The army of the later Empire has had a bad press, being widely regarded as a motley collection of half trained , poorly equipped, incompetent peasant farmers associated with increasing numbers of Germanic barbarians whose very presence diluted and degraded the once proud Roman military machine.

This was an army that, even with its vastly increased numbers, proved ultimately incapable of preventing the destruction of the Empire, so it clearly cannot have been anywhere near as effective as its early Imperial counterpart, or so the theory runs.

In any case, the later Empire was, as Gibbon saw it, a period  of decline and fall, and the army must surely have shared in the general degeneration of Roman life and society.

Actually, this is to misunderstand both the nature and ethos of the later army and that of the society of which it formed a part. The fourth century world may have been very different from that of the Julio-Claudians but that difference does not necessarily imply decline. Such moral judgements seem inappropriate. First century society was not intrinsically 'better' than fourth century society (one only has to think of Britain's fourth century villas and their mosaics) and the later Roman army was just as effective in pursuing the defensive goals for which it was designed as its earlier counterpart was in fulfilling Rome's expansionist mission.

In fact, the main reason for the evolution of the later military system was precisely because the early Imperial army had proved incapable of dealing with the changed conditions and the new threats to the Empire which emerged during the later second and third centuries. Had the system not adapted to change, the Empire would undoubtedly have collapsed considerably earlier than it actually did.

Obviously the later army  doesn't seem to compare very favourably with the expansionist force of earlier days since by its very nature it was a defensive army and defensive armies do not seem to hold much glamour. In addition, in appearance there is no doubting that it seems to have more in common with the Bayeux tapestry than with Trajan's column and one cannot argue with the fact that this has much to do with the increasing proportion of 'barbarian' troops serving Rome in the fourth century.

Yet appearances can be deceptive:

• A defensive army is not necessarily incompetent or weak - how expansionist is the British army today?

• Developing  'barbarisation' does not necessarily imply a decline in standards - even the much derided frontier troops or 'limitanei' were capable of putting up stiff resistance  when the occasion arose, the commissariat was still able to organise the provision of military supplies on a colossal scale while so-called barbarian troops proved disciplined, skilled and conspicuously loyal in their service to the Roman state.
  

• In fact, the recruitment of soldiers from beyond the frontiers was only the culmination of a long-standing Roman tradition of employing the fighting skills of others to their advantage and it is hard to see where sufficient fighting men of the right quality could have been obtained in an Empire which by this time may well have lost much of its fighting edge thanks to its success in spreading the benefits of civilisation and which may well have been experiencing demographic problems. In any case, they did not become preponderant until the very end of the fourth century.
  

• Lack of body armour may make later Roman troops appear 'primitive' next to their earlier counterparts, but this only represents the normal fighting kit of troops from beyond the frontiers which then spread to those recruited from within the Empire. This does not mean that the later Empire was somehow in terminal technological decline - artillery was still very much in use at this time, while heavily armoured cavalry, the cataphracti and clibanarii , were employed in field armies throughout the Roman world and testified to the continuing skill of Roman armourers.
  

Changing Ethos and Strategy.

• First Century
  The ethos of the army of the Principate had been one of aggression and expansion. Problems on  the frontiers were dealt with by military action, often resulting in conquest, and any Emperor worth his salt was expected to be a "Propagator Imperii". Legionary bases and auxiliary forts were simply temporary winter quarters for mobile, offensive armies, and the confident belief was that, sooner or later, the Pax Romana would embrace the entire world.
     One of the first cracks in the confident outlook of divine mission came with the Clades Variani of AD 9 when 3 legions and associated auxiliaries were massacred in the Teutoberger Wald by Germanic barbarians. Clearly, Roman arms were not necessarily invincible, and although further conquests were made, Roman expansion began to slow down. The world outside the Mediterranean basin was looking increasingly unattractive and valueless and Rome was now in any case close to the natural limits of expansion that could be supported by her existing economic and administrative resources.
  
  
• Second Century.

