The army took recruits wherever it could get them, and there are clear signs that many people went to drastic lengths to avoid service. In real terms army pay was of less value than it had been in the first and second centuries, while discipline and punishment remained brutal. Repeated laws punished the practice of self-mutilation to avoid being conscripted – potential recruits cut off their thumbs so that they would be unable to hold a sword or shield properly. There was a famous case of an equestrian who had done this to his sons during Augustus’ reign, so the practice was not new, but it does seem to have become more common. The frequency of legislation dealing with the problem suggests that the laws were not effective. A letter from a clergyman to a garrison commander in Egypt asked that he exempt a widow’s son from conscription or, failing that, at least enlist him in the local limitanei so that he could stay near home instead of sending him away to join the comitatenses. The fourth-century army did not have massive resources of manpower and this factor played a significant part in shaping its operations.
Whatever the actual size of the fourth-century army, more units certainly meant more officers to command them. Regimental commanders were usually called tribunes, although other titles such as praepositus were also used. Such a post gave a man considerable status – even if, once again, his pay and social importance were somewhat less than those of an equestrian officer in the second century. In comparison to the early Principate, the Late Roman army was top heavy in its command structure. It is more than likely that the desire to reward supporters with high rank – and in some cases their own independent commands – had as much to do with the multiplication of units as any practical concerns. We hear of unattached tribunes, sometimes serving in a staff capacity, and it is more than likely that there were significantly more men with commissions than there were units for them to command. Some tribunes may well have been promoted after service in the ranks. A much more common route was to be commissioned from special units at the imperial court. The candidati (candidates) acted as personal bodyguards of the emperor during court ceremony. The protectores domesticiserved as junior staff officers either with the emperor or a Master of Soldiers. Ammianus was one of these. Promotion was officially the emperors’ prerogative, but in practice he had to rely on recommendations of senior officers, officials and courtiers. There was no formal system for training these officers or for selecting on the basis of talent. As always in Roman society, patronage played an important role in determining a man’s career.”
Modern scholars conventionally refer to the comitatenses as mobile field armies, and often they go further and dub them elite troops. They are seen as a necessary response to the greater external threats facing the empire. In the past, wars requiring the removal of troops from one frontier zone weakened defences there and left the region vulnerable to attack. In the fourth century the limitanei remained permanently in place. They were not as numerous as the forces on the frontiers in the early empire and could not hope to defeat major incursions. Yet their bases were strongly fortified, as were towns and cities, and they were expected to hold out for as long as they could and harass the enemy. A sizeable army of comitatenses could then be sent to the region to confront the invader. In essence, they formed the mobile reserves that it is claimed the earlier deployment lacked.”
Much of this analysis has been called into question, particularly the sense in which they acted as reserves. Nothing of this sort is ever implied by the ancient sources. No army could ever move faster than an infantryman could march, and more often than not its speed was reduced to that of the plodding draught oxen that pulled its baggage train and carried its food supply. Given the size of the empire, talk of reserves makes little sense, since unless they were fairly close to the theatre of operations then it would take them a very long time to get there. In spite of such criticism, the `mobile field army’ tag has stuck, so that it is worth making a few points about the actual deployment and use of the cornitatenses.
Unlike the lilnitanei, the colnitatenses did not occupy permanent garrison posts. When not on campaign they were stationed within the provinces and not on the frontiers. However, they were not kept concentrated as large army corps ready to take the field at a moment’s notice, since this would have made it difficult to supply them. Temporary camps, the men living in tents or roughly built shacks, were unhealthy in winter, and in any case it was dangerous politically to keep armies concentrated during the winter months of inactivity in case they rebelled. The army did not build large bases in this period, and even many of the existing legionary fortresses designed to accommodate 5,000 men in the second century were now abandoned or substantially run down. At the beginning of the sixth century the historian Zosimus claimed that Diocletian had kept the empire secure by stationing the whole army along the frontiers in strongly fortified posts. `Constantine abolished this security by removing the greater part of the soldiery from the frontiers to cities that needed no auxiliary forces.’ Once there, he claimed that the soldiers became a burden on the communities and were themselves softened by the pleasures of urban life.”
When not on campaign – and even at the most intensive periods of operations these were rarely conducted over winter – the comitatenses were dispersed in towns and cities within the provinces. It is doubtful that more than a brigaded pair of units were often stationed in one place. This spread the burden of feeding them and made it harder for the units to join together and back a usurper. In addition, it provided trained soldiers to man the city walls in the case of a sudden threat from a foreign enemy or a rival for imperial power. There was nothing especially new about stationing troops in or near towns. This had been common in the eastern provinces in the first and second centuries, although there was a well-established literary cliche that maintained it had an enervating effect on them.
