by Ian Hughes
Discussions with Placidia and Valentinian followed. Although unattested, alongside Aetius’ use of the Huns as a bargaining counter it is possible that the prestige accorded to his mother’s family also played a part. If the sources are correct in saying that she was rich, it is probable that she had strong ties to the western Senate. In that case, Placidia would not wish to simply execute Aetius, as this could provoke unnecessary resentment in the Senate. Placidia would need the wholehearted support of the Senate to ensure a smooth transition of power to her son Valentinian.
Aetius accepted the position of a high-ranking military officer in the new regime.58 In return, he negotiated a treaty with the Huns. They were paid a large amount of gold as compensation for their travel and losses, and after an exchange of hostages and oaths the Huns departed for home. The concept that Aetius also ceded territory in Pannonia to the Huns at this time is mistaken and belongs to a later episode in his life.59
However, Placidia would never forget that Aetius had supported John against her son: the two would remain political opponents and she would not hesitate to manipulate him in her attempts to ‘foil the ambitions of any general who showed signs of becoming too powerful for her liking’.60
Having secured the West, Placidia and Valentinian travelled to Rome where on 23 October 425 Valentinian was proclaimed emperor by Helion, cousin of Theodosius II. Valentinian was now just six years old. Theodosius had wanted to preside at the ceremony himself, but had been prevented by illness.61 Upon securing the throne Placidia made a man named Felix the new magister militum praesentalis, at the same time giving him the title patricius.62 Nothing is known about Felix before this appointment. The lack of information suggests that he was not an outstanding military leader. As a result, it is likely that his appointment was due to him being well known and trusted, at least by Theodosius II in Constantinople, and that he was seen as a loyal individual who would not pose a threat to Placidia.
One of the first decisions of the new regime was to send Castinus into exile rather than have him executed.63 Castinus’ decision to maintain a low profile during John’s reign saved him from death.64 In the hope of gaining support, and in recognition that Rome had still not fully recovered from the sack by the Goths, in early 426 one of the main taxes on the city of Rome and the Senate, the aurum oblaticium, was remitted by the new emperor.65
The senior post given to Aetius was, in the circumstances, an extremely important one, and one in which he would be allowed to demonstrate his military abilities to the full. He was made magister militum per Gallias.66 However, it should be noted that at some time in the past, and presumably during John’s usurpation, Aetius’ father Gaudentius had been killed by an uprising of the troops in Gaul.67 Although the promotion was a major advancement for Aetius, it is possible that Placidia and Valentinian may have been hoping that Aetius’ ambition would be held in check, or that he might even be assassinated, due to animosity felt by the troops for his father being carried over to Aetius. Aetius’ acceptance of the risky appointment can be seen as his accepting that he needed to prove his loyalty to the imperial court, even at the risk of his own life. Just as importantly, the appointment also removed Aetius from the court in Italy, allowing Felix time to consolidate his power as magister militum without interference from Aetius and his relatives.
Aetius’ rise to power had been remarkably fast. In under two years he had gone from being the cura palatii of an usurper to the magister militum of an accepted emperor. However, his meteoric rise had one major drawback: he was now in political opposition to Galla Placidia, who would find it hard to forgive him for his support of John, and of the other two major players in the west: Felix, the new magister militum praesentalis, and Boniface. In return for his undying loyalty throughout her troubled time, Galla Placidia promoted Boniface to comes domesticorum et Africae (Count of the Household and of Africa), ordering him to stay in Africa – possibly to ensure that the vital province remained peaceful and trouble free.68 Both of these men had more influence than Aetius with Galla Placidia. Yet in reality she was forced to play the three men off against each other in an attempt to maintain a level of independence for herself and Valentinian, and it would not be long before Aetius was thrust once again into the centre of imperial politics. In the meantime, he prepared for life as the army commander in Gaul.
Chapter 4
The Late Roman Army1
COMMAND HIERARCHY
Modern historians analyzing the Notitia Dignitatum, and especially those with a military background, have tended to relate the linear, rigid hierarchy apparent in its pages with the modern armies of the ‘advanced’ nations.2 This fails to take into account the fact that the Notitia may be depicting an ideal, and reality is rarely this organized. For example, in many cases the posts depicted may have been only intended to be occupied for a short period under a specific emperor, or they may have been created by an emperor for a personal favourite but then left vacant after the individual retired from service.
Unfortunately, it has been relatively common in the past to take the titles given in the Notitia and link them to the titles given to individuals in other sources. On the other hand, it should be remembered that ancient writers were not usually interested in applying strict military rankings according to an accepted hierarchy: in many cases the writer was more concerned that the terminology follow the flow of the narrative or the metre of the poem to consider such niceties. As a result, the use of titles such as magister peditum, magister equitum, magister utriusque militiae and so forth may all actually represent the same post, depending upon the authority, accuracy, mode (history, poetry and chronicle) and sources used by the ancient author in question. Furthermore, there are numerous examples in the sources of individuals with the title magister militum vacans. Although sometimes given to military commanders conducting specific campaigns beyond the remit of the usual magistri, such as the campaigns in Spain under Aetius, the title could also be used simply to reward a loyal follower with the prestige and privileges attached to the post of magister militum.
Furthermore, the Notitia describes the standing army in time of peace. During a military campaign ad hoc command structures would have been utilized, rather than a strict adherence to the peacetime organization.3 Some of the ‘promotions’ may then have continued in existence as a form of reward from the emperor for a job well done. However, these did not fit easily into the command structure represented in the Notitia and may account for some of the confusion both in the document and in the existing written sources.
There are two final warnings with regard to the Notitia. The first is that the original was a working document. The political and military complexities inherent in the West in the early fifth century mean that, almost without doubt, the Notitia was out of date when the final list was completed. This explains why there are doublets and omissions from the text.4 The second is that this is not a document from the fifth century: it is a copy of a document, and as such may contain many errors and omissions that were not in the original.
Yet with all of these caveats, the Notitia is still of vital importance to the historian, since it is one of the very few documents that, whatever its failings, gives an insight into the intended working of the Roman military and bureaucratic infrastructure.
REFORMS
The reign of Constantine I (306–337) saw the praefecti praetoriani lose their military powers. These tasks were assumed by the newly created magister peditum (‘master of the infantry’) and magister equitum (‘master of the cavalry’). By at the latest the midfourth century a third, subordinate magister, plus attendant troops, was added in the West, the magister militum per Gallias (Master of the Troops in Gaul). These three magistri controlled the comitatenses (on the troop types and other ranks, see below) and in theory had authority above the provincial duces (‘dukes’), who controlled the limitanei in the provinces. However, the duces retained the right to correspond directly with the emperor, so the commonly perceived linear hierar
chy is not actually present.
Since the new magistri commanded large bodies of troops, for his own security the emperor retained control of the elite scholae via the magister officiorum.5 These troops would act as the nucleus for an army that would defeat any magister unwise enough to raise the standard of revolt.
Below the various magistri was an assortment of comes (counts) and duces (dukes). Attempts to maintain the illusion of a linear hierarchy are doomed to failure, as it seems likely that the designations comes and dux were given by different emperors depending upon the individual circumstances surrounding the appointment.
There is one last proviso that needs to be remembered. The fluid nature of the ranks and titles allotted within Roman Empire was not hidebound by modern preconceptions of ‘amateur’ and ‘professional’ soldiers. The result was that in an emergency almost any imperial official could act as a military commander, regardless of whether he was of the ‘correct’ rank, or even if he was a soldier.6
With these observations in mind, it is clear that great care needs to be taken before deductions are made based on the specific titles held by military commanders. Due to the improvisational nature of many appointments, it may be possible to draw up a table showing the chain of command under one or possibly two consecutive emperors. To attempt to do the same for the entire fourth and fifth centuries would be impossible.
UNIT HIERARCHY7
During the third and fourth centuries the army was divided into two main components: the comitatenses and the limitanei. The comitatensis, usually translated as ‘field army’, was stationed to the rear of the borders and was, at least theoretically, a ‘mobile’ army that could be used at any point within the empire. The comitatenses were further divided into two, with the ‘ordinary’ units of comitatenses being slightly lower in rank than the palatina or ‘palace’ troops. Furthermore, some units appear to have been divided into seniores and iuniores, with the seniores having superior status.
The limitanei were troops stationed permanently on the limes (borders). The limitanei are given different names in the sources, being known as riparienses, ripenses, castellani or burgarii. Although in theory these names may indicate a differentiation in the nature of the troops, what differences there were between them remains unknown. Whatever the nature of the troops, the terms legiones and cohortes appear to have been used for the infantry, with ala and equites being used for the cavalry.8
To add to the complexity, over the course of time many of the limitanei were taken from the frontiers and transferred to the ‘field army’, being given the title pseudocomitatenses, and units could be punished by being downgraded from comitatenses to limitanei.
Slightly outside this two-tier system were the scholae, guard cavalry units under the personal control of the emperor. However, since during the course of the fifth century the western emperor no longer went to war in person, over time the scholae gradually lost their martial status and became ceremonial units. Finally, the emperor himself chose individuals from among the scholae to form the candidati, his personal forty-man bodyguard.9
DISTRIBUTIO NUMERORUM
Included within the Notitia Dignitatum is a document with the title Distributio Numerorum (‘Distribution of the Army Units’). This attempts to list the distribution of the army throughout the West.10 A close analysis of the Notitia, and especially of the Distributio, has suggested that between the years 395 and 420 nearly fifty per cent of the western ‘field army’ was lost.11 This was due mainly to the civil wars at the end of Theodosius I’s reign (379–395) and the civil wars and invasions from 401 until the settlement of the Goths in Aquitaine in 418–419.12 The losses in manpower, due to fighting, desertion and possibly the disbanding of rebellious units, combined with the loss of tax revenues caused by the continuous wars, decimated the ‘field-armies’. To plug the gaps, Stilicho and Constantius III had transferred many units of the limitanei to the field army as pseudocomitatenses.13 The result was that some areas of Gaul and Africa were virtually stripped of limitanei. Although the consequences of these actions were to be dire, this is only apparent in hindsight. The treaties with tribes along the frontiers in Gaul, and especially the Franks, plus the fact that Africa was not exposed to any serious military threats, meant that the withdrawal of troops did not result in any immediate repercussions.
Despite these problems attempts have been made to estimate the size of the Roman army in the fifth century, and especially of the ‘field armies’ that faced the barbarian invaders, and sometimes each other, in the field. This is one estimate of the size of the different armies:
WEST
Western Illyricum 13,500
Gaul 34,000
Praesental 28,500 (by the time of Aetius there was usually only one magister in the west)
EAST
Thrace 24,500
Illyricum 17,500
Oriens 20,000
Praesental 42,00014
It should be remembered that these are ‘paper’ strengths according to the Notitia Dignitatum. They do not take into account the fact that the Notitia was by the mid 420s becoming obsolete. It is possible that many of the units of the limitanei that were listed may have had no existence outside the Notitia, especially in Africa and Spain, since at least some of these troops had been called to the ‘field army’ during the earlier rebellions and invasions and never returned to their places of origin. Further, there would be constant attrition due to losses in battle, by accident, from troops being ill or hospitalized, or even troops being deployed on non-military duties away from the main body of the army. The figures also do not take into account difficulties with recruitment. It is possible that in extreme cases the troops available for combat could be as low as half- to one-third this number, although this may be taking too negative an approach.
NON-ROMAN UNITS
The terminology used for non-Roman units is extremely confusing. This is in part because the Romans appear to have used the same terms for units of widely differing origin. Furthermore, as the sources do not use the terms in a strict, military sense, but often use whichever word fits with the style or meter of the work being written, certainty is impossible.15
Foederati
A major example of this is the use of the term foederati. Originally the term was applied to non-Romans serving in regular army units under the terms of a foedus (treaty), possibly such as that signed by Constantine I (306–337) with the ‘Ripuarian’ Franks, or that signed by Julian (355–363) with the ‘Salian’ Franks.16 Many of the barbarian units listed in the Notitia probably began their existence as wholly barbarian units, yet it should be noted that these were ‘regular’ Roman units, trained, supplied and equipped by the empire. As time passed they almost certainly lost their ‘barbarian’ status due to the enlistment of non-barbarian troops in their ranks.17
However, over time the term foederati came to be applied more loosely. As a result, when the words foederatus or foederati are used by the sources, the full context needs to be determined before any deductions are made concerning the exact nature of the troops being described. Therefore, for the purposes of this book, the term foederati will be used exclusively with regard to barbarian troops serving under their own leaders in non-Roman units, whether these are ‘internal’ forces such as the Goths or ‘external’ forces such as the Huns.
Bucellarii
During the reign of Theodosius I (379–395) the magister militum Stilicho and the praefectus praetorio Orientem Rufinus were probably the first individuals to begin recruiting bucellarii, barbarian troops serving as personal bodyguards to military and political leaders as well as the emperor.18 Indeed, the employment of bucellarii quickly became fashionable and Stilicho was forced to pass a law restricting the size of these ‘personal’ armies.19 As a result, during the lifetime of Aetius such troops were still limited in number. It is only later, and especially in the reign of Justinian I (527–565), that large numbers began to be recruited by powerful individuals.
The t
endency of late Roman military commanders to supplement their forces with barbarian mercenaries is understandable, although it was contrary to the aristocratic belief that Rome should be defended by Romans. Magistri such as Stilicho and Aetius may have understood that employing mercenaries was in many ways beneficial to Rome. Rather than entering the army, Roman civilians were able to continue their trades and pay the taxes needed to employ mercenaries. Furthermore, the troops employed by the empire were denied to the barbarian leaders, reducing their effectiveness whilst simultaneously increasing the power of the defending forces.
UNIT STRENGTH20
Along with the change in structure came a change in the organization of the units within the army. Although figures remain conjectural and different authorities have different ideas, modern estimates of the strength of the various units is roughly as follows: Guard units (scholae21 etc.) 500 men; auxilia palatinae 800–1,200 men; legions (comitatenses) 1,000–1,200men; legions (limitanei) 3000 men; limitanei/riparienses 300–500 men; cavalry (comitatenses) 480–600 men; cavalry (limitanei) 350–500 men. However, the units entitled milliary in the Notitia possibly numbered c.750–1,000 men, as they had in the third century, and some legionary units may have retained their earlier strength of c.5,000 men until much later.22 An additional factor is that, in an army covering such a vast area as the Roman Empire, there would also be regional differences between units.23 Finally it should be remembered that on campaign units could easily be as low as two-thirds strength due to illness, injury and the need to leave a depot garrison behind at the unit’s base for administrative and recruitment purposes.24 Any numbers given for Roman troops in battle must remain hypothetical.