  Gradually, momentum dwindled, temporary halting places hardened into stone built fortresses, the 8 legions poised on the Rhine to overrun 'free' Germany  were reduced to 6 and then to 4, and the limits of advance crystallised into fixed frontiers, delimited first by roads, these then becoming festooned with watchtowers and palisades and finally, in some places, stone curtain walls. The word 'limes' which had once meant a road into enemy territory now came to mean a frontier parallel to enemy territory. Conquest was turning into consolidation and aggression into protection.

      This is not to say that the Romans were hiding behind these frontier works. Far from it, in fact, since the basic military strategy was one of preclusive security, to exclude the enemy from Roman territory completely. This was achieved by maintaining effective intelligence gathering beyond the frontiers thereby facilitating rapid deployment of  the auxiliary forces stationed along the frontier line and of the legions located somewhat further back with the aim of confronting the enemy on their territory, well beyond the limits of direct Roman control. (This is assuming, of course, that the often preferred diplomatic option was not being pursued).

• Third Century.

Although preclusive security remained the ideal down to Diocletian and beyond, the varying threats to Roman territorial and political integrity which emerged in the third century made this an increasingly unattainable goal. At the same time as Rome dissipated her energies in civil war, her enemies beyond the frontiers were growing ever more powerful, with the revived Persian Empire inflicting a series of humiliating defeats in the East and, in the West, new confederations of barbarians such as the Germanic Alamanni (or 'all men') and Franks making damaging inroads into Roman territory, one of the worst being in 276 when perhaps 70 Gallo-Roman cities were plundered and devastated.

Such incursions occurred with alarming regularity along the length of the frontiers during this period, revealing the inability of the existing military arrangements to cope with the new situation. Legions and auxiliaries had put down roots and become relatively static; once the barbarians had broken through the relatively thinly held frontier line, there was nothing to prevent them from plundering soft targets in the interior provinces at will. The massive walls of Aurelian, built in the 270's to defend Rome herself, are eloquent testimony to the new insecurity.

Yet a new approach was developing. Third century Emperors spent little time at Rome, often originating in frontier provinces and spending most of their careers where the action was. Naturally enough they tended to take their court around with them and attached to this was the comitatus or 'companions', at first (perhaps under Gallienus in mid-century) an assemblage of cavalry units possibly intended to form a highly mobile central reserve which would enable the Emperor to respond rapidly to any new incursions (or usurpations?) that might arise This may well mark the first stirrings of a new military strategy for the Empire.

• Fourth Century.
  Some historians believe that Gallienus' force represents the precursor of the mobile field armies of the fourth century, later developed by Diocletian but brought to full fruition during the reign of the Emperor Constantine. Diocletian was certainly a crucial agent of recovery, responsible for restoring Rome's frontiers, but it appears to have been Constantine who finally abandoned the old idea of preclusive security in favour of a more flexible frontier-in depth approach to frontier control.

• Later Roman Military organisation.
  
  Constantine spent much of his reign in military conflict, and out of his experience was "forged the New Model Army of the Later Roman Empire". Gone was the traditional division between legionaries (citizen heavy infantry) and auxiliaries ( non-citizen light infantry and cavalry) , to be replaced by a new distinction between the elite troops of the Emperor's mobile field army, the comitatenses and the lower status ( though not necessarily less effective) limitanei or frontier troops.

• The Field Army
  

  The Constantinian field army of comitatenses, first officially recorded in 325, comprised both cavalry and infantry under new commanders known as the Magister Equitum and the Magister Peditum.. At this stage it retained its close physical link with the person of the Emperor,( although it should not be confused with the elite units of the Imperial bodyguard, the Scholae Palatini) but very quickly it became obvious that the extent of the Empire meant that one central force could not deal effectively with simultaneous threats on several fronts.
  
  The result was that we find detachments of comitatenses being stationed away from the Emperor and turning into regional field armies , the most significant being those of Gaul, the East, and Illyricum. These were mobile troops who were not based in permanent forts, but usually billeted in towns, a local rapid reaction force ready to respond to any barbarian threat on their 'patch'. They were commanded by professional soldiers known as 'comites' or counts, now that the old provincial governors had  lost their military powers, and they proved  a credible and effective response to the needs of the time