants who were, in many areas, theoretically tied to their particular estates by imperial legislation, and over whom the landlords had rights which can look very like the rights of owner over slave. It was for instance possible to be described as servus et colonus (both a slave and a colonus), and slaves could also be tenants.

Recent historiography, under the influence of the reinterpretation of late Roman legislation on the subject, has radically questioned the traditional view of the coloni, which seemed to suggest a move towards a tied peasantry in the later Roman empire and thus underpinned ideas of immobility and decline; such a view was fundamental to Marxian ideological approaches to the period.²¹ In contrast, according to a recent discussion, ‘the ‘colonate’ was not a generalized condition of rural dependency’, but rather arose in the context of the late Roman tax system and the desire by the authorities to track not only the ownership of the land itself but also the labour force. The state was interested in maximizing tax revenue and restricting any reduction in the agricultural labour force.²² Coloni were registered by their landlords for tax purposes and the state had an interest in ensuring that tax liabilities were met. Hence, as we can see from the fifth-century Theodosian Code and the Justinianic Code a century later, late Roman emperors passed a mass of legislation which on the face of it sought to restrict the freedom of movement of coloni and tie them to the land; they also tended to use the familiar language of slavery. If these laws are taken at face value and were successful, we would have to conclude that the late empire was a time of real repression, in which the population was reduced to virtual serfdom.²³ Indeed, the concept of being ‘free’ became difficult to define, and the difference between slave and free may often have been slight or non-existent in practical terms: by the time of Justinian, for instance, tenants known as adscripticii (bound to the soil) were treated in the legal texts more or less as if they were slaves.²⁴ Yet the impression we get of these classes from other sources is far from being one of total repression and alienation,²⁵ and social mobility was surprisingly common at higher levels. There was thus evidently a large gap between theory and practice.

It is important to realize that late Roman law often followed, rather than led, social practice. The frequently repeated and often contradictory pronouncements of emperors do not signify authoritarian intrusions on the lives of individuals so much as vain attempts to regulate a situation which might in practice be beyond their control. Complexities and inconsistencies abound, not to mention those caused by the very processes of recording and codification. The legislation on coloni grew out of the difficulties experienced in collecting tax, and the state legislated essentially to control and trace the labour force on which the tax was due. Not surprisingly, given the ways of late Roman government, this legislation developed only gradually, and piecemeal, during the fourth century, and uncertainty as to the relation of slave and colonus in individual areas, and inconsistency between geographical regions, were among the results of the untidy process that was adopted. Legislation dealing with the status of coloni was also introduced at differing rates in different geographical areas, in Illyricum and Palestine not until the end of the fourth century. Furthermore, as the evidence of the Egyptian papyri suggests, there were many possible variations at the level of actual arrangements between landlord and tenant; loans, effectively mortgages, from large landowners to small were common, and defaulting borrowers were subjected to coercive measures from the lenders which were of more immediate concern than any imperial legislation. In general, it seems questionable whether conditions for the lower classes had in practice significantly deteriorated since the early empire. The condition of the poor, whether urban or rural, remained hard at all times. There had indeed been over the imperial period a progressive intensification of penalties applied to those convicted under the law, with an ever-widening division between the treatment of the rich and powerful and the cruel treatment (torture, chains, mutilation) meted out to the poor.²⁶ But the same process coincided during our period with a new consciousness of ‘the poor’ as a class, no doubt inspired by Christian teaching, which found expression, as far as the urban poor were concerned, in various forms of Christian charity,²⁷ while saints’ lives attest to the role of the local bishop in alleviating economic distress in the country areas, and especially in providing food in times of famine.

The economic changes which took place in the late empire were not of a revolutionary nature. This was still a basically agrarian society, and much of the land was owned by large landowners and worked by tenants, whether slave or free. Again, though comparisons with medieval feudalism have been tempting, and important for Marxist historians, there was no simple chronological transition from late Roman coloni to serfdom.²⁸ It would also be a mistake to suppose that peasants in earlier centuries had had much possibility or inclination de facto to move away from their area, or that they had not also been dependent. As for the lower classes in the towns, it is equally difficult to get a fair picture of their lot when so much evidence is anecdotal and when so many of the literary sources are liable to exaggerate for their own purposes. Naturally it is easy, as in most periods, to find evidence in the sources of both urban and rural poverty, especially in relation to tax debts, but again one should be cautious about generalizing too much on the basis of this evidence. On the whole, change was slow: local and unforeseen factors such as famine, pestilence or the like constantly threatened an agrarian economy with few obvious technological advances, but were also part of the expected range of possibilities and could therefore be contained; external factors such as warfare were of course another matter.