Yet in the earlier centuries army units had normally lived in their own barracks within or near to cities. This does not seem to have been the case with the cornitatenses – although admittedly the archaeological evidence for the layout of most towns in the fourth century is extremely limited. Instead, they were billeted in civilian houses, and legal documents talk of officials painting on the door posts the number of men and the unit from which they should come. Throughout history billeting has frequently caused friction between soldiers and civilians. From a military point of view, the dispersal of units in small groups over many separate dwellings was not conducive to good discipline. In their purposebuilt barracks the units of the first- and second-century army had been concentrated in one place under the close eye of their officers. They were provided with good sanitation, exercise and bathing facilities, as well as hospitals for their sole use, and had parade grounds and training areas readily available. Facilities in cities and towns were far more limited and not for the exclusive use of the military. Even the largest city might well struggle to find good stabling for the 500-1,000 horses mustered by a couple of cavalry regiments.
Distributing the comitatenses in cities was the easiest solution for the government, but was scarcely the best way to keep them in good condition. Military training was and is not a simple thing, permanently instilled once it is learned, but something that must be constantly repeated. As important was physical fitness – essential for the marches required on campaign, let alone actual combat. Both were harder to maintain when the army was split and billeted in civilian settlements. There was also inevitably a delay in concentrating the units before a campaign could be begun. The army may have maintained some pack and baggage animals permanently – and if so these were an extra burden to the communities on top of the cavalry mounts – but still needed to requisition or purchase many more to carry its food and other stores in the field. All this took time and faced the added complication of much of it being controlled by bureaucracies entirely separate from and with different priorities to the army.
It is reasonable enough to call the comitatenses field armies. They were more mobile and should usually have been more effective fighting units than the limitanei, at least for large-scale operations. Yet, on the whole, there was nothing particularly innovative about them, certainly nothing that would justify calling them elite. The physical requirements were the same as for service in the earlier army, and later in the century even these would be reduced. Over the course of time, some units of limitanei were attached permanently to the field forces, receiving the halfway status of pseudocomitatenses. This suggests that there was no stark distinction in the military potential of the two grades of troops. In the end, troops were as effective as their training, leadership, tactics and equipment allowed. Only the Persians came close to matching the army tactically, and barbarian armies were markedly inferior. The factory-produced equipment of the fourth-century army has a more functional look than the armour and weapons of the early empire. Both the pilum (heavy javelin) and short gladius (sword) had fallen out of use. Instead, the standard weapons were a long-bladed spatha (sword) – previously only used by cavalrymen – and simpler spears suitable for both thrusting and throwing. As an individual, the Roman soldier was still a well-equipped fighting man. Infantry tactics were probably a little less aggressive, but remained effective. The fourth-century army won the great majority of the battles it fought.
The standards of training and quality of leadership of the fourth-century Roman army inevitably varied – as indeed they had done in earlier periods. On average standards may have been a little lower, but it remained the only professional army in the known world. The demise of the 5,000-man legion removed one command level useful for operating and controlling very large armies in the field. It also meant that it was harder to support large numbers of specialists – engineers, architects, siege specialists, artillerymen etc. – within the army and pass on their experience to successive generations. The comitatenses had no permanent bases to act as depots and to maintain records of personnel, their postings, equipment and mounts. Records were still kept, but had to be moved continually if they were to remain useful to the unit. The same was true of soldiers. New recruits, convalescents or detached men returning to normal duty would have to travel to rejoin the parent regiment wherever it happened to be. The frequency with which a unit went on campaign would have steadily worn it down. Some men would be lost to enemy action, many more to sickness or detachment. This meant that large numbers of men may have existed and been rightfully paid by the state, but would not actually be present with their unit when it went on campaign. The probability that most units were heavily under strength most of the time makes it all the harder to estimate their full theoretical complement.”
We should not exaggerate the efficiency of the fourth-century Roman army, but neither should we forget just how unique it was in its day. For all the complexity of arranging supplies, the Romans did possess a system for organising these things on a massive scale. The fourth-century army was far from perfect, but still enjoyed marked advantages over all its opponents. We need always to remember that it had been shaped by almost a century spent fighting itself. Diocletian and Constantine did not create an army according to a coldly logical design, changing the military system of the second century because it was obsolete. Instead, they patched together a unified force from the badly dislocated debris left by generations of civil wars. Their first priority was always to protect themselves from challengers, and every other consideration was secondary to this. The immediate threat of civil war never went away and continued to dominate the thoughts of their successors. Given this context, the fourth-century army was highly effective, and it is worth now looking at it in action.