Cities, Taxes, and Empire

The Roman Empire on the eve of the Hunnic ascendancy remained what it had always been: an empire of cities.³ That is not to claim that the empire was urban in the modern sense of the word: in Rome, as in all pre-industrial societies, the overwhelming proportion of the emperors’ subjects lived and worked in the countryside, and it was agricultural land and those who worked it that were the main source of the taxes upon which the Roman state depended.⁴ Hence although, in the early fourth century, the Emperor Constantine had introduced a tax on mercantile profits (the so-called collatio lustralis or chrysargyron), in the late fifth century, the eastern emperor Anastasius felt free to abolish that tax without seemingly having to worry too much about the financial implications of his largesse.⁵ Moreover, in economic terms, the dividing line between city and countryside was not nearly as clear-cut as is sometimes supposed. The great imperial capital of Constantinople, for example, maintained extensive stretches of agricultural land within the embrace of its formidable walls, and in the economically most developed regions – such as Egypt (where perhaps up to a third or so of the population lived in what were formally termed cities) – the levels of artisanal production and commercialised exchange in the countryside rendered any economically functional distinction between the countryside as a zone of primary production, and the city as a centre for secondary and tertiary economic activities largely superfluous.⁶

Rather, the Roman Empire was quintessentially urban in terms of ideology and political culture, in that those settlements that were formally and legally recognised as cities served as the key nodal points of imperial administration and control. It was in the city (known in Latin as the civitas or in Greek as the polis) that the local governor resided with his officers and staff. Importantly, there he was joined by members of the locally dominant landowning families (drawn from the surrounding area known as the territorium or chôra), who were enrolled onto the city councils (curiae or boulai) of the empire, to which many administrative and fiscal responsibilities had traditionally been delegated, effectively rendering the Roman Empire of the first and second centuries what has been depicted as a “patchwork quilt” of self-governing communities.⁷ By the end of the fourth century, cities had lost some of their autonomy, but the basic model still held true.⁸

Indeed, in Rome’s northern and western provinces, where cities had been largely introduced as a direct product of Roman imperialism, the drawing in of regional elites to the local city, the enrollment of their members into the ranks of the council, their co-option into the imperial system, and their associated acquisition of Greco-Roman cultural values and political identity, had been central to how Rome as an empire had become provincially embedded by the start of the fourth century.⁹ To the east, an urban infrastructure had been inherited from the world of the Greeks and the Hellenistic kingdoms, from whom the Romans themselves had acquired their city-focused culture.¹⁰ Hundreds of years earlier, the Greek philosopher Aristotle had declared that man was by nature “a political animal” (politikon zôon), meaning that it was the city or polis that was a person’s natural habitat and most fulfilling abode.¹¹ By the fourth century, this position had become a deeply entrenched assumption in the minds of members of the Roman regional elites from the furthest reaches of Britain to the banks of the Euphrates.

The city was, therefore, the seat of government, the focus for elite ambitions, and a locus for the transmission and reproduction of high culture. In that sense, it was differentiated from other “dense” settlements (as human geographers would call them) both administratively and socially. As a corollary to this, they were also differentiated from other settlements in terms of their monumental architecture and appearance, in that they were expected to possess the great public monuments which, to the classical mind, characterised the city: processional paved roads, a properly ordered agora or forum to serve as a public marketplace, bath houses, places of worship, perhaps an acropolis, and (increasingly from the third century) impressive city walls. The city provided, in short, a specific type of monumentalised public space in which the Roman elite male could deport himself after the extrovert manner of the gentleman of antiquity, living his life in the constant gaze of his peers, his equals, and his rivals.¹²

