Much has been written about the advantages which one side had over the other in the battle of Hastings; in fact, in terms of the arms used there was very little difference. The state of war and its technology spread beyond any single state, county, duchy or kingdom. The English and the Normans had more in common than they had differences in 1066. On the Bayeux Tapestry, one can not easily tell English from Norman by either the arms they carry or the armour they wear. Indeed, to show a difference the artist often resorted to the use of distinguishing hairstyles: short for the Normans, long for the English. Both sides wore similar helmets and armour, both used similar swords and spears.
Historians have been clear for a long time now that the English army was well armed and well organised. Any lingering ideas that readers might retain of native troops armed only with clubs or similarly crude weapons, called together haphazardly and acting as a rabble rather than a trained army, should be dismissed at once. It seems probable that every army called into being in England by the kings consisted largely of trained men, if not quite professionals in our modern sense.
The main composition of the fyrd or army consisted of the household warriors and landed retainers of the king and of the great men who owed him allegiance. They might be seen as ‘royal war bands’ rather than national armies in a modern sense. The royal force, or indeed a local force acting for the king, could call upon shire levies too. Local shire levies seem to have been prepared to act in an emergency, for example, against invasion. Men from Somerset and Devon turned out against Harold Godwinson when he returned to England against the Confessor in 1052. But even shire levies consisted largely of the middling to higher social ranks, armed men mostly with some experience of war. This is not to say that no men of lower rank participated. Ceorls, not the lowest of the low, did join the fyrd, as we see at Maldon, and the poem called the Carmen speaks of peasants at Hastings.1 But it is not likely that large numbers of untrained troops were used. The whole process of assembling an army was geared towards ensuring the reverse: bringing together selected men, chosen because they would be useful in war.
Housecarls were military retainers, probably introduced by the Danish Cnut, and they were to be found in the household of an earl or ealdorman as well as that of the king. Such men appear to have been paid wages, and may in a sense be seen as mercenaries or stipendiaries, as probably were the Danes employed in the royal fleet in 1015.2 But housecarls were sometimes granted lands and a place to live. They were not so much mercenaries or some sort of standing army as they were household men.
Military households are a common feature of the medieval world, both before and after Cnut. The meaning of ‘housecarls’ after all is household servants, and this is essentially what they were. There were 15 acres of Wallingford where housecarls dwelled, presumably employed as a garrison there, and Domesday Book records various other examples.3 We shall look at the English system in due course, and find that it provided well-trained men quite capable of fighting any force of the day.
Increasingly too it is becoming apparent that although there were some social distinctions between English and Norman society, and although armies were raised by slightly different methods, that the differences were not as great as once thought. This is not the place for a lengthy discussion on feudalism and what we mean by it. It was not a medieval word for a start. For our purposes, what can be said is that although the actual process of raising forces was not quite the same, the underlying rationale was not so very different. Both powers, for example, raised some troops on the basis of the land held by individual warriors, and the service owed in respect of that land.
There were some differences in the composition of the armies. Two stand out and will need to be considered in detail. Firstly, at Hastings, the Normans had archers in some numbers whereas the English appear to have had few. Secondly, the Normans made considerable use of cavalry whereas the English, although they had horses and were experienced riders, seem not normally to have fought as cavalry. In the end these two factors might be said to have made the difference between two well-matched forces which stood against each other in the field for practically the whole length of a long autumn day. It is important to examine the reasons for these points of contrast.
ARMS AND ARMOUR
Let us begin by examining what in general the two sides had in common, the normal arms and armour of a fighting man in the middle of the eleventh century. Archaeology has provided very few useful objects from this period, virtually none from England and Normandy. Until this is remedied, we rely heavily on the images in the Bayeux Tapestry.
There has been some discussion over the date and provenance of this remarkable work. Suffice it for the present to say that there is general consensus, about which we see no need to quarrel, with a date very shortly after the battle, in the late eleventh century probably between 1077 and 1083, and also with its being made in England and possibly at Canterbury. The artist is anonymous, and he was not present at some of the events portrayed, but by and large the more research goes into the Tapestry, the more respect historians have for the accuracy and care of his work.
There are some details over which one is uncertain of the artist’s intent, and there may be some errors since he was probably not a man with military experience, so we must beware of accepting everything without question, especially when the Tapestry seems at odds with surviving contemporary chronicles; but for the most part we can accept the Tapestry as the best evidence for the arms and armour of the time.
Let us examine what the typical soldier on the Tapestry is wearing, and what arms he is carrying. Firstly, let us consider those troops portrayed as fighting on foot. Occasionally on the Tapestry one gets hints of clothing worn under the armour. This obviously existed, but from its nature is impossible to describe with any certainty for this period. Almost certainly both head and upper body would have been covered with enough clothing to give some padding effect to metal armour.
