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

Saturday, November 15, 2014

Battle of Satala - Persia Invades

Persian Cavalry  (radpour.com)

Satala - Battle of the Iberian War

The Iberian War was fought from 526 to 532 between the Eastern Roman Empire and Sassanid Persian Empire over the eastern Georgian kingdom of Iberia.

Tensions between the two powers were further heightened by the defection of the Iberian king Gourgen to the Romans. According to the historian Procopius, Kavadh I tried to force the Christian Iberians to become Zoroastrians, who in 524/525 under the leadership of Gourgen rose in revolt against Persia, following the example of the neighboring Christian kingdom of Lazica. Gourgen received pledges by Justin I that he would defend Iberia; the Romans indeed recruited Huns from the north of the Caucasus to assist the Iberians.

Violence escalated at various points where the power of the two empires met: in 525 a Roman fleet transported an Aksumite army to conquer Himyarite Yemen and in 525/526, Persia's Arab allies, the Lakhmids, raided Roman territories on the edge of the desert. By 526–527, overt fighting between the two empires had broken out in the Transcaucasus region and upper Mesopotamia, and the Persians continued to exert pressure on the Romans to obtain funds from them. Following the emperor Justin I's death in 527, Justinian I succeeded to the imperial throne. 

The early years of war favored the Persians: by 527 the Iberian revolt had been crushed, a Roman offensive against Nisibis and Thebetha in that year was unsuccessful, and forces attempting to fortify Thannuris and Melabasa were prevented from doing so by Persian attacks.

The Roman - Persian Frontier
Some 30,000 Persian troops invaded Byzantium's Armenian 
provinces resulting is the Battle of Satala.

In spring 530, the Persian attack in Mesopotamia met with defeat at the Battle of Dara. A Roman army of 25,000 men under the great Roman General Belisarius marched out to meet 40,000 Persians including large units of the Immortals. Belisarius crushed the Persians killing 8,000.

At the same time as Dara, the Persians had gained ground in the Caucasus, having subdued Iberia and invaded Lazica. The Persian shah, Kavadh I decided to take advantage of this and sent an army into Byzantium's Armenian provinces. For this task, he chose the general Mihr-Mihroe (Mermeroes).
Mihr-Mihroe began assembling his forces near the Byzantine border fortress of Theodosiopolis (Erzurum). According to Procopius, his army composed mostly of levies from Persian-ruled Armenia and Sunitae from the northern Caucasus, as well as 3,000 Sabirs
The Byzantine commanders were Sittas, who had just been promoted from magister militum per Armeniam to magister militum praesentalis, and his successor in the former post, Dorotheus

Sittas was the husband of Comito, the elder sister of the Empress Theodora, and possible father of the later empress Sophia.  He enters history in the reign of Emperor Justin I as a doryphoros ("bodyguard") in the guard of Justinian, then magister militum per Orientem.

As soon as news of the ongoing Persian preparations reached them, they sent two of their guards to spy on them. One was captured, but the other returned with information that allowed the Byzantines to launch a surprise attack the Persian camp. The Persian army scattered with some loss, and after looting their camp, the Byzantines returned to their base.

Satala Fortress East Gate
Yes, I said the same thing, "That's a gate? How do they know?"
I guess when the Persians destroy a fortress they really destroy a fortress.  

From Livius.org

The Fortress of Satala

(From Livius)  When the emperor Vespasian added Commagene to the Roman empire (72 CE), the upper Euphrates became a frontier zone; across the river were the Parthian Empire and the buffer state Armenia. The main road along the Roman border (limes) was from Trapezus on the shores of the Black Sea to Alexandria near IssusSeleucia, and Antioch near the Mediterranean in the south. The legionary bases along this highway were Satala (for the recently founded Sixteenth legion Flavia Firma), Melitene (XII Fulminata), Samosata (VI Ferrata), and Zeugma (IIII Scythica). 

Satala was chosen because it commanded not only the Euphrates, but also the road from central Anatolia to Armenia (essentially the modern E88/E80).