We hear many complaints in this period from the town councillors, the curiales, about their difficulties in continuing to finance urban life, and a process of increased imperial intervention in the affairs of cities, especially their financial affairs, has generally been seen; this went along with corresponding changes in the traditional ways of urban government (further, Chapter 7).²⁹ This was another example of how in trying to deal with a situation, imperial legislation actually made it worse; emperors attempted to prevent the curials from leaving their duties while giving them the opportunities to do just that. The ‘flight of the curials’ was an indicator that the old style of urban administration was giving way. It is less clear when or exactly how this change occurred, though it seems to have happened later in the west. Under Anastasius (491–518) it was decreed that the governing body for cities should be drawn from ‘the bishop, the clergy, the honorati (office-holders), the possessores (landowners) and the curiales’.³⁰

It was from the curial class that office-holders were drawn, and for whom imperial office-holding was highly attractive. Imperial service was now the lucrative way to advancement: councillors themselves were by now often men of modest means who could not shoulder their former burdens, and when public monuments or statues were erected in eastern cities such as Aphrodisias it was usually by men who held imperial ranks and offices. Naturally this produced some statements of nostalgia, as well as serious tensions for individuals, and in the mid-sixth century John the Lydian reminisced about the time, now in the past, when councils still ruled the cities.³¹ It was often the local bishop who stepped into the breach, and who became not merely an authoritative leader in the town but in many cases the apex of a much reduced municipal organization; see below, Chapter 7. Jones rightly emphasizes the very large number of posts (dignitates) that had to be filled on a regular basis, and the law codes assume that curiales frequently endeavoured to escape their lot and better themselves in the administration, the church or the army. This class as a whole was the subject of what Jones calls ‘a vast and tangled mass of legislation’, whereby the state attempted ineffectually to prevent the seepage and maintain the councils on whom the cities depended. This legislation did not attempt to address the overall problem, but was issued piecemeal and in response to local conditions, and with no likelihood of general enforcement.³² Earlier attempts to return curiales to their cities if they had managed to secure a post in the administration failed, and in principle after 423 individuals could no longer escape their obligations in this way. Similarly, the fifth-century emperors were still attempting to stop the loophole opened by Constantine when he freed clergy from curial obligations, as was Justinian, when in 531 he allowed ordination of curiales only if they had spent fifteen years in a monastery first, and were willing to surrender a substantial part of their estate (CJ I.3.52). But despite the complaints with which the sources abound, with the exception of the not inconsiderable part now played by Christian charity one may suspect that the condition of the people remained much the same.

But if there was no economic revolution, certain new factors did become operative, including, on the one hand, settlement on a large scale and, on the other, the growth of the church as a major economic institution in its own right, with profound implications ranging from the role of bishops as urban and rural patrons and the diversion of resources into church building to the growth of monasteries and their potential impact on the local economy. It was factors such as these, combined with the centralizing tendency of the late Roman state and the severe damage caused in some areas and to some towns by invasion and war, which disturbed the balance of landholding and wealth and which inevitably brought profound change.

The Classes of Late Antique Society

The senatorial class of the west had been a major beneficiary of the disturbances which took place in the third century, and was, moreover, at least in part itself a product of the patronage of Constantine and his successors. Constantine extended the size of the senatorial class, and by severing its necessary connection with the city of Rome, opened it up to new membership in the east, making possible the development of a new senate centred on Constantinople.³³ However, the traditional aristocracy in the west retained its wealth and its prestige, and one of the main features of the late fourth- and early fifth-century west is the enormous wealth, by which we mean the enormous landholdings, of the western senatorial class.³⁴ Perhaps because of unsettled conditions in many areas, it had become possible to acquire vast estates – the size of towns, we are told. A landowner would also expect to have at least one town house in which he lived in extreme luxury, as we learn from Ammianus’ famous (and scathing) description of the fish ponds and table delicacies of the Roman nobility of the late fourth century.³⁵ Owning estates on this scale was a business in itself, even if the landlord was an absentee. According to Ammianus,

A journey of fair length to visit their estates or to be present at a hunt where all the work is done by others seems to some of them the equivalent of a march by Alexander the Great or Caesar.

