Let us recall a story told earlier. In the winter of AD 187–188, a man called Maternus, whom the ancient sources describe as a ‘deserter’ and ‘bandit’, set out for Rome at the head of an improvised army with the intention of assassinating Commodus and replacing him as emperor. Such was the ‘insecurity throughout Gaul and Spain’ and the ‘countless numbers’ mobilized that a major military operation was necessary to suppress the revolt. Maternus himself was betrayed, captured and beheaded.
About 20 years later, in AD 206–207, we hear that a man called Bulla or Felix was at large in Italy at the head of a band of some 600 men. For two years he ‘continued to plunder Italy under the very noses of the emperor and a multitude of soldiers’. He is reputed to have sent a message to the Roman authorities saying, ‘Feed your slaves to stop them becoming bandits.’ And later, after capture, when asked by his Roman interrogator, the Praetorian Prefect Papinian, ‘Why did you become a bandit?’, to have answered, ‘Why are you a prefect?’(7)
Scattered references to banditry pepper ancient accounts of the anarchy. Philip the Arab (AD 244–249) stationed units in Italy as a defence against ‘robbers and pirates’. Probus (AD 276–282) campaigned against a bandit-chief called Lydius, whose army of peasant outlaws had evicted the Roman authorities from much of the countryside of Lycia and Pamphylia in southern Turkey. The Isaurians, whose mountain territory lay immediately to the east, were also, it seems, out of control. The ancient sources demonize the Isaurians as bandits, pirates and serial rebels, reporting numerous campaigns against them down the centuries. Diocletian, in AD 284–285, ordered the suppression of rebels known as bagaudae. This is the first of many references to bagaudae in the ancient sources. They mounted a series of rebellions in Gaul, Spain and possibly Britain between the late 3rd and mid 5th centuries AD. Even when not specifically referred to, we can sometimes assume their identity. Ammianus Marcellinus, for example, our best historical source for the 4th century AD, refers mysteriously to ‘many battles fought in various parts of Gaul’ early in the reign of Valentinian (AD 364–375), but these he considers ‘less worthy of narration’ than fights with Germanic tribesmen, it being ‘superfluous to describe them, both because their outcome led to nothing worthwhile, and because it is unbecoming to prolong a history with ignoble details.’(8) Here, surely, are bagaudae waging guerrilla war: something likely to bore and irritate an aristocratic army officer and military historian like Ammianus.
Later, in the 5th century, references to bagaudae become frequent and explicit. Risings are recorded in Gaul in AD 407–417, 435–437, 442 and perhaps 448; in Spain in 441, 443, 449, 454 and 456; and possibly in Britain in 408. These references imply revolts serious enough to warrant intervention by imperial troops – and therefore notice from historians. The further implication is that banditry was extensive and endemic in much of the countryside of the Western Roman Empire during the 5th century. Eric Hobsbawm has argued that rural ‘social banditry’ is, in fact, a normal state of affairs in pre-capitalist class societies where those who work the land are exploited by landlords and governments. Mostly it exists only as a nagging irritant, for the bandits rarely amount to more than one in a thousand of the rural population, and a typical band numbers no more than 10 or 20 hiding out in some remote spot and supporting itself by preying on big estates, tax collectors and rich travellers. Recruited largely from the rural underclass of escaped slaves and serfs, army deserters and fugitives from the law, and impoverished crofters unable to make a living on the land, the bandits still remained part of peasant society, enjoying a measure of protection and support – even sometimes being lionized as champions of the poor. As such, if conditions in the countryside worsened considerably, and if the state’s repressive power weakened, social banditry could flare into peasant revolt.
Bulla’s two recorded statements reveal radical consciousness. Occasional references to the practices of bagaudae reveal the same. The senator and poet Rutilius Namatianus records that his relative Exuperantius, in suppressing bagaudic revolt in Gaul in AD 417, had ‘restored the laws and brought back liberty, and did not allow the Bretons to be slaves of their own domestics’.(9) The world, it seemed, had momentarily been turned upside down. It was not law and liberty as such that needed restoring, of course, only the law and liberty of the ruling class. That much is clear from another reference. A comedy written by an unknown author around the same date makes jokes about life beside the Loire, where men live by popular laws, peasants make speeches, capital sentences are pronounced under an oak tree and recorded on bones, and ‘anything goes’: a satire, apparently, not on a world without law or liberty, but on one without landlords, tax-collectors and police.