The two main pieces of armour were a helmet and a byrnie, or hauberk. The eleventh-century helmet was made of either leather or metal, and was conical in shape: Normans, English and Scandinavians all wore much the same kind. Leather helmets were not much more than caps but offered protection against the weather as well as against weapons. Better armed soldiers, and especially wealthier ones, would have had a metal headpiece. This could be made of one piece of metal, as is proved by a surviving example in the Museum of Armour at Vienna. Most of the helmets on the Tapestry appear to be of the type made of separate plates of metal, riveted together on to a framework of metal strips. In one helmet from the early Saxon period, the plates of the helmet were made from horn.4 The most important feature of the helmet of our period was the nasal. This was normally made from a metal strip in the centre at the front, projecting down beyond the level of the helmet so that it covered the nose, thereby offering some protection for the eyes and face.
Some Tapestry helmets appear to have projected at the back in order to cover the neck. One must assume this to be an occasional rather than a regular feature. A helmet found in York from the Viking period has a curtain of mail fixed to cover the neck.5 The head could also be covered, under the helmet, with chain mail. This might be in the form of a hood attached to the main mail coat, or as a separate piece. The word ‘hauberk’ itself derives from Frankish ‘halsberg’, meaning neck protection, so the hood may have been an original feature.
The hauberk became a symbol of status, only a man of some rank would own and use one. When his horse was killed at the battle of Dreux in 1014, Hugh of Maine buried his hauberk, put on a shepherd’s cloak and carried shepherd’s gear on his shoulder as disguise in order to escape.6 Indeed, the unit of land which provided military service was known as the ‘fief de hauberk’. The byrnie, or hauberk, or mail coat, was shaped rather like a tunic or tee-shirt, and had to be donned by slipping it over head and body.
Both infantry and cavalry wore a similar mail coat, as seen on the Tapestry. William of Poitiers confirms that the second line of the Norman army, consisting of better armed infantry, wore hauberks.7 Modern efforts to reproduce this form of armour are illuminating in terms of giving some idea of the difficulties and advantages to be gained from the garb. It feels quite heavy to lift, and it was thought a feat of strength that the Conqueror on one occasion returned to camp smiling, having carried his own and the rather large William fitz Osbern’s hauberk on his shoulders for some distance. And yet, once put on, the mail coat balances on the shoulders and is less restrictive to movement than one might suppose.8
The hauberk is constructed from circles of metal looped through each other. There are various possible variations on this method of production, some coats given double or even triple layers of protection; some being soldered, some riveted. It is also possible to assist the comfort and usefulness of mail by varying the size of the metal rings: smaller rings at the edges make it fit more snugly to the body. Mail provided some protection from certain blows, more from slashing than thrusting efforts; but it could always be pierced, for example, by a direct shot from an arrow, bolt, spear or lance, and it did not cover the whole body, so that there were always vulnerable spots – the face, the hands, the lower part of the legs.
There are some points to examine with regard to the armour of cavalrymen. Their hauberks were much the same as those of the infantry, but two possible differences need to be considered: the appearance of a rectangular piece on the chest, which in the Tapestry and elsewhere seems to be applied to cavalry rather than infantry armour; and the question of how the tunic shape would need to be adapted in order to make riding possible and comfortable.
There is some interest in a feature of some mail coats in the Tapestry, and elsewhere, which have a rectangular shape over the breast. They do not appear on all examples on the Tapestry, and feature most prominently in a section before the battle. They are not worn always by the great men. It could be that the artist only included when he felt like it, what was actually a normal feature.
One explanation of this rectangle, which seems to be edged with leather, is that it was an extra plate attached in order to give added protection to the chest. Illustrations other than the Tapestry look as if this could be the case, where the piece is coloured differently and looks like a single piece of plate armour. However, on the Tapestry there appears to be mail within the rectangle, though it could still be a separately made piece.
The best explanation seems to be that this was the way the hauberk was made easier to put on, having an enlarged opening which could be closed up once it was on, like buttoning up a shirt at the neck once in place. At any rate, a story that, on arrival in England, the Conqueror put on his hauberk the wrong way round does suggest that the tunic was not uniformly constructed front and back. The suggestion is made that the rectangle, clearly visible in the illustration above, was a flap of mail, part of which needed to be secured by tying in place. The only question then is why it appears to be rectangular, an L-shaped flap might seem more likely. One suggestion does not necessarily exclude the other: it is possible that an extra piece of armour was tied on over the neck opening.
Another suggestion is that it might have been a contraption of leather straps to allow the shield to be tied in place. One Tapestry illustration shows a Norman grasping part of a rectangular arrangement of straps, which looks exactly like the rectangle over the chest, but is clearly independent of the mail.9
The other question is whether hauberks were made differently for mounted men. Ian Peirce believes that they may have been trousered, and certainly that is how some appear on the Tapestry. This would have been extremely uncomfortable for mounted men and very inconvenient. It is more likely that they were always, for both infantry and cavalry, tunic-shaped. The tapering of rings towards the edges would have made the metal cling round the limbs and give a trousered appearance.