(Hist. 28.6)

Late Roman aristocrats commissioned luxury items for their grand houses, including manuscripts, silverware and ivory, and kept alive the tradition of artistic production in the classical manner. These were members of the aristocracy and holders of the highest official honours, depicted on elaborately carved ivory panels in all their panoply of office. They owned grand houses and could afford to commission the best artists and craftsmen. In the sixth century great Roman families such as the Anicii had large estates and lands in Egypt and the east, and remained patrons of luxury items; they also intermarried both with the new Germanic rulers and within imperial circles in Constantinople. The upkeep of the estates of such families required armies of retainers and an elaborate system of production and supply of goods. Owners were interested in profits, and had perforce to devote a good deal of time simply to keeping things going. Some of it was occupied in dealings for mutual benefit with others in a similar position, transactions which reinforced the gift element and the importance attached to display which were typical features of the late Roman economy. Sulpicius Severus and Paulinus of Nola provide evidence in their writings of typical gifts from one landowner to another of such commodities as oil and fowl, a practice also known from Sidonius in fifth-century Gaul, and from bishops and kings in the Merovingian period; Pope Gregory the Great was no different in this respect from a secular landowner of earlier times. Landlords were certainly involved in production, and engaged in long-distance transport; it has been argued that both might take place within an exchange system involving either simply his own estates, or those of himself and his friends. If so, this was less an economic activity than a patronal relationship. Even the widespread appearance overseas of African pottery during our period may be partly a product of this mutual exchange rather than the result of new commercial market or production systems. However, recent studies have argued forcefully for the commercial activity of such families, and there has been greater emphasis on trade and profit as motivation for the long-distance movement of goods; the interpretation of the huge amount of evidence from late Roman pottery is critical here.³⁶ It seems obvious, as Wickham argues, that elite demand provided the stimulus for such activity. If it is true that the amount of land in the hands of great proprietors (the po-tentes) increased, and if we accept more recent views about the nature of their economic activity,³⁷ a very different picture emerges of economic and political factors in the late empire than the traditional one.

The enlargement and transformation of the senatorial class, greatly increased in numbers from the time of Constantine on, rendered the old equestrian class otiose; the latter eventually disappeared as its former offices were progressively renamed and redefined as senatorial. Nor was it enough to be called simply vir clarissimus (the standard senatorial rank in the early empire). Valentinian I in AD 372 laid down a senatorial hierarchy ranging from clarissimi to spectabiles and (at the top) illustres; these grades were attached to the holding of particular offices, and other privileges of rank, such as seats allotted at the Coliseum in Rome, also followed. The senate of Constantinople, on the other hand, differed from that of Rome since it was an artificial creation; while the Roman senate comprised families of vast wealth and pretensions to aristocratic lineage (even if in many cases they did not go further back than the third century), its counterpart in Constantinople was filled with new men. This feature in the long run helped its future continuance; being based on Constantinople itself, and before the late fifth and sixth centuries generally lacking the enormous estates of its counterparts in Rome, the eastern senate was better able to avoid tensions which developed between the Roman senate and the imperial government. But eastern senators also enjoyed substantial privileges, and their role as members of the traditional landowning class, allegedly preyed upon by the rapacious emperor, is emphasized in the Secret History of Procopius, who identified with their interests. Like their western counterparts, eastern senators were no doubt in a good position to evade the special tax (collatio glebalis or follis) which had belatedly been imposed on the senatorial class by Constantine. In the example of the senatorial class, we can in fact see the combination of tradition and innovation which is typical of the late empire; for while, on the one hand, the late Roman senate was essentially a service aristocracy which differed considerably from the senate of the early empire, it did not occur to anyone not to maintain existing social patterns, so that many of the outward signs of senatorial status and privilege were retained or even enhanced. In such circumstances, the Christianization, and in particular the conversion to asceticism, of members of leading senatorial families, which began to occur in Rome in the late fourth century, seemed to present a threat to status, wealth and tradition, and therefore met with considerable opposition.³⁸