The cities of the Roman Empire necessarily varied enormously in size and scale. This diversity in terms of size was particularly marked in the fourth and fifth centuries. Some cities, such as the great megalopoleis of the East, comprising Antioch, Alexandria, and Constantinople, covered areas substantially over one hundred hectares and possessed populations of perhaps up to half a million. Others, particularly in the more underdeveloped West, could be tiny in comparison, with populations of perhaps only a few hundred. Nevertheless, to contemporaries, all these were cities, and were to be differentiated as such from towns or villages that lacked their defining characteristics and official designation, even if these towns or villages were actually larger than the cities concerned. Broadly speaking, the largest cities tended to be those at which emperors and their courts resided: most obviously Rome and Constantinople, or those cities closer to the frontiers that were turned into imperial capitals in the third and fourth centuries to face down burgeoning military threats from beyond imperial territory (such as Trier, Ravenna, or Vienne [in Gaul]).¹³ So, for example, the anonymous author of the fourth-century gazetteer of the empire known as the Expositio totius mundi et gentium (Account of the Whole World and Its Peoples) wrote of the occasional imperial residence and capital of Antioch thus: “Take the city of Antioch, in which all delights are plentiful, especially those of the circus. Why, you ask, all delights? Because this is the seat of the emperor, and everything is necessary on his account.”¹⁴

Many of the cities of the Roman Empire of the fourth century, however, were clearly substantial institutions and – crucially – ones which were not, typically, an organic part of the surrounding landscape in which they stood. As a result, they could only maintain their population levels, their monumental character, and their prosperity by means of the tightest possible control over the agrarian resources of the surrounding countryside, and by means of the active support and the active subvention of the Roman state. Many of these cities, as we have seen, were the product of Roman imperial will and were only maintained in an alien landscape by imperial fiat.

This was most evident with respect to Rome and Constantinople, the inhabitants of which were critically dependant on the imperial shipment of vast quantities of grain, at enormous difficulty and expense, from North Africa and Egypt respectively.¹⁵ Without such long-distance grain supplies, these great imperial cities would have found it impossible to feed themselves and would have been obliged to contract dramatically (as each would do, in the sixth and seventh centuries respectively). But such arrangements, encountered in their most spectacular form in the cases of Rome and Constantinople, would also appear to have been repeated hundreds of times over (albeit at a reduced level) in relation to the other cities of the empire.¹⁶ Thus, writing in the sixth century, the Greek historian Procopius informs us that the annual shipment of grain from Egypt, via Alexandria, to Constantinople was meant not only to meet the needs of Constantinople, but also that such grain left over after the needs of the imperial city had been met was automatically assigned to what he simply describes as “the cities of the East.”¹⁷ He does not describe this process in any detail, as to him it was a perfectly standard feature of imperial life.

At the same time, all cities were fundamentally reliant on the fiscal infrastructure of the Roman state in a secondary sense, in that the role of cities as functioning centres of commercialised production and exchange, and their status as centres of elite residence, were critically dependent on the existence of a highly monetised economy, created and supported by the monetised fiscal demands of the Roman state. The state demanded that its subjects pay tax in coin to support the army and bureaucracy, and minted vast quantities of coin accordingly.¹⁸ This in turn enabled farmers and landowners to sell their goods at market to pay their taxes.¹⁹ At the same time, however, it allowed landowners, by selling the produce of their estates in return for coin, to escape the narrow confines of rural life and instead, by living at some remove from the physical source of their wealth, to maintain the quintessentially urban existence that Roman elite culture demanded. Cities, taxes, and empire thus went hand in hand, and, as we shall see, both the fiscal and urban infrastructures that Rome had generated and sustained would have difficulty surviving when the empire itself began to fragment under barbarian pressure.

Transforming the Countryside

As already indicated, there were, of course, regional differences in the urban economies and cultures of the Roman Empire of the fourth century. Urbanism was most firmly rooted in the Greek-speaking East. In Rome’s northern and western provinces, by contrast, although landowners everywhere aspired to an urban existence, they appear to have spent more time in their rural villas than elsewhere. In certain of Rome’s frontier territories, moreover, such as north-western Gaul, there are indications that the chronic military insecurity of the third century had dealt a blow to urban life from which some cities were never to recover.²⁰ This stands in marked contrast to the rest of the empire, where the fourth century would witness rising population levels and urban growth. In general terms, the fourth and fifth centuries would witness urban expansion to the south and east, and growing evidence of stagnation and decline to the north and west.²¹ These regional trajectories were driven above all by objective military circumstances, to which cities and their populations were highly sensitive.

