The Fall of the Roman Empire: A New History

by Peter Heather

  The Limits of Government

  IN THEORY, the emperor was the supreme authority when it came to issuing general legislation, and in individual cases he had the right to modify the law, or break it, as he chose. He could condemn to death, or pardon, with a single word. To all appearances, he was an absolute monarch. But appearances can be deceptive.

  Valentinian, a long-time soldier before his accession, had first-hand experience of supervising the Rhine frontier; based at Trier, he was close enough to investigate promptly any untoward incident. But a problem arising in Africa was a very different matter. The first Valentinian knew of the Lepcis episode was the sudden arrival at his court of two diametrically opposed accounts of it, one brought by the first legation from the provincial assembly, the other from Romanus via the magister officiorum, Remigius. Trier placed Valentinian about 2,000 kilometres away from the scene of the action. As he couldn’t leave the Rhine frontier to investigate one relatively minor incident in a rather obscure corner of North Africa, all he could do was send a representative to sort out the facts for him. If that person fed him misinformation, as was the case here, and ensured that no alternative account reached the imperial ears, the emperor was bound to act accordingly. The essential point that emerges from Lepcisgate is that, for all an emperor’s power, in both theory and practice, Roman central government could only make effective decisions when it received accurate information from the localities. The regime of Valentinian liked to style itself as the protector of the taxpayer from the unfair demands of the military. But, thanks to Palladius’ false report, the emperor’s actions in the case of Lepcis Magna had entirely the opposite effect.

  A leap of imagination is required to grasp the difficulty of gathering accurate information in the Roman world. As ruler of just half of it, Valentinian was controlling an area significantly larger than the current European Union. Effective central action is difficult enough today on such a geographical scale, but the communication problems that Valentinian faced made it almost inconceivably harder for him than for his counterparts in modern Brussels. The problem was twofold: not only the slowness of ancient communications, but also the minimal number of lines of contact. The Lepcis problem was exacerbated not only by the snail’s pace of such communications as there were, but also by the sheer paucity of points of contact: two in the first instance (the ambassadors, plus Remigius representing Romanus’ view), supplemented by a third when Valentinian sent his fact-finding mission in the person of Palladius. Once Palladius verified Romanus’ view, that was two against one, and Valentinian had no additional sources of information. In the world of the telephone, the fax and the internet, the truth is much harder to hide. Beyond the immediate vicinity of his base on the Rhine frontier, Valentinian’s contacts with the city communities that made up his Empire were sparse and infrequent.

  Insight into the problem is provided by another extraordinary survival from the later Roman Empire: papyrus documents preserved through the centuries by the dry heat of the Egyptian desert. (As fate would have it, most of the archive ended up in the John Rylands Library in Manchester, a city famous for its rainfall.) These particular papyri, purchased by the great Victorian collector A. S. Hunt in 1896, come from Hermopolis on the west bank of the Nile at the boundary between Upper and Lower Egypt. One key letter got separated from the rest, ending up in Strasbourg. When identified as part of the same collection, it became clear that these were the papers of a certain Theophanes, a landowner from Hermopolis and a fairly high-level Roman bureaucrat of the early fourth century. In the late 310s he was legal adviser to Vitalis who, as rationalis Aegypti, was the finance officer in charge of the arms factories and other operations of the Roman state in the province. The bulk of the archive refers to a journey Theophanes made from Egypt to Antioch (modern Antakya in southern Turkey, close to the Syrian border), a regional capital of the Roman east, on official business, sometime between 317 and 323. The papers don’t provide a narrative of the journey – we can only guess what the aim of the mission may have been – but something in its own way more valuable: packing lists, financial accounts and dated itineraries which, between them, bring Roman official travel vividly to life.6