The shield portrayed on the Bayeux Tapestry is mostly the long, tapering, kite-shaped type. The same type of shield seems to have belonged to both infantry and cavalry. It is shown held over the arm, and slung round the neck when not in use. It was not to have a long history after this period. For infantry the tendency was towards a slightly shorter and more manoeuvrable shield with a straight-edged top. For cavalry the kite shield must have been unwieldy and awkward, and it is not surprising that it was replaced by various types of smaller shield, some similar to infantry shields, some round. In fact, some of the soldiers on the Tapestry, notably English infantry, one of whom seems to be Gyrth Godwinson, are using a smaller shield, which is circular and more convex in shape.
The straps shown on the Tapestry seem of simple design, about a quarter of the way down the shield, from side to side, so that when held in action the top of the shield could comfortably protect the face. Some of the shields are shown with bosses placed at the centre of the broader part, and some have a few rivets – four, six, nine, even eleven – probably to hold together the planks of wood making up the shield, perhaps to help fasten on a leather covering, and perhaps also to hold the straps. The boss, like the rivets, was of iron and some shields had a metal rim, some an iron bar inside for the hand-grip. Various woods could provide the basic material for a shield, but alder and willow were the most popular in England.10
The Tapestry shields are decorated but, unless the artist was ignorant of such designs, were not heraldic in nature. It seems significant that the body of Harold Godwinson could not be recognised after the battle; had he been wearing distinguishing heraldic arms this would not have been the case.11 The Tapestry artist is probably to be trusted.
Individual and family arms, passed by descent and with modifications, were just about to begin their appearance at this time, probably first in France. The earliest manuscript with apparent heraldic shields on is the Stephen Harding Bible, a Cistercian work of the early twelfth century, and even that is an imaginary depiction of biblical scenes. Some Tapestry designs no doubt did show who were their possessors, as did some of the standards, but they seem to be purely individual affairs.
Some of the standards displayed on the Tapestry have a broader significance, for example, the dragon (the wyvern of Wessex) held for Harold Godwinson. We know that Viking leaders commonly used a raven standard, which had religious significance as the mark of Odin. These had an ancient and sometimes religious as well as national importance. They represent the whole force behind the banner in some sense.
The arms of the English infantryman were the common weapons of such soldiers over centuries: sword, spear and less commonly, axe. The sword was already the weapon par excellence of the noble warrior. Its manufacture had developed over the centuries, and would continue to improve in design. But by the eleventh century, swords were well advanced and even in some areas manufactured in bulk.
The early medieval sword at its best was perhaps the Viking sword, a weapon developed for cutting rather than thrusting, with the emphasis on the sharp edges. At least in legend, it was said that such a sword was so sharp that on one occasion a man was sliced: ‘so cleanly in two as he sat in his armour, that the cut only became apparent when, as he rose to shake himself, he fell dead in two halves’. Most of the main weapons of this period depended upon the skill of the smith who had first worked the iron. Forging methods had improved particularly from about AD 900, and the blade could now be made longer and lighter.
Frankish swords were also improving, and the makers – either individuals or ‘factories’ – seem to be Frankish. Sword makers’ names were sometimes engraved on the blade: the name ‘Ulfberht’ appeared in the tenth century, ‘Ingelri’ from about 1050 for the type X sword with a broad, flat blade with a rather rounded point, and type XI, also eleventh century, with the name ‘Gicelin’, which was rather longer and narrower in style.
There were several processes for producing the blade, which was normally double edged. The handle or hilt resulted from a careful assembly of parts: a guard to protect against blows sliding up the blade, with a pommel which served the dual purpose of fixing the handle on to the blade, and balancing the heaviness of the blade so that the hand could be comfortably at or near the pivotal point in terms of weight.
The spear had two main components: a wooden shaft and a metal head. It could be used for thrusting, but was certainly at times employed for throwing, as can be seen on the Tapestry.It was generally the weapon of the ordinary soldier, but nobles could also have one. At Maldon, Dunnere, a ceorl, ‘shook his throwing spear’; but the commander, Byrhtnoth, himself ‘brandished his slender spear’ when he spoke to the men. It is interesting, and suggestive of the importance of the weapon in general, that there are eight Old English words for a spear.12 The shaft was commonly made of ash, indeed ‘ash’ was one of the eight words meaning spear, but various woods might be used.
The cavalry lance on the Tapestry looks very much like the infantry spear, and perhaps at this time there was no great difference. One might expect the lance to be longer, but there is no sign yet of the elaborate hand guards and heavier design which would mark the later cavalry lance. It is true that spears themselves might be long. The Aberlemno stone depicts such a spear.13 In 1016 at Sherston we hear of ‘spear and lance’, which suggests a longer infantry weapon.14 And Wace did say that when the horseman dismounted to fight on foot he broke his lance in half, obviously suggesting that the cavalry weapon was longer, but then Wace was writing in the twelfth century. The shaft of the cavalry lance, like that of the spear, appears to be quite straight, with no special holding point.
The lance was not yet used for a concerted charge, that is, held couched underarm by all the men charging together with lances levelled so that its force came upon impact. On the Tapestry one does see the lance sometimes couched, but more often it is held overhead, and can be both thrust and thrown. It has been argued that it might have been used in this manner in response to the problem of riding against an enemy on higher ground. But it seems more likely that as yet cavalry did not operate as one large unit in a charge together, and that the way of using the lance was a matter of individual preference. There was more improvisation and less organisation about charges than would be the case a century later.