Likewise, there were significant regional differences in the underlying structure of the agrarian economy, beyond those dictated by variations in climate, soil and crop. Again, in broad terms, distinctions are commonly drawn between east and west. To the east, where the urban focus of elite culture was most pronounced, rural society was characterised to a high degree by the existence of nucleated village settlements notable (in parts of Syria, for example) for their institutional cohesion and strong sense of identity. To the west, patterns of settlement tended to be much more dispersed, and were increasingly given coherence primarily by local networks of villas.²² Both east and west, however, the interests and needs of the countryside were subordinated to those of the city. In the context of a world that was often perched on a knife edge between sufficiency and famine, this could have devastating implications for the lives of countryfolk.²³ This emerges with horrific clarity from the writings of the Roman medical author Galen. In his treatise “On Wholesome and Unwholesome Foods,” written in the second century, Galen draws a distinction between the effects of bad harvests on the rich and the poor, and on city dwellers and the inhabitants of the countryside. Describing one particular year of bad yields in Asia Minor, Galen wrote of how:

Galen goes on to describe how, predictably, many of these wretches soon died. The lives of the empire’s rural subjects were also framed and determined by the ambitions of those same city dwellers in the sense that, as noted earlier, the countryside was the main source of elite wealth: the city councillors and senators of the late Roman world may chiefly have been men of the city, but they were also landowners who were constantly on the lookout to increase the extent and productivity of their rural estates.

There is every indication that the fourth century was a period when such estates (worked by both slave and “free” labour) were on the rise. The ever more widespread dissemination from the reign of the Emperor Constantine onwards of a new gold currency – the solidus – greatly facilitated the monetisation of the Roman economy, and with it the opportunities for commercialised production and exchange from which large estates (with the advantages they offered in terms of economies of scale and the more rational organisation of labour) were well placed to profit.²⁵ In particular, feeding the expanding urban populations of the fourth century could be highly lucrative for the owners of large estates, not least given the fact that, as members of the local city council to whom the organisation of civic food supplies was traditionally delegated, such men could effectively rig the market to line their own pockets. In the city of Antioch, for example, members of the council – in cahoots with local merchants – are reported to have deliberately withheld grain at a time of famine in order to send its market price skyrocketing. As ever, the poor man’s hunger was the rich man’s opportunity.²⁶

Above all, the fourth century witnessed a dramatic increase in the size of the imperial bureaucracy. In response to a “crisis of under-governance” that had bedevilled emperors amid the military crisis of the third century, rulers from Diocletian (284–305) onwards had considerably expanded the numbers of those provincial administrators and officials (both military and civilian) directly employed by the Roman state. These new posts were primarily recruited for from amongst the ranks of the leading families of the cities of the provinces, whose members were increasingly enrolled into the senatorial order both east and west, and thus came to enter into a new and more direct relationship with imperial power that served to bolster their own resources of authority, wealth, and prestige. The holders of these new governmental posts, in short, formed the kernel of what would become a new imperial aristocracy of service, whose members exercised authority and intermarried at a transregional level, thus allowing their power to expand beyond the territorial confines of their home towns.²⁷ As the social and economic clout of members of this newly transregional elite snowballed, so too did they increasingly invest their authority and wealth in land, leading to a growing concentration of landownership across the Roman world, which in turn resulted in the creation of what has been vividly described as “an increasingly proletarianised peasantry.”²⁸ It should also be noted that some of the best evidence we have for rural slavery in the Roman world comes from precisely this period.²⁹