  Being on official business, Theophanes was able to use the same public transport system that carried Symmachus to Trier, the cursus publicus, which comprised neatly spaced way-stations combining stables – where official travellers could obtain a change of animals – and (sometimes) travel lodges. The most immediately striking documents are those dealing with Theophanes’ itineraries: daily listings of the distances he managed to cover. Having begun the journey to Antioch on 6 April at the town of Nikiu in Upper Egypt, he eventually rolled into the city three and a half weeks later on 2 May. His daily average had been about 40 kilometres: on the first part of the journey, through the Sinai desert, he made only about 24 kilometres a day, but speeded up to about 65 once he hit the Fertile Crescent. And on a breakneck final day into Antioch, scenting the finishing line, his party covered over a hundred. The return journey took a similar time. Bearing in mind that Theophanes’ official status allowed him to change horses whenever necessary – so there was no need to conserve equine energy – this gives us a benchmark for the bureaucratic operations of the Roman Empire. We know that in emergencies, galloping messengers, with many changes of horse, might manage as much as 250 kilometres a day. But Theophanes’ average on that journey of three and a half weeks was the norm: in other words, about 40, the speed of the oxcart. This was true of military as well as civilian operations, since all the army’s heavy equipment and baggage moved by this means too.

  The other striking feature of Theophanes’ journey is its complexity. As might be expected, given such rates of travel, only the top echelons of the Roman bureaucracy tended to travel outside their immediate province – hence, lower-level officials wouldn’t know their counterparts, even in adjacent regions. Egypt, for most purposes, ran itself, so Theophanes didn’t usually need to know people in Antioch, and neither, for that matter, did he know people anywhere else en route. Vitalis armed him, accordingly, with letters of introduction to everyone who mattered on the way, some of which he didn’t use (which is how they come to survive in the archive). Given contemporary rules of etiquette, you had to think ahead and take with you a range of appropriate offerings: courtesy demanded that an exchange of gifts – sometimes valuable ones – inaugurate any new relationship. The accounts record items destined for such a fate, such as lungurion (coagulated lynx musk), an ingredient of expensive perfumes.7 Large amounts of cash also had to be carried, probably supplemented in Theophanes’ case by letters of authorization allowing him, as an official traveller, to draw funds from official sources. Hence such travellers would often need protection, hiring armed escorts where necessary. Theophanes’ accounts record food and drink bought for soldiers who accompanied them during the desert legs of the journey in Egypt.

  The packing lists also make highly illuminating reading. Theophanes obviously needed a variety of attires: lighter and heavier clothing for variations in weather and conditions, his official uniform for the office, and a robe for the baths. The travel lodges of the cursus publicus were clearly very basic. The traveller brought along his own bedding – not just sheets, but even a mattress – and a complete kitchen to see to the food situation. As this suggests, Theophanes did not travel alone. We don’t know how many went with him, but he was clearly accompanied by a party of slaves who dealt with all the household tasks. He generally spent on their daily sustenance just under half of what he spent on his own. This battered bundle of papyrus documents at Manchester is full of such gems of detail. Just before leaving civilization to cross the desert again, the party bought 160 litres of wine for the home journey. This cost less than the two litres of a much rarer vintage that Theophanes had with his lunch on the same day. At another point, the accounts record the purchase of snow, used to cool the wine for dinner. What emerges is an arresting vision of the complex and cumbersome nature of o
fficial travel.

  In reality, then, places were much further away from one another in the fourth century than they are now. As I sit here writing, it’s about 4,000 kilometres from Hadrian’s Wall to the Euphrates, and so it was in Theophanes’ day. But at Theophanes’ rate of progress – even giving him a higher average daily rate of 50 kilometres (not counting the days spent crossing the desert) – a journey that overland would now take a maximum of two weeks would in the fourth century have taken something close to three months. Looking at the map with modern eyes, we perceive the Roman Empire as impressive enough; looked at in fourth-century terms, it is staggering. Furthermore, measuring it in the real currency of how long it took human beings to cover the distances involved, you could say it was five times larger than it appears on the map. To put it another way, running the Roman Empire with the communications then available was akin to running, in the modern day, an entity somewhere between five and ten times the size of the European Union. With places this far apart, and this far away from his capital, it is hardly surprising that an emperor would have few lines of contact with most of the localities that made up his Empire.

