Conclusion: A Simple Answer

`It will be enough for me, however, if these words of mine are judged useful by those who want to understand clearly the events which happened in the past and which will (human nature being what it is), at some time or other and in much the same ways, be repeated in the future.' - Thucydides, writing at the very end of the fifth century BC.

The Western Roman Empire ceased to exist in the fifth century. Even those scholars who talk of transformation admit this simple fact. The Eastern Roman Empire lasted for another thousand years until it was overrun by the Turks. Even at its height it could never hope to dominate the world. It was a power, rather than a superpower. The sixth century demonstrated that it lacked the capacity to recapture the lost western provinces. In the seventh century the Arab conquests stripped it of even more territory. It continued to exist as just one amongst many powers in the known world, and some of these were geographically larger and both militarily and financially stronger. Even so, none could be said to have replaced the Roman Empire or matched its former size and power.

None of this happened quickly, but viewed in the long term it cannot be seen as anything other than decline and - in the case of the Western Empire - fall. It was a long process and no single event, lost war or decision can be said to have caused it. The basic question remains of why this occurred, and whether the most important cause was internal problems or external threats. Throughout their history the Romans had always fought a lot of wars against very varied opponents. They had suffered some serious defeats, but had always recovered. There was never any question that such defeats could cause the collapse of the empire. Yet this did happen in the west in the fifth century and therefore we must ask whether the threats faced by the Late Roman Empire were greater than those of earlier periods. This in turn raises two basic possibilities. Either one or more individual enemy was more formidable, or there were simply so many simultaneous threats that the empire could not cope.

It is usually asserted that the Sassanid Persians were far more formidable than the Parthians, or indeed any enemy the Romans had faced for centuries. They certainly won more victories over the Romans than the Parthians. On the other hand, the levels of Persian aggression varied enormously and there were long periods of peace. Some Persian kings needed the wealth and glory offered by a successful war with Rome. Usually this was necessary to secure their own hold on power. The largest Roman armies of the period were those sent east to face the Persians and massive resources were expended on frontier fortifications. Having said that, only border territory was ever actually lost to Persia and even this was on a fairly modest scale. The idea that from its first appearance in the third century Persia was an especially deadly opponent - even a rival superpower - remains firmly entrenched in the minds of scholars. It is a belief that is very hard to reconcile with the evidence, but this does not mean that it will not continue to be asserted.

Groups from the tribal peoples of Europe eventually took control of the Western Empire. However, it is extremely difficult to see major change in the military efficiency of the tribal peoples of Europe from Julius Caesar's day to that of Stilicho's or Aetius'. To some degree larger tribal confederations appeared, but we should never exaggerate the degree of unity. It is convenient to talk of the Franks or the Goths, in spite of the fact that these remained divided into many separate and sometimes mutually hostile tribes. At no stage before the creation of the barbarian kingdoms inside the provinces was there a single king of all the Franks or any other people. Attila united both his own people and allied and subject races to a remarkable degree. Yet, once again, he was unable to take much territory from the Romans and was essentially a raider and extortionist on a grand scale. Other powerful barbarian leaders had emerged in the past and, like Attila, they had proved unable to pass on their power to a successor. The Huns were a frightening enemy, but it is worth remembering that their power had been broken before the final collapse of the Western Empire and that they had anyway devoted most of their attentions to the Eastern Empire.

There is no good case for claiming that the enemies of the Late Roman Empire were simply more formidable than those of earlier periods. This also makes it harder to argue that the Roman Empire had to adapt in the third century to face new and more dangerous threats, most of all the Sassanid `superpower'. Does this mean that it was the sheer quantity rather than the scale of individual threats that was the problem? There certainly do seem to have been more major wars in the third and subsequent centuries than in the early Principate. In particular, raiding by barbarian groups in Europe is much more prominent in our sources. Such predatory attacks, often on a small scale, were not new. In the past they had always increased in scale and frequency whenever the frontier defences were perceived to be weak. An impression of vulnerability encouraged attacks and this makes it hard to judge whether an increase in raids and invasions was the consequence of a rise in barbarian numbers and strength or a result of Roman weakness. It is clear that all of Rome's enemies, including the Persians, exploited the empire's frequent internal disputes and civil wars.