This concentration of landownership necessarily took different forms in different regions, but everywhere a fundamental restructuring of agrarian social relations is evident across the fourth and early fifth centuries (one which the imperial authorities would effectively set in legal stone through the development of the institution of the “adscript colonate,” which served to bind agricultural labourers and their families to the expanding large estates).³⁰ In Egypt, the documentary papyri reveal a proliferation and expansion of directly managed properties, which would culminate in the sixth century in the crystallization of extensive landed enterprises such as that of the Apion family recorded around the Middle Egyptian city of Oxyrhynchus, from which a rich cache of documentary texts survives. These texts reveal the Apion estates (which first become visible in the early fifth century) to have been highly commercialised, with agriculture carefully managed and geared towards maximising the family’s income in gold. The estates combined landholdings in and around villages (kômai) and urban property in the city of Oxyrhynchus, with entirely estate-owned labour settlements (epoikia), to which local labourers and their families were drawn. There, these agricultural workers were rewarded with wages in the form of cash, credit, share of crop, or access to plots of land in return for their commodified labour. These estates formed only one part of the Apion property portfolio, and we have evidence for the existence of higher levels of estate management beyond the Oxyrhynchite in both Alexandria and Constantinople, where members of the family held high office.³¹

The spread of similarly structured estates is recorded in the epigraphic evidence for Palestine at around the same time, which again attests to a proliferation of estate-owned epoikia.³² Likewise, inscriptions from the rich coastal zone of western Asia Minor reveal a wave of fourth-century estate expansion, with landowners extending their control over both individual parcels of land (agroi) and entire villages (chôria) worked by a combination of resident labourers (paroikoi) and slaves.³³ A similar pattern appears to have pertained in Syria, where the churchman (and future patriarch of Constantinople) John Chrysostom denounced the exploitation of peasants drawn onto the estates of the powerful.³⁴

At the same time, entire villages (kômai) were passing into the private ownership of such men: the Antiochene rhetor Libanius, for example, contrasted those “large villages” (kômai megalai) in the vicinity of his native city belonging to many masters, to others which belonged to a single lord. Local military “bigwigs,” he complained, were attempting to extend their patronage and control over both.³⁵ Certainly, by the sixth century, we know of entire villages – from Palestine to Armenia – that were being purchased outright by the wealthy, or which were being awarded by the imperial government to its favoured clients.³⁶

In the village-dominated world of the East, the rise of the great estate thus combined the creation of estate labour settlements with the extension of aristocratic control over what were often deeply rooted and long-standing village communities. In the West, by contrast, where settlement patterns were more dispersed, the process of aristocratic enrichment took a slightly different form, and is most visible archaeologically in what has been termed a fourth-century “villa boom” discernible from Britain to Sicily, whereby villa complexes simultaneously proliferated and expanded, becoming at once both more sophisticated and highly capitalised.³⁷

Indeed, by the late fourth century, members of the western senatorial order (now dominated by members of the new service aristocracy) had become proverbial for the productivity and extent of their estates. The contemporary historian Ammianus Marcellinus, for example, related how they would “hold forth unasked on the immense extent of their family property, multiplying in the imagination the annual produce of their fertile lands which extend, they boastfully declare, from farthest east to farthest west.”³⁸ For the early fifth century, the Greek account of the life of Saint Melania records that she and her husband owned property throughout the western provinces: in Italy, Spain, Sicily, Africa, Mauretania, Britain, and, the hagiographer adds, “the other regions.”³⁹