We would like to know more about the original settlement, but it will be hard to reconstruct it, because the site was destroyed in the mid-third century by troops of the Persian Sasanians (who had replaced the Parthians after 224), and was rebuilt several times. What can be seen today, belongs to the sixth-century reconstruction by the emperor Justinian (527-565). The circuit of the walls is too small to offer accommodation to a first- or second-century legion. The original military settlement is still lost.

Satala, Northern "Wall"

Satala, East "Wall"

Hardly anything is known about Satala's civil settlement, although it is certain that in the fourth century, there was a Christian community.
The site was fortified again in 529 by the Emperor Justinian. His historian Procopius writes:
The city of Satala had been in a precarious state in ancient times. For it is situated not far from the land of the enemy and it also lies in a low-lying plain and is dominated by many hills which tower around it, and for this reason it stood in need of circuit-walls which would defy attack. Nevertheless, even though its surroundings were of such a nature as this, its defenses were in a perilous condition, having been carelessly constructed with bad workmanship in the beginning, and with the long passage of time the masonry had everywhere collapsed. But the Emperor tore all this down and built there a new circuit wall, so high that it seemed to overtop the hills around it, and of a thickness sufficient to ensure the safety of its towering mass. And he set up admirable outworks on all sides and so struck terror into the hearts of the enemy.
This fortress survived for almost a century, but was eventually captured and destroyed in 607/608 by the Persian Sasanian king Khusrau II the Victorious (590-628).
Persian King Kavadh I

Battle of Satala

Once the Persian commander Mihr-Mihroe had finished assembling his army of 30,000 men in Armenia he invaded Byzantine territory. 

The Persians bypassed the Roman fortress of Theodosiopolis on the border and headed for Satala. While not much made of this action it is a major importance to this campaign.

In reading between the lines, the fortress was bypassed because it was far too powerful to successfully attack and capture in a reasonable amount of time.  Also, there is no way the Persians would allow a large Roman garrison to operate in their rear.  So a considerable number of Persian troops would have been left behind to hold the Romans inside the fortress.

It was a 126 mile march from Theodosiopolis to Satala through some very stark countryside.  Securing the Persian lines of communication would have involved detaching even more units from the main invading force.

After a long march the Persians set up their camp some distance from the city walls. How many Persians were there we do not know.  The historian Procopius (who was not there) says the Romans had 15,000 men and were outnumbered about two to one. That would mean the Persians did not detach any troops.  That would be wrong.

But the Romans did show up with a very powerful force that might have been in that range.  The Romans feared to take on the for more numerous Persians in open combat. Most the the Romans stayed with Dorotheus inside the city walls.  Sittas took 1,000 cavalry into the hills overlooking the fortress.
On the next day, the Persians advanced and began to surround the city, preparing for a siege. 

At this point, Sittas and his 1,000 man cavalry detachment charged down from the high ground of the hills into the rear of the Persians. The charge created a huge cloud of dust that made it hard to estimate the Roman numbers. The Persians thought they were facing the main Roman army. They quickly gathered their forces and turned to meet them.

With the Persians turned to face Sittas, the Romans in the city under Dorotheus led his own men to attack what was now the "new" rear of the Persian army. 

Despite their bad tactical position, facing attack from both front and rear, the Persian army resisted effectively, due to its greater numbers. At one point, however, a Byzantine commander, Florentius the Thracian, charged his unit into the Persian center and managed to capture Mihr-Mihroe's battle standard. Although he was killed soon after, the loss of the flag caused fear among the Persian ranks. Their army began to retreat to their camp, abandoning the battlefield.

The large Persian force may have retreated to their camp, but the Romans had no interest in following up their victory with additional attacks. That implies that the Persians were still well organized and to be feared. On the other hand, the Persians had been seriously mauled and no longer wanted to fight.

The next day, the Persians departed and returned to Persian Armenia, unmolested by the Byzantines, who were satisfied with their victory over a far larger force. 

This victory was a major success for Byzantium, and was followed by the defections of a number of Armenian chieftains to the Empire, as well as by the capture or surrender of a number of important fortresses, like Bolum and Pharangium. 

Negotiations between Persia and Byzantium also resumed after the battle, but they led nowhere, and in spring 531 war resumed, with the campaign that led to the Battle of Callinicum .

Possible Route of the Persians
It was a 126 mile march for the Persians from the Roman border
fortress of Theodosiopolis (modern Erzurum) to Satala (Sadak).
The Persians would have had to detach considerable numbers of
troops from the main army to secure their lines of communications
and surround the Roman garrison at 

Eastern Roman Army Re-enactors

By Procopius of Caesarea
(500 to 560AD)

Editor  -  Procopius is wonderful to read.  He gives us huge amounts of detail and first person accounts to events that were not seen by historians again for centuries. In the case of Satala, Procopius was not there.  I have reprinted his account because it does provide a certain amount of information.  But it is obvious by the lack of detail that Procopius is reporting to us events that were told to him by others.

And Cabades sent another army into the part of Armenia which is subject to the Romans. This army was composed of Persarmenians and Sunitae, whose land adjoins that of the Alani. There were also Huns with them, of the stock called Sabiri, to the number of three thousand, a most warlike race. And Mermeroes, a Persian, had been made general of the whole force. When this army was three days' march from Theodosiopolis, they established their camp and, remaining in the land of the Persarmenians, made their preparations for the invasion.

Now the general of Armenia was, as it happened, Dorotheus, a man of discretion and experienced in many wars. And Sittas held the office of general in Byzantium, and had authority over the whole army in Armenia. These two, then, upon learning that an army was being assembled in Persarmenia, straightway sent two body-guards with instructions to spy out the whole force of the enemy and report to them. And both of these men got into the barbarian camp, and after noting everything accurately, they departed. And they were travelling toward some place in that region, when they happened unexpectedly upon hostile Huns.

By them one of the two, Dagaris by name, was made captive and bound, while the other succeeded in escaping and reported everything to the generals. They then armed their whole force and made an unexpected assault upon the camp of their enemy; and the barbarians, panic-stricken by the unexpected attack, never thought of resistance, but fled as best each one could. Thereupon the Romans, after killing a large number and plundering the camp, immediately marched back.
Not long after this Mermeroes, having collected the whole army, invaded the Roman territory, and they came upon their enemy near the city of Satala. There they established themselves in camp and remained at rest in a place called Octava, which is fifty-six stades distant from the city. 
6th Century Byzantine Cavalry 

Sittas therefore led out a thousand men and concealed them behind one of the many hills which surround the plain in which the city of Satala lies. Dorotheus with the rest of the army he ordered to stay inside the fortifications, because they thought that they were by no means able to withstand the enemy on level ground, since their number was not fewer than thirty thousand, while their own forces scarcely amounted to half that number. 

On the following day the barbarians came up close to the fortifications and busily set about closing in the town. But suddenly, seeing the forces of Sittas who by now were coming down upon them from the high ground, and having no means of estimating their number, since owing to the summer season a great cloud of dust hung over them, they thought they were much more numerous than they were, and, hurriedly abandoning their plan of closing in the town, they hastened to mass their force into a small space. 

But the Romans anticipated the movement and, separating their own force into two detachments, they set upon them as they were retiring from the fortifications; and when this was seen by the whole Roman army, they took courage, and with a great rush they poured out from the fortifications and advanced against their opponents. They thus put the Persians between their own troops, and turned them to flight. 

However, since the barbarians were greatly superior to their enemy in numbers, as has been said, they still offered resistance, and the battle had become a fierce fight at close quarters. And both sides kept making advances upon their opponents and retiring quickly, for they were all cavalry. 

Thereupon Florentius, a Thracian, commanding a detachment of horse, charged into the enemy's centre, and seizing the general's standard, forced it to the ground, and started to ride back. And though he himself was overtaken and fell there, hacked to pieces, he proved to be the chief cause of the victory for the Romans. For when the barbarians no longer saw the standard, they were thrown into great confusion and terror, and retreating, got inside their camp, and remained quiet, having lost many men in the battle; and on the following day they all returned homeward with no one following them up, for it seemed to the Romans a great and very noteworthy thing that such a great multitude of barbarians in their own country had suffered those things which have just been narrated above, and that, after making an invasion into hostile territory, they should retire thus without accomplishing anything and defeated by a smaller force.

Remains of the walls and Eastern gate of Satala's late Roman fortress.
From Mavors.org

At that time the Romans also acquired certain Persian strongholds in Persarmenia, both the fortress of Bolum and the fortress called Pharangium, which is the place where the Persians mine gold, which they take to the king. It happened also that a short time before this they had reduced to subjection the Tzanic nation, who had been settled from of old in Roman territory as an autonomous people; and as to these things, the manner in which they were accomplished will be related here and now.
As one goes from the land of Armenia into Persarmenia the Taurus lies on the right, extending into Iberia and the peoples there, as has been said a little before this[19], while on the left the road which continues to descend for a great distance is overhung by exceedingly precipitous mountains, concealed forever by clouds and snow, from which the Phasis River issues and flows into the land of Colchis. 

In this place from the beginning lived barbarians, the Tzanic nation, subject to no one, called Sani in early times; they made plundering expeditions among the Romans who lived round about, maintaining a most difficult existence, and always living upon what they stole; for their land produced for them nothing good to eat. Wherefore also the Roman emperor sent them each year a fixed amount of gold, with the condition that they should never plunder the country thereabout. And the barbarians had sworn to observe this agreement with the oaths peculiar to their nation, and then, disregarding what they had sworn, they had been accustomed for a long time to make unexpected attacks and to injure not only the Armenians, but also the Romans who lived next to them as far as the sea; then, after completing their inroad in a short space of time, they would immediately betake themselves again to their homes. 

And whenever it _so_ happened that they chanced upon a Roman army, they were always defeated in the battle, but they proved to be absolutely beyond capture owing to the strength of their fastnesses. In this way Sittas had defeated them in battle before this war; and then by many manifestations of kindness in word and in deed he had been able to win them over completely. For they changed their manner of life to one of a more civilized sort, and enrolled themselves among the Roman troops, and from that time they have gone forth against the enemy with the rest of the Roman army. They also abandoned their own religion for a more righteous faith, and all of them became Christians. Such then was the history of the Tzani.
Gravestone of a Roman legionary
soldier and his wife from Satala.

Beyond the borders of this people there is a cañon whose walls are both high and exceedingly steep, extending as far as the Caucasus mountains. In it are populous towns, and grapes and other fruits grow plentifully. And this canon for about the space of a three days' journey is tributary to the Romans, but from there begins the territory of Persarmenia; and here is the gold-mine which, with the permission of Cabades, was worked by one of the natives, Symeon by name. 

When this Symeon saw that both nations were actively engaged in the war, he decided to deprive Cabades of the revenue. Therefore he gave over both himself and Pharangium to the Romans, but refused to deliver over to either one the gold of the mine. And as for the Romans, they did nothing, thinking it sufficient for them that the enemy had lost the income from there, and the Persians were not able against the will of the Romans to force the inhabitants of the place to terms, because they were baffled by the difficult country.
At about the same time Narses and Aratius who at the beginning of this war, as I have stated above,[20] had an encounter with Sittas and Belisarius in the land of the Persarmenians, came together with their mother as deserters to the Romans; and the emperor's steward, Narses, received them (for he too happened to be a Persarmenian by birth), and he presented them with a large sum of money. 

When this came to the knowledge of Isaac, their youngest brother, he secretly opened negotiations with the Romans, and delivered over to them the fortress of Bolum, which lies very near the limits of Theodosiopolis. For he directed that soldiers should be concealed somewhere in the vicinity, and he received them into the fort by night, opening stealthily one small gate for them. Thus he too came to Byzantium.

Little remains of Satala. Here is a small portion of their ancient Roman aqueduct.

The "Eternal Peace"

After six years of combat the Iberian war was basically a draw.  The Romans had won the Battles of Dara and Satala, but Persian losses were so high in the Battle of Callinicum it was effectively a Pyrrhic victory.  After Callinicum the Romans captured some forts in Armenia, and effectively repulsed a Persian offensive.

Justinian's envoy, Hermogenes, visited Kavadh immediately after the Battle of Callinicum to re-open negotiations but without success. Justinian therefore took steps to bolster the Roman position, trying, at the same time, to engage Kavadh diplomatically. Kavadh died shortly afterwards, and in spring 532 new negotiations began between the Roman envoys and the new Persian king, Khosrau I, who needed to devote his attention to secure his own position. 

The two sides finally came to an agreement, and the Eternal Peace, which lasted less than eight years, was signed in September 532. Both sides agreed to return all occupied territories and the Romans to make a one-off payment of 110 centenaria (11,000 pounds of gold). 

The Romans recovered the Lazic forts, Iberia remained in Persian hands, but the Iberians who had left their country were allowed to remain in Roman territory or to return to their native land.

The Iberian War had ended.  But in only eight years yet another new Roman-Persian war broke out in Lazica.

(Procopius - History of the Wars Book I XV)      (Sittas)      (Livius.org - Satala)

(Mavors.org)      (Iberian War)      (Satala-530)

The Eastern Roman and Persian Empires

Saturday, November 1, 2014

Infantry vs. Cavalry : The Byzantine Infantry Square


Editor  -  Below is a wonderful 1988 article by Professor Eric McGeer on Eastern Roman military tactics.  It is refreshing to read material by true experts on a subject.

Eastern Roman military history had suffered from a near total lack of proper histories written by those who witnessed the events.  We historians have to fill in the lack of detailed information with what we know from similar events. In this case I can say that the Byzantine infantry units have not been given proper credit by historians.

Byzantine infantry have lived in the shadow of the Roman Legions. But the Byzantine Army stood centuries longer than the legions of Rome. They must have been doing something right.

Professor McGeer details how Byzantine units formed and maintained complex infantry squares against attacking Arab cavalry. This required a great deal of training for the officers so they would be able to organize multiple units to act together while under enemy fire. But also the average soldier and his unit would must have had considerable training to firmly hold their assigned position during the madness of battle.

For over 800 years the Byzantines had to face down an seemingly endless stream of fanatical Islamist armies bent on conquest. Simply they held their own and often defeated and drove back their enemies.  Winning is not an accident. The soldiers and their officers were trained well.


The Byzantines encountered many different nations on the battlefield during their long history. The surveys of foreign peoples in the military manuals amply illustrate the Byzantines' readiness not only to analyze the tactics and characteristics of their enemies, but also even to learn from them when necessary. Their recognition of the need to study and to adapt themselves to the unfamiliar methods of warfare practiced by their enemies pays witness to the intellectual and practical character of the Byzantine approach to war.

The recorded observation of enemy skills and tactics was a feature which the Byzantines added to the long tradition of military science inherited from classical Antiquity. The study of war was energetically renewed in tenth-century Byzantium, as the number of important manuscripts and texts dating from this period clearly demonstrates.

This renewal of military science was largely in response to the increasing danger from the Arabs, whom the Byzantines had come to consider their most formidable enemies. It is always a difficult problem to determine what relation there was between traditional theory and contemporary practice in the Byzantine military texts — to what extent did the tenth-century strategists combine theory with practice to create formations and tactics which would be effective against the Arabs?

The analysis of the battle formation and tactics prescribed for infantry in the Praecepta militaria (ca. 965) sheds interesting light on this question. The choice of this subject will provide the opportunity to examine the underestimated role and importance of infantry in Byzantine armies of the period, as well as to see how the author of the Praecepta relied on earlier sources and his own observations to develop a formation and set of tactics for Byzantine infantry facing Arab cavalry.

Dealing with Arab Cavalry
Starting in September, 629 AD the Eastern Roman Empire came in contact with an enemy like none they had faced before: rapidly moving fanatic Islamist armies from the deserts of Arabia.  
Unlike the slower moving conventional armies of the Persian Empire, the Arab cavalry forces were extremely nimble and moved swiftly over the harsh conditions of the Roman desert frontiers. The Byzantines had to develop new tactics to both hold off the aggressive Islamist armies and also try to retake lost territories. 

Other tenth-century treatises show that the Byzantines also used a square for the same purposes, but the Sylloge is the first text in which a square is prescribed as the standard battle formation for Byzantine infantry. According to this text, one employed the infantry square to act as a mobile base for cavalry, either to follow in support of a successful cavalry attack on the enemy or to offer an immediate place of refuge in case the cavalry met with defeat. The author of the Praecepta followed the Sylloge closely as the blueprint for the basic deployment and tactics for infantry supporting cavalry in battle, occasionally even quoting his main source.

But at the same time, it must be said that he read the Sylloge critically and realistically, leaving aside all of its painstaking calculations of the manpower in the infantry force or of the measurements of the infantry square. In selecting only material which he knew conformed with the types of soldier and equipment at his disposal, the author of the Praecepta sought to give an up-to-date account of the Byzantine army". The real departure from the Sylloge, however, begins with his systematic description of the infantry force and the situations which it might have to confront on the battlefield.

The first and second chapters of the Praecepta treat the numbers, equipment, deployment and tactics for the infantry force attending the cavalry. The infantry were divided into twelve ταξιαρχίαι of one thousand men each.

A single ταξιαρχία included four types of infantryman in the following quantities :
  • four hundred όπλΐται (heavy infantrymen, armed with spear and sword, and protected by corslet, cap and shield),
  • three hundred τοξόται (archers, « called ψιλοίby the ancients »),
  • two hundred άκοντισταίand σφενδοβολισταί(light spearmen and slingers),
  • and one hundred μοναυλάτοι (heavy infantrymen who carried an exceptionally thick and solid spear, the μοναύλιον).

The author informs us that the heavy infantrymen were to be picked out from both Byzantines and Armenians (who formed a particularly ferocious contingent in the armies of Nikephoros Phokas), while the lighter άκοντισταίwere supplied by « Russians » (or by other foreigners).
10th Century Byzantine
Varangian Guard

The following description of the infantry square will be understood more easily with reference to the accompanying diagram. The author first instructs that the infantry be deployed in a « double-ribbed » (τετράγωνοςδιττή « called a τετράπλευροςby the ancients »), with three ταξιαρχίαι on each of the four sides. What exactly he means by « double-ribbed » becomes clear when he presents the battle order of the infantrymen in each ταξιαρχία. 

They stood one hundred men broad and seven men deep, that is, two lines of όπλϊται in front of three lines of τοξόται, backed in turn by two lines of όπλϊται, thus creating what the author calls an αμφίστομοςπαρά ταξις « a double-faced formation ». Such a deployment ensured that the rear lines of όπλϊται could protect their comrades' backs by turning around to face any enemy who had managed to break into the square. 

Each line to face any enemy who had managed to break into the square. Each line of one hundred men was commanded by a έκατόνταρχοςstanding in the middle, while two πεντηκόνταρχοι stood on the right and left wings of the line.

Intervals (χωρία) were allowed between the ταξιαρχίαι to permit twelve to fifteen cavalrymen at a time to ride through into or out of the square.  Twelve such intervals could be created in the square, but if the enemy infantry far outnumbered the Byzantine, then the corners of the square could be closed off and only eight intervals would remain, two in each of the four sides of the square. It was the duty of the άκοντισταί standing behind the ταξιαρχίαι to which they belonged, to watch over the intervals and rush forward to block them off whenever the enemy attacked.

The square offered the Byzantines important advantages as a battle formation in enemy lands. Facing four ways, it could not be outflanked or attacked from behind, always an urgent consideration when dealing with the Arabs. In providing immediate refuge for defeated cavalry, it prevented mass and prolonged flight which was usually the makings of real disaster for an army far from home. 

Furthermore, the author tells us that during battle the wounded and the exhausted could find shelter inside the square, while extra infantrymen could be assigned to bring water to the combatants to relieve their thirst, or stones and arrows to the slingers and archers so as to avoid these soldiers having to leave their places in search of more ammunition. Many of these factors suggest strongly that the author was well aware of the psychological advantages inherent in such a formation, not least the enhanced sense of collective security among men who know that their sides and backs are protected, that they can be saved if wounded and relieved if overcome with thirst or exhaustion. 

It must not be over looked, either, that a square facing four ways prevents easy flight by its very shape. For men about to face an all-out cavalry charge on their position, the lack of alternatives was probably the only reason why many of them decided to stay and fight when they would much rather have run away. Deployed as we have seen them, how were the Byzantine infantry to join battle with the enemy ?

Arab Cavalry in World War I.

Infantry versus infantry encounters are treated very briefly. If the enemy were not very sophisticated and simply attacked in a broad line, the άκοντισταίand the μοναυλάτοι on the two flanks of the square not directly engaged were to pour out round the enemy's flanks in a semi-circular movement and crush their line between them. If the enemy infantry were also deployed in a square (as Leo tells us the Arabs often were), then the άκοντισταίand the μοναυλάτοι inside the Byzantine square were sent to the aid of their comrades on whichever side of the square the enemy had attacked. 

These very sparse directions indicate that the author considered a purely infantry battle to be very unlikely, and, as a result, was far more occupied with infantry versus cavalry confrontations —both how to resist enemy cavalry with his infantry and how to destroy enemy infantry with his own cavalry, spearheaded by the mighty κατάφρακτοι.

It was when the enemy had defeated or scattered the Byzantine cavalry and intended to follow up on their success with an assault on the remaining force that the Byzantine infantry came into their own. 

The Arab cavalry posed two problems which the infantry square was designed to counter The first problem was that of their light skirmishers (to whom our author refers as Άραβϊται), who were mounted on very swift horses and used their great speed to ride round the square in hopes of luring the Byzantines into breaking ranks, whereupon they would suddenly wheel about to catch them off guard. 

But if these skirmishers were left at a distance or ignored, their effectiveness was much reduced, since they would never dare close with a strongly defended infantry formation, nor could they surprise the Byzantines with attacks from the flanks or rear. For their part, the Byzantines had no hopes of coming to grips with the elusive Arab raiders and thus could only remain in formation, undeceived by their enemies' feigned attacks and withdrawals.

The Infantry Square
Above is a Turkish attack on an Austrian infantry square in 1788.
The infantry square can be traced to ancient times. The formation was described by Plutarch and used by the Romans, and was developed from an earlier circular formation. In particular, a large infantry square was utilized by the Roman legions at the Battle of Carrhae against Persia, whose armies contained a large proportion of cavalry.
The formation was constituted as a hollow square. It was vital for squares to stand firm in the face of a cavalry charge, but they were not static formations. Astute commanders could, in suitable terrain, manoeuvre squares to mass archer fire on enemy formations.
Attacking cavalry would attempt to "break a square" by causing it to lose its cohesion, either by charging to induce poorly disciplined infantry to flee before making contact, or by causing casualties through close-range combat.
Combined attacks by infantry and cavalry would also have the same effect - the defending infantry unit would be placed in the difficult position of either forming square and being shot to pieces by archers of the attacking infantry, or being ridden down by the cavalry if it decided to remain in line while trading volleys with the attacking infantry.

A Byzantine Infantry Square

The second problem was posed by the more intimidating Arab regular cavalry, or even, as it appears, by their heavy cavalry, to whom the task fell to make a direct attack on the Byzantine infantry. As it became clear which side of the square the enemy planned to attack in strength, the Byzantines bolstered their lines accordingly. The two πεντηκόνταρχοι (one on the left wing of the line, the other on the right) in one of the two rear lines of όπλϊται led their fifty men forward through the intervals into the front lines of their ταξιαρχία, making them three deep in όπλϊται. 

At this point, the one hundred μοναυλάτοι in the ταξιαρχία also came forward through the intervals into the front lines, now four deep. This manoeuvre, taught to the soldiers in training, not only provided for the prompt  reinforcement of the front lines where necessary, but also served to deceive the enemy as to the real depth of the front lines which they were about to attack. As I interpret the Praecepta, it would appear that the όπλϊται and the μοναυλάτοι anchored the butt ends of their spears against the ground and aimed the points at an angle into the chests of the enemy warhorses, creating, in effect, a « chevaux de frise » four men deep. 

The exceptionally thick and solid μοναύλιον was designed to withstand the impact of an enemy armoured cavalry charge, for as the author says, « even if the three-deep spears of the όπλϊται are smashed by the enemy κατάφρακτοι, then the μοναυλάτοι, being firmly set, stand their ground bravely, receiving the charge of the κατάφρακτοι and turn them away »29. Once embroiled with the όπλϊται and μοναυλάτοι in front of them, the enemy cavalrymen were then set upon by the άκοντισταί who circled in from the flanks of the square not under attack. These light and thus more agile soldiers could take advantage of the restricted mobility of the enemy cavalrymen engaged at close quarters and pick them off one by one by striking them from behind 
or from their unprotected right sides. 

The Arab cavalry ran up against this thicket of spears after riding through a hail of arrows launched by the nine hundred archers stationed behind the spearmen on any one side of the square. If indeed the όπλΐται and the μοναυλάτοι were crouched over their fixed spears, the archers would have been able to shoot over their heads all the more easily, even to within very short range as the enemy drew near. 

Most unfortunately, our author does not give any details as to how archers stood, how they were commanded, or what their rate of shot was expected to be in battle. But their close cooperation with spearmen in repulsing enemy cavalry must have been judged indispensable if one takes into account the vast number of arrows the army was instructed to have on hand. Each archer carried one hundred arrows himself and received fifty more from the store of arrows carried by the pack-animals in the army's baggage train. 

This plentiful supply was doubtless intended to guarantee that the archers would not run out of arrows during battle, and we have already seen that extra men were detailed to keep up a steady supply to them during the fighting. It seems clear enough from this evidence that the Byzantine generals wanted a constant and efficient performance from their archers to take a heavy toll on the enemy cavalrymen well before they reached the infantry lines. 

A Cavalry Charge
It took nerves of steel and discipline to stand shoulder to shoulder with your fellow soldiers and face a charge of enemy cavalry looking to chop you into small chunks. Your safety rested on all the other troops in your unit doing their job and holding firm. 
This video of a cavalry charge from "Kingdom of Heaven" gives you a flavor of what that must have been like.

Such, then, was the system developed and presented by the author of the Praecepta in order to satisfy the defensive requirements of Byzantine expeditionary armies. He adapted a simple, symmetrical formation outlined in an earlier source to the types of infantrymen in his army, and gave each type of infantrymen one specific task to perform in this defensive system. The formation and tactics which he developed were intended to solve two main difficulties for Byzantine infantry facing Arab cavalry — the swift attacks and counterattacks of the light skirmishers and the concentrated attacks of the regular or heavy cavalry. 

But as long as the Byzantine infantry could force the Arabs to fight on their terms, by making them attack from directly in front against a concentrated and reinforced defence, then the chances of success were probably very good. 

Later Byzantine strategists were not averse to making what changes were necessary to maintain the shifting balance between infantry and cavalry on the battlefield. Thirty years or so after the Praecepta was written, a second version of this work was written and included in the Tactica of Nikephoros Ouranos. 

Here we find a slight, but telling, adjustment in the system by which the Byzantine infantrymen deepened their lines before receiving the enemy charge. Instead of advancing one of the two rear lines of όπλΐται through the intervals into the front ranks; as we saw in the original Praecepta, the second version of this work by Nikephoros Ouranos (the victor at Spercheios in 996) instructs every second file of men in the ταξιαρχία to step sideways into the file beside it, thus making a file of men seven deep into a file of men fourteen deep. It will be observed from the diagram of this manoeuvre that the width of the ταξιαρχία is reduced by one file only. 

This adjustment was probably intended to secure two further advantages over the earlier system — that the Byzantine infantry could make their formation even deeper than before and that they could do so in less time. It is therefore tempting to conclude from this adjustment that as heavy cavalry came into greater use (as did the Byzantine κατάφρακτοι in the armies of Nikephoros Phokas and John Tzimiskes) the infantry were constantly obliged to find the means to stop them, resorting to deeper and deeper formations and to specialised soldiers such as the μοναυλάτοι, and that these countermeasures were periodically revised to keep pace with fresh developments. 

The close attention to infantry tactics in the military manuals consulted here reminds us that the Byzantines by no means neglected this component of their army in the middle period. The use of infantry was essentially defensive — in battle, on the march and in protecting encampments or fortresses — but was nevertheless indispensable in support of cavalry. In a broader context, the development of infantry tactics from the Sylloge to the Praecepta to the Tactica of Nikephoros Ouranos strengthens the argument that in this period the Byzantines did attempt to combine theory with practice and to pass their conclusions on for further thought. It is no coincidence that by the end of the tenth century their position along the Arab frontiers was much stronger than it had been one hundred years before. 

Eric McGeer 
Université de Montréal (Département d'Histoire) 


Arab cavalry charge in the movie Lawrence of Arabia.