There may be other reasons for Roman weakness and we need to consider these. Unfortunately, for so many of the theories about longterm problems we lack the basic information either to confirm or deny them. There are no good figures for the population of the empire at any period and, therefore, we cannot say with any certainty that this was in long-term decline. Similarly, we must study the economy without any adequate statistics. It seems more than probable that levels of trade and prosperity fell from the end of the second century onwards and never again achieved the levels of the early Principate. However, sources at best hint at such trends, and some scholars will interpret these glimpses of the past in radically different ways. The same is true of the traditional picture of a Late Roman world where the burden of tax was oppressive and fell disproportionately heavily on the poor, who were already oppressed by their rich landlords. Land fell out of cultivation and the rural population was reduced to the level of serfs. None of this is implausible, but that is also true of other models and it is impossible to prove any of them. Far more data - the bulk of which must come from archaeology if it can ever be found - is needed before we can speak with some confidence on these topics. The same is true of claims about climate change and other wider problems.

The type of evidence we have, as well as the interests of scholars, has meant that a good deal of the work on Late Antiquity has focused on economy and society, law and government, intellectual life, culture and religion. Studies tend to concentrate on broad themes and inevitably this emphasises continuity rather than change. By comparison, narrative history has all too often been neglected and certainly has made only a minor contribution to most scholars' mental picture of the period. There are exceptions and study of the frontier relations and foreign wars has often been more traditional in style, since a narrative or chronological element is obviously essential. At the same time civil wars and internal conflict have not received such detailed and coherent treatment. This is odd, for these are the one aspect of the empire's internal problems for which we have considerable evidence.

It is worth once again emphasising that from 217 down to the collapse of the Western Empire there were only a handful of periods as long as ten years when a civil war did not break out. Some of these conflicts were very brief and some were confined to a small region - the usurpers who were proclaimed and then suppressed, or rejected and murdered by their own men after a reign lasting just a few weeks. Challenges for imperial power were sometimes resolved without serious fighting. On the other hand, some conflicts were fought on a very large scale and lasted for years. It is easy to remember Constantine as the great emperor who united the entire empire under his control, but we should not forget that he was a usurper who fought or prepared for civil wars for the first half of his reign.

Civil war and challenges to the imperial throne were common occurrences. Every adult emperor from Septimius Severus onwards experienced at least one such conflict during their lifetime. Usurpers never wanted to destroy or change the empire. These were not conflicts about ideology, but purely for political power. A small minority of the losers in these wars were allowed to keep their lives, although only a tiny handful were permitted to continue in a public career. In the vast majority of cases such conflicts only ended with the death of one of the rivals. Usurpers were the most direct and personal threat faced by any emperor and tended to be treated accordingly. It was normal for an emperor to abandon a war against a foreign enemy to deal with a Roman rival.

Usurpers did not act alone. They needed supporters and the most important of these expected rewards including promotion and riches if the rebellion was successful. If a usurper was suppressed, then many of his backers were likely to suffer with him. Punishment was often extended to their families, especially those holding any office or whose wealth made them appealing targets for informers. In this way even a localised rebellion could mean life, death, imprisonment or ruin to people in distant provinces who had not been involved in it in any direct way. This was a world of patronage, where the powerful exerted themselves to secure benefits for relatives and friends. Such webs of favour and gratitude could become very dangerous for all concerned at times of internal conflict.

All usurpers needed military backing to succeed. Emperors from Augustus onwards tried to keep their soldiers loyal through solemn oaths and regular donatives. On the whole, the army tended to stay loyal to an established dynasty unless the emperor seriously alienated them. Few usurpers could count on similar loyalty. Losses were considerable in some civil wars, as the army wasted its strength fighting against itself. Soldiers fighting an internal struggle could not simultaneously operate on one of the frontiers. Time and again substantial parts of the army were drawn away and Roman military dominance across its borders reduced or utterly shattered. Successive civil wars dislocated the army's administrative and logistical structures, its training patterns, recruitment and also its discipline, which suffered whenever licence was given in an effort to win loyalty. Ordinary soldiers could usually expect to change sides to join the victors after a failed rebellion. This was not so easy for more senior officers.