It is sometimes suggested that, perhaps in contrast to the forms of direct management encountered in the East, members of the late Roman aristocracy in the West tended simply to lease out their land to tenant farmers in return for economically static rental incomes. The evidence for this claim, however, is meagre. Rather, both eastern and western sources indicate (at least with respect to the best quality land) a marked aristocratic preference for the direct management of estates, such as we encounter in the documentary papyri from Egypt. Regimes of direct management, for example, are taken entirely for granted in the agronomic texts that survive from Late Antiquity in both Latin and Greek.⁴⁰ Likewise, the fifth-century Gallic landowner Paulinus of Pella (who also inherited property in Greece) recorded how, as a young dominus, he had personally driven on his estate workers.⁴¹ Both east and west, such landowners were also characterised by their commercial drive and their lust for gold. The fifth-century historian Olympiodorus of Thebes, for example, provides estimates for the annual incomes of the leading western senatorial families which reveal a marked preference for hard cash, whilst an eastern law contained in the Theodosian Code exempted from the tax on mercantile profits those landowners who marketed the produce of their estates (with or without the help of intermediaries): such men, the emperor declared, “should be thought of not so much as merchants, but rather as skilled and zealous masters.”⁴² Again, such economic strategies would come under pressure in the West in the fifth century as objective military and political conditions combined to curtail opportunities for commercialised agriculture and monetised exchange. Before attempting to understand that process, however, we must first turn to the world beyond Rome, beginning with social and economic conditions to the East.

The World beyond Rome

The Roman-Persian frontier in the fourth and fifth centuries bisected a Mesopotamian world that essentially formed a linguistic, cultural, social, and economic whole. As a result, it should not surprise us that social and economic conditions on the Persian side of the frontier should have broadly mirrored those encountered in Roman territory. Like Roman Syria, Sasanian Assyria (Iraq) was highly urbanised. Moreover, like the Roman Empire, the Sasanian Empire in general was a highly ranked society characterised by stark disparities of wealth.⁴³ Indeed, it is possible to argue that over the course of the fourth and fifth centuries, East Roman and Sasanian social relations were drawing closer together. The outright ownership of villages and control of their inhabitants that aristocrats in Syria, Palestine, and Egypt were aspiring to in this period, for example, would appear to have been something that members of the great Iranian noble families had long been able to take for granted. If peasantries in the East were becoming increasingly vulnerable and insecure over the course of the fourth and fifth centuries, it was perhaps in the core territories of the Sasanian Empire that their state of dependence and subjugation was at its most complete. The Roman historian Ammianus Marcellinus (who saw active military service in Persia) claimed that members of the Sasanian nobility enjoyed rights of life and death not only over their slaves, but also over what he termed the “undifferentiated masses” (plebs obscura).⁴⁴

The societies to Rome’s immediate north, by contrast, although also ranked societies, dominated by an archaeologically visible military elite, were characterised by much shallower social hierarchies than those encountered in either the Roman or Sasanian Empires. Both in the realm of the West Germanic peoples beyond the Rhine and also in the territory occupied by the Goths and other East Germanic and Slavonic peoples between the Danube and Crimea, the evidence points to a landscape primarily inhabited by sedentary agriculturalists, whose first instinct was to remain wedded to the land that they worked unless obliged to migrate by objective military or climatic conditions.⁴⁵ Although, at the level of village society, domestic slavery clearly existed, this was a world dominated by free peasant producers to an extent that was inconceivable in much of the territory ruled by Rome.⁴⁶

This comes across very clearly with respect to the Goths. Warlords and their retinues played an important role in fourth-century Gothic society. The most successful or forceful of these men were capable of projecting their authority over extensive areas. Military kingship itself, however, which would become so important to the Goths once settled in Roman territory, appears to have been a relatively embryonic social institution even in the years immediately following the crossing of the Danube in 376, and the “weapon-bearing free” embraced a much broader swathe of society than those who clustered around such characters as aspired to it.⁴⁷

Thus whilst we do have archaeological evidence for sites of elite residence in Gothic-settled territory for the period before the 370s (some of it revealing signs of Roman architectural influence), and there are indications that disparities of wealth and power were widening by virtue of economic and cultural contact with Rome, what is most striking about the archaeology of Gothic settlement between the Danube and Crimea from the late third to mid-fourth centuries is the extent to which it bears witness to the existence of a vibrant peasant society rooted in an economy devoted largely to subsistence agriculture (although there are considerable indications of craft specialisation and the possible use of Roman coinage as a medium of exchange).⁴⁸

This impression is reinforced by a fourth-century text, the Passion of Saint Saba, recording the sufferings of a Gothic convert to Christianity, who was martyred on Gothic territory beyond the Danube in 372. The account is laced with vignettes concerning the nature of village society and the relationship between the villagers and the locally dominant warlord, who is depicted as living beyond the village, and the interventions of whose followers are depicted as unwelcome, destabilising, and – from the perspective of the locals – best ignored or, if not ignored, subverted.⁴⁹ Thus when the holy Saba is commanded to publicly abjure his faith by eating the flesh of pagan sacrificial victims, the villagers offer to get him out of a tricky situation by secretly swapping the polluted meat with something more innocuous (a suggestion he inevitably refuses). There are incidental references to the existence of a village council and of gatherings of a broader body of villagers. The accoutrements of village life make fleeting appearances: wagons, wooden roof-beams, a woman preparing dinner. Only when the holy man insults the Gothic lord in the presence of his retainers was his fate sealed and his martyrdom secured.

Huns, Goths, Romans, and the Demise of the Imperial Economy

Naturally, those Goths who had settled on the Pontic steppe in southern Russia, were obliged to adapt to ecological and cultural conditions very different to those encountered to the immediate north of the Roman Empire’s Danubian frontier. Here, many historical societies have found it useful to differentiate between those nomadic or transhumant groups inhabiting the open steppe, who tend to support themselves from the rearing of stock (primarily cattle and horses), and the sedentary agriculturalists of the woodland zone (hence the Slavonic poljane “field dwellers” and drevljane “wood dwellers”).⁵⁰ As philologists have suggested, the Gothic division into Greutungi (derived from a Germanic root for “sandy soil”) and Tervingini (from a root for “tree”) that we find in the fourth century mirrors this distinction and thus would appear to have been a product of ecological realities.⁵¹ From the world of the steppe, the Gothic language would also acquire the word for horses bred for warfare (Old Gothic [h]ros), and technical terms for warlords leading units made up of a hundred and a thousand men (hundafaths and thusundithafs respectively – words reflecting the organisation of nomadic military formations).⁵²

As these loan words would suggest, the encounter between the Germanic peoples and the Huns would appear to have primarily affected the former at the level of military and elite culture. This is clear, for example, from the Hunnic tradition of skull-binding that the Gothic elite adopted under nomad influence; from their adoption of Hunnic names; and from the deportment on horseback of their leading men in battle, which echoed the martial traditions of the steppe.⁵³ Likewise, with respect to the West Germanic peoples, we should note the burial of horses encountered in conjunction with the gravesite of the late fifth-century Frankish king Childeric, a style of elite burial attested from the Rhine to Manchuria, which we can probably trace back to the era of Hunnic ascendancy.⁵⁴ In economic terms, the important point to note is that, in spite of nomadic overlordship, the Germanic peoples’ preference for the lifestyle of the sedentary agriculturalist remained intact.

This fact would have major implications for the Roman Empire. As noted elsewhere in this volume, the Hunnic elite of the fourth and fifth centuries (like that of the Avars and Turks in the sixth and seventh centuries thereafter) brought with them from the world of China a taste for tribute and a predisposition to rule from cities.⁵⁵ As a result, theirs would be an essentially parasitic relationship with the Roman state, predicated upon the pumping out from imperial coffers of as much gold as possible. However devastating the military consequences of Hunnic aggression, therefore, Hunnic demands were ultimately dependent upon the smooth operation of the Roman fiscal economy. Like any hungry parasite, the Huns required a healthy host, and this meant preserving Roman fiscal, urban, and economic structures substantially intact (just as the Arabs would preserve and feed off Roman economic and fiscal structures in the Near East in the seventh century).⁵⁶

Those Germanic warlords and their allies who crossed into Roman territory in the 370s and the early years of the fifth century, by contrast, primarily wanted good-quality agricultural land, and were determined to acquire it, by negotiation or force. The hard-line eastern bishop Synesius of Cyrene, for example, was highly critical of how, in the aftermath of the battle of Adrianople, the emperor Theodosius I bought peace with the Goths by “generously making them gifts of land.”⁵⁷ Likewise, the Chronicle of Hydatius records how, around 411, the invading Alans, Vandals, and Sueves in Spain “apportioned to themselves by lot tracts of the provinces to inhabit” (“sorte ad inhabitandum sibi provinciarum dividunt regiones”).⁵⁸ In Gaul, Italy, and Africa, the settlement of Goths, Franks, Burgundians, and Vandals would on every occasion ultimately be associated with the establishment of these invaders as landowners. As the Burgundian evidence in particular makes clear, this meant the acquisition of land not only on the part of the “barbarian” leadership, but also on the part of the military rank-and-file.⁵⁹

Good-quality land always has a master, and lands belonging to the imperial government, or to absentee landowners, are likely to have been confiscated outright. There are indications, for example, that the landed wealth of the Merovingian dynasty, concentrated around the “Paris basin” in northern Gaul, owed much to what had once been imperial estates that had passed into Frankish control. In Vandal Africa too, much land was confiscated from aristocrats resident in Italy. Alternatively, land could be taken from those who were still present. The enrichment of the Burgundian rank-and-file, for example, was evidently achieved at the expense of local Roman landowners, who were obliged to cede portions of their estates in return for protection and peace. A similar division of estates occurred in southern Gaul, Italy, and Spain, to the benefit of the Goths.⁶⁰

An important ramification of the barbarian settlement of the fifth century, therefore, was an undermining of the economic dominance of members of the late Roman aristocracy in the West. By virtue of the same phenomenon, the economic and social significance of small- or medium-scale landowners, such as the Burgundian soldiery, was bolstered (although directly managed large estates continued to exist). But for the structures of the underlying fiscal economy in those regions that passed out of direct Roman control, the consequences were to be far more pronounced. For, as noted in the first section of this essay, the highly monetised and highly urbanised economy of the Roman Empire of the fourth century was primarily sustained and given cohesion by the monetised fiscal demands of the Roman state. By minting and circulating vast quantities of coinage with which to pay the army and the civil service, and by demanding that taxes be paid in coin, the Roman state had catalysed and facilitated a much broader and deeper monetisation of the economy. At the same time, it had used its wealth to support cities, and had rendered practicable the urban aspirations of members of the provincial elite.

By virtue of the barbarian “land grabs” of the fifth century, however, local society in the West would become increasingly dependent militarily not on a standing army primarily paid in coin, but rather on bands of barbarian warriors, the majority of whom were rewarded with land in return for their military service. As a result, both the fiscal and monetary structures inherited from the Roman state began to fade away, as the collection of the land tax in coin became both less necessary and more politically problematic, and as post-Roman rulers responded by minting coin in significantly diminished volume. As coinage became scarcer, trade began to contract, and as trade contracted, so too did the urban lifestyle of surviving members of the Roman elite become increasingly difficult to maintain, leading in most regions to a final aristocratic flight from civitas to villa that would be complete by the seventh century. The ultimate effect of the barbarian settlement in the West, therefore, would be a dismantling of much of the sophisticated economic infrastructure that had come to characterise the Roman world.⁶¹ If Attila and the Huns had wanted an ever-larger slice of the fiscal cake of the late Roman state, many of his erstwhile followers ended up effectively closing down the metaphorical bakery. This was not, however, so much the result of the greater destructiveness of the Germanic invaders as compared to the Huns, but rather, it was an expression of the differing priorities and contrasting economic cultures that they had brought with them into Roman territory.