A similar spirit was at large in the countryside of Roman North Africa. Here wandering bands of poor peasants were inspired by the simple rural piety of the Donatist Church. Known as circumcelliones in reference to the shrines of martyrs where they gathered, they had their own communal organization, collective rituals, and distinctive style of dress. Many courted martyrdom – even a bitter enemy like St Augustine conceded that ‘they lived as robbers but died as circumcelliones and were honoured as martyrs’.(10) Sometimes, crazed by religious enthusiasm, whole groups would commit suicide by jumping off a cliff or setting fire to themselves. Others plundered villas and churches. Some organized collective resistance to debt-collectors. One source records bands ofcircumcelliones forcing rich men out of their carriages and making them run behind while their slaves took their seats. The combination of social and religious radicalism attested by the ancient writers appears to make the circumcelliones precursors of the millenarian sects of the Middle Ages.
If rural banditry and revolt were as endemic as they seem in the Late Roman Empire, it is not difficult to suggest reasons. In AD 238 the villagers of Scaptopara in Thrace petitioned the emperor to complain that soldiers and officials attending an annual festival nearby regularly demanded accommodation and supplies without payment. So heavy was the burden that many villagers had abandoned their ancestral farms and let their fields return to waste. Earlier complaints had prompted the provincial governor to issue an edict ordering that the villagers be left alone. But it had provided only temporary respite, and soon the problem was as bad as ever. Little wonder: soldiers and officials took their lead from their masters. Two contemporary historians, Dio Cassius and Herodian, describe the financial ruthlessness of the military monarchy and its consequences. ‘Nobody in the world should have money but I,’ Dio has Caracalla (AD 211–217) exclaim, ‘so that I may bestow it on the soldiers.’ Later, facing the censure of his mother for excessive expenditure, he is said to have exhibited his sword and declared, ‘Be of good cheer, mother, for as long as we have this, we shall not run short of money.’(11) Herodian compares the ravages of Maximinus Thrax (AD 235–238) to a barbarian invasion: ‘Maximinus, after reducing most of the notable houses to poverty … began to lay hands on the public treasuries. He expropriated whatever public moneys there were – funds which had been collected for the grain-supply, or for distribution to the people, or earmarked for shows or festivals. Dedications in temples, statues of gods, honours to heroes, and whatever embellishment there was of a public nature, or adornment of a city, or material out of which money could be made – he melted all of it.’(12) Dio and Herodian lament especially the spoliation of aristocratic estates, but, in speaking up for the property rights of their class, they illuminate the exploitation of all. Yet more informative are surviving Roman law codes.
The Theodosian Code (published in the East in AD 438) was a fairly comprehensive record of laws issued in the previous 40 years, and a more patchy record of laws from the time of Constantine onwards. The Justinianic Code (AD 529) and the Justinianic Digest (AD 533) superseded the earlier code, eliminating obsolete laws and amending, abbreviating and reordering those that remained in force; thus, like the Theodosian Code, these documents also preserve many laws dating back to the early 4th century AD. The law codes are key sources for official policies, the detail of administration, the social structure, and aspects of everyday life in town and country. It is from the law codes that we learn that many of the peasants had become coloni: though terms and conditions varied from place to place, generally, it seems, this meant they were serfs, that is, men who were tied to the service of a particular estate. For the state, this had the advantage that taxes and labour services could be more easily levied. For the landlords, it denied tenants their principal means of redress – seeking a better tenancy elsewhere – and thus allowed the rate of exploitation to be ratcheted up. But if the law, in recording the rise of serfdom, implies growing oppression in the Late Roman countryside, it also, in its frequent references to agri deserti, alerts us to one of the fatal consequences of this. Agri deserti meant ‘abandoned fields’: land that it had once been profitable to cultivate now too burdened with tax, rent, debt and labour service to be viable. A.H.M. Jones, the great historian of the Late Roman Empire, estimated that perhaps the poorest 20 per cent or so of land went out of use. He argues that, while some land may have been abandoned due to shortage of agricultural labour or in some areas to insecurity, in the main it was caused by the high and increasing rate of taxation, which reduced the net return on marginal land to vanishing point. Archaeology supports the notion of a slow-working agricultural depression in the Late Empire, beginning as early as the 2nd century in places, becoming more widespread in the 3rd, gathering strength through the 4th, and bringing whole regions to ruin in the 5th.