The later Roman empire was characterized by a high degree of competition for status and access to wealth and privilege, which we can see operative also in the centralized bureaucracy. Since posts in the imperial service could be highly lucrative, and released the holder from burdensome existing obligations, the bureaucracy drew off talent from the ranks of the curiales in the cities even as imperial legislation, conscious of economic and administrative needs, sought to keep them in their places. One of the most persistent of modern myths about the late empire is that of a top-heavy and rigid bureaucracy which wielded the hand of repression yet whose size made it unsustainable in relation to the existing resources of the empire. In fact the empire was engaged in a constant balancing act between what was perceived to be necessary and what was possible. There was in practice a high degree of social mobility, and the court and the office-holders had a natural tendency to proliferate. The nomenclature and emoluments of the imperial service paralleled those of the army; office-holders held titles of military equivalence and received military stipends. This had little to do with modern concepts of efficiency, though the government had at least an interest in filling the administration with people it deemed suitable; at the same time it also needed to maintain the numbers of curiales in each city (who were also the obvious candidates for openings in the imperial service), since on them fell financial responsibility and tax obligations at local level. The double bind in which the government found itself was further complicated by the willingness of individuals effectively to buy their way into the administration, and that of the government to sell offices within it – the attraction for the purchaser being the emoluments that went with the position, and often the possibilities for favour and extortion that it carried in addition. To a modern observer this suggests corruption,³⁹ and it was easy for contemporaries to abuse the system and for others to complain about them. We should remember, however, that for all its impressive state apparatus, the late Roman empire was still a very traditional society. Many official posts were sinecures, or unimportant in themselves, and John the Lydian, just such an official, has left a vivid account of what it felt like to serve in one of the great offices of the state in the sixth century.⁴⁰ In practice the combination of patronage, aristocratic prestige, the need to fill the ranks of the bureaucracy and at the same time its tendency to swell because of the advantages it offered to those lucky enough to secure a position implied a constant balancing act only partly evident at the time.

The practice of selling offices in the imperial administration provides a particularly delicate example: on the one hand, the late Roman and later the Byzantine governments were concerned to stop the abuse of the practice, while at another level each used it as a financial tool and mechanism for selection of officials. In 439 an oath was exacted from all those appointed to provincial governorships that they had not paid to secure office:

we ordain that men appointed to provincial governorships should not be promoted by bribe or payment but by their own proven worth and your [i.e., the prefect’s] recommendation; let them testify on oath that in gaining their responsibilities they have neither made any payment nor will they make any subsequently.

(CJ IX.27.6 pr.)

Yet later emperors actually sold offices: Zeno, for instance, raised the price for the governorship of Egypt from 50 lb of gold to nearly 500 (Malchus, fr. 16, Blockley). Justinian again tried to stop the practice, repeating the earlier demand for an oath from those appointed. Novel 8 (535) forbade the practice of suffragium, i.e. the buying of office, for provincial governors (though there were fixed costs that governors were expected to take on), and the secular officials were required to take an oath that was strongly Christian and orthodox in character.⁴¹ Bishops and clergy also commonly made payments for office; again, Justinian legislated against this, but later took the alternative course of regulating such payments.⁴² The whole constituted a finely balanced ecosystem with both economic and political implications for the emperors, and they were reluctant as well as unable to upset it too drastically. As in most complex societies, they also had to balance the tension between two needs: to manage the expectations of the elite and to keep the machinery of government working. Justinian promoted a ‘reform’ agenda in his self-presentation, but whether efficiency was even a possible aim in practice, as distinct from official and legal pronouncements, is a fundamental question. ‘Efficiency’ is a modernizing concept, and it is unclear that the system of payment made the administration any less efficient than it would otherwise have been. Procopius claims that Justinian himself was selling offices again within a year of his edict of 535, and his successor, Justin II (565–78) legislated again in 569 to stop governors from buying their office.⁴³ The corollary, and the underpinning of the sale of offices, was, of course, the desire of the office-holders themselves to recoup the moneys paid while they held office, the prospect of which had been a powerful attraction to purchase in the first place. Regulation was therefore probably the best option for the government, and we can also see the close connection between their attempts at regulation and the changing processes by which officials, especially provincial governors, were appointed.