Each civil war cost the empire. Anything gained by the winning side inevitably had to be taken from other Romans and a prolonged campaign was likely to involve widespread destruction within the provinces where fighting occurred. Almost as important as the physical price of civil war was its impact on attitudes and behaviour from the emperor down. Personal survival became the first objective of every emperor and shaped all of their decisions and the very structure of the empire. In the quest to protect themselves successive emperors gradually reshaped the empire itself and, ironically enough, often made themselves more rather than less vulnerable.

The biggest change was the marginalisation of the senatorial class in the third century and, along with them, the city of Rome as a real rather than merely spiritual capital of the empire. Senators - and most of all a handful of distinguished men and those trusted with senior provincial commands - were for a long time the only possible rivals for imperial power. At first the major military provinces were divided up so that no one man commanded too large an army. By the end of the third century senators had virtually ceased to hold military rank of any kind. They had also all but ceased to become emperor.

Emperors could now come from a far wider section of the empire's population. Any connection with the imperial family - even spurious claims to be the illegitimate son of an emperor - was sufficient to make a claim. In the past Rome's emperors had had to be wary of only a small number of senators, men who were known to them personally and whose careers meant that they spent many years in and around Rome. Now a rival could be almost anyone. They did not need political connections or family reputation, simply the ability to persuade some troops to back them. Many emperors were equestrians, and almost all were army officers or imperial officials.

The trend towards smaller provinces continued. In addition, military and civil power were made separate. This helped to protect an emperor against challengers, but made it far harder to get things done. In particular, it was very difficult to raise and supply a large enough army to deal with a serious problem on the frontiers. From the emperor's point of view this was comforting, since the same army could easily have been turned against him by a rival. At times extraordinary commands were created so that one commander could deal with a problem, but emperors had to be wary of offering such power to a potential usurper. More often emperors chose to go themselves and take personal command of a campaign. From the middle of the third century onwards Roman emperors spent much of their time performing tasks that would once have been dealt with by an imperial legate. Again, it is worth emphasising that it was not the scale of the problems that had increased, but the ability of the empire to employ its resources to deal with them.

An emperor could not be everywhere at once. If he was unwilling to trust anyone else with sufficient power to deal with a distant problem, then it would simply not be dealt with at all. Time and again this sense of neglect by central government prompted a region to rebel and proclaim its own emperor. One solution was to have more than one emperor. The tetrarchic system is often praised, but its success was always limited and no one was able to repeat the dominance of Diocletian for any great length of time. In a way, the acceptance that more than one emperor would exist offered usurpers the prospect of advancing to supreme power in stages. It also tended to encourage regionalism as separate military and civil hierarchies developed in different parts of the empire. Each group was naturally inclined to give priority to its own aims and problems, and often proved reluctant to assist other parts of the empire.

Emperors had always travelled in some state, surrounded by members of their household, bureaucrats and guards. This increased massively in scale during the third century. All wanted to have sizeable military forces under their direct control. If the field armies were intended to perform a strategic role, then this was first and foremost to guard against Roman rivals. Emperors surrounded themselves with more and more attendants and personal bodyguards, and made court ceremonial increasingly elaborate. In part this was to dignify and secure the rule of men who had often seized power in brutal fashion comparatively recently. It was also intended to protect the emperor's person. Assassination was less common in the fourth century than the third. At the same time all of this tended to isolate the emperor. It made it harder for him to know personally even his more senior officials and commanders, let alone the vastly inflated number of bureaucrats who now worked in the imperial administration. Control over the activities of the men who represented imperial authority throughout the provinces was extremely limited.

All emperors lived with the fear of usurpation. It shaped their behaviour and also that of all of the officials and officers who served under them. A career in the imperial service offered the prospect of legal privileges and wealth, gathered both through pay and, even more, from bribes and payments for services. The most successful achieved very high rank with all the patronage and influence this brought. A small minority were even able to reach imperial rank. However, alongside the advantages came serious risks. Any suspicion that an individual was plotting against the emperor was likely to be punished severely. The same was true of anybody associated with a failed usurper or their supporters. In a system where careers were routinely advanced by personal recommendation, such networks of patronage inevitably put many individuals in danger. Personal survival and personal success and profit were the foremost aims of most officials.

The imperial bureaucracy in the Late Roman Empire was certainly far larger than in the first and second centuries. The army may or may not have been bigger, but certainly consisted of far more small, independent units. Size on its own does not mean that either of these institutions was more efficient. There were far more administrators than could readily be supervised, especially since they formed part of a bureaucracy that was both divided and confused in its structure. The imperial administration raised funds and resources to support both itself and the army. Such short-term expedients as debasing the coinage suggest that at times this supply proved inadequate. However, on the whole the system seems to have functioned in the third and fourth centuries, at the very least to a minimal necessary level. It still left plenty of room for inefficiency and corruption, and such wastage may well have been on a massive scale. Most individual members of the bureaucracy did their job well enough to keep the system functioning and prevent their peculation becoming too blatant. Some may genuinely have been both honest and competent!

Civil wars were most common in the third century, but remained frequent afterwards. The state developed in ways intended to protect emperors from internal rivals, but singularly failed to do so. Personal survival had always been an important concern for all emperors since the creation of the Principate. Augustus had fought his way to power through a series of civil wars. Assassination plots and open rebellion were threats faced by each of Rome's rulers from the very beginning. Augustus was a monarch, but created a system in which his power was carefully veiled. Since he was not formally a king, there was no clear institution to arrange the succession. Some have seen this as a fatal flaw in the system of the Principate - effectively, an accident waiting to happen. Others would go further and see the Augustan system as a `millstone', revered by tradition that prevented proper reform of the empire in the third and fourth centuries.'

This cannot explain the quite staggering difference between the Principate and the Late Roman Empire. There was civil war for a year after the death of Nero in 68 and another longer conflict after Pertinax was murdered in 193. Claudius, Domitian and Marcus Aurelius each faced a challenge from a rebellious governor, although all of these revolts swiftly collapsed. Assassination plots and attempted coups at Rome were a little more common, although some of these may have been imagined by nervous emperors or invented by their ruthless subordinates. The early Principate was not wholly free from the reality or threat of internal conflict, but for more than two hundred years it still suffered only rarely from these. This is also in marked contrast to the last half-century of the Republic. If the system created by Augustus was so seriously flawed, then only remarkable luck could explain this. With Gibbon, we might stop `inquiring why the Roman empire was destroyed', and instead `be surprised that it lasted so long'.

It stretches credibility to see two centuries of largely unbroken internal peace as a mere fluke, especially when they were followed by a longer period when civil war was so very frequent. It is true that each fresh bout of internal conflict weakened imperial authority and the institutions of the state and therefore made future usurpations and rebellions more likely. Yet, once again, it cannot have been solely chance that such a cycle did not develop earlier. In the third century the empire largely lost the Republican facade so carefully constructed by Augustus. He and his successors ruled through the Senate. As a body this had no real political independence, but sensible emperors took care to respect its dignity. More importantly, they employed senators in virtually all important posts, effectively ruling the empire through them.

It may seem odd in this day and age to praise a system based on an aristocratic elite, consisting of men who were amateurs in the modern sense. Yet the system had many advantages in the Roman context. It provided a manageable group of senior soldiers and administrators - an emperor could know all of these men and their families. Only a minority were potential rivals and these could be closely observed. Public life remained focused on the fixed location of Rome itself, making it easier to sense the mood of the aristocracy. Emperors in the first and second centuries were able to trust selected senators to control substantial armies and large provinces. Only rarely - usually during times of major conflict with Parthia - was it necessary to appoint a commander to control more than one province and this did not automatically lead to an attempt at usurpation. In the first and second centuries emperors were able to delegate and did not feel obliged to direct campaigns in person. Rome was the centre of the empire in more than just a spiritual sense. We do not need to idealise the senatorial legates of the early period. Some were incompetent, a few untrustworthy and probably quite a lot were more or less corrupt. In all these respects they seem at the very least no worse than the senior officials of the Late Roman Empire. Politically, the small senatorial class was simply easier for an emperor to control. Reliance on the Senate was a Republican tradition, but actually made sound sense.

The governments of ancient states had limited ambitions and did not concern themselves with major programmes of health, education or the detailed day to day regulation of markets, industry and agriculture. For all its size and sophistication, the Roman Empire was not fundamentally different in this respect. It raised revenue and other resources and made use of these in a range of ways. The army was the biggest single cost, but there was also the maintenance of many buildings, some ports and a vast road network, as well as the subsidised or free doles of food to the population of Rome and later Constantinople. None of these duties of the empire ground to a halt in the third or fourth century. However, this does not mean that they were functioning well.

The Roman Empire did not fall quickly, but to use this as proof that its institutions were essentially sound is deeply misguided. The empire was huge and faced no serious competitors. Persia was the strongest neighbour, but there was never a prospect of a Persian army reaching the Tiber. Rome was massive, heavily populated and rich. This remained true even if the population and economy were in decline. It had a transport system of all-weather roads and busy commercial routes by river, canal and sea on a scale unmatched again in Europe until recent centuries. Although we may note the difficulties emperors had in making their will felt in distant provinces, their capacity to do this at all was still far greater than the leaders of any other people. The Roman army was a large, sophisticated, permanent and professional force backed by an extensive logistical system. Like the empire itself, it was different from anything else in existence in the known world. The Romans possessed many great advantages over all of their competitors. None of these rivals had the power to push the empire over in the third or fourth centuries. The empire was huge and did not need to operate at the highest levels of efficiency to succeed. It possessed massively greater resources, technological and other advantages. There was also the probability that somewhere along the line some officers and officials would do their job at least moderately well. This meant that the Romans were likely to prevail in the long run. None of its enemies were capable of inflicting more than a limited defeat on the Romans.

None of this meant that the cost of repeated civil war was not felt. It is not difficult to make the case that the majority of emperors in the first and second centuries had the wider good of the empire as their main ambition. All were concerned with personal survival, but this had not become the overwhelming priority it would be for their successors in later eras. That is not to say that the later emperors were more selfish, but simply that they could never be as secure. Many may have had the best of intentions to rule well, but the government of the empire became first and foremost about keeping the emperor in power - and at lower levels, about the individual advantage of bureaucrats and officers.

The Late Roman Empire was not designed to be an efficient government, but to keep the emperor in power and to benefit the members of the administration. Many of these could enjoy highly successful careers by the standards of the day without ever being effective in the role that they were theoretically supposed to perform. Sheer size prevented rapid collapse or catastrophe. Its weakness was not obvious, but this only meant that collapse could come in sudden, dramatic stages, such as the loss of the African provinces to the Vandals. Gradually, the empire's institutions rotted and became less and less capable of dealing with any crisis, but still did not face serious competition. Lost wars were damaging, but the damage was not fatal to the empire itself. As an example, from 376-382 the Romans could not lose the war against the Goths, but they still struggled to win it. Even defeats at the hands of the Persians did not deprive the empire of major or essential resources.

The Roman Empire continued for a very long time. Successive blows knocked away sections of it, as attackers uncovered its weaknesses. Yet at times the empire could still be formidable and did not simply collapse. Perhaps we should imagine the Late Roman Empire as a retired athlete, whose body has declined from neglect and an unhealthy lifestyle. At times the muscles will still function well and with the memory of former skill and training. Yet, as the neglect continues, the body becomes less and less capable of resisting disease or recovering from injury. Over the years the person would grow weaker and weaker, and in the end could easily succumb to disease. Long decline was the fate of the Roman Empire. In the end, it may well have been `murdered' by barbarian invaders, but these struck at a body made vulnerable by prolonged decay.