  Moreover, even if his agents had somehow maintained a continuous flow of intelligence from every town of the Empire into the imperial centre, there is little that he could have done with it anyway. All this putative information would have had to remain on bits of papyrus, and headquarters would soon have been buried under a mountain of paperwork. Finding any particular piece of information when required would have been virtually impossible, especially since Roman archivists seem to have filed only by year.8 Primitive communication links combined with an absence of sophisticated means of processing information explain the bureaucratic limitations within which Roman emperors of all eras had to make and enforce executive decisions.

  The main consequence of all this was that the state was unable to interfere systematically in the day-to-day running of its constituent communities. Not surprisingly, the range of business handled by Roman government was only a fraction of that of a modern state. Even if there had been ideologies to encourage it, Roman government lacked the bureaucratic capacity to handle broad-reaching social agendas, such as a health service or a social security budget. Pro-active governmental involvement was necessarily restricted to a much narrower range of operation: maintaining an effective army, and running the tax system. And, even in the matter of taxation, the state bureaucracy’s role was limited to allocating overall sums to the cities of the Empire and monitoring the transfer of monies. The difficult work – the allocation of individual tax bills and the actual collection of money – was handled at the local level. Even here, so long as the agreed tax-take flowed out of the cities and into the central coffers, local communities were left – as the municipal laws we examined in Chapter 1 imply – to be autonomous, largely self-governing communities.9 Keep Roman central government happy, and life could often be lived as the locals wanted.

  This is a key to understanding much of the internal history of the Roman Empire. Lepcisgate illustrates not so much a particular problem of the later Empire, but the fundamental limitations affecting Roman central government of all eras. To comprehend the operation of government fully, the logistic impossibility of day-to-day interference from the centre must be considered alongside the imperial centre’s absolute legal power and unchallenged ideological domination. It was the interaction of these two phenomena that created the distinctive dynamic of the Roman Empire’s internal functioning. Given that it was administratively impossible for central government to control everything, anything to which it did add its stamp of authority carried an overwhelming legitimacy, if put to the test. What tended to happen, therefore, was that individuals and communities would invoke the authority of the centre for their own purposes. At first sight, this could suggest that the imperial finger was constantly being stuck into a whole host of local pies, but such an impression is misleading. Outside of taxation, emperors interfered in local affairs only when locals – or at least a faction of local opinion – saw an advantage to themselves in mobilizing imperial authority.

  We have already seen this pattern at work in the early imperial period. As the Spanish inscriptions (pp. 38–9) show us, Roman-style towns existed right across the Empire as a consequence of local communities adopting municipal laws drawn up at the centre. In particular, the richer local landowners had quickly appreciated that securing a constitution with Latin rights was a path to Roman citizenship, which would qualify them to participate in the highly lucrative structures of Empire. The story had its shadier side, of course. A grant of Italian status was so valuable to the leaders of the community involved that they were willing to do whatever it took to win the privilege, often by courting patrons at the centre who would put in a good word for them with the emperor of the day. This kind of relationship between centre and locality was the bedrock on which the Empire was built.10

  This relationship also applied to individuals who used the ‘rescript’ system. Rescripts allowed you to consult the emperor – in practice, his legal experts – on a matter of legal detail. Using the top half of a piece of papyrus, anyone could write to the emperor about an issue on which he wanted a decision. The emperor would then reply on the bottom half. You couldn’t use the system to get him to settle an entire case – only to raise a technical point of law that might dictate its outcome. Again, we’re indebted to a unique papyrus survival for an indication of how extensively the system was used. In the spring of AD 200, the emperors Severus and Caracalla were installed in the city of Alexandria in Egypt. A papyrus, now to be found at Columbia University, records that the emperors answered five rescripts (the replies were posted publicly) on 14 March, four on 15 March, and another four on the 20th.11 So even if we allow emperors lots of annual holidays, at least a thousand people a year could cite an imperial opinion in their private legal disputes.