A number of the English soldiers are shown using a battle axe. This seems to be already a somewhat antique weapon. It had been a favourite of Scandinavian armies, and it does still appear from time to time in later conflicts. Its use in England may also have depended chiefly on Scandinavian influence, the two-handed weapon does not seem to have been used before the later Anglo-Saxon period.15 An axe could still be handed to King Stephen at the battle of Lincoln in 1141. In the twelfth century, by which time its use was rare, it was seen as the typical Scandinavian weapon by one chronicler, writing of the defender of the river crossing at Stamford Bridge using ‘his country’s weapon’.16
The axe was a fearsome and devastating weapon, but not easily wielded and less flexible than a sword. An axeman required a good deal of space around him in order to be effective, and this could leave the line vulnerable to charging cavalry. It must also have been a particularly tiring weapon to wield through a long day of battle. Nevertheless, it was favoured by some of the English.
One cannot be certain about the Tapestry’s accuracy on size. The art form of the age did not attempt exact reproduction in size. In people, size probably denoted rank rather than height. If it is to be trusted in this respect, then the Tapestry seems to show two types of axe: a smaller hand axe, which might have been thrown, and the long-shafted battle axe, which could only be used in the hand. There is a particularly vivid example of the latter held at the very front of the English line at the moment when it came face to face with the Norman cavalry. Another view shows Leofwin Godwinson wielding an axe, and it may have been primarily an aristocratic weapon.
At the other end of the social scale, the bow in war was usually the weapon of the lowly, though this had not always been the case in Scandinavia. Most of the archers shown on the Tapestry are small and not well armoured, both signs of humble rank. Only wooden bows made from a single piece of wood appear on the Tapestry. Given the size problem with the art, and the dwarf size of the archers portrayed, it is quite probable that the bow was in fact about longbow length. This would be necessary to give it sufficient impact. So the bow used by the Normans was similar to a longbow, a single stave tapered at the extremities, held by a string.17
The armour of the cavalryman, which for Hastings means the Norman cavalryman, differed little if at all from that of the well-armoured infantryman: they also wore helmet and hauberk and carried a kite shield. The only instance where there might be a difference was in the styling of the hauberk. In later periods it was common to split the hauberk at front and back towards the lower edge, so that it could spread out like a skirt. This would mean that the rider could sit astride his mount without having uncomfortably to sit upon a tunic of mail. This is probably the method used in 1066.
The cavalryman’s weapons were similar to those of the infantryman: the same type of sword and a lance which was not unlike a spear. As we have seen, the more specialised cavalry lance was not yet developed. However, the rider would not use the battle axe, a purely infantry weapon, nor in this instance did he use a bow. The aristocracy saw bows as weapons for hunting, but in war it was the weapon of those of lower social rank. However, during the pursuit after the battle, one Norman is shown mounted and drawing a bow.
Perhaps the main weapon of the cavalryman was his horse. Its weight, speed and impact could have great effect. Before long, the Frankish cavalry would be proving its worth against different style armies in the East during the crusades, when those seeing it in action for the first time were greatly impressed. Anna Comnena thought that ‘the first charge of Frankish cavalry was irresistible’; the Frank could ‘drill his way through the walls of Babylon’.18
The cavalry horse was a special animal, more valuable than any other horse. It had to be specially bred and specially trained. It was large without being too clumsy. In battle, as the Tapestry makes abundantly clear, warhorses were stallions. The fact that the Normans went to all the trouble involved in bringing their own horses by sea suggests how important the special animals were. This process is demonstrated on the Tapestry, and described in the Carmen.19
Warhorses had bridles, reins and large saddles, raised at front and rear to give a solid seat. In the Tapestry the saddles appear to be held by a strap round the horse’s chest, and sometimes a strap under the animal’s belly may also be seen. As yet the horses themselves do not appear to have been protected by armour. By this time stirrups were well established, which made fighting from the saddle more feasible and the couched lance effective.
Glover thought that the Norman horseman was ‘for the most part a mounted javelineer’, but the pictures of the Tapestry showing a lance held in couched position suggests this is too restrictive on methods. By the time of the First Crusade, lances were used in the couched position. Anna Comnena describes the southern Norman Robert Guiscard tucking a lance ‘under his arm’ as he prepares for action.20 The rider wore spurs, not a new invention, but important for control in the tense situation of battle.
Both the English and the Norman systems of raising troops were effective. In the normal course of warfare, large armies were rarely required. Often the need was to defend on a wide geographical scale, and therefore the bulk of the troops raised were distributed to key points on frontier or coast according to the threat of the moment. The nature of England’s geography, and the history of Viking raids and invasions, meant that naval defence was a necessity, so fleets and seamen had also to be assembled. It has been argued that the lithsmen of late Anglo-Saxon England composed a permanent force, and they made up a fleet. But this is uncertain, their usual function seems to be that of hired men for specific purposes and temporary rather than permanent, usually being paid off at the end of their period of use.21
Throughout northern Europe it was expected that all mature men would be prepared to fight in defence of their country or state or whatever power held their allegiance. But for larger forces, which needed to be kept in the field a length of time, there was always a problem in taking men away from their usual employment. Armies habitually relied not on forces raised in temporary emergencies, but on more or less professional soldiers.