One of the main hallmarks of the late Roman administrative system was patronage and the use of influence. Recent work has increasingly emphasized the importance of patronage in understanding ancient society as a whole, especially in the context of the Roman empire, and the patronage system of the later Roman empire could involve not just networks of obligation, but also coercion and even violence.⁴⁴ Patronage has existed and does exist in many – perhaps even all – societies, but is present typically where the protection offered by the state is weak, where the social bonds are loose or where there is change and competition for place in the new scheme of things. In late antiquity, ‘traditional patrons found themselves supplanted as patrons, but also impeded as landlords – and as tax collectors – by men with local authority, secular or religious. Their protests were echoed with legislation representing the fiscal interests of the central government.’⁴⁵ For when new actors – bishops, state officials – entered a stage on which patronage already operated on every level, and when the interests of the poor, the landlords and the state diverged, any existing equilibrium was broken. In such conditions the poor and the helpless looked where they could for protection. The state for its part made repeated attempts to declare this form of protection (patrocinium) illegal, on the grounds that it represented an evasion of the responsibilities of those who were its subjects, and an illicit appropriation of authority by those who took it on, not to mention the extra tips which they doubtless imposed. A law of 415 allowed the churches of Constantinople and Alexandria to retain villages which had come under their protection, provided that all taxes were paid and other obligations fulfilled,⁴⁶ but later emperors such as Marcian and Leo continued to try to end the practice, and Leo attempted to forbid all patronage contracts from 437 in Thrace and from 441 in the east.⁴⁷ Again, the practice itself and the government’s inability to deal with it demonstrate not so much endemic corruption as the huge challenge presented to the bureaucratic system in relation to the vast and fragmented areas which it was attempting to control.

Financing the State

Behind many of the problems which the state experienced, and which gave rise to these social difficulties, lay the need for tax revenue and the difficulties of collection. Many scholars have believed that the level of taxation was higher in late antiquity – so high in fact as to contribute substantially to increased extortion and consequent decline. However, while we have in the sources a number of figures for tax revenue and budgetary expenses, it is far from clear whether they are reliable or not, or how far they may have changed over the period we are considering. At least in the late empire there was now a regular tax period (‘indiction’) for which levels were fixed, and the system had been changed to take into account both labour force and quality of land. Constantine had imposed special taxes on senatorial wealth and on commerce, and had thus at last brought these sectors into the tax net. But taxation still fell mainly on the land and on agricultural production, and the government had little recourse in response to loss of land and shortage of manpower except to try to close loopholes and harangue the population through often repeated legislation such as that seeking to restrict movement of coloni and help landlords to keep their tenants and maintain the tax revenue:

Whereas in other provinces which are subject to the rule of our serenity a law instituted by our ancestors holds tenants down by a kind of eternal right, so that they are not allowed to leave the places by whose crops they are nurtured or desert the fields which they have once undertaken to cultivate, but the landlords of Palestine do not enjoy this advantage: we ordain that in Palestine also no tenant whatever be free to wander at his own choice, but as in other provinces he be tied to the owner of the farm.

(CJ XI.51.1)

The tax collector looms large in contemporary literature as a hated and dreaded figure, and the danger to those who could not pay was very real. Paphnutius, a hermit near Heracleopolis in the Thebaid, met a former brigand who told him how he had once come upon a woman who had suffered in this way, and asked her why she was crying; she replied:

‘Do not ask me, master; do not question me in my misery but take me anywhere you wish as your handmaid. For my husband has often been flogged during the last two years because of arrears of taxes amounting to three hundred gold coins. He has been put in prison and my beloved three children have been sold as slaves. As for me, I have become a fugitive and move from place to place. I now wander in the desert but I am frequently found and flogged. I have been in the desert now for three days without eating anything.’

‘I felt sorry for her,’ said the brigand, ‘and took her to my cave. I gave her the three hundred gold coins and brought her to the city, where I secured her release together with that of her husband and children.’

(Russell, Lives of the Desert Fathers, 95)

The repeated laws show, however, how little the government could actually do to enforce collection of revenue. The taxes were highly regressive: small peasant proprietors paid the same as great landlords for the same amount of land. And despite Constantine’s reforms, the traditional emphasis on the land still led to a failure to tap major sources of wealth, whether from trade or, importantly, from senatorial incomes. In the latter case, especially, it was in part the nature of the tax laws themselves which enabled senators to amass colossal fortunes while the government went short. Emperors themselves shared the traditional view that exemption from taxation was a privilege to which rank and favour allowed one rightfully to aspire, and thus their grants of exemption were not simply a way of gaining popularity but an expression of this traditional attitude. Cancelling arrears was another common device, in the face of real inability to enforce the law, for political reasons or in response to these traditional attitudes, and there was little conception of budgeting for the future. On the other hand, as Jones points out, the eastern government at any rate seems to have been able to collect very substantial sums on a continuous basis;⁴⁸ this was despite the outflow of large sums of gold to buy peace with Persia, or for ‘subsidy’ payments to barbarian groups.⁴⁹ The sums expended for both purposes could be very high, and the practice continued over the whole period – from 434 annual payments of gold agreed by Rome to the Huns stood at 700 lbs, and rose to 2,100 lbs in 447, with a payment of 6,000 lbs to cover arrears; according to Priscus this required a much higher tax burden on the senatorial class in the east which some could hardly pay. Surviving hoards of gold solidi from across the Danube are testimony to these payments. As for commerce, the chrysargyron (gold and silver tax, so-called because it had to be paid in gold and silver, usually, in practice, gold) was consistently unpopular and was abolished in AD 499 by the Emperor Anastasius as the collatio glebalis (follis), levied on senators, had been by the Emperor Marcian.⁵⁰ Taxation was an ideologically charged issue, and emperors who raised taxes, even if like Justinian they did so for military purposes, are uniformly criticized in contemporary sources.

The late Roman taxation system was a complicated and unwieldy affair, full of inequities and far from perfectly administered. Apart from the cost of the imperial court or courts and the administration, the state’s major item of expenditure went towards the maintenance of the army, and this was also the most difficult to organize. If it were true that Diocletian had really doubled, let alone quadrupled, the size of the army, as well as increasing the bureaucracy the economic problems of the later empire would indeed have been insuperable. Jones put the problem neatly in his famous statement that the late empire had too many ‘idle mouths’, i.e., non-producers, who had to be paid for from the diminishing resources of the empire.⁵¹ But few historians today would be as confident as Jones was in 1964. As we have seen already (Chapter 2), Diocletian is more likely to have regularized the status quo than actually doubled the army in size, and it must be regarded as doubtful whether even that figure could be maintained after the late fourth century. Even at 400,000 plus (other recent estimates would put it higher, as noted earlier), the late Roman army was still an extremely large force, and such an army must certainly have represented a great drain on resources. An elaborate system of requisitioning and supply had to be in place to get the items needed to the troops who needed them. Since the later third century and under the Diocletianic system, much of the army’s pay had been collected in kind, by means of a cumbersome set of arrangements which one is surprised to find working at all; in fact, while the method of calculation varied from province to province, it was possible from time to time to reduce the demand on a particular province, as with Achaea, Macedonia, Sicily, Numidia and Mauretania Sitifensis in the fifth century. However, the regular censuses necessary to keep the registers of land and population accurate tended not to be held, and great discrepancies could thereby arise. Once collected, the goods had to be transmitted to the necessary unit – a further process requiring complicated organization.

Other forms of taxation were also of great importance throughout the imperial period, especially the grain and oil requisitions for the food supply of Rome, a system Constantine also extended to his new foundation of Constantinople (Chapter 1). Since the Republic, the Roman government had considered it a priority to ensure the food supply for the capital, and had maintained free corn and bread doles for the purpose.⁵² The grain came largely though not only, from North Africa and Sicily, where in each case its provision had a major impact on the local economies.⁵³ In the case of Rome, the loss of North Africa to the Vandals caused severe disruption, but the distributions went on and were eventually taken over by the church. In the east, the grain and oil annona, as it was called, accounted for a high proportion of long-distance transportation between Alexandria and Constantinople, and can be traced in the pottery evidence.⁵⁴ Egypt continued to supply Constantinople, but the link was abruptly broken by the Persian invasion of Egypt in the early seventh century.⁵⁵ There were other problems: the tickets on which the actual distributions were made had come to be passed on by sale or inheritance; the government tried at times to regulate these practices too, but as time went on, as with other late Roman taxes, the match between those theoretically qualified and those actually receiving the dole had already become less and less close. Nevertheless the movement of goods in connection with tax requirements was a central factor in the continuance of long-distance exchange across the Mediterranean and an important contributor to the maintenance of Mediterranean unity.

The army was also a powerful factor affecting the circulation of coin and could act as an economic stimulus as well as a drain for the state. By the beginning of our period, payment in kind was beginning to be commuted into gold, especially in the west, at varying rates of commutation which at times had to be regulated by the government, and in the sixth century payment in gold (pounds or solidi) was the norm. Gold raised in taxes from the central provinces made its way to where army units were located, and soldiers spent their pay locally. State spending also followed the development of imperial residences and provincial centres such as Milan and Ravenna in the west and Antioch in the east. Maintenance of the cursus publicus was another example where state spending acted as a local stimulus.⁵⁶ However, the development of Constantinople as the eastern capital and its own massive needs deprived the west of eastern resources; it was no longer possible to maintain a paid army along the Rhine and upper Danube in the fifth century or the lower Danube in the sixth. Britain too was left to its own devices and the army withdrawn in 410. The establishment of rule from Constantinople in North Africa after Belisarius’s victory over the Vandals in 534 required major investment, as did the Justinianic wars in Italy and against the Persians. The presence of the army in a particular area, with all that it required for its maintenance – not just the pay and supplies of the troops, but also a good road system and transport and local support systems – could be a powerful economic stimulus and was no doubt one of the factors behind the undoubted prosperity and density of settlement in the fifth century in the south-eastern frontier areas, even in barren and difficult places such as the Hauran and the Negev, where there were many small military settlements in addition to the major fortresses. But by the sixth century much of the defence of the south-eastern part of the frontier area from Transjordan to Arabia had been left to Arab allies (Chapter 8), and barbarians were frequently used in the regular armies.