The evidence for most parts of the empire remains anecdotal and impressionistic, but that for Roman Britain is based on systematic surveys of excavation data. A sample of 78 Romano-British villas revealed a steady climb to a peak in occupation around AD 325 – the true ‘golden age’ of villa civilization in Britain, when many sites had been expanded into great country houses with suites of dining-rooms, bedrooms and baths opening on to colonnaded gardens. Thereafter, however, there was decline: some villas had been abandoned as early as AD 350, more than a third by 375, and well over two-thirds by the end of the century. Around AD 400, in fact, all building-work had ceased on villa sites, and within a decade or so elite occupation had been completely terminated. Occasional claims for ‘continuity of occupation’ turn out to be unfounded: at a handful of sites there is evidence that former country houses were converted into working farms; there is not a single site in the entire British archaeological record where elite occupation at a villa can be demonstrated as late as AD 425.
Decline is also apparent among the villages, hamlets and farmsteads where the Romano-British peasantry dwelt. A survey of 177 sites spread randomly across the country revealed that, between a 2nd century AD peak and the late 4th century, some 37 per cent of settlements were abandoned. A more detailed survey of 317 sites in the Severn Valley and Welsh Marches region showed a drop of 27 per cent over a similar period. Shortly after AD 400, moreover, virtually all Romano-British rural sites appear to have been abandoned; only gradually over the succeeding decades does a new rural settlement pattern emerge in the archaeological record. All of the evidence – increasing peasant resistance, complaints about exploitation, references to agri deserti, the end of the villas, the abandonment of native rural settlements – confirms the impression of a generalized and steadily worsening agricultural depression in the Late Roman Empire.
Little wonder, then, that the decurionate was in crisis. The law codes list a series of edicts making service on local town councils compulsory and imposing stiff sanctions on absentees. Traditionally the gentry – legally defined by regular censuses as men endowed with estates and other property of a certain minimum value – had served willingly on local councils, often competing energetically for elected office and distinguishing themselves in acts of public benefaction. The provincial towns of the 2nd century AD, with their basilicas, temples, baths and fine houses, were monuments to the civic-mindedness of this class. But as the military monarchy battened on to civil society, as taxes, requisitions and corvées fell more heavily, as resentment and resistance mounted, public service became more burden than honour, and many men withdrew, eschewing politics, administration and the law, retreating into the country, where they managed their estates and embellished their villas. The state pursued them there, demanding a return to duty. When some avoided service by joining the army, the civil service or the clergy, the law thundered against them: such persons were to be tracked down, dragged back, forced to perform their municipal duties. Centrally appointed town governors –curatores – were dispatched to administer the municipalities. Boards of ten – decemprimi – were formed of the leading local notables – principales – and given authority to impose on their more junior colleagues and enforce attendance and service. But edicts and orders from on high constitute intentions, not reality, and the archaeology of the towns of the Roman Empire argues for failure and a steady decline in urban life.
Across the empire, the great age of urban public building came to an end with a final flourish in the early 3rd century AD. At Leptis Magna in Libya are to be found the ruins of some two dozen major public buildings erected between the reigns of Augustus and Severus. Every imaginable type of Roman public architecture is represented: forums, basilicas, temples, baths, theatres, amphitheatres, circuses, monumental arches, public fountains and shopping malls. The ruins are dominated by the Augustan-period theatre, the Hadrianic public-baths, and the imperial forum, colonnaded street and harbour-works of Septimius Severus. Then it all stopped. Almost nothing was built thereafter, with the notable exceptions of a circuit of defensive walls enclosing most (but not all) of the existing city in c. AD 250–350, and the conversion of some pagan temples into Christian churches in the later 4th century AD. Leptis, moreover, is typical. Except for refurbishments, conversions and defence-works, urban building virtually ceased in most provincial towns after the early 3rd century AD. The evidence from 17 Romano-British towns has been collated and analysed. It reveals that the towns were growing from the late 1st to the early 3rd century AD, with heavy expenditure on public buildings in AD 75–150, followed by the construction of elite town-houses in AD 150–225. The middle years of the 3rd century, however, were a time of crisis, with a collapse in civil construction and diversion of resources into building town walls. There was some recovery in the early 4thcentury, but this was limited and faltering, with no return to the boom conditions of the 2nd century, and virtually all construction work had ceased by AD 400. Population levels mirrored this decline in building: they reached a peak in the early 3rdcentury, retained this level for about a century, but then declined dramatically – down more than a quarter by AD 350, more than a half by AD 375, and collapsing to a mere eighth of peak levels by the end of the century.
We cannot know how successful were the Late Roman state’s efforts to dragoon reluctant gentry into taking their seats in local council chambers. What we do know is that the infrastructure of the towns in which those chambers lay was degrading from the early 3rd century onwards. Indeed, in the course of the 4th century, most towns in the Western Roman Empire lost their distinctive urban characteristics. Typically, by c. AD 400, a Late Roman ‘town’ comprised a circuit of strong walls, often enclosing only a small inner enclave, within which most public buildings were in a state of ruin, and most private houses abandoned or taken over by lower-class ‘squatters’. Often there is evidence for some sort of administrative centre, one or two churches, a few grand residences, perhaps a large warehouse or two, a small garrison of soldiers, and a much-reduced plebeian population of artisans, traders and labourers. Heaps of stinking refuse and sewage frequently clogged the empty houses, yards and back-streets of abandoned districts. Populations of a few hundred were probably common.
Town and country were in crisis. Municipal aristocrats rarely came to town, were no longer interested in public building, and more often than not failed to maintain their residences. But at their country seats, too, there were problems: not at first perhaps, but certainly from the mid 4th century onwards, landowners discovered that the burdens on estates meant dwindling returns. Soon the great country houses were too expensive to maintain: frescoes flaked and were not replaced; mosaics were holed and badly patched; the water-channel got blocked and the bath-house ran dry. Beyond, in the villages, there was grinding poverty and sullen resentment. Some of the outlying farms were in ruins, the fields overgrown; others were short of labour, animals, equipment, and the resources and will to make good. Many peasants had simply fled, disappearing perhaps to another estate, perhaps to eke out a living in the wilderness, perhaps to join the outlaws and hold up travellers on the remoter roads. There were, in short, three worlds of Late Antiquity: the world of the imperial grandees, of emperors, generals, courtiers and bishops; the world of the provincial gentry, of ruined towns, crumbling villas and bankrupt estates; and the world of the peasantry, of tax, rent, debt, corvées, and a desperate struggle to survive on the margins of existence. Because these three worlds were linked, because grandees needed gentry to manage their empire, and peasants to create its wealth, the great edifice of Roman imperial power was, by the 4th century, resting on a foundation of crumbling sand.
The Western Roman Empire endured for as long as it did only because its myriad discontents could never be organized into a revolutionary force capable of challenging the military-bureaucratic complex. Decurions might abscond to their country seats and there grow bitter as ancestral estates decayed. But, divided between a thousand local-government districts, their outlook was ever a parochial one, and the decurionate never coalesced into a national class. Peasants might flee to join a band of outlaws, or even, on occasion, rise in revolt en masse and, for a while, drive the bailiffs, tax-collectors and recruiting-officers away. But peasants, too, lacked national organization; without some catalyst of unity – the leadership of an urban class or the inspiration of a revolutionary ideology – they remained preoccupied with farm, fields, the village, the eternal cycle of the seasons. The socio-economic base of the Late Empire rotted away, therefore, largely unseen, beneath the gaze of history, and only indirectly were the effects felt amid the froth of events – in a renewal of dynastic strife, in religious conflict, and in the growing threat posed by the barbarians hammering at the gates.
Behind the internecine conflict that tore apart the House of Constantine was the old problem of regionalism in an overextended empire. Constantine attempted to satisfy the need for local emperors without the attendant danger of usurpation by restricting power to members of his own immediate family. Conflict erupted nonetheless. The emperor’s eldest and most accomplished son, Crispus, was suddenly and inexplicably executed by his father in AD 326. Fausta, the emperor’s wife, followed shortly after (probably charged with adultery). Though he had advanced his three remaining sons and a nephew to supreme power by the time of his death in AD 337, his carefully crafted partition of the empire between the four was immediately rejected by the army, which refused to recognize any but the three sons of Constantine as emperors. Territory was reapportioned: Constantine, the eldest, took Britain, Gaul, Spain and the Rhine frontier; Constans, the youngest, Italy, Africa, the Balkans and the Danube frontier; and Constantius, the East. But civil war soon broke out between Constantine and Constans, and when the former was defeated and killed, the latter amalgamated his brother’s territories with his own (AD 340). Ten years later, in AD 350, Constans was overthrown by Magnentius, one of his own senior commanders, but the Balkans revolted against the usurper and declared allegiance to Constantius, who promptly invaded from the East, defeated Magnentius, and reunited the empire under his sole authority (AD 353). Finally, Constantius II had two of his own leading subordinates arrested and executed, one in AD 354, the second in AD 355. Unsurprisingly perhaps, the last of the sons of Constantine emerged from almost two decades of intermittent civil war with a well-deserved reputation for paranoia and ruthlessness.
The only prominent surviving male member of the House of Constantine apart from the emperor himself was now Julian, the son of a half-brother of Constantine the Great killed by the army in AD 337. Julian himself had been saved by his youth (he was only six at the time). Constantius, preoccupied with the defence of the East against the Sassanids, needed a colleague to guard the Rhine frontier, and, though ever suspicious, came to depend heavily on his cousin. Julian was made Caesar (junior emperor) in Gaul and Germany in AD 355. His mission was to restore the damage after deep and devastating barbarian raids, and then to take the offensive in order to break the power of the Germanic ‘rogue state’ of the Alemanni. He won a spectacular victory at the Battle of Strasbourg in AD 356, and proved himself also an efficient administrator, restoring city defences, establishing military supply-bases, and cracking down on corruption in the Roman administration. We have a detailed account because the historian Ammianus Marcellinus, formerly one of Julian’s officers, was an admirer for whom Julian’s career was especially heroic, edifying and pivotal. In Ammianus’s detailed narrative we gain insight into the mechanics of making emperors. Julian’s success consolidated his base – and heightened the suspicions of his master. When Constantius II demanded the dispatch from Julian’s army of four crack units – as a test of loyalty? to weaken a potential usurper? or simply because he needed them? – Julian (or so Ammianus leads us to believe) wished to comply. But Julian’s supporters refused – it would weaken them politically against Constantius, and militarily against the Germans. Yet to refuse was to revolt. Thus, in a classic confrontation over the deployment of scarce military resources, fuelled by mutual suspicion, Constantius and Julian were propelled towards civil war. In AD 360 Julian was hailed emperor by his supporters. Constantius refused recognition as a co-equal, and both sides prepared for war. Only the death of Constantius the following year prevented a clash of arms. The elevation of Julian, in the event, passed peacefully.
His short reign amounted to a doomed reaction against the Constantinian order. The sons of Constantine had pursued a Christianizing policy, banning pagan sacrifices, closing temples, granting tax immunity to the clergy, funding the building of churches. Also, like their father, they intervened frequently in Church affairs in the interests of unity. Their ideal was a single Church preaching a uniform message. Their fear was that a divided Church might become a vehicle for political opposition. But schism was a perennial feature of the Early Church. Sometimes, notably among the Donatists in North Africa, sectarian faction gave expression to popular discontent and threatened to erupt into class warfare. More often, splits in the Church reflected rivalries within the ruling class. This was surely the case in the bizarrely obscure, yet intractable, theological dispute between Arians and Catholics in the East. The argument revolved around the relationship between God the Father and God the Son, and whether the latter was of the same priority and substance as the former. The priest Arius had argued that God the Father came first. His bishop, Alexander of Alexandria, had denied this. Constantine had attempted to heal the rift at the great Council of Nicaea in AD 325. Attended by around 250–300 bishops, a moderate majority had united around a compromise formula destined to become one of the cornerstones of Catholic theology: the Nicene Creed. But the compromise spawned a party of extreme rejectionists on either side, and bitter disputes raged in the Eastern Church until the end of the 4th century AD.
The controversy was fuelled by the fast-growing wealth of the Church. Control over valuable assets was at stake. The patronage of the House of Constantine quickly transformed the Church into a major legal authority, a political power, and a privileged corporation with a huge property portfolio. The great monuments of the age were usually churches. Major foundations included: in Rome, the Mausoleum of Helena, the Lateran Basilica, and St Peter’s on the Vatican Hill; in Constantinople, the Church of Santa Sophia; and in Jerusalem, the Church of the Holy Sepulchre. Constantinople, built in AD 324–330, was of special significance. It was a ‘New Rome’ with its own Senate, but it was conceived and built as a thoroughly Christian city. It was not just a monument to its founder, a celebration of victory over a pagan enemy (Licinius), a fortified palace complex, and a city with an excellent harbour that was highly defensible, strategically located, and destined to become one of the greatest in world history; it was also a symbol of the new Christian Empire that it had been Constantine’s work to create.
The rise of the Church elevated the bishops to the imperial aristocracy. Increasingly they were recruited from among the grandees, and certainly all were able to take their place among them. Bishops, positioned at the head of a rigid ecclesiastical hierarchy, managed large estates, were numerous at Court, and dominated their home towns, where they conferred benefactions, largesse and favours. Clashes between Donatists and Catholics, Arians and Alexandrians, were, in part at least, struggles between aristocratic factions for wealth and power.
So, too, were clashes between Christianity and paganism. The new religion was associated with the court, the army, and the new men who formed an aristocracy of merit around the emperor – with, that is, the military-bureaucratic complex that controlled the empire. Throughout Roman history new nobles had been resented by old, aristocrats of office by aristocrats of birth. It was true in the conflict between patricians and plebeians in the 5th and 4th centuries BC, and again in the civil wars between Caesarian ‘new men’ and the Senate in the 1st century BC. It had fuelled the opposition to the military monarchs and soldier-emperors of the 3rd century AD. Now, in the 4th, the conflict took the particular form of religious strife, in which the old order – traditional landowners and hereditary peers – rallied to the defence of the pagan cults. They found their champion in the emperor Julian (AD 361–363).
His short reign, though, must be judged a miserable failure. His attempt to restore the old religion by withdrawing the privileges of the Church and lavishing largesse on the temples amounted to little more than a temporary blip. The programme was curtailed by the emperor’s early death and the strong Christian reaction evoked in his successors. Julian came to grief – as so many of his predecessors had done – in the East. Resuming Constantius’s interrupted war, he marched his army deep into Mesopotamia, but was defeated by the Sassanid strategy of scorched earth and guerrilla warfare (AD 363). During a night attack on the Roman camp, Julian led out a sortie without waiting to put on his armour and was shot and mortally wounded, dying a few hours later. The House of Constantine was finally extinguished.
Embroiled in a losing war, the generals at first chose the stopgap emperor Jovian to lead them, fearing that the split between mainly Christian appointees of Constantius and mainly pagan appointees of Julian might blow the high command apart; now was not the time for theological dispute. Jovian dutifully cobbled together a peace agreement with the Sassanids and pulled the army out of Mesopotamia. The following year he died (whether naturally or not is unclear), and the generals assembled again to choose an emperor. Valentinian, the son of a peasant who had risen to become a general, was selected, a man who, according to Ammianus, ‘hated the well-dressed and educated and wealthy and well-born’.(13) He took as co-emperor his younger brother Valens, giving him responsibility for the defence of the East while he took charge in the West. Both men were firm Christians. Their elevation meant that the pagan revival was dead. Religious compromise was thrown to the winds. Militant Christianity – and the military-bureaucratic complex – resumed its forward march. The old order was sidelined again by the dominant power-nexus of court, army, Church, and the aristocracy of merit.
For more than ten years, moreover, Valentinian (AD 364–375) and Valens (AD 364–378) suppressed internal opposition and defended the frontiers of the empire. When Valentinian died – in an uncontrollable fit of rage, it is said, while negotiating with some barbarian leaders – he was succeeded by his sons Gratian (AD 375–383) and Valentinian II (AD 375–392). But Gratian was a teenager, Valentinian a child, and they were in fact ciphers for their uncle, Valens: figurehead rulers to secure the House of Valentinian against usurpers in the West. Then, suddenly and cataclysmically, the regime was destroyed in a huge and terrible battle against the Goths at Adrianople on the lower Danube in AD 378.
Adrianople was the greatest Roman defeat since Cannae. But whereas, in its ascent to global power, Rome had rebounded from Cannae to win final victory against Hannibal, now, in its decline, it would never recover from Adrianople. The haemorrhaging of men that it suffered on that bloody field would be fatal. The Late Roman counter-revolution had put massive strain on the empire’s economic and social foundations. As civil society decayed, tax revenues and reserves of men dwindled, and the military balance swung, slowly but inexorably, away from Rome and in favour of the increasingly well-organized and well-equipped barbarian confederations of Central Europe. The empire, in short, was bleeding to death. Adrianople was the moment when the crisis matured and history turned.
As the Huns advanced westwards across South Russia, the resistance of the (eastern) Ostrogothic and (western) Visigothic kingdoms collapsed. Large masses of displaced people, led west by the Visigothic chieftains Fritigern and Alavivus, appealed to the Eastern Roman Emperor Valens for admission to the Balkans. Their request was granted: they could settle on abandoned land in frontier Thrace in return for military service. In the late autumn of AD 376 the Goths were ferried across the Danube. Some immediately went east to serve in the Roman army. Others were settled around the city of Adrianople in central Thrace. Most, however, were left in refugee-camps in northern Thrace, without adequate food supplies, and prey to exploitation by corrupt Roman officials. The starving Goths were traded dog-meat in return for selling their families into slavery. The rate was one dog per slave.
Anger sometimes boiled over. There were armed clashes. As order broke down, the shattered remnants of the Ostrogothic people, led by the chieftains Alatheus and Safrax, crossed the Danube into Roman territory, swelling the numbers of refugees – and Gothic warriors. When the corrupt Roman military commander in Thrace murdered the escort of the two Visigothic leaders as they dined with him during negotiations, revolt exploded across the refugee-camps and beyond. The Gothic settlers at Adrianople and the Goths sold into slavery joined the revolt. So, too, did some of the provincials, including Thracian miners who had recently been rounded up and returned to work by the Roman authorities. The whole of Thrace was soon under the control of the insurgents. Landlords, tax-collectors and corrupt contractors fled. Local villas were plundered for food.
Valens arrived on the scene with the bulk of the Eastern Roman army in May AD 378. Though reinforcements had been promised from the West, he was confident he could win alone. Leaving their heavy baggage at Adrianople, the Romans advanced the 13 km to the Gothic camp, a great defensive wagon laager. The day was hot. When the Romans arrived, they were tired, thirsty and hungry. The land around the Gothic camp had been scorched, so there was no food to be had. There was then more delay, as abortive negotiations were dragged out by the Goths – deliberately, it seems, for the men defending the laager were awaiting the return of their cavalry, away foraging when news had come of the Roman advance.
Finally it began. The account in Ammianus is confused, but it seems that Valens ordered a massed assault on the Gothic laager, and as this became bogged down, more and more of his men were fed in, until no reserves were left. Then, with the entire Roman army committed, the Gothic cavalry reached the battlefield and charged into the exposed enemy flanks and rear. Surrounded and slowly compressed into an ever-smaller space, with neither room to manoeuvre nor opening to escape, the emperor Valens and some two-thirds of the Eastern Roman army perished on the battlefield.
The Goths declined to attack either Adrianople or Constantinople – ‘I am at peace with walls,’ Fritigern is reported saying – and the revolt was contained in Thrace. But the loss of manpower had inflicted irreparable damage. Never again would a Roman citizen army fight a major battle in Europe. Henceforward the defence of the West would depend on the services of barbarian mercenaries. Adrianople had revealed the rotted condition of Roman imperial society: the alienation of its own people; the corruption of its officials; and the sinking numbers, efficiency and morale of its fighting forces. In the chaos and carnage of that baking early summer’s day, the world changed. Soon, the East, with its vastly superior estate, would cut the West adrift, leaving it starved of resources to die a lingering century-long death. But it would be the barbarian war-bands recruited to fill the gaps in the battle-line left by Adrianople that would be the instrument of termination.