There were two main sources of such men. Firstly, all kings, dukes, counts, earls and great men of this kind had their own households, perhaps the most significant element of which was the military household. This consisted of trained and often experienced soldiers. They lived with their lord, they accompanied him in war and peace, they defended and protected him. Their loyalty was demanded, often with an oath to guarantee it, his protection and maintenance of them was expected. In battle such men would fight as an integrated group, and group loyalty and comradeship was also a feature of such troops.
The second type of professional soldier were mercenaries or stipendiaries. They served for pay, and were hired for the purpose. There was often little distinction between mercenaries and allies; men who agreed to fight for you and expected rewards and who were paid in some sense. Both the Anglo-Saxon and Norman systems allowed room for hiring experienced soldiers.
The line between mercenary and loyal household man could be thin; one might easily become the other. Because troops were paid certainly did not mean that you could expect them to be unreliable and disloyal. They often prided themselves on giving good service, and on many occasions outshone other men in their dogged fighting for their lord. Sometimes such men also came in groups and were hired under a leader, but the age of captains of large mercenary troops was in the future.
Such professional troops were the backbone of armies, but in order to fulfil all the demands of war, other men were required: for garrison work and field armies when the threat was greater, when, for example, invasion was anticipated or undertaken. Northern Europe as a whole had developed systems which obliged the more solid citizens and farmers to assume military duties. The systems varied, rather as creatures vary with evolutionary development, but the purpose was the same, and the methods at least in such a close area as northern Europe were quite similar.
The differences between the English and the Norman systems are often stressed, but in truth the similarities were more numerous and perhaps more important. Thus in one way or another the primary link was between landholding and military obligation. Maldon in Essex had to aid the royal host by sending a horse for the army. Under King Aethelstan, a law read ‘every man shall provide two well-mounted men for every plough’, and that king also took ‘no small mounted force’ with him against the Scots.22 Even to the possession of horses, this is not unlike the demands made on Normans of a similar rank.
This matter has been distorted by the long historical debate over feudalism. Historians have strained to define it, not surprisingly because in a sense it only exists in their imaginations. This in effect means that any historian can define what he or she sees as feudalism. Our only need is to try and see what actually existed. And of course something did exist: there was a system in each area.
If we avoid talking about feudalism, we may reasonably examine how these systems operated without too much heartache. However, it needs to be made clear that, if not a system, there were certainly areas of land which had been given out in both England and Normandy on the understanding that military service was attached to them. In England, king’s thegns probably held in this way. An early law stated that ‘if a noble who holds land neglects military service, he shall pay 120s and forfeit his land’.23
In England as in Normandy, men’s first allegiance was to their immediate lord, whether he was the king, the duke or another. Before the word fee or fief was used, Church lands in particular had often been granted as benefices, with military obligations attached. English bookland had originally been ecclesiastical. Such land had been given to the Church without any services attached, but much of it was later recovered in order to support military obligations. And English loanland compared closely in function to a benefice, neither being necessarily military in origin. In Domesday Book ‘feudum’ or fief was sometimes the word used for loanland.24
The English had developed a method of raising forces which depended upon land assessment. This had been deliberately initiated for the purpose of raising armies during the years of threat from Scandinavia. In essence, land was assessed by a unit known as the hide. This was referred to by Bede as a unit for a family, and no doubt had a real basis in the minds of its inventors, but like all units of assessment (compare, say, the rating system in England of recent times, or the council tax going through the same process now) it became a unit in its own right not exactly related to any other unit of land or money.
The hidage assessment needed to be reviewed and altered from time to time and so moved further and further away from its origins. Some areas received privileged ratings, others developed after the system had begun and so on. But it proved such a useful method that it was retained for centuries, and was still an important unit in Domesday Book even after the Norman Conquest. Roughly speaking, so many men were demanded for military service according to how many hides an estate had been assessed at. If this was gradually altered to allow men to pay so many shillings instead of performing personal service, it still led to the raising of forces.
The common soldier of the royal army was not, in England any more than in Normandy, the lowest social being in the state. A recent work defines the royal fyrd or army as ‘a royal levy composed of privileged landowners and their own retainers, reinforced by the king’s military household and stipendiary troops’.25 All this confirms a long-held suspicion among historians that neither army at Hastings was probably very large. There is no way of being certain, but figures such as the 1,200,000 given by the Carmen for Harold’s force can certainly be ignored.26
Probably the main group of people called up were the thegns, men of some high social rank in terms of the populace as a whole. The term thegn underwent a change of meaning because the thegn underwent a change of status at the Conquest. The thegn of the England of Edward the Confessor was the ordinary man of rank throughout the realm. However, there were gradations in the rank of thegn itself.
It is becoming clearer that a king’s thegns were a special group. These owed their military obligation to the king directly. Therefore, they are strikingly like the lords who owed military service to the Norman duke for their lands. We need not debate in detail the relation between hides and military service, and there are several problems and uncertainties about it. There may not have been a nation-wide arrangement, but certainly in some areas there was a definite link between the land one held, valued in hides, and the amount of military service given: typically one well-equipped soldier from every five hides. Possession of five hides was also seen as a qualification for the rank of thegn.
It used to be said that English thegns owed their military service because of their rank, while Norman lords owed their service for their lands; but the distinction has become increasingly difficult to maintain. Both groups held lands and both groups owed service. It is doubtful that they made the distinction over purpose that historians have when called to the colours. They came because it was their duty to do so. A Norman or an Englishman would need a good excuse to evade a call from duke or king to join an army of invasion or a defence force against invasion.
The main difference between the composition of the English and Norman armies was neither the system of raising nor the type of man who responded, but the circumstances which made particular demands on either side. Harold was calling on his normal national and local system to face invasion; in 1066 to face a double invasion, which made for rather exceptional demands. We shall see this in practice when we examine the events of that year. But it meant that Harold raised a field army from his own resources and from as many shires as he could use, as well as a fleet. His northern earls also raised forces in the emergency from their resources and from the local shires.
William’s position was quite different. He had to persuade men to join him on a dangerous expedition which was beyond the normal demands of military service. He therefore had to search more widely for men to come, and needed to cope with the problem of keeping a force beyond the usual time limits. Only the offer of extraordinary rewards could succeed. It used to be thought that William had a fully organised feudal system by which to raise his forces. Now this is less clear. It is certainly true that, as in England, there were Normans who owed military obligations. But a good deal of the duchy was held in alods, without obligations. The duke, like the English king, also relied on household troops and mercenaries. In a campaign against the Angevins we hear of him paying fifty knights.27
One problem of troops raised through obligation related to land was the attachment of an understanding of the time limit for it, whether it be forty days, or two months, or whatever. If an army were to be kept in the field for a considerable length of time, there must be additional rewards offered, and therefore even troops which came to fulfil an obligation might stay to be paid, and could in a sense also be considered as mercenaries.
There was also the important and related problem of provisioning an army for a prolonged period: food and the necessities could be collected for distribution, but in the end there would be a resort to foraging. Harold found the problem of keeping a force in the field difficult in 1066. By September ‘the provisions of the people were gone, and nobody could keep them there any longer’.28 Harold in fact disbanded his fleet before the vital battles occurred, having to recall it in the emergency. For William the problem of fighting overseas made more acute the problems of both long service and provisions.
With the glittering if uncertain rewards offered by the possible conquest of England, and thanks to the high military reputation he had won by 1066, William was able to attract into his ranks a variety of men who might term themselves allies or who had some link to him but who could not be described as obligated: men, for example, from Flanders, the county of his wife, Brittany, Maine and other parts of France.
William also put efforts into not only collecting ships but also building anew. The Tapestry portrays both the making of ships and the loading of provisions, including wine, for transport over the Channel.29 Success in war depended as much on the ability to keep a force in the field and to feed it, as it did upon tactical brilliance. In 1066 both Harold and William showed themselves to have abilities above the average in these respects.
Both sides had fleets and sailors available. The Norman fleet was hastily raised, collected and built. The Viking founders of Normandy had maritime expertise, and Normandy has a long coastline with many inhabitants who made their living from the sea. But Normandy had not engaged in naval war and had no naval traditions, apart from its Viking past. William’s fleet was a temporary expedient. It served its purpose, since William’s primary need was for transport.
The English had the more difficult task of using a fleet for defence, but they had a well-ordered naval organisation going back over many years to King Alfred. In 1008 Aethelred II had ordered the building of new ships, one from every 310 hides.30 On other occasions it seems land assessed at 300 hides was expected to provide a ship. The English ships, like those of the Normans, were probably derived chiefly from Viking models. There is a lengthy description of one ship given to Edward the Confessor by Earl Godwin:
A loaded ship, its slender lines raked up
In double prow, lay anchored on the Thames,
With many rowing benches side by side,
The towering mast amidships lying down,
Equipped with six score fearsome warriors.
A golden lion crowns the stern. A winged
And golden dragon at the prow affrights
The sea, and belches fire with triple tongue.
Patrician purple pranks the hanging sail, …
The yard-arm strong and heavy holds the sails.31
And John of Worcester recorded ‘a skilfully made galley with a gilded prow’ given to Harthacnut.32
Men were obliged to serve at sea in much the same way as in the army on land. A number of ports had special obligations in this respect. The Isle of Wight seems often at this time to have been used as a base for the south coast. But in September 1066 Harold had to disband his fleet because it had been kept active too long, and William luckily, or perhaps through good planning, was able to cross without opposition. Harold did recall the fleet, and it posed some threats again to William’s security in England.
THE KEY DIFFERENCES
We have already suggested that there were two vital differences between the armies which met at Hastings: the Norman possession of groups of archers, and the Norman use of cavalry. Each of these factors helped to decide the outcome of the battle, and deserve careful consideration.
Why the English did not have many archers at Hastings is not easy to explain, but it does seem to be true. None of the sources give any indication that English archers played any important role in the battle. None of the reliable literary sources mention English archers. On the Bayeux Tapestry there is one solitary archer among the English ranks, but not portrayed as in any way significant in the fighting, and his bow looks rather unimpressive in size. Yet the word bow itself is an English word, appropriately meaning ‘flexible’ or ‘that which bends’.33
Before 1066 there is not a great deal of evidence about English battles, let alone about the composition of English armies, but there is sufficient to believe that the English did possess archers, as indeed probably to a greater degree did the Scandinavians, so that unlike the matter of cavalry it was not a question of any fundamental difference in the way battles were normally fought. Indeed, whereas the difference over cavalry troops is explicable, the lack of archers in Harold’s army remains puzzling.
The eighth-century Franks’ Casket (from Northumbria) shows a man using a bow in defence of his home. At the beginning of the tenth century, an archer killed a noble by shooting him with an arrow through the window as he relieved himself in the privy. At Maldon in 991, ‘bows were busy’, though it is not said on which side. Elsewhere in the poem there is mention of a Northumbrian hostage, Aescferth the son of Eglaf, who ‘did not shrink back at the war-play,/rather he sent forth arrows swiftly –/sometimes he hit a shield, sometimes pierced a warrior,/time and again he dealt out wounds’.34 Riddle number twenty-three in the Exeter Book is about a bow, spelled backwards, which must be guessed: ‘Wob’s my name, if you work it out;/I’m a fair creature fashioned for battle’, which clearly suggests that these weapons were not only for hunting. A letter of Aldhelm confirms this, with ‘the warlike bowman in the midst of battle’, though he is quoting from a classical source.35
Snorri Sturlusson, the thirteenth-century Icelandic author of the Heimskringla sagas, has the English using bows in the battle of Stamford Bridge. There is no reason that this should not be right, but Snorri’s version of the battle is late, garbled and unreliable, and without confirmation it is dangerous to accept his version as correct. In his account, the Vikings had bows and arrows, while the English attacked with ‘spears and arrows’, with the Norse king being killed by an arrow in the throat.36 There is perhaps some confirmation in an addition made to the C version of the Anglo-Saxon Chronicle, not about the battle, it is true, but about crossing the river, when ‘an Englishman shot an arrow’ at the defender of the bridge, though without dislodging him.37
Two main answers to the puzzle of why the English used archers on previous occasions but not at Hastings seem possible. The first is that, excluding Snorri, these other occasions were much earlier and possibly what had once been a customary method of fighting had become more or less obsolete. There is no more proof for this than for a second contention, but the second seems more likely.
Archers came from certain regions, normally where there were woods or forest. This sort of region made for the use of archery on a larger scale than elsewhere, providing easy sources of material for equipment and plentiful targets for hunting. It is noticeable in later history how often one hears of archers coming from, say, the Forest of Dean or Sherwood Forest. Methods of fighting also tended to be regional, and archery may well have been more popular in Northumbria than Wessex: one notes how many of the early instances of evidence come from the region of the old northern kingdom. Later on, Gerald of Wales wrote that Gwent was noted for its archers whereas other parts of Wales were better known for producing spearmen.
We know very little about the composition of the English armies at either Gate Fulford or Stamford Bridge. Possibly archers fought in one or other or both of these, in which case those same would not have been available at Hastings. As relatively lowly infantry they would be recruited in the local levies. At Hastings, Harold gathered men from as wide an area as he could in the brief space between Stamford Bridge and his fight with William, but the local levies for Hastings inevitably came largely from the south and probably not for the most part from regions especially noted for archery.
In other words it may just be an accident of events that deprived the English of archers at Hastings: the unusual circumstance of having three important battles within the space of a few months and the location of the third and last of them being in the south. We know that the Weald did later produce archers, and perhaps some from there were at Hastings. It may be also that Harold was more concerned about his élite troops, the swordsmen and axemen. Only these men who came from the wealthier social class would possess horses and be able to ride from Stamford Bridge to Hastings. The infantry would be what the local levies provided, and the evidence of the battle suggests that they were chiefly spearmen.
The Normans certainly possessed and used archers at Hastings. Norman archers were also reputed to be lowly. There is a story in William of Jumièges where Bernard the Philosopher puts on poor man’s clothes and then takes up the appropriate weapon, a bow, in order to gain the attention of Duke Richard II by pretending to shoot him. He was arrested, but later pardoned.
The Normans had archers in sufficient numbers to use them in tactical groups. At Hastings, Norman archers were put at the front of the army in the opening phase, played a significant role in the battle, and accounted for the life of Harold Godwinson. Nor was this the first occasion that we hear of Norman archers; they were used, for example, at Varaville in 1057.