  Equally important, once a rescript had been sent back to the provinces, the emperor lost control over it, so a piece of paper carrying his name and his authority was on the loose. Hardly surprising, then, that these imperial replies were used in all kinds of unintended ways. The fifth-century Theodosian Code (see pp. 124-5) cites a number of scams: cases where the imperial answer had been physically detached from the original question and then used to answer another, others where letters extracted for one case had been applied to another, and still others where letters had been extracted under false pretences.12 Roman lawyers were as inventive as their modern counterparts, and subject to much less control. Not only does the rescript system show us an imperial authority that was essentially reactive, but abuses of it also make clear that distance could allow the suitor to make unintended use of the potent weapon represented by a legal ruling with the emperor’s name on it.

  In addition to the rescript system, emperors were also deluged with requests of a more general kind, which they might or might not respond to positively. They could either launch their own inevitably slow investigation, or accept the petitioner’s inherently biased version of the truth. This usually meant the deployment of imperial power to more or less random effect: the emperor chose either to believe or not to believe the petitioner, and acted accordingly. The impact this had on day-to-day affairs depended upon the lengths that citizens in local communities would go to in order to exploit that power.

  Any picture of Roman government, then, has to bear in mind that, for all their legal and ideological authority, emperors’ control was limited. All the same, such was their monopoly of authority that their approval was constantly being solicited by the citizenry. Consequently, the imperial centre was both powerful and strictly constrained.

  IN THE MIDDLE years of the third century, this inherently limited governmental machine was suddenly forced to confront an entirely new range of problems, all traceable to the rise of Sasanian Persia. As we have seen, the immediate problem was solved by the military, fiscal and political restructuring of the Empire. It has long been customary to argue, h
owever, that, while rescuing the Empire from these difficulties, the changes it put in place doomed it to decline and collapse in the longer term. After Diocletian, according to this view, the Roman agricultural economy was substantially overtaxed. Peasants were forced to surrender so much of their produce that some of them died from starvation. The new tax levels, it is argued, also ruined the landowning classes who had built and run the towns of the Empire since its formation. In fact, the whole imperial edifice came to be dominated by constraint rather than consent, symbolized in a repressive, bureaucratic machine staffed, as one influential view put it, by many ‘idle mouths’, a further burden on the taxpayer. On the military side, the enlarged army may have done its job in the short term; but manpower shortages within the Empire forced fourth-century emperors to draw increasingly on ‘barbarian’ recruits from across the frontier. As a result, the Roman army declined in both loyalty and efficiency. All in all, this line of argument goes, while the initial Persian crisis was overcome, it had required such an effort that the financial, political and even military strength of the Empire was visibly draining away.13

  These views remain deeply entrenched. The present generation of scholarship has demonstrated beyond reasonable doubt, however, that such a stance greatly underestimates the economic, political and ideological vitality of the late Roman world.

  The Price of Survival

  ANCIENT AGRICULTURE suffered from two limitations. First, before the invention of tractors, the productivity of any piece of land was hugely dependent on how much labour was available to work it. Second, ancient farmers, while employing their own sophisticated techniques for maintaining fertility, were unable greatly to increase their output of foodstuffs in anything like the way that the use of chemical fertilizers has made possible in the modern era. This in turn acted as a brake on population levels, since human numbers tend to increase up to a limit imposed by the availability of food. In addition, transport was hugely expensive; Diocletian’s Prices Edict (see p. 65) records that a wagon of wheat doubled in price for every fifty miles it travelled. In these fundamental ways, the Roman economy was at every era trapped at not much above subsistence levels. Until very recently, scholars have been confident that the higher tax-take of the late Roman state aggravated these conditions to the extent that it became impossible for the Empire’s peasant population to maintain itself even at existing low levels.