During the eleventh century, the Normans gained a strong reputation for their performance on the battlefield. While they were distinguished for their craftiness and cunning spirit, they were also known for their cruelty, bloodthirstiness, and destruction as they conquered throughout Europe. However, the point that I wish to raise in this article has to do with Norman battle tactics and their ‘invincibility’ on the battlefield: Did the Normans exhibit any innovation in the battlefields of Normandy, England, Italy, Sicily, or the Balkans?
To answer this question, I will compare two of the most famous battles of the Normans from two very different operational theatres – Hastings (1066) and Dyrrhachium (1081) – based on several points: What were the size and composition of the Norman armies and how were they recruited? How suited was each location for the cavalry charge of the Normans? What similarities do we see in the battle formations and main tactics of the Norman armies? What was the effect of the Norman heavy cavalry charge upon units of heavily-armed and disciplined infantry?
Advertisement
Norman expansion in eleventh-century Europe was a movement of enormous historical importance, which saw men and women from the duchy of Normandy settling in England, Apulia, Calabria, Sicily, and the Principality of Antioch. The Norman establishment in the South is particularly interesting because, by the middle of the eleventh century, they had become the undisputed masters of the entire southern Italian provinces of Apulia and Calabria, and by 1060 they were planning the invasion of Muslim Sicily, a conquest that would take them no less than three decades to complete. The capital of the Byzantine province of Longobardia (comprising Apulia and Calabria), the city of Bari, fell to Duke Robert Guiscard, a Norman military leader, in 1071 after a three-year siege.
Now, in the words of Anna Comnena, Guiscard “was always thinking out some more ambitious project.” He would challenge the Byzantine emperor Alexius Comnenus by invading the province of Illyria, in what is today Albania, and laying siege to the strategic port of Dyrrhachium. Whether he wanted to lay claim to the Imperial crown, we will never know, but the deeper reasons behind the Normanno-Byzantine clash in the shores of Illyria lay in the geopolitical expansion of the Norman aristocracy in the second half of the eleventh century, in the Byzantine involvement in rebellious activities against Robert Guiscard in Apulia, and in the quest to establish a principality for Guiscard’s eldest son Bohemond who – theoretically being a bastard – had no realistic chances of succeeding his father.
Advertisement
The armies
The question of numbers can be a tough one indeed, for contemporary chroniclers who reported the size of an army or opposing armies in a battlefield are rarely reliable in the figures they provide. Their estimates would have been affected by the inherent tendency of such narratives to exaggerate; by bias towards friends or foes; by the time at which a chronicler might be writing his account; and, unless he had taken part in an expedition, by his reliance on oral testimony from eyewitnesses, which always bears a great risk of miscalculation or inflation.
The sources for the Hastings campaign do offer several figures to modern historians, ranging from William of Poitiers’ 50,000 to the Carmen de Hastingae Proelio’s 150,000. Historians have reached some sort of consensus with a number between 7,000 and 8,000 effective soldiers being considered most likely, bearing in mind that William would certainly have left troops to guard the fortifications and the ships at Pevensey. This force was accompanied by some 2,000-2,500 horses, being transported from St-Valéry to Pevensey in a number of vessels that chroniclers put between 696 (Wace) to 3,000 (William of Jumièges), although it is unlikely that they exceeded 800. Therefore, with the cavalry numbering no more than 2,500, the rest of the force consisted of probably 4,000 foot of various sorts and around 1,500 archers and crossbowmen.
The main difference between the two opposing armies at Hastings was that, while the Normans had a proportion of mounted troops that constituted their elite unit, the English army was comprised almost entirely of combatants who either fought on foot, or travelled on horseback but chose to dismount and fight on foot. Theirs was a core force of around 800-1,000 housecarls (elite fighters, of whom some lived at the court or hall of the king or earls and received a wage), supplemented by some 6,500 men of the select fyrds of several counties of the south and central England (men performing military service in return for the land they held, based on a five-hide unit), and the great fyrds of Sussex and, perhaps, Kent (every able-bodied man called for service by the king in case of emergency). We have no clear idea of their numbers, but bearing in mind that Harold did not have time to replenish his losses after Stamford Bridge or to wait for the fyrds from the west and north, historians put the Saxon army’s numbers at slightly higher than that of the Normans, some 8,000.
The Normans were also superior to the Saxons in terms of another weapon, the bow, the use of which required a great degree of professional training and skill. Harold’s army included a very small number of archers and slingers, probably because the survivors lacked the horses to keep up with the rapid advance of the Saxon army south after Stamford Bridge; there is just a solitary figure shown in the Bayeux Tapestry. According to William of Poitiers, the hail of arrows launched against the Saxons in the first stage of the battle had some effect in ‘killing and maiming many,’ with the tightly-massed housecarls and thegns providing a good target for the Norman bowmen and arbalists. But the Saxon shield-wall and the elevated ground would have protected the ranks of the English from heavy casualties, and the counterattack by the latter with all kinds of missile weapons – if William of Poitiers is to be believed – swept back the Norman archers!
Advertisement
Crossing over to the operational theatre of the Adriatic Sea, trying to assess the size of the expeditionary force that sailed from Otranto in the spring of 1081 is a challenging task due to the contradictory estimates of the Byzantine and Norman sources. Anna Comnena attests that the Normans mustered an expeditionary force of some 30,000 men, with 150 ships of all types carrying them across the Adriatic with around 200 men and horses on board each ship. Thirty thousand men is surely an exaggerated figure, no doubt used by the princess to enhance her father’s victory over the Normans in the eyes of her audience. Geoffrey Malaterra’s figure of 1,300 knights, ‘as those present have testified’, probably represents the elite core of Guiscard’s army, if we bear in mind Malaterra’s comments that the largest portion of the army was an ‘imbecille vulgus.’ Other sources, like Orderic Vitalis, put the figure up to no more than 10,000 men; Peter the Deacon notes 15,000 men, while Romuald of Salerno talks about 700 horsemen. Whether the Normans had brought any archers with them is not certain, but even if they did, their presence was not felt during the battle that day.
Contrary to Hastings, the Normans at Dyrrhachium were heavily outnumbered both in cavalry and in overall numbers, as the Byzantines had not only mustered elite mercenaries such as the Varangians, the <i>vestiaritae<i>, the Frankish <i>Maniakatoi<i> and the turcopoles (probably some 2,000 men for each unit), but the Emperor had also brought with him units from the Balkans, both horse and foot, some 5,000 strong or more, along with the heretic Manicheans and allied troops from Armenia. The total number of the troops under Alexius Comnenus’ command would have been around 18,000-20,000 men, although the cavalry-infantry ratio is difficult to establish.