Judging by their sophisticated tactical employment at Hastings, archers were a fully integrated part of Norman forces, probably having been employed over a considerable period of time. Both the Franks and the Vikings had archery traditions, so the Normans could have inherited either or both. The bow shown on the Tapestry is always the wooden bow. The likelihood is that this was of length sufficient to be compared to a longbow, and was certainly of the same construction. Actual bowstaves which have been found confirm the point that longbow staves were common over a long period: some forty were found from the late Roman period, one from the seventh century, others from the Viking period. A reconsideration of the artwork of the Tapestry suggests that many of the bows there were probably the height of a man.
Literary sources make it clear that the Normans also possessed crossbows. There is no reason not to believe this possible. The crossbow has a long history, from ancient China to the Romans and the Franks. We have no detailed description and no depiction on the Tapestry, but probably in the eleventh century these would be of relatively primitive construction, perhaps using a forked piece of wood as the basis. It was, nevertheless, thought a powerful weapon. The author of the Carmen says that shields were of no use against crossbow bolts.38
The crossbow was mechanical in operation, with a trigger. It shot a shorter, heavier missile than the wooden bow did, called usually a quarrel or a bolt. One advantage of crossbows was that men could be more easily trained to use them. We can only guess why crossbows are not illustrated on the Tapestry. The most likely explanation is that they were rare in England; we hear nothing of them before the Conquest, though it has been suggested that one is pictured on a seventh-century Pictish carving.39 Since the Tapestry was designed by an English artist, we may guess that he was unfamiliar with this weapon.
As to the Norman possession of cavalry, this is more explicable. The English again had, to a degree, the capacity to use horses in war, and occasionally before 1066 they had done so. Likewise the Scandinavian conquerors; when Cnut crossed the Thames in 1016 it was ‘with many men on horses’.40 When Harold was sent by the Confessor against the Welsh in 1063, he went with ‘a small troop of horsemen’; later he met his brother Tostig who commanded a small ‘equestrian army’. It is even true that an Englishman of rank was expected to ride: ‘a noble should be on a horse’s back’. An early scene on the Bayeux Tapestry shows Harold riding to Bosham with his retainers, called ‘milites’, and they are mounted. There are Englishmen in Domesday Book referred to as ‘milites’, though it is not clear what their military function was. It is not for war, and the horses are probably not warhorses, but the similarity to Norman mounted warriors is unmistakable.
Similarities were marked by the English themselves, and their word for the French ‘chevalier’ is the one which has established itself in the language: ‘cniht’ meaning retainer as the origin, of course, for ‘knight’.41 But the English ethos of war, like that of the Vikings, was an infantry one. Swordsmen and spearmen, in the Scandinavian case even archers, were the admired warriors. The known English burials of warriors demonstrate that the weapons of an infantryman were those considered precious enough to take into the other world, reflecting their values in life. The English warriors expected to stand side by side, comrades together around their leader, to hack, or thrust, and trade blows. The Carmen poet wrote ‘the English scorn the solace of horses and trusting in their strength they stand fast on foot’; and he also stated that at Hastings the English had horses which they left to the rear when forming for battle.
The shield-wall, or war-wall, was a poetic description of this put into practice: a solid line of men standing shoulder to shoulder to face the enemy. At Sherston in 1016, the best men were placed in the front line, which was probably normal practice. At Hastings, the English were described as all on foot and in close ranks.42 The ideal was to advance towards victory or to die.
Cavalry, with its capacity to ride off to safety, was not an admired kind of troop. A telling phrase in the Anglo-Saxon Chronicle relates to the flight of the English under Earl Ralph at Hereford in 1055, ‘because they were on horseback’. In the poem of the battle of Maldon, Byrhtnoth told his men each to let go of his horse, ‘to drive it far off and to march forwards’. This attitude lasted well into the age of cavalry, when it could often be a mark of courage to abandon the horse and its promise of escape, and to stand alongside your men.43
There is plenty of evidence that the English had horses and used them for war, sometimes that they even did fight from horseback. There are early, again northern, stone carvings of cavalry in battle.44 There are countless examples of horses being used for transport, and no doubt at all that the English aristocracy could ride and did ride. Harold’s quick journey from Stamford Bridge to Hastings was not made on foot. The attempts of some ill-informed individuals to do the trip on foot to test its feasibility seem rather pointless. There was no serious problem in covering the distance on horse as Harold and his household and professional or noble troops clearly did. There are also plenty of examples of horses used in raids and for pursuits after a battle. The horses possessed by English warriors were clearly held in the vicinity and used as required.
But actual cavalry fighting in battle is another matter. It required three things at least which the English lacked: a desire to fight in this way; trained warhorses for use in battle; trained men used to fighting from the saddle. The Normans possessed all three, perhaps derived from Frankish connections. A ninth-century proverb in Francia claimed that training for cavalry should begin before puberty, since after that it became difficult to acquire the knack. Young men, such as Robert de Grandmesnil, trained to fight on horseback in William’s household. Such young men were no doubt the ‘boys’ encouraged by Bishop Odo at Hastings on the Tapestry.45
Certainly the English had horses and bred