The gold solidus introduced in the early fourth century remained standard for centuries henceforth, but later emperors were unsuccessful in their attempts to reintroduce silver coinage on a stable basis. Inflation also continued, as can be seen from prices given in papyri, probably because the government mint-ed too much of the small base-metal coinage. By the fifth century, however, there was effectively nothing between the gold solidus and the tiny copper denominations, whose rate, valued against the solidus, was constantly changing. The later fifth-century emperors, in particular Anastasius (491–518), who introduced a new large follis (498), were more successful in introducing some stability; here it is interesting to find that Anastasius’ reform of the coinage continued trends already to be seen in the Vandalic and Ostrogothic coinages of the west.⁵⁷ The circulation and issuing of coinage was highly regionalized, with a system of local mints, and the gold and bronze currencies operated separately; nevertheless the economy was still highly monetized, and money changers became common, with banking also a significant element. The sixth-century church of San Vitale in Ravenna was built by a rich banker, Julius Argentarius, at a cost of 26,000 solidi.

Thus, as we have seen, the late Roman taxation system was highly complex and had many problems. Its most important part, the supplies for the army, was also the most difficult to organize. The regular censuses necessary to keep the registers of land and population accurate tended not to be held, and great discrepancies could arise. Once collected, finally, the goods had to be transmitted to the necessary unit – a further process requiring complicated organization. Commutation of tax in kind into gold, especially in the west, at varying rates of commutation which at times had to be regulated by the government, was a further factor, and in the sixth century payment in gold (pounds or solidi) was the norm. It is extremely difficult to assess the economic consequences of this state of affairs. On a priori grounds alone, the collection and distribution of taxes in kind, which continued in part well into the fifth century, and the constant fluctuation of value against the solidus of the base-metal/copper coinage (pecunia) would seem likely to have had a depressive effect on the existence of a market economy.

The Economics of Urban Change

There was, however, urban change at varying rates in different parts of the empire, and the degree of continuity or decline in urban contexts is one of the most contentious issues in current scholarship. This will be discussed further in Chapter 7; for now, it is enough to say that while there was great variety between east and west and within individual provinces, evidence from sites all over the empire suggests that the typical late Roman urban life, with peristyle houses and open public spaces, was by some time in the sixth century (the exact date postulated varies and is often a matter of controversy – the change came much earlier in the western provinces) giving way to an urban fabric which suited a different kind of social and economic life. Public buildings were not repaired, or fell out of use, their stone being used as spolia for purposes including fortification. Houses were subdivided; shops or workshops encroached on colonnaded streets. In the Balkans, urban sites began to give way to fortified hill settlements, while on the eastern frontier sites described by Procopius in his Buildings, such as Resafa/Sergiopolis near the Euphrates, were characterized by their military provision and fortifications and their large churches and episcopal buildings rather than by public streets or grand houses.⁵⁸ But it is difficult to generalize: the large urban site of Scythopolis in Palestine for example continued to flourish into the Umayyad period until it was hit by a major earthquake in 749, and the large number of ‘dead cities’ of northern Syria were able to maintain a prosperous lifestyle not based on state investment or the presence of large landowning magnates. Similar changes can be detected in different parts of the empire, as continues to be shown by a huge mass of recent and current work, but at very different rates and for different local reasons. They cannot be explained in simple or universally applicable economic terms, but rest on much deeper changes in society, and especially on the changing role of elites and degree of maintenance of the late Roman administrative structure.

Long-distance Trade and Exchange

Over the last few decades the study of the vast amount of pottery evidence, and in particular the diffusion of amphorae, the containers of the late Roman world, has transformed our understanding of specialized production and exchange. The pioneering work of John Hayes in the early 1970s provided a typology of late Roman pottery which enabled archaeologists to date excavated material far more securely and to log stratigraphic evidence; these findings were also extended into the study of late Roman amphorae, particularly by Italian archaeologists. The evidence from ancient shipwrecks sometimes provides valuable dating material of this kind.⁵⁹ With the growth of Constantinople and the diversion of Egyptian grain to the eastern capital, an eastern axis, Carthage/Constantinople, became important, and long-distance exchange continued into the fifth century, without serious break after the Vandal conquest of North Africa in 439;⁶⁰ this eastern axis continued in existence throughout the period, as the importance of Constantinople reached its peak, until the Persian and then the Arab invasions cut the connections between Constantinople and Egypt.

Many questions suggest themselves as a result of these conclusions, of which the following are only the most obvious. For example, whether these findings actually represent trading links (the evidence of pottery will not tell us the why of transmarine exchange, only the how). What if anything can they tell us about the impact of the post-Roman kingdoms in the west on the Mediterranean economy? Finally, to what extent does this evidence tally with that of urban change to suggest that a significant weakening of the Mediterranean system can be located in the later sixth to early seventh centuries, i.e., after the Justinianic attempt at reconquest and before the Arab invasions?

All these issues are the subject of ongoing debate and considerable disagreement, not least because they raise ideological issues about trade and the nature of the ancient economy. Carandini, Panella and their colleagues see the evidence as reflecting trading patterns in a market economy. Among the questions still to be settled is that of the overall economic impact of the Vandal conquest, including that of the effect on North Africa (and the other conquered areas) of the cessation of Roman taxation. Wickham rightly underlines the importance for North Africa of the grain requisitions for Rome, which would have had the effect of requiring a highly developed navigation and export system from which other products could also benefit, and whose cessation was therefore likely to have serious effects. According to this argument, the enforced grain exactions for Rome and Constantinople called forth a considerable level of production that was itself non-commercial, but which served to underpin commercial networks.⁶¹ According to Wickham, the eastern Mediterranean saw an agrarian boom in the fifth-sixth centuries and an ‘active commercial exchange network which linked Egypt, the Levant, and the Aegean in overlapping ways’; this system collapsed in three generations after 600, in the face of political change, but was ‘pretty stable’ until then.⁶²

The Fall of the Roman Empire

There is no simple way to characterize the late Roman economy, or the actual effect on society of the government’s attempts to control it. Certain trends are evident, not simply the profound impact of barbarian invasion and settlement during this period, but also more general developments such as the tendency towards the amassing of vast amounts of land by individuals, the return to taxation in coin (gold) instead of in kind, the growing gulf between east and west and the difficulty experienced by the government in ensuring the collection of revenues and staffing its own administration. The establishment of the post-Roman kingdoms in the west and the effects of the wars of reconquest had major economic repercussions; these will be discussed in Chapter 5. In the east, by contrast, there is evidence of population increase and of intensified agriculture and cultivation in areas such as the limestone massif of northern Syria and even in such unpromising areas as the Hauran and the Negev; this evidence will be discussed further in Chapter 7. On the other hand, by the late sixth century, following the effects of war and perhaps also plague, the Roman military presence in the east was clearly becoming harder and harder to maintain. Clearly it is misleading if not impossible to generalize over so wide an area and so eventful a chronological span. As we saw, older historiography connected a highly negative view of the supposed rigidity, corruption, and over-taxation of the later Roman empire with the reasons for the fall of the Roman empire, and modern historiography also abounds in confident value-judgements about decline and the end of antiquity, many of which rest on unacknowledged assumptions about the late Roman system. The current vogue for comparing the end of empires which has arisen since the collapse of the Soviet Union and more recently in relation to America, routinely draws on the ‘end of the Roman empire’ for comparisons, and has served to entrench these older views.⁶³ Much of the literature on the fall of Rome poses the question ‘Why?’ in terms of radical alternatives: either barbarian invasions or internal collapse; interestingly, for all his generally gloomy view of the late Roman system, A.H.M. Jones, opted for the former.

The late Roman administrative and economic system was certainly cumbersome and had many defects. Lacking modern communications, the state could neither operate efficiently nor respond easily to change. The government resorted all too easily to empty and hectoring legislation; officials did what they could, and often enriched themselves; the people learnt how to cheat the system. There is nothing surprising in that, and the power of inertia was also great. What is impressive, and remarkable in such a context, is rather that this highly traditional and very complex society did manage to survive so well for so long.