Dedicated to the military history and civilization of the Eastern Roman Empire (330 to 1453)

"Time in its irresistible and ceaseless flow carries along on its flood all created things and drowns them in the depths of obscurity."

- - - - Princess Anna Comnena (1083–1153) - Byzantine historian

Sunday, July 7, 2013

The Birth of the Eastern Roman Army (395 AD)

January 17, 395 AD
The Birthday of the Eastern Roman Army

The birthday of the army of the Eastern Roman Empire.  It is a rare thing for a historian to be able to point to an exact day for a major historical change.

But on January 17, 395 Theodosius I (r. 379-95), the last Emperor of a united Roman Empire died.  The day before on January 16th, Emperor Theodosius commanded Roman troops stationed from Mesopotamia to Morocco to England to Bulgaria.  But at some point on the 17th a sole commander-in-chief of the Roman military machine died.

The death of the Emperor led to the final split of the Empire into two political entities, the West (Occidentale) and the East (Orientale).

For many decades to come the Eastern Roman Army would not have looked or acted much different from its Western counterpart fighting off the barbarian invasions in Gaul and Italy.  Any changes in unit structure, uniforms and tactics would have been very gradual.  The Eastern Roman military evolution would have been based on changes the economy and the types of enemies they faced.

Emperor Theodosius I
The last Emperor of a united Empire.
The Roman Legion would fade and Eastern infantry units would evolve to be more defensive in nature in order to man fortresses and strong points against invaders.  Eastern Roman Cavalry units would mirror their Persian enemies and would grow to become the mailed fist of the army in combat.

The general military organization of the East would roughly stay the same until the Arab invasions of the 600s.  With  the permanent loss of Syria, Palestine, Egypt and Carthage to the Arabs in the 7th century the Eastern Roman Army reorganized by themes to better react to local incursions by Muslim forces.

The Eastern Roman Army is thus the intermediate phase between the Late Roman Army of the 4th century and what could be called the Byzantine Army of the 7th century onwards.  Though it should be noted, they easterners always called themselves Romans up to the fall of Constantinople in 1453.

The division into two sections recognized a growing cultural divergence. The common language of the East had always been Greek, while the West was Latin-speaking.

This was not per se a significant division, as the Empire had long been a fusion of Greek and Roman cultures and the Roman ruling class was entirely bilingual. But the rise of Christianity strained that unity, as the cult was always much more widespread in the East than in the West, which was still largely pagan in 395.

Constantine's massive reconstruction of the city of Byzantium into Constantinople, a second capital to rival Rome, led to the establishment of a separate eastern court and bureaucracy.

The Army of Theodosius I (395)


The size of the Eastern army in 395 is controversial because the size of individual regiments is not known with any certainty. Plausible estimates of the size of the whole 4th-century army (excluding fleets) range from c. 400,000 to c. 600,000. This would place the Eastern army in the rough range 200,000 to 300,000, since the army of each division of the empire was roughly equal.
Late Roman Empire Infantry

The higher end of the range is provided by the late 6th-century military historian Agathias, who gives a global total of 645,000 effectives for the army "in the old days", presumed to mean when the empire was united. This figure probably includes fleets, giving a total of c. 600,000 for the army alone.

Agathias is supported by A.H.M. Jones' Later Roman Empire (1964), which contains the fundamental study of the late Roman army. Jones calculated a similar total of 600,000 (exc. fleets) by applying his own estimates of unit strength to the units listed in the Notitia Dignitatum. Following Jones, Treadgold suggests 300,000 for the East in 395.

But there are strong reasons to view 200,000 soldiers as more likely:
  1. Jones' assumptions about unit strengths, based on papyri evidence from Egypt, are probably too high. A rigorous reassessment of the evidence by R. Duncan-Jones concluded that Jones had overestimated unit sizes by 2-5 times.
  2. The evidence is that regiments were typically one-third understrength in the 4th century. Thus Agathias' 600,000 on paper (if it is based on official figures at all) may in reality have translated into only 400,000 actual troops on the ground.
  3. Agathias gives a figure of 150,000 for the army in his own time (late 6th century) which is more likely to be accurate than his figures for the 4th century. If Agathias' 4th- and 6th-century figures are taken together, they would imply that Justinian's Empire was defended by only half the troops that supposedly defended the earlier empire, despite having to cover even more territory (the reconquered provinces of Italy, Africa and S. Spain), which seems inherently unlikely.
The discrepancy in army size estimates is mainly due to uncertainty about the size of limitanei regiments, as can be seen by the wide range of estimates in the table below. Jones suggests limitanei regiments had a similar size to Principate auxilia regiments, averaging 500 men each. More recent work, which includes new archaeological evidence, tends to the view that units were much smaller, perhaps averaging 250.

There is less dispute about comitatus regiments, because of more evidence. Treadgold estimates the 5 comitatus armies of the East as containing c. 20,000 men each, for a total of c. 100,000, which constitutes either one-third or one-half of the total army.

About one third of the army units in the Notitia are cavalry, but cavalry numbers were less than that proportion of the total because cavalry unit sizes were smaller. The available evidence suggests that the proportion of cavalry was about one-fifth of the total effectives: in 478, a comitatus of 38,000 men contained 8,000 cavalry (21%).

In 395 AD the Roman Empire is forever divided into two nations
under two Emperors, two Senates and two armies and two navies. 

Click on chart to enlarge
High command structure of the East Roman army c. AD 395. Commands and army sizes based on data in the Notitia Dignitatum Orientis. Eastern magistri militum, in command of comitatus armies, reported direct to the Emperor. Duces are shown reporting to their diocesan magister militum, as suggested by Jones and Elton. Locations given indicate usual winter quarters in this period.

Command structure

The later 4th-century army contained three types of army group: (1) Imperial escort armies (comitatus praesentales). These were ordinarily based near Constantinople, but often accompanied the emperors on campaign. (2) Regional armies (comitatus). These were based in strategic regions, on or near the frontiers. (3) Border armies (exercitus limitanei). These were based on the frontiers themselves.

The command structure of the Eastern army, as recorded in the Notitia Dignitatum, is represented diagramatically in the organisation chart (above).

By the end of the 4th century, there were 2 comitatus praesentales in the East. They wintered near Constantinople at Nicaea and Nicomedia. Each was commanded by a magister militum ("master of soldiers", the highest military rank) Each magister was assisted by a deputy called a vicarius.

There were 3 major regional comitatus, also with apparently settled winter bases: Oriens (based at Antioch), Thraciae (Marcianopolis), Illyricum (Sirmium) plus two smaller forces in Aegyptus (Alexandria) and Isauria. The large comitatus were commanded by magistri, the smaller ones by comites. All five reported direct to the eastern Augustus. This structure remained essentially intact until the 6th century.

Late Roman Empire Cavalry
About half the Eastern Roman Army continued to be recruited in
the Latin-speaking Danubian regions of the Eastern Empire.
Latin remained as the operating language of the army into the late 6th century.


Regiments were classified according to whether they were attached to the comitatus armies (comitatenses) or border forces (limitanei). Of the comitatenses regiments, about half were palatini (literally: "of the palace"), an elite grade.

The strength of army regiments is very uncertain and may have varied over the 5th/6th centuries. Size may also have varied depending on the grade of the regiment. The table below gives some recent estimates of unit strength, by unit type and grade:


Estimated size of regiments in 4th-century army
unit type
(inc. palatini)
unit type
(inc. palatini)
Ala120-500Auxilia800-1,200 or 400-600400-600

The overall picture is that comitatenses units were either c. 1,000 or c. 500 strong. Limitanei units would appear to average about 250 effectives. But much uncertainty remains, especially regarding the size of limitanei regiments, as can be seen by the wide ranges of the size estimates.

Scholae  -  The Scholae Palatinae were elite cavalry regiments that acted as imperial escorts. At the end of the 4th century, there were 7 scholae (3,500 men) in the East. They were outside the normal military chain of command as they did not belong to the comitatus praesentales and reported to the magister officiorum, a civilian official. However, this was probably only for administrative purposes: on campaign, the tribunes (regimental commanders) of the scholae probably reported direct to the emperor himself. 40 select troops from the scholae, called candidati from their white uniforms, acted as the emperor's personal bodyguards.

Comitatenses  -  Comitatenses cavalry regiments were known as vexillationes, infantry regiments as either legiones or auxilia. About half the regiments in the comitatus, both cavalry and infantry, were classified as palatini. They were concentrated in the comitatus praesentales (80% of regiments) and constituted a minority of the regional comitatus (14%). The palatini were an elite group with higher status and probably pay.

The majority of cavalry regiments in the comitatus were traditional melee formations (61%). These regiments were denoted scutarii, stablesiani or promoti, probably honorific titles rather than descriptions of function. 24% of regiments were light cavalry: equites Dalmatae, Mauri and sagittarii (mounted archers). 15% were heavily armoured shock charge cavalry: cataphracti and clibanarii

Limitanei  -  In the limitanei, most types of regiment are present, including the old-style alae and cohortes of the Principate auxilia.

The limitanei, meaning "the soldiers in frontier districts" (from the Latin phrase limes, denoting the military districts of the frontier provinces established in the late third century).

They reflect the organization of the Late Roman army and subsequently the Byzantine Empire. They were light infantry similar to spear men and served as a policing force to patrol Rome's distant, far-flung border regions and when necessary, to delay advancing enemy forces until counter-attacks could be arranged. They are historically significant in that their appearance, as part of a plan of military reforms enacted in the late 3rd century, was able to extend the life of the Roman Empire by pushing back the great barbarian invasions of late antiquity. They worked in conjunction with the comitatenses.


In 395, the army used Latin as its operating language. This continued to be the case into the late 6th century, despite the fact that Greek was the common language of the Eastern Empire. This was not simply due to tradition, but also to the fact that about half the Eastern army continued to be recruited in the Latin-speaking Danubian regions of the Eastern Empire.

An analysis of known origins of comitatenses in the period 350-476 shows that in the Eastern army, the Danubian regions provided 54% of the total sample, despite constituting just 2 of the 7 eastern dioceses (administrative divisions): Dacia and Thracia. These regions continued to be the prime recruiting grounds for the East Roman army e.g. the Emperor Justin I (r. 518-27), uncle of Justinian I, was a Latin-speaking peasant who never learnt to speak more than rudimentary Greek.

The Romanized Thracian (Thraco-Roman) and Illyrian inhabitants of those regions, who came to be known as Vlachs by foreigners in the Middle Ages, retained the Roman name (Romanians) and the Latin tongue.

Shield insignia of regiments under the command of the Magister Militum Praesentalis II of the East Roman army c. 395 AD. Page from the Notitia Dignitatum.

The insignia of the Eastern scholae, from the Notitia Dignitatum. 
The Scholae Palatinae (literally "Palatine Schools", in Greek: Σχολαὶ, Scholai), were an elite military guard unit, usually ascribed to the Roman Emperor Constantine the Great as a replacement for the equites singulares Augusti, the cavalry arm of the Praetorian Guard.
The Scholae survived in Late Roman and later Byzantine service until they disappeared in the late 11th century, during the reign of Alexios I Komnenos. 

Late Roman Empire Infantry

(Late Roman Army)          (Byzantine Army)

(Theodosius I)          (Eastern Roman Army)


Anonymous said...

Why do you think they shifted from the rectangular to the round shield?

Gary said...

Somewhere there is someone far smarter than I am who knows the answer to that question. It might be as simple as change for the sake of charge or fashion.

Anonymous said...

I think it may be, because that became mose usefull for mobility , cause the ancient formation of the roman legion is usefull on massive cavalry charges as the goths or the persians used in those days . That may be the reason .
Not fashion .