But how did William and Robert Guiscard manage to overcome the limitations of the eleventh-century military institutions in Normandy and Italy? For the former, the most adequate answer seems to be that William proposed to each of his men a minimum knight and ship ‘contribution’ to the English campaign, based on each one’s possessions, with the men being free to contribute more, depending on their desire for rewards if the campaign was successful. Comparing this with the Norman expedition in 1081, Robert Guiscard gathered (according to Malaterra) a ‘poorly-armed mob’, while Anna Comnena describes these foot soldiers (Greeks and Lombards) as ‘over-age and under-age, pitiable objects.’ Probably, these troops were ducal levies conscripted to serve overseas as a result of Guiscard proclaiming the <i>arrière-ban<i>, although no term of this kind appears in the primary sources. But since the lords of the realm had no feudal obligation to serve in Illyria, we presume that a very similar pattern to the 1066 negotiations must have been followed, although the chronicler material is silent about any formal or informal talks between the Duke and his magnates.
Advertisement
The battlefields
The topography of a battlefield and the exact position in which an army leader deploys his forces to face his enemy play a prominent role in the outcome of the battle. A careful choice or a clever exploitation of the terrain can diminish the numerical advantage of the enemy, hinder the mobility of the enemy’s cavalry units, and prevent any encircling manoeuvres. Hence we encounter the following questions: How suited were the battlefields of Dyrrhachium and Hastings for the battle-tactics of the Normans? And who held the strategic initiative in choosing the battlefield in each case?
The city of Dyrrhachium was very well defended, built on a long and narrow peninsula that ran parallel to the coast but with a marshy and swampy lagoon separating it from the mainland. There were also two fortified outposts situated on the opposite mainland area, both centred around two churches. Between the landing of the Norman army in June 1081 and the arrival of the Byzantines in mid-October, Robert Guiscard had pitched his camp just north of Dyrrhachium, between the city and the lagoon.
Alexius Comnenus’ initial plan was to perform a surprise night-attack on the Norman camp from two sides, sending the contingents of Serbs and Turks through the marshes towards Robert’s rear, while he pressed for a frontal attack-raid. However, during the night of 17-18 October, the Normans had already moved out of their camp and to the opposite side of the lagoon. Whether it was simply a coincidence, or if the Byzantine plan had somehow reached Guiscard’s ears, we will never know for sure. By the next morning, the Normans had placed themselves between the lagoon and Alexius’ army, having the lagoon to their rear and the sea on the right.
The terrain in that part of the Illyrian mainland served three purposes. Firstly, it was flat enough for the core unit of the Norman knights to apply their heavy cavalry attack. Secondly, with the sea on their right and the lagoon at their back, the Normans had their flanks covered from any encircling manoeuvres of the Byzantine or turcopole cavalry, especially if we consider the fact that Guiscard had placed his lightly-armed and inexperienced levies on the flanks, holding his heavy cavalry in the centre of the formation. Finally, the specific location where the Normans were deployed offered no escape route, especially if we consider that Guiscard had ordered his ships to be burned the previous day, thus echoing Vegetius’ principle that ‘trapped men draw extra courage from desperation, and when there is no hope, fear takes up arms’.
Advertisement
The terrain at Hastings also offered several advantages to an army that was on the defensive while fighting on foot. The position where Harold deployed his forces was at a place called Senlac, identified by Orderic Vitalis as a Norman-French adaptation of the Old English word Sandlacu, which means “sandy water”. This location consists of a hill called Caldbec, some 1100 yards long and 150 yards broad, with the greater part of the ridge being 250 feet high. The south slope facing the Normans was rather gentle, while the north on the other side was very steep, which prevented any chase by the Norman cavalry in case of a retreat. Also, the dense vegetation on the west edge of the hill would have offered significant protection to the Saxon army’s right flank from cavalry attacks. The hill sinks to a marshy bottom, to climb up again on the opposite hill of Telham at 441 feet, from where the Normans descended to deploy their forces. The topography of the battlefield offered William no alternative except a frontal uphill attack with his cavalry.
This location seems ideal for an army like the Saxon one, consisting almost entirely of foot-soldiers and whose tactics compelled it to stand on the defensive and in a tight formation. It was the latter, along with the elevated ground and the marshy bottom of the hill, that matched the Saxon lack of heavy horse – their army would have been easily overrun by the Norman cavalry if they had deployed on the plain. But was a defensive battle Harold’s original plan? Following the chronicler accounts of Florence of Worcester and the Anglo-Saxon Chronicle, it was Harold who had the intention of surprising William, but instead it was the latter that turned the tables on Harold and seized the initiative.
The tactics
Both Hastings and Dyrrhachium are considered two of the most characteristic examples of the use of the rectangular 3×3 battle formation in the eleventh century – the three linear formations of skirmishers, infantry, and cavalry, that also formed three divisions under three commanders (left flank, centre division, and right flank). If we turn to Dyrrhachium first, although we have little idea of the composition of each of the three battles of the Norman army, it is most likely that Guiscard would have kept the bulk of his elite cavalry units in the centre of his formation, and probably behind the heavy infantry for better protection against enemy missile weapons, while it is possible that he would have also had a corps of skirmishers with him that would have proved valuable during the four-month siege of the city of Dyrrhachium.
Fortunately, we are much better informed about the three lateral divisions of the Normans: Robert Guiscard commanded the main force at the centre of the formation; his son Bohemond, the second-in-command, took charge of the left wing; while the right wing, closer to the sea, was entrusted to Amicus II of Molfetta and Giovenazzo. Great similarities can be found to William’s deployment of his forces at Hastings in three lines: archers and, if we believe William of Poitiers and the Carmen, crossbowmen as well; infantry, with only a small number of them having mail coats; while the heavy cavalry was kept back to deliver the knock-out blow. Less clear is the composition of each of the corps, and we assume from William of Poitiers’ writing that the centre division, by far the largest of the three, was composed of Norman contingents with – presumably – William in command, while on the left flank were the units from Anjou, Poitou, and Maine, and the French and the Flemings were on the right.
The Norman division in echelons also served its purposes in the cases of Hastings and Dyrrhachium. First, the archers played the role of modern artillery in softening up the defences of the enemy’s tight infantry formation, bringing anger, frustration, fear and death to the enemy foot soldiers; we should keep in mind that a large number of both the Saxon and the Byzantine armies comprised of inexperienced militia. The infantry would then advance and engage the enemy in hand-to-hand fighting, while they also served as a protective shield for the knights, who would then advance to take advantage of any cracks in the enemy tight formation. Both William and Robert wanted to make full use of their elite unit of heavy cavalry for the final stage of the battle where they would deliver their shock blow to the – hopefully – frightened and disorganized enemy units.
We should note at this point that until the early tenth century, the classical formation of a mixed army, both in the European East and West, placed the infantry units at the centre of the formation, where the general also stood, while the cavalry took its place in the flanks to protect the most vulnerable points of an infantry and look for ways to envelop the enemy. By the middle of the tenth century, however, the cavalry in Byzantium was positioned inside a novel square infantry formation – a double-ribbed hollow square which was, practically, a camp converted into a battle formation, manoeuvring through the intervals on the sides to open the battle, and therefore reclaiming its position as the force de frappe in the hands of the Byzantine generals.
The kind of troops William was about to face in the field at Hastings consisted of heavily-armed housecarls and thegns packed together with the militia from the fyrd, in what seemed from a distance to be a single shield-wall between eight to ten shields deep, although it probably would have consisted of more sub-units, like the maniples of Vegetius’ Roman phalanx that were deployed in a chess-like formation. Since William placed the bulk of his elite cavalry in the centre of his formation and not in the flanks – as recommended by Vegetius and applied by Henry I at Tinchebray four decades later – it is clear that the Duke considered an attack on the well-protected Saxon wings as futile, preferring a coordinated infantry-cavalry charge at the centre where he could see Harold’s banner.
But the situation at Dyrrhachium seemed different for Robert Guiscard in many respects. Firstly, the Byzantine army had several units of skirmishers and cavalry, although the latter was of inferior quality, both in terms of experience and equipment, compared to the Norman chivalry, and certainly bore no resemblance to the crack units of the kataphrakts or the other tagmatic units of the tenth century that had dominated the battlefields of Syria, Mesopotamia and Bulgaria. Secondly, the Normans were outnumbered by many times; and thirdly, although the Norman Duke forced a pitched battle upon the Emperor, it was Alexius who had the chance to arrange his forces accordingly. The latter, having faced Norman forces in the field before, and probably having read the writings of Leo VI regarding the fighting tactics of the ‘Φράγκοι’ (the “Franks”), knew what to expect of his enemy. Hence, his response was to put his elite heavy infantry of the Varangian Guard in the centre-front line of the whole formation, projecting a few yards forward – obviously to take advantage of their thick infantry formations against the anticipated Norman cavalry charge – while units of skirmishers were ordered to move through their lines and release volleys of arrows before retiring.
The fighting
The Battle of Hastings was opened by the skirmishers, who moved forward to shoot at the enemy; but, falling short of what was expected of them, their attack produced poor results due to the fact that they had to shoot uphill – their arrows were easily repulsed by the housecarls, whose shield-wall must have served its purpose quite well! It is likely that Robert Guiscard also had lightly-armed peltasts and archers in his army, but according to our chronicler material their presence was not felt during the battle that day.
At Dyrrhachium, the battle opened with the Duke sending a cavalry detachment, probably of the elite knights of his division, to charge against the Varangian Guard in the centre of the Byzantine lines, and then to feign a disorderly retreat to entice their opponents into breaking their ranks. But because the Emperor had put archers and peltasts immediately behind them (like at Hastings, when the infantry of William’s army was met with a heavy shower of arrows, javelins, lances, and other ‘primitive casting weapons’), this cavalry attack produced poor results, with the Varangians staying put and their defensive formation unshaken.
The second stage, however, proved to be the most crucial for both battles. At Dyrrhachium, while the Norman cavalry detachment was engaged in moderate skirmishing with the Varangians and the supporting peltasts, the rest of the Norman army had managed to cover most of the distance between them and the Byzantines, and, at that important point of the battle, the infantry levies and light cavalry of the Norman right division (which was led by Amicus II) launched an attack against the Varangians’ exposed left flank. The latter resisted stoutly, receiving reinforcements by the Byzantine left and centre divisions, which resulted in the attackers being routed. And it was at that point that disaster struck for Alexius’ army.
With their right wing in a disorderly retreat, thus leaving the main division exposed to flanking enemy movements, this would have seemed like a perfect opportunity for Alexius Comnenus to strike a serious blow to Robert Guiscard’s army and perhaps even win the field. But, as had happened at Hastings – where, after the first infantry and cavalry charge, Harold’s men from his right-wing broke ranks to pursue their enemies downhill and were cut to pieces – the same fate befell the elite Varangians fifteen years later. Indeed, after the failure of the skirmishers to make any impression on the English, Duke William ordered his infantry and knights forward against the Saxon shield-wall, resulting in a terrible clash that saw the Norman left wing’s panicky retirement, affecting the entire Norman army.
What happened next proves one significant point: no tactical body of infantry can go after a retreating enemy when lacking the necessary training, discipline, and adequate support from cavalry units. Furthermore, the fact that any unit of heavy infantry is projected forward from the main army leaves it with its flanks exposed to enemy attack, thus significantly increasing the chance of encirclement by the enemy cavalry. If the Norman sources, and especially William of Poitiers, are to be believed, the English were greatly weakened by the fact that many of their comrades abandoned their positions to pursue the Normans downhill. Whether, however, this move came simply as a result of indiscipline, or if the Saxon officers decided that this was the right moment to counterattack, we will never know with certainty. Were Harold and his great earls not aware that infantry should not follow a retiring enemy who is on horse? The same question should be asked about the Varangian ‘hot-headed’ pursuit of the Norman right wing at Dyrrhachium and their eventual encirclement. Is that proof of indiscipline on the part of these elite mercenaries, as Anna Comnena categorically writes in her Alexiad? Or, rather, does it indicate that the Emperor issued orders for an all-out counterattack, but the rest of the units simply failed to keep up the pace with the Varangians, resulting in them being cut off from the main army?
After the crisis in the English position at Hastings, we reach the second of the well-known incidents of the battle that decided its outcome. The debate on whether there was indeed a feigned retreat of William’s cavalry in the second stage of the battle has been fought as hard as the two enemy armies fought at Hastings, and we need not engage with this issue more – I am ready to believe that there was, indeed, a feigned flight of the Norman chivalry, and the fact that the men under William would have fought in a conroi and would have been trained to attack the enemy in this manner for many years meant that they were perfectly capable of enacting this challenging manoeuvre. With units of the English being enticed to leave their defensive formation and march down the slope to chase the ‘retreating’ Normans, the latter got exactly what they wanted – a chance to ride them down as they turned their horses around.
The exposure of the main Byzantine army to the heavy cavalry attack of the Norman knights proved disastrous for Alexius Comnenus as well, as he had lost the protective shield that could have repulsed a Norman frontal attack: the unit of the Varangian heavy infantry. Robert Guiscard had not yet used his elite cavalry units, except for the feigned retreat during the opening stages of the battle, and he saw that this was the right time to throw them against his enemy. This all-out attack by the Norman cavalry proved effective in shattering Byzantine morale and discipline, with the entire front disintegrating rapidly.
Why did the Normans win?
With the contemporary chroniclers so often portraying the Normans’ victory as promised by God, I would like to return to the main point that was raised at the beginning of this article about Norman battle-tactics and their ‘invincibility’ in the battlefield. Writing in line with the argument first presented by David Bates nearly three decades ago, I agree that the Normans did not exhibit any innovation in the battlefields of Normandy, England, Italy, Sicily, or the Balkans in the eleventh century. By studying the Norman battle-tactics of the second half of the eleventh century in the aforementioned theatres of war, we can clearly see that in all cases the Normans relied on the charge of their heavy cavalry and the shock impact that this would have had on their enemies. This was the case not just in Dyrrhachium and Hastings, but in other battles like Civitate in 1053, where the bulk of Pope Leo IX’s army’s units were largely comprised of Lombard infantry levies, which gave way almost immediately after the first impact with the Norman knights.
The Normans used their cavalry charge in the operational theatre of Sicily to compensate for their numerical inferiority against the Muslim armies of the local Kalbite dynasty, and the reinforcements arriving from Tunisia, three times in seven years (1061, 1063, and 1068). But heavy cavalry attacks were common in Frankish warfare, and the Normans simply implemented what they had experienced in France for decades. And we also have evidence that the Byzantines were well aware of the charge of the Frankish chivalry, judging by the writings of Leo VI in the early tenth century. This is the case for the feigned retreat as well, as the examples of Hastings (1066), Dyrrhachium (1081), and Messina (1061) clearly demonstrate; the Normans must have known this tactic for centuries, as it appears to have its roots either in mock battles that were being reported throughout France and Flanders at least as early as the Carolingian period, or in the influence of the Armorican cavalry tactics dating from the fifth century.
Studying the battles of Hastings and Dyrrhachium, we come across several common traits. The Normans in southern Italy were coming from the same institutional background as their fellow-countrymen from the north, and thus we would expect them to introduce similar – but not identical – forms of lord-vassal relations, and the customs of tenure, military service, and inheritance established in Normandy and other parts of France. Therefore, in terms of military service, this meant a society dominated by an elite who fought on horseback and could demand the services of a vassal majority of troops that were ill-trained, ill-equipped, and fought on foot. This would explain why in all major battles fought by the Normans – except for Hastings – the Normans were greatly outnumbered by their enemies, not only in terms of heavy cavalry but in overall numbers as well. It was simply impossible to raise and maintain a large number of knights for a campaign overseas without any special arrangements being made, and even that did not guarantee a strong contingent of them, as we saw in the case of Dyrrhachium.
The Norman reliance on their heavy cavalry brings out another significant point in their overall strategy. They would choose a relatively broken, hilly, or marshy terrain, which was also dominated by a river or an uphill castle, to diminish the numerical advantage of their enemies and the mobility of their cavalry, while their smaller units could operate with more ease. That was the case against the much larger armies of the Muslims in Sicily, who put large numbers of infantry but also of heavy cavalry on the field (although their exact numbers are almost impossible to know). In the two major battles of Castrogiovanni (1061) and Cerami (1063), the Norman leaders deployed their forces on the banks of the rivers Dittaino and Cerami respectively, and near the castles from which the battles got their name, thus using the terrain to hamper the movement of the enemy cavalry.
Even though there was no well-established model to follow in the case of battle-arraying, armies of this period were divided into three or four units called battles or divisions (acies), being lined up one behind the other. The acies’ formation made the front shorter but increased the depth of the (cavalry) formation, thus giving greater impetus to break the resistance of the enemy’s centre formation. This technique was applied by Guiscard and his brother Roger against the Muslims of Sicily for two reasons: firstly because of their numerical inferiority (as in all the cases, they were outnumbered by many times by the Muslims), and also due to the troops they were facing in battle – no doubt many heavily-armed cavalry, but also large numbers of infantry militia with poor equipment and low morale.
But there are also several examples from this period where the divisions were put into the field directly facing the enemy, such as Civitate (1053), Tinchebray (1106), Hastings (1066), and Dyrrhachium (1081). It made more sense for an army to be arrayed into three parallel divisions, all facing the enemy, when an army consisted of more than one nation, if it had more than one general, or if they wished to increase the length of their formation. That can be seen at Civitate (1053) where the Normans divided their forces into three main divisions, the centre commanded by Humphrey of Hauteville, the right wing by Richard of Aversa, and the left by Robert Guiscard. However, the cases of Hastings and Dyrrhachium stand out by their combination of the 3×3 battle array, which was made possible by the fact that William and Robert Guiscard had both brought with them large numbers of foot soldiers. In no other battle of the second half of the 11th century would the Normans have enough numbers in both infantry and cavalry to deploy their forces in a 3×3 formation.
So, can we see the Norman victories at Hastings and Dyrrachium as a confirmation of the Norman superiority of cavalry over heavy infantry? From what I have pointed out in this article, the answer is clearly negative! But the basic point that we can draw out is that heavy cavalry units alone could make no impression upon well-equipped and disciplined foot soldiers who kept their formation unbroken. Such a unit, on the other hand, was clearly a defensive formation which, like the Varangian Guard or the Saxon thegns, had to be sufficiently deep and dense, and which needed the support of units of cavalry and archers because, although it could repel the cavalry charges, its speed and ability for manoeuvres made any counterattack almost unthinkable. Both Harold’s and Alexius’ men may have been brave, but they were not trained to form a phalanx and move out as a unit like the Ancient Greek hoplites, the Roman legionaries, or the Macedonian phalangitai. Thus, let us ask ourselves: What if the Saxons had not attempted to counterattack but rather stayed in close formation? Or if the ‘hot-headed’ Varangians had not attacked the retreating Norman right wing?
The death of Harold proved to be the catalyst that tipped the scale in favour of the Normans and led to the Saxon resistance crumbling rapidly thereafter. We must remember that this was a development of tremendous importance for the future of the English kingdom. At Dyrrhachium, even though Alexius was hotly pursued and surrounded by the Normans, he managed to escape and establish a rallying point at Thessaloniki. His death would have brought the state to the brink of a renewed civil war, just like the aftermath of the Battle of Manzikert had done ten years before, and the future of the Byzantine Empire would have been very different.
But before we completely dismiss the Norman reputation for distinctive martial prowess, we should ask whether it was simply by good fortune that a band of Norman bandits conquered England, half of the Italian peninsula, and Sicily in just half of a century, and seriously threatened the Byzantine Empire more than once? In the South, bands of Normans were employed by every rival camp (whether the Lombards, the Byzantines, or Rome) precisely because they were the best. And they were the best because they depended on strong leadership and unified command (with William, Robert Guiscard, Roger Hauteville, and Bohemond of Taranto being the undisputed leaders in their respective operational theatres), along with a combination of elite stipendiary knights, acting in co-ordination with recruited foot-soldiers, which were all supported, when necessary, by fleets.
Georgios Theotokis: Ph.D History (2010, University of Glasgow), specializes in the military history of the Eastern Mediterranean in Late Antiquity and the Middle Ages. He has published numerous articles and books on the history of conflict and warfare in Europe and the Mediterranean in the Medieval and Early Modern periods. His latest book is Twenty Battles That Shaped Medieval Europe. He has taught in Turkish and Greek Universities; he is currently a postdoctoral researcher at the Byzantine Studies Research Centre, Bosphorus University, Istanbul.
A version of this article was first published in Medieval Warfare magazine’s special issue on 1066: The Battle of Hastings. Click here to buy the issue.
By Georgios Theotokis
During the eleventh century, the Normans gained a strong reputation for their performance on the battlefield. While they were distinguished for their craftiness and cunning spirit, they were also known for their cruelty, bloodthirstiness, and destruction as they conquered throughout Europe. However, the point that I wish to raise in this article has to do with Norman battle tactics and their ‘invincibility’ on the battlefield: Did the Normans exhibit any innovation in the battlefields of Normandy, England, Italy, Sicily, or the Balkans?
To answer this question, I will compare two of the most famous battles of the Normans from two very different operational theatres – Hastings (1066) and Dyrrhachium (1081) – based on several points: What were the size and composition of the Norman armies and how were they recruited? How suited was each location for the cavalry charge of the Normans? What similarities do we see in the battle formations and main tactics of the Norman armies? What was the effect of the Norman heavy cavalry charge upon units of heavily-armed and disciplined infantry?
Norman expansion in eleventh-century Europe was a movement of enormous historical importance, which saw men and women from the duchy of Normandy settling in England, Apulia, Calabria, Sicily, and the Principality of Antioch. The Norman establishment in the South is particularly interesting because, by the middle of the eleventh century, they had become the undisputed masters of the entire southern Italian provinces of Apulia and Calabria, and by 1060 they were planning the invasion of Muslim Sicily, a conquest that would take them no less than three decades to complete. The capital of the Byzantine province of Longobardia (comprising Apulia and Calabria), the city of Bari, fell to Duke Robert Guiscard, a Norman military leader, in 1071 after a three-year siege.
Now, in the words of Anna Comnena, Guiscard “was always thinking out some more ambitious project.” He would challenge the Byzantine emperor Alexius Comnenus by invading the province of Illyria, in what is today Albania, and laying siege to the strategic port of Dyrrhachium. Whether he wanted to lay claim to the Imperial crown, we will never know, but the deeper reasons behind the Normanno-Byzantine clash in the shores of Illyria lay in the geopolitical expansion of the Norman aristocracy in the second half of the eleventh century, in the Byzantine involvement in rebellious activities against Robert Guiscard in Apulia, and in the quest to establish a principality for Guiscard’s eldest son Bohemond who – theoretically being a bastard – had no realistic chances of succeeding his father.
The armies
The question of numbers can be a tough one indeed, for contemporary chroniclers who reported the size of an army or opposing armies in a battlefield are rarely reliable in the figures they provide. Their estimates would have been affected by the inherent tendency of such narratives to exaggerate; by bias towards friends or foes; by the time at which a chronicler might be writing his account; and, unless he had taken part in an expedition, by his reliance on oral testimony from eyewitnesses, which always bears a great risk of miscalculation or inflation.
The sources for the Hastings campaign do offer several figures to modern historians, ranging from William of Poitiers’ 50,000 to the Carmen de Hastingae Proelio’s 150,000. Historians have reached some sort of consensus with a number between 7,000 and 8,000 effective soldiers being considered most likely, bearing in mind that William would certainly have left troops to guard the fortifications and the ships at Pevensey. This force was accompanied by some 2,000-2,500 horses, being transported from St-Valéry to Pevensey in a number of vessels that chroniclers put between 696 (Wace) to 3,000 (William of Jumièges), although it is unlikely that they exceeded 800. Therefore, with the cavalry numbering no more than 2,500, the rest of the force consisted of probably 4,000 foot of various sorts and around 1,500 archers and crossbowmen.
The main difference between the two opposing armies at Hastings was that, while the Normans had a proportion of mounted troops that constituted their elite unit, the English army was comprised almost entirely of combatants who either fought on foot, or travelled on horseback but chose to dismount and fight on foot. Theirs was a core force of around 800-1,000 housecarls (elite fighters, of whom some lived at the court or hall of the king or earls and received a wage), supplemented by some 6,500 men of the select fyrds of several counties of the south and central England (men performing military service in return for the land they held, based on a five-hide unit), and the great fyrds of Sussex and, perhaps, Kent (every able-bodied man called for service by the king in case of emergency). We have no clear idea of their numbers, but bearing in mind that Harold did not have time to replenish his losses after Stamford Bridge or to wait for the fyrds from the west and north, historians put the Saxon army’s numbers at slightly higher than that of the Normans, some 8,000.
The Normans were also superior to the Saxons in terms of another weapon, the bow, the use of which required a great degree of professional training and skill. Harold’s army included a very small number of archers and slingers, probably because the survivors lacked the horses to keep up with the rapid advance of the Saxon army south after Stamford Bridge; there is just a solitary figure shown in the Bayeux Tapestry. According to William of Poitiers, the hail of arrows launched against the Saxons in the first stage of the battle had some effect in ‘killing and maiming many,’ with the tightly-massed housecarls and thegns providing a good target for the Norman bowmen and arbalists. But the Saxon shield-wall and the elevated ground would have protected the ranks of the English from heavy casualties, and the counterattack by the latter with all kinds of missile weapons – if William of Poitiers is to be believed – swept back the Norman archers!
Crossing over to the operational theatre of the Adriatic Sea, trying to assess the size of the expeditionary force that sailed from Otranto in the spring of 1081 is a challenging task due to the contradictory estimates of the Byzantine and Norman sources. Anna Comnena attests that the Normans mustered an expeditionary force of some 30,000 men, with 150 ships of all types carrying them across the Adriatic with around 200 men and horses on board each ship. Thirty thousand men is surely an exaggerated figure, no doubt used by the princess to enhance her father’s victory over the Normans in the eyes of her audience. Geoffrey Malaterra’s figure of 1,300 knights, ‘as those present have testified’, probably represents the elite core of Guiscard’s army, if we bear in mind Malaterra’s comments that the largest portion of the army was an ‘imbecille vulgus.’ Other sources, like Orderic Vitalis, put the figure up to no more than 10,000 men; Peter the Deacon notes 15,000 men, while Romuald of Salerno talks about 700 horsemen. Whether the Normans had brought any archers with them is not certain, but even if they did, their presence was not felt during the battle that day.
Contrary to Hastings, the Normans at Dyrrhachium were heavily outnumbered both in cavalry and in overall numbers, as the Byzantines had not only mustered elite mercenaries such as the Varangians, the <i>vestiaritae<i>, the Frankish <i>Maniakatoi<i> and the turcopoles (probably some 2,000 men for each unit), but the Emperor had also brought with him units from the Balkans, both horse and foot, some 5,000 strong or more, along with the heretic Manicheans and allied troops from Armenia. The total number of the troops under Alexius Comnenus’ command would have been around 18,000-20,000 men, although the cavalry-infantry ratio is difficult to establish.
But how did William and Robert Guiscard manage to overcome the limitations of the eleventh-century military institutions in Normandy and Italy? For the former, the most adequate answer seems to be that William proposed to each of his men a minimum knight and ship ‘contribution’ to the English campaign, based on each one’s possessions, with the men being free to contribute more, depending on their desire for rewards if the campaign was successful. Comparing this with the Norman expedition in 1081, Robert Guiscard gathered (according to Malaterra) a ‘poorly-armed mob’, while Anna Comnena describes these foot soldiers (Greeks and Lombards) as ‘over-age and under-age, pitiable objects.’ Probably, these troops were ducal levies conscripted to serve overseas as a result of Guiscard proclaiming the <i>arrière-ban<i>, although no term of this kind appears in the primary sources. But since the lords of the realm had no feudal obligation to serve in Illyria, we presume that a very similar pattern to the 1066 negotiations must have been followed, although the chronicler material is silent about any formal or informal talks between the Duke and his magnates.
The battlefields
The topography of a battlefield and the exact position in which an army leader deploys his forces to face his enemy play a prominent role in the outcome of the battle. A careful choice or a clever exploitation of the terrain can diminish the numerical advantage of the enemy, hinder the mobility of the enemy’s cavalry units, and prevent any encircling manoeuvres. Hence we encounter the following questions: How suited were the battlefields of Dyrrhachium and Hastings for the battle-tactics of the Normans? And who held the strategic initiative in choosing the battlefield in each case?
The city of Dyrrhachium was very well defended, built on a long and narrow peninsula that ran parallel to the coast but with a marshy and swampy lagoon separating it from the mainland. There were also two fortified outposts situated on the opposite mainland area, both centred around two churches. Between the landing of the Norman army in June 1081 and the arrival of the Byzantines in mid-October, Robert Guiscard had pitched his camp just north of Dyrrhachium, between the city and the lagoon.
Alexius Comnenus’ initial plan was to perform a surprise night-attack on the Norman camp from two sides, sending the contingents of Serbs and Turks through the marshes towards Robert’s rear, while he pressed for a frontal attack-raid. However, during the night of 17-18 October, the Normans had already moved out of their camp and to the opposite side of the lagoon. Whether it was simply a coincidence, or if the Byzantine plan had somehow reached Guiscard’s ears, we will never know for sure. By the next morning, the Normans had placed themselves between the lagoon and Alexius’ army, having the lagoon to their rear and the sea on the right.
The terrain in that part of the Illyrian mainland served three purposes. Firstly, it was flat enough for the core unit of the Norman knights to apply their heavy cavalry attack. Secondly, with the sea on their right and the lagoon at their back, the Normans had their flanks covered from any encircling manoeuvres of the Byzantine or turcopole cavalry, especially if we consider the fact that Guiscard had placed his lightly-armed and inexperienced levies on the flanks, holding his heavy cavalry in the centre of the formation. Finally, the specific location where the Normans were deployed offered no escape route, especially if we consider that Guiscard had ordered his ships to be burned the previous day, thus echoing Vegetius’ principle that ‘trapped men draw extra courage from desperation, and when there is no hope, fear takes up arms’.
The terrain at Hastings also offered several advantages to an army that was on the defensive while fighting on foot. The position where Harold deployed his forces was at a place called Senlac, identified by Orderic Vitalis as a Norman-French adaptation of the Old English word Sandlacu, which means “sandy water”. This location consists of a hill called Caldbec, some 1100 yards long and 150 yards broad, with the greater part of the ridge being 250 feet high. The south slope facing the Normans was rather gentle, while the north on the other side was very steep, which prevented any chase by the Norman cavalry in case of a retreat. Also, the dense vegetation on the west edge of the hill would have offered significant protection to the Saxon army’s right flank from cavalry attacks. The hill sinks to a marshy bottom, to climb up again on the opposite hill of Telham at 441 feet, from where the Normans descended to deploy their forces. The topography of the battlefield offered William no alternative except a frontal uphill attack with his cavalry.
This location seems ideal for an army like the Saxon one, consisting almost entirely of foot-soldiers and whose tactics compelled it to stand on the defensive and in a tight formation. It was the latter, along with the elevated ground and the marshy bottom of the hill, that matched the Saxon lack of heavy horse – their army would have been easily overrun by the Norman cavalry if they had deployed on the plain. But was a defensive battle Harold’s original plan? Following the chronicler accounts of Florence of Worcester and the Anglo-Saxon Chronicle, it was Harold who had the intention of surprising William, but instead it was the latter that turned the tables on Harold and seized the initiative.
The tactics
Both Hastings and Dyrrhachium are considered two of the most characteristic examples of the use of the rectangular 3×3 battle formation in the eleventh century – the three linear formations of skirmishers, infantry, and cavalry, that also formed three divisions under three commanders (left flank, centre division, and right flank). If we turn to Dyrrhachium first, although we have little idea of the composition of each of the three battles of the Norman army, it is most likely that Guiscard would have kept the bulk of his elite cavalry units in the centre of his formation, and probably behind the heavy infantry for better protection against enemy missile weapons, while it is possible that he would have also had a corps of skirmishers with him that would have proved valuable during the four-month siege of the city of Dyrrhachium.
Fortunately, we are much better informed about the three lateral divisions of the Normans: Robert Guiscard commanded the main force at the centre of the formation; his son Bohemond, the second-in-command, took charge of the left wing; while the right wing, closer to the sea, was entrusted to Amicus II of Molfetta and Giovenazzo. Great similarities can be found to William’s deployment of his forces at Hastings in three lines: archers and, if we believe William of Poitiers and the Carmen, crossbowmen as well; infantry, with only a small number of them having mail coats; while the heavy cavalry was kept back to deliver the knock-out blow. Less clear is the composition of each of the corps, and we assume from William of Poitiers’ writing that the centre division, by far the largest of the three, was composed of Norman contingents with – presumably – William in command, while on the left flank were the units from Anjou, Poitou, and Maine, and the French and the Flemings were on the right.
The Norman division in echelons also served its purposes in the cases of Hastings and Dyrrhachium. First, the archers played the role of modern artillery in softening up the defences of the enemy’s tight infantry formation, bringing anger, frustration, fear and death to the enemy foot soldiers; we should keep in mind that a large number of both the Saxon and the Byzantine armies comprised of inexperienced militia. The infantry would then advance and engage the enemy in hand-to-hand fighting, while they also served as a protective shield for the knights, who would then advance to take advantage of any cracks in the enemy tight formation. Both William and Robert wanted to make full use of their elite unit of heavy cavalry for the final stage of the battle where they would deliver their shock blow to the – hopefully – frightened and disorganized enemy units.
We should note at this point that until the early tenth century, the classical formation of a mixed army, both in the European East and West, placed the infantry units at the centre of the formation, where the general also stood, while the cavalry took its place in the flanks to protect the most vulnerable points of an infantry and look for ways to envelop the enemy. By the middle of the tenth century, however, the cavalry in Byzantium was positioned inside a novel square infantry formation – a double-ribbed hollow square which was, practically, a camp converted into a battle formation, manoeuvring through the intervals on the sides to open the battle, and therefore reclaiming its position as the force de frappe in the hands of the Byzantine generals.
The kind of troops William was about to face in the field at Hastings consisted of heavily-armed housecarls and thegns packed together with the militia from the fyrd, in what seemed from a distance to be a single shield-wall between eight to ten shields deep, although it probably would have consisted of more sub-units, like the maniples of Vegetius’ Roman phalanx that were deployed in a chess-like formation. Since William placed the bulk of his elite cavalry in the centre of his formation and not in the flanks – as recommended by Vegetius and applied by Henry I at Tinchebray four decades later – it is clear that the Duke considered an attack on the well-protected Saxon wings as futile, preferring a coordinated infantry-cavalry charge at the centre where he could see Harold’s banner.
But the situation at Dyrrhachium seemed different for Robert Guiscard in many respects. Firstly, the Byzantine army had several units of skirmishers and cavalry, although the latter was of inferior quality, both in terms of experience and equipment, compared to the Norman chivalry, and certainly bore no resemblance to the crack units of the kataphrakts or the other tagmatic units of the tenth century that had dominated the battlefields of Syria, Mesopotamia and Bulgaria. Secondly, the Normans were outnumbered by many times; and thirdly, although the Norman Duke forced a pitched battle upon the Emperor, it was Alexius who had the chance to arrange his forces accordingly. The latter, having faced Norman forces in the field before, and probably having read the writings of Leo VI regarding the fighting tactics of the ‘Φράγκοι’ (the “Franks”), knew what to expect of his enemy. Hence, his response was to put his elite heavy infantry of the Varangian Guard in the centre-front line of the whole formation, projecting a few yards forward – obviously to take advantage of their thick infantry formations against the anticipated Norman cavalry charge – while units of skirmishers were ordered to move through their lines and release volleys of arrows before retiring.
The fighting
The Battle of Hastings was opened by the skirmishers, who moved forward to shoot at the enemy; but, falling short of what was expected of them, their attack produced poor results due to the fact that they had to shoot uphill – their arrows were easily repulsed by the housecarls, whose shield-wall must have served its purpose quite well! It is likely that Robert Guiscard also had lightly-armed peltasts and archers in his army, but according to our chronicler material their presence was not felt during the battle that day.
At Dyrrhachium, the battle opened with the Duke sending a cavalry detachment, probably of the elite knights of his division, to charge against the Varangian Guard in the centre of the Byzantine lines, and then to feign a disorderly retreat to entice their opponents into breaking their ranks. But because the Emperor had put archers and peltasts immediately behind them (like at Hastings, when the infantry of William’s army was met with a heavy shower of arrows, javelins, lances, and other ‘primitive casting weapons’), this cavalry attack produced poor results, with the Varangians staying put and their defensive formation unshaken.
The second stage, however, proved to be the most crucial for both battles. At Dyrrhachium, while the Norman cavalry detachment was engaged in moderate skirmishing with the Varangians and the supporting peltasts, the rest of the Norman army had managed to cover most of the distance between them and the Byzantines, and, at that important point of the battle, the infantry levies and light cavalry of the Norman right division (which was led by Amicus II) launched an attack against the Varangians’ exposed left flank. The latter resisted stoutly, receiving reinforcements by the Byzantine left and centre divisions, which resulted in the attackers being routed. And it was at that point that disaster struck for Alexius’ army.
With their right wing in a disorderly retreat, thus leaving the main division exposed to flanking enemy movements, this would have seemed like a perfect opportunity for Alexius Comnenus to strike a serious blow to Robert Guiscard’s army and perhaps even win the field. But, as had happened at Hastings – where, after the first infantry and cavalry charge, Harold’s men from his right-wing broke ranks to pursue their enemies downhill and were cut to pieces – the same fate befell the elite Varangians fifteen years later. Indeed, after the failure of the skirmishers to make any impression on the English, Duke William ordered his infantry and knights forward against the Saxon shield-wall, resulting in a terrible clash that saw the Norman left wing’s panicky retirement, affecting the entire Norman army.
What happened next proves one significant point: no tactical body of infantry can go after a retreating enemy when lacking the necessary training, discipline, and adequate support from cavalry units. Furthermore, the fact that any unit of heavy infantry is projected forward from the main army leaves it with its flanks exposed to enemy attack, thus significantly increasing the chance of encirclement by the enemy cavalry. If the Norman sources, and especially William of Poitiers, are to be believed, the English were greatly weakened by the fact that many of their comrades abandoned their positions to pursue the Normans downhill. Whether, however, this move came simply as a result of indiscipline, or if the Saxon officers decided that this was the right moment to counterattack, we will never know with certainty. Were Harold and his great earls not aware that infantry should not follow a retiring enemy who is on horse? The same question should be asked about the Varangian ‘hot-headed’ pursuit of the Norman right wing at Dyrrhachium and their eventual encirclement. Is that proof of indiscipline on the part of these elite mercenaries, as Anna Comnena categorically writes in her Alexiad? Or, rather, does it indicate that the Emperor issued orders for an all-out counterattack, but the rest of the units simply failed to keep up the pace with the Varangians, resulting in them being cut off from the main army?
After the crisis in the English position at Hastings, we reach the second of the well-known incidents of the battle that decided its outcome. The debate on whether there was indeed a feigned retreat of William’s cavalry in the second stage of the battle has been fought as hard as the two enemy armies fought at Hastings, and we need not engage with this issue more – I am ready to believe that there was, indeed, a feigned flight of the Norman chivalry, and the fact that the men under William would have fought in a conroi and would have been trained to attack the enemy in this manner for many years meant that they were perfectly capable of enacting this challenging manoeuvre. With units of the English being enticed to leave their defensive formation and march down the slope to chase the ‘retreating’ Normans, the latter got exactly what they wanted – a chance to ride them down as they turned their horses around.
The exposure of the main Byzantine army to the heavy cavalry attack of the Norman knights proved disastrous for Alexius Comnenus as well, as he had lost the protective shield that could have repulsed a Norman frontal attack: the unit of the Varangian heavy infantry. Robert Guiscard had not yet used his elite cavalry units, except for the feigned retreat during the opening stages of the battle, and he saw that this was the right time to throw them against his enemy. This all-out attack by the Norman cavalry proved effective in shattering Byzantine morale and discipline, with the entire front disintegrating rapidly.
Why did the Normans win?
With the contemporary chroniclers so often portraying the Normans’ victory as promised by God, I would like to return to the main point that was raised at the beginning of this article about Norman battle-tactics and their ‘invincibility’ in the battlefield. Writing in line with the argument first presented by David Bates nearly three decades ago, I agree that the Normans did not exhibit any innovation in the battlefields of Normandy, England, Italy, Sicily, or the Balkans in the eleventh century. By studying the Norman battle-tactics of the second half of the eleventh century in the aforementioned theatres of war, we can clearly see that in all cases the Normans relied on the charge of their heavy cavalry and the shock impact that this would have had on their enemies. This was the case not just in Dyrrhachium and Hastings, but in other battles like Civitate in 1053, where the bulk of Pope Leo IX’s army’s units were largely comprised of Lombard infantry levies, which gave way almost immediately after the first impact with the Norman knights.
The Normans used their cavalry charge in the operational theatre of Sicily to compensate for their numerical inferiority against the Muslim armies of the local Kalbite dynasty, and the reinforcements arriving from Tunisia, three times in seven years (1061, 1063, and 1068). But heavy cavalry attacks were common in Frankish warfare, and the Normans simply implemented what they had experienced in France for decades. And we also have evidence that the Byzantines were well aware of the charge of the Frankish chivalry, judging by the writings of Leo VI in the early tenth century. This is the case for the feigned retreat as well, as the examples of Hastings (1066), Dyrrhachium (1081), and Messina (1061) clearly demonstrate; the Normans must have known this tactic for centuries, as it appears to have its roots either in mock battles that were being reported throughout France and Flanders at least as early as the Carolingian period, or in the influence of the Armorican cavalry tactics dating from the fifth century.
Studying the battles of Hastings and Dyrrhachium, we come across several common traits. The Normans in southern Italy were coming from the same institutional background as their fellow-countrymen from the north, and thus we would expect them to introduce similar – but not identical – forms of lord-vassal relations, and the customs of tenure, military service, and inheritance established in Normandy and other parts of France. Therefore, in terms of military service, this meant a society dominated by an elite who fought on horseback and could demand the services of a vassal majority of troops that were ill-trained, ill-equipped, and fought on foot. This would explain why in all major battles fought by the Normans – except for Hastings – the Normans were greatly outnumbered by their enemies, not only in terms of heavy cavalry but in overall numbers as well. It was simply impossible to raise and maintain a large number of knights for a campaign overseas without any special arrangements being made, and even that did not guarantee a strong contingent of them, as we saw in the case of Dyrrhachium.
The Norman reliance on their heavy cavalry brings out another significant point in their overall strategy. They would choose a relatively broken, hilly, or marshy terrain, which was also dominated by a river or an uphill castle, to diminish the numerical advantage of their enemies and the mobility of their cavalry, while their smaller units could operate with more ease. That was the case against the much larger armies of the Muslims in Sicily, who put large numbers of infantry but also of heavy cavalry on the field (although their exact numbers are almost impossible to know). In the two major battles of Castrogiovanni (1061) and Cerami (1063), the Norman leaders deployed their forces on the banks of the rivers Dittaino and Cerami respectively, and near the castles from which the battles got their name, thus using the terrain to hamper the movement of the enemy cavalry.
Even though there was no well-established model to follow in the case of battle-arraying, armies of this period were divided into three or four units called battles or divisions (acies), being lined up one behind the other. The acies’ formation made the front shorter but increased the depth of the (cavalry) formation, thus giving greater impetus to break the resistance of the enemy’s centre formation. This technique was applied by Guiscard and his brother Roger against the Muslims of Sicily for two reasons: firstly because of their numerical inferiority (as in all the cases, they were outnumbered by many times by the Muslims), and also due to the troops they were facing in battle – no doubt many heavily-armed cavalry, but also large numbers of infantry militia with poor equipment and low morale.
But there are also several examples from this period where the divisions were put into the field directly facing the enemy, such as Civitate (1053), Tinchebray (1106), Hastings (1066), and Dyrrhachium (1081). It made more sense for an army to be arrayed into three parallel divisions, all facing the enemy, when an army consisted of more than one nation, if it had more than one general, or if they wished to increase the length of their formation. That can be seen at Civitate (1053) where the Normans divided their forces into three main divisions, the centre commanded by Humphrey of Hauteville, the right wing by Richard of Aversa, and the left by Robert Guiscard. However, the cases of Hastings and Dyrrhachium stand out by their combination of the 3×3 battle array, which was made possible by the fact that William and Robert Guiscard had both brought with them large numbers of foot soldiers. In no other battle of the second half of the 11th century would the Normans have enough numbers in both infantry and cavalry to deploy their forces in a 3×3 formation.
So, can we see the Norman victories at Hastings and Dyrrachium as a confirmation of the Norman superiority of cavalry over heavy infantry? From what I have pointed out in this article, the answer is clearly negative! But the basic point that we can draw out is that heavy cavalry units alone could make no impression upon well-equipped and disciplined foot soldiers who kept their formation unbroken. Such a unit, on the other hand, was clearly a defensive formation which, like the Varangian Guard or the Saxon thegns, had to be sufficiently deep and dense, and which needed the support of units of cavalry and archers because, although it could repel the cavalry charges, its speed and ability for manoeuvres made any counterattack almost unthinkable. Both Harold’s and Alexius’ men may have been brave, but they were not trained to form a phalanx and move out as a unit like the Ancient Greek hoplites, the Roman legionaries, or the Macedonian phalangitai. Thus, let us ask ourselves: What if the Saxons had not attempted to counterattack but rather stayed in close formation? Or if the ‘hot-headed’ Varangians had not attacked the retreating Norman right wing?
But before we completely dismiss the Norman reputation for distinctive martial prowess, we should ask whether it was simply by good fortune that a band of Norman bandits conquered England, half of the Italian peninsula, and Sicily in just half of a century, and seriously threatened the Byzantine Empire more than once? In the South, bands of Normans were employed by every rival camp (whether the Lombards, the Byzantines, or Rome) precisely because they were the best. And they were the best because they depended on strong leadership and unified command (with William, Robert Guiscard, Roger Hauteville, and Bohemond of Taranto being the undisputed leaders in their respective operational theatres), along with a combination of elite stipendiary knights, acting in co-ordination with recruited foot-soldiers, which were all supported, when necessary, by fleets.
Georgios Theotokis: Ph.D History (2010, University of Glasgow), specializes in the military history of the Eastern Mediterranean in Late Antiquity and the Middle Ages. He has published numerous articles and books on the history of conflict and warfare in Europe and the Mediterranean in the Medieval and Early Modern periods. His latest book is Twenty Battles That Shaped Medieval Europe. He has taught in Turkish and Greek Universities; he is currently a postdoctoral researcher at the Byzantine Studies Research Centre, Bosphorus University, Istanbul.
Click here to read more from Georgios Theotokis
Related Posts
Subscribe to Medievalverse