The significance of Adrianople, 378 A.D.

[Spartan JKM]

We've been having a great discussion (myself, ThiudareiksGunthigg, and Stalins Ghost) regarding the nature of the famed battle of Adrianople and the origins of the stirrup on another thread; but Marcus Scaurus (you change your username Red Baron?) made a good point - we began to deviate from the thread's original content, that of why the Romans succeeded militarily for so long, so we should begin anew.

This is how I basically see it. There were no stirrups used at Adrianople by the Goths, Alans and their other briganded friends; there is just no concrete evidence to support their use. These horsemen, like all horsemen growing up riding and fighting on horseback, could ride adeptly without stirrups. Stirrups would be a necessity for the mounted archers whomaneuvered while implementing firepower upon thier enemies, most notably the Mongols, to achieve what they excelled at, but the saddles these horsemen had were efficient enough for stability. Despite the wonder as to why the stirrup wasn't in Europe earlier, being it was in the eastern lands perhaps 1/2 millennia earlier than when it was used prevalently in Europe, there is simply no evidence to support their use. I don't see how it could have 'gotten by anybody' until the 7th century, when in fact it did arrive.

Another element which improved the capabilities of cavalry was the horse-shoe. Rough terrain could be manipulated on shod horses without long periods of time afterwards for the animals to heal in soft pastures. Anyone know when this came along?

The battle of Adrianople was significant because it irrevocably changed the conditions on which barbarians and Romans would deal with each other in the future, and Rome might have been sacked sooner if not for the diplomatic astuteness of Theodosius I; his offer to the vagabonding Goths in the Balkan regions of allied status (foederati) with Rome, while retaining their own sovereignty, proved very efficacious until his death in 395 A.D. There was something more appealing for many barbarian chieftains about leading their warriors to fight along the ranks of the Roman military system than merely leading their own federated mobs (somewhat) into battle against Romans. They were probably thinking along the lines of stature, such as gaining war ministries in the empire. Moreover, Theodosius paid them higher than the Roman troops (reputedly), and showed them greater consideration. Trained units were also transferred from Egypt to replenish Thedosius' ranks.

The result of the battle of Adrianople also seemed to signify the reality to the Eastern Empire's military that efficient cavalry must be adopted more fully. Part of Theodosius' foederati was perhaps 40,000 conscripted Gothic warriors, many being cavalry. This must have been instrumental in the vanquishing of Theodosius' rivals, Magnus Maximus and Eugenius by 394 A.D. Actually, Alaric was part of the campaign that crushed Eugenius.

Now, I would like to pose three major questions (Tim will just have to paste over his intelligible posts form earlier ) I will refrain from addressing anything in depthly on the first two right now until, or if, the thread gets going.

Was the battle of Adrianople the advent of the cavalry cycle - the so-called tactical revolution in the relation between cavalry and infantry which would influence the period when heavy cavalry, typified by the Medieval knight, dominated the battlefields of Europe for another 1,000+ years?

Contrarily, was the battle simply a battle lost catastrophically by the emperor Valens because of his tactical and intelligence failures, combined with a swift tactical surprise from the Gothic horsemen? But remember, it wasn't an ambush akin to what Hannibal achieved at Trasimene, Scipio's 'Burning of the Camps' (night ambush), or the Teutoburger Wald. The clash at Adrianople was fought on an open plain.

Did the Gothic cavalry contemporaneously send the Roman horsemen into flight and aid their infantry in the destruction of their enemy? Or was it really a battle won by Gothic infantry over Roman infantry, who were already thrown into disorder (terribly tight-packing of their ranks) by the initial salvo of Gothic cavalry, who all left the field chasing Roman cavalry, thus being merely a preliminary supplement?

Some reputable modern sources state unequivically 'no' to the cavalry-cycle issue. I agree; this battle led to a Roman catastrophe due to faulty intelligence and tactical surprise from the Gothic cavalry. The arms and tactics employed were nothing new from both sides. This wasn't like Crecy of 1346.

I think, basically, both armies numbered between 20-30,000 men, with Valens having an edge in quality and quantity (slight edge in numbers). But Valens thought he was approaching an enemy force, primarily Fritigern's footmen housed in their wagonlaager (wagon-fort), composed of slightly more than 1/2 of what it actually was. Fritigern had sent for his allied horsemen, out foraging 'far away', and was 'begging' a truce with Valens, sending out envoys of 'low origin' initially, whom Valens refused to see. Valens wanted to be sent a suitable chieftain to execute a lasting treaty. Fritigern then sent a 'common' soldier as a herald, requesting that picked men of noble rank among the Romans be sent to him as hostages, and he would tolerate the anger felt by his warriors because of his virtual overture of peace. But, in actuality, he was purposely delaying, hoping that amid the pretended truce his cavalry might return. Valens approved, but may have been also buying time to deploy efficiently (Marcellinus doesn't say this on the part of Valens). This makes sense, as the wagonlaager was, presumably, well situated and difficult to assault, and Valen's men were weary and the day was extremely hot.

Fritigern also set the surrounding land afire, thus exposing further with time the Romans to such horrid conditions in the 'fiery summer', exhausting them by heat and thirst. Valens had already imposed a forced march on his men to get here. Hunger began to set in for the Romans, too.

Marcellinus tell us of Valens' emissary, one Richomeres,

"...As he was on his way to the enemy's rampart, the archers and targeteers, then under the command of one Bacurius of Hiberia and Cassio, had rushed forward too eagerly in hot attack, and were already engaged with their adversaries; and as their charge had been untimely, so their retreat was cowardly; and thus they gave an unfavorable omen to the beginning of the battle..."

This untimely action, perhaps one of insubordination, is believed to have been instigated by the right (we don't know for sure) Roman cavalry. Then, like a 'thunderbolt does near high mountains', the cavalry under Alatheus and Saphrax arrived, and threw themselves into the 'confusion'.

This is the Marcellinus' overview of the entire battle, Book 31.13 (Loeb Classic Library),

"All the Goths united, namely, the Theruingi under the command of King Fritigern, and the Greuthungi led by Alatheus and Saphrax, and engaged with the Romans in the open, routed their cavalry, and put to flight the infantry, thus left unprotected and crowded together, with enormous losses; Valens was killed, but his body could not be found."

Just a pedantic observation - 'All the Goths united', with no specific mention of all of the Gothic cavalry leaving the field :hmmm:. Thus, in my opinion, the revised theory that this was a battle of technically infantry over infantry is not incontrovertible. But Marcellinus' account does carry some holes.

Right from the beginning, Marcellinus states,

"On every side armor and weapons clashed..."

A few sentences later,

"...the lines dashed together like beaked ships, pushing each other back and forth in turn..." (could infantry alone, of no more quantity, cause this?)

He then tells us that the left cavalry wing of the Romans, deserted by some of their comrades, was hard pressed by the enemy's numbers, crushed, and overwhelmed, the infantry now unsupported. Again, he does not say the Gothic cavalry left the field. But he doesn't say the stayed either. Gosh, I wish he had specified!

He continues,

"...Because of the clouds of dust the heavens could no longer be seen, and echoed with frightful cries. Hence the arrows whirling death from every side always found their mark with fatal effect...But when the barbarians, pouring forth in huge hordes, trampled down horse and man...no room for retreat..."

Now, I hope the 'picking and choosing' doesn't signify a taking out of context, but think it is plausible, though the word 'cavalry' doesn't come up from Marcellinus after the initial charge of the Gothic horsemen, that phrases such as, "On every side", and, "Because of the clouds of dust the heavens could no longer be seen", and, "arrows whirling death from every side", and, "barbarians, pouring forth (not 'out', as out of their wagon-fort), in huge hordes, trampling down horse and man", and, finally, "no room for retreat", can not mean only the Gothic infantry itself surrounded the Romans, which has been a major belief of the current appraisals of the battle (Thomas Burns and Stephen Williams, to name a couple). These beliefs have been accepted, and are very credible (I have skimed Williams' excellent Theodosius: The Empire at Bay) on many issues (IMHO). This is definitley not a black & white issue. Much of the Gothic cavalry vanquished the Roman horsemen with great celerity, mainly because of the element of surprise, but some seemingly swung inwards to aid in the envelopment of the body of infantry, already thrown into disarray. Men alone do not "trample down horse and man" (do they?).

Perhaps I am being too academic, and ancient accounts can suffer from 'static' with the translations down the timelines. One of the eminent and panoramic works of Roman history, especially of this period, says this of the battle of Adrianople,

"On the 9th of August 378 A.D., a day which has deserved to be marked among the most inauspicious of the Roman Calender, the emperor Valens, leaving under a strong guard his baggage and military treasure, marched from Hadrianople to attack the Goths, who were encamped about 12 miles from the city....The event of the battle of Hadrianople, so fatal to Valens and to the empire, may be described in a few words: the Roman cavalry fled; the infantry was abandoned, surrounded, and cut to pieces. The most skillful evolutions, the finest courage, are scarcely sufficient to extricate a body of foot encompassed on an open plain by superior numbers of horse..."

-Edward Gibbon, The Decline and Fall of the Roman Empire, Vol. III, Ch. XXVI.

Gibbon did could not make scientific extrapolations from examining the field etc., and certain sources he used have been declared advanced since his time (he wrote his masterpiece throughout the 1770s-1780s ), but he drew from the same sources our contemporaries do. Actually, he is constantly used for a source. Another is Theodor Mommsen, whose famed work I don't have with me.

Thanks, Spartan JKM :original:

[Gaius Baltar]

Great article, better than going to the bookstore.

[ThiudareiksGunthigg]

I've posted it to TWC before, but you can find my article on Adrianople here.

I'll post a reply to the original post later, but for now I'll simply say that (i) we see no differences in the way the Romans utilised or responded to cavalry before and after Adrianople and (ii) we don't see the rise of the medieval-style of heavy cavalry until 600-700 years after 378 AD.

It was a significant battle, no doubt about that. But it was not significant in relation to the tactics, troop types or equipment apart from in the minds of some old-fashioned and outdated writers of about 50+ years ago.

[Spartan JKM]

Thank you both. ThiudarG, your article of the events leading up to and the battle itself is superior writing. Thanks.

I think this connection with Adrianople to Medieval feudalism/chivalry seems, indeed, to be a myth.

It may have been primarily with Charles Martel and his Franks who began encouraging cavalry to the degree that created and influenced European chivalry. Threats came from all directions upon their kingdom, and infantry, though effective in a set-piece battle if properly utilized, was too slow to deal with potential enemies from different directions. Cavalry is simply far more mobile, whether for raiding and/or repulsing raids.

I agree this doctrine, not a swift process, probably did not start with Adrianople.

Tim, I'm taking a good look at Stephen William's and Gerard Friell's (Friell is the archaelogical expert of the duo) Theodosius: The Empire at Bay; they totally agree with what you are presenting about Thomas Burns' assessments regarding Adrianople.

Pg. 19,

"In terms of military history, it has been commonly assumed that the outcome of the battle demonstrated the superiority of cavalry over infantry; which lasted until the late middle ages and the development of more effective projectile weapons. In our view this is a flawed approach, arising from a misunderstanding of the battle itself and the nature of the opposing forces, which ignores the vast bulk of evidence to the contrary, both in the later fourth century and in other periods. Adrianople was an infantry battle, decided by a series of tactical moves and accidents, and was no different in its troop composition and maneuvering from most of the set-piece warfare of the day..."

The book is excellent, far form unscholalry; most likely, these are the people we truly learn from. But no one is infallible, and I just want to do more research until I am sold on it being an 'infantry battle'. BTW, I got M. Pavan's name from their bibliography, where they basically stated he believed Adrianople was significant for the cavalry issue. Pavan's work they cited from was published in 1979.

Bernard Montgomery states in his History of Warfare (1968), Pg. 127,

"...the legions were crushed together in the centre by the cavalry of the Goths from the left and their infantry in front...the battle of Adrianople is also of great significance as the first victory of heavy cavalry over infantry. The success of the Parthians and the reforms of Gallienus had hinted at the future, but it was the Gothic horsemen who first crossed the threshold of medieval warfare..."

From one Hoffman Nickerson (2003), Warfare in the Roman Empire and the Middle Ages, Pgs 61-62,

"The importance of the campaign of Adrianople is not its immediate consequences, which as we have seen were local, temporary and slight. The point is that after that campaign the Roman army was reorganized in a form big with consequences for the future...the tactical revolution in the relations between cavalry and infantry was now complete. Cavalry was now the chief arm. Vegetius seems to have written shortly after Adrianople, and in Roman tactics as he describes them it is the cavalry whom (in normal cases) are called upon to deliver the decisive attack...It is impossible to exaggerate the importance of the tactical revolution which made cavalry the chief arm..."

Nickerson does continue by stating that infantry, "normally, for the most part because of the fear of men in general felt by horses", is and must be the chief arm. He also says the the supremacy of cavalry was due in part to the "certain decline in the discipline of infantry", and begins to stress the mobility factor.

Thanks, Spartan JKM :original:

[Freddie]

I can't really see how Gothic horses can be compared to medieval-style heavy cavalry. The manner in which they lived (nomads) dictates that the horses they bred were more likely to smaller and faster rather then big and strong. Surely speed was of importance in those days and the necessity for breading stronger horses only become an issue when riders started to wear heavier armour and uses heavier weapons.

[Daneboy]

I've posted it to TWC before, but you can find my article on Adrianople here.

we don't see the rise of the medieval-style of heavy cavalry until 600-700 years after 378 AD.

By "medieval-style cavarlary" you exclude the kataphract? The Romans adopted them to fight the sassanids, as far as I know.

Even earlier, during the reign of Hadrian, the Romans fielded Kataphracts. Though, overall, I believe that they were to expensive to field in large numbers, specially in the age of Adrianople, where the economi were suffering.

[ThiudareiksGunthigg]

By "medieval-style cavarlary" you exclude the kataphract? The Romans adopted them to fight the sassanids, as far as I know.

Actually, no I don't exclude them at all. But these popularist, amateur books that claim Adrianople was some vast tactical turning point clearly do.

The thesis that Adrianople marked some sort of turning point in military history would need to demonstrate that the Roman Army changed significantly after 378 AD. But it didn't. Given the trouncings that the Sassanid cavalry had given the Romans repeatedly in the Third Century, the Roman Army had changed to make cavalry in general, and armoured cavalry in particular, far more prominent.

The idea that this happened after Adrianople is hopelessly incorrect. The army with which Valens faced Fritigern was already cavalry-heavy, thanks to the much earlier reforms of Gallenius, Diocletian and Constantine. The defeat at Adrianople didn't begin or even accelerate the increase in the tactical importance or status of cavalry in Valens' army. That had taken place long before, which is why Valens' cavalry outnumbered that of the Goths and why they were definitely better equipped and trained. Valens' army was cavalry heavy thanks to the reforms of the previous 75 years. Later armies didn't change the numbers, armour or significance of their cavalry arms.

But there is a difference between Sassanid heavy cavalry, their Roman equivalent and the central tactical role played by Eleventh and Twelfth Century knights. The Sassanid kataphracts, the later Roman clibinarii and even the later Frankish and Lombaric scholae were all an adjunct to other unit types. But by the late Eleventh Century these European heavy cavalry units had developed to the point where they were the tactical focus of a pitched battle. They weren't just skirmishing troops, flanking guards or mopping-up units (as at Adrianople) - they were the central hammer blow that won or lost a battle.

And that happened many centuries after Adrianople.

The idea that Adrianople changed the Roman attitude to cavalry is total and complete nonsense. Roman cavalry had become more central, more substantial and more esteemed long before 378 AD. We see no difference in the numbers, use, deployment, status or effectiveness of Roman cavalry after 378 AD. In fact, we see no change in their deployment or their tactical significance in the slighest. We see no changes at all to the already well established increase in the numbers and status of cavalry that pre-dates Adrianople by about a century.

But by the Eleventh Century we've seen a gradual change in the way heavy calvary was used. It wasn't an adjunct to infantry, as it had been in ancient warfare and as it had been at Adrianople. Now the combination of high saddles, heavy armour, strong horses, long, curved shields and couched lances made heavy knights the primary tactical blow.

THAT was the birth of the medieval knight. And it happened several centuries after Adrianople.

Adrianople was not a tactical turning point.

[Stalins Ghost]

The way I see it, the flaws in the Roman army at Adrianople were strategic as well as tactical. Firstly, Valens made the mistake of marching on the Gothic force in the sweltering Thracian heat, which, after a 12 mile march to the camp, would quite simply tired his men out- how were they supposed to repel cavalry when they were tired?

Next he completely failed to consider the inclusion of Gothic cavalry at the battle, consequently, and rashly he wasted his own elite cavalry lead by Cassio and Bacurius on skirmishing duties while his infantry lined up for battle. Bare in mind that these would have most likely been the Scutarii- the elite Roman cavalry. Already occupied with their attack on the Gothic formation, who they were struggling to defeat- it is thought that they were beginning to retreat from their rash attack: "their retreat was as cowardly as their advance was rash" (Ammianus), the force of Greuthungi and Alan cavalry made light work of them. All this happened before the Roman infantry had deployed. Whether facing cavalry or infantry, the undeployed formation is a weak formation- that much is tactical sense. Ammianus goes on in his passage to describe the Roman left as meeting with success- even reaching the Gothic encampment. However, without its crucial flank protection, it was vunerable, and was destroyed. The Lanciarii and Matiarii were the only units that held their ground, in what I think exemplifies the ability of the Late Roman forces, but it just wasnt enough. The combination of events and mistakes made by the rash Roman leadership crushed what should have been a stunning Roman victory,

Influence of Adrianople? On a tactical level- marginal. On a strategic and political level, it was immense. Germanic tribes now saw that the Romans could be forced onto the back foot. Not through sheer military success- even by 450 the Romans could deal some large scale successes one way or another, but it showed the Germanic peopels that by entering the Empire through semi"legal" terms via diplomacy and arm twisting, they could have an impact.

[scottishranger]

Does anyone else share the opinion that if Spartan JKM turned on his reputation then he would almost rival the might Garbsbardar. But anyways another fine, exceptionally fine essay by Spartan JKM.

Thanks, Spartan JKM

No thank you, you are saving me tons of money having to go to the bookstore. :wink:

I am afraid I don't have much else to add to the discussion that hasn't been already said, seeing as I focus a bit more on the Later Medieval Ages than Rome. Still a great read.

[ThiudareiksGunthigg]

It may have been primarily with Charles Martel and his Franks who began encouraging cavalry to the degree that created and influenced European chivalry. Threats came from all directions upon their kingdom, and infantry, though effective in a set-piece battle if properly utilized, was too slow to deal with potential enemies from different directions. Cavalry is simply far more mobile, whether for raiding and/or repulsing raids.

Actually, cavalry is not more rapid strategically than infantry. With the exception of steppe-style horse armies (with their strings of remounts and capacity for forced marches), cavalry actually travels at the same pace as infantry over long distances. If anything, infantry can be more fast-moving in an emergency than cavalry - you can inspire men to march through the night if it means they will (i) survive or (ii) win a victory, but there's no way to motivate or inspire dead tired horses, no matter how many great speeches you make to them.

Bernard Montgomery states in his History of Warfare (1968), Pg. 127,

"...the legions were crushed together in the centre by the cavalry of the Goths from the left and their infantry in front...the battle of Adrianople is also of great significance as the first victory of heavy cavalry over infantry. The success of the Parthians and the reforms of Gallienus had hinted at the future, but it was the Gothic horsemen who first crossed the threshold of medieval warfare..."

Whatever you may say about Monty as a general, he was not a professional academic historian. When he was writing, this was the commonly accepted wisdom amongst generalists at that time.

From one Hoffman Nickerson (2003), Warfare in the Roman Empire and the Middle Ages, Pgs 61-62,

"The importance of the campaign of Adrianople is not its immediate consequences, which as we have seen were local, temporary and slight. The point is that after that campaign the Roman army was reorganized in a form big with consequences for the future...the tactical revolution in the relations between cavalry and infantry was now complete. Cavalry was now the chief arm. Vegetius seems to have written shortly after Adrianople, and in Roman tactics as he describes them it is the cavalry whom (in normal cases) are called upon to deliver the decisive attack...It is impossible to exaggerate the importance of the tactical revolution which made cavalry the chief arm..."

I've never heard of this guy but the title and the fact that this is published by Dover Books seems to indicate that this is another generalist overview. He certainly gets several things badly wrong in this quote:

(i) The importance of the campaign was definitely not 'local, temporary and slight' considering the long-term impact of the eventual settlement of the Goths in Pannonia.

(ii) His assertions about how 'the Roman army was reorganized in a form big with consequences for the future' after Adrianople is totally wrong. The late Roman Army certainly did reorganise so that cavalry formed a much greater proportions of the troops, a variety of cavalry types (including heavy cavalry) were developed and cavalry units rose in status to make up most of the Army's elite units.

But this definitely did not happen after Adrianople - it happened long before 378 AD. The rise in the numbers, variety and status of the cavalry arm began under Gallienius (260-268 AD) and increased under Diocletian (284-305 AD) and Constantine I (306-337 AD). Valens' army was the result of these earlier reforms and changes.

On the other hand, we don't see any substantial changes to the Roman Army after 378 AD at all. The only major change was the introduction of more horse archer units as a result of contact with the Huns and Alans in the Fifth Century, but the changes that this Nickerson guy attributes to the aftermath of Adrianople actually took place about 100 years before the battle.

(iii) Vegetius may have been writing after Adrianople, but the status he gives to cavalry had nothing to do with that battle. If anything Vegetius tended to draw on much earlier military manuals and his depiction of the Roman Army is a weird combination of contemporary Fourth Century information and stuff that dates back to 400 years earlier.

As for 'It is impossible to exaggerate the importance of the tactical revolution which made cavalry the chief arm', I'd say it's impossible to exaggerate how much this Nickerson needed to do his homework properly.

[Spiff]

Moved to the Musaeum, another excellent article.

[Spartan JKM]

Thank you Spiff; I am quite flattered that the majority of threads I start wind up here in the Musaeum.

The way I see it, the flaws in the Roman army at Adrianople were strategic as well as tactical...

I agree. Unfortunately, Ammianus Marcellinus' narrative is not conclusive enough to deduce without some questions as to what really happened, such as not specifying which side, if just one, the Gothic cavalry attack come from. It is tenable that Valens committed a strategic blunder by fighting before Gratian arrived, his nephew whom he was apparently jealous of; it also seems bad intelligence led him to underestimate the quantity of the army of his Gothic foe. Attacking the Gothic field fortification while the Gothic powerful cavalry was uncommitted allowed the first blow by the Gothic cavalry to be very effective - the paramount value of surprise. The Roman infantry, at a point probably deployed in a crescent around the wagonlaager, was extremely vulnerable once they lost their protection from their horsemen on their flanks.

I think the Gothic cavaly played the pivotal role to set in motion the debacle that afflicted Valens that terrible day, but I agree with the revised overview that this was not a revolution in warfare - a sign of cavalry defeating infantry to spell the beginning of a cavalry cycle that saw cavaly as Europe's primary military arm. Afterall, to cite one example, the great battle of Tours, fought three and a half centuries later, was won by Charles Martel's 'wall of ice' against the slashing horsmen of the Moorish army; thus the Franks deployed in a compact phalanx (clearly guessing right as to what was coming), essentially, which the Moorish horsemen could not fully pierce.

Actually, cavalry is not more rapid strategically. With the exception of steppe-style horse armies (with their strings of remounts and capacity for forced marches), cavalry actually travels at the same pace as infantry over long distances. If anything, infantry can be more fast-moving in an emergency than cavalry - you can inspire men to march through the night if it means they will (i) survive or (ii) win a victory, but there's no way to movitate or inspire dead tired horses, no matter how many great speeches you make to them.

Well, that doesn't strain credibility from the context you are providing, but horses are simply faster than footmen, possessing far more endurance. Gallienus and Diocletian certainly provided conditions which enabled cavalry to act independently as a major element of the army, rather than as an adjunct whose full potential was never harnessed, which was evident in the earlier centuries. You are correct, though: strategically, unless the army was exclusive of infantry (ie, steppe armies) it would be foolish to advance one's cavalry faster than the infantry. And men certainly are going to be more conforming to horses amid battle.

Basically, and this is what I was mainly referring to, the cavalry were the eyes and ears of the army beginning in the 3rd century A.D. The army reforms, apparently beginning with Gallienus, allowed the expanding structure of the empire to be defended at its frontiers. When advancing into hostile territory cavalrymen could scout ahead, look for signs of the enemy, good routes, good camp locations, river fords, and sources of victuals. These vexillationes could screen the army from the eyes of enemy scouts, and could serve as flank and rear guards or foragers. On the defensive the cavalry units could locate and observe enemy raiding parties, perhaps impeding their progress. All in all, the reforms begun by Gallienus (if that claim can be sustained) saw the mobile cavalry constantly in the field, and ready to nip revolts in the bud before they could establish too firm a hold.

However, I don't think cavalry, despite these changes, ever assumed a greater importance than the infantry. For the most part, the Romans preferred infantry to cavalry because infantry could be trained to retain formation in combat, while cavalry tended to flee when faced with danger. Heavily armored (cataphractoi, clibanarii) units did increase, especially in the east, as ThiudarG stated, to face the heavily mounted Persians, whom they would prove to be capable of defeating, though it seemed difficult to gain much advantage from a tactical success in these lands. Empire borders have to exist somewhere, and the Danube, Rhine, and Euphrates were seemingly logical points for Rome (though rivers can also serve as transport channels for invaders).

The most significant change in the structure may have been of the division of troops of field armies (comitatenses) and those stationed to garrison the border regions (limitanei). This seperation was the work of Constantine the Great, but it was more a natural reactionary process than an innovation, due to the rising circumstances of external threats. The emphasis on cavalry and the usefulness of maintaining a standing mobile army established itself as the constant civil strife and the barbarian threats became rife on all fronts. The comitanses could protect the emperor from rising usurpers, and they were never tied down to a particualr region. Basically, they were the main strategic component of an empire-wide defence system.

The limitanei differed only in status to their field counterparts; they were regular units of trained soldiers. They patrolled and garrisoned the frontiers, and could cope with small-scale warfare. A major raid would entail the calling in of the closest comitanses.

One major change wrought from Adrianople was not a major re-restructuring of the Roman arms, but the conciliation of the Goths on the part of Theodosius I. As has been stated, the trained units lost at Adrianople could not be soon replaced, and the encroaching Goths could not be destroyed, so he enlisted many of them into the army, but still under the leadership of their own chieftains (foederati. Remember, these wandering peoples were not looking to destroy Roman power more than simply finding lands to settle in. This caused a 'barbarization' of the army (a controversial term), as more and more of its officers and men were drawn from the 'uncivilized' peoples, most notably the Germanic tribes. It has been largely discredited, however, that Rome fell because of the 'barbarization' of the army. It did polarize things, but Rome basically fell from within (the short and impossibly simple way to put it, if you will); the centralization which maintained a well-equipped and professional army desiccated over time (Adrian Goldsworthy, The Complete Roman Army, Pg. 214). We have to keep in mind, too, that the Byzantines pretty much upheld the Roman sophistication in engineering and logistics, but adopted efficient heavy cavalry and light infantry into an organized system... but that is a deep, evolving event in itself (the themata system, then more mercenaries etc. being utilized etc.)

This is certainly a fascinating, but deep, complicated issue.

Thanks, Spartan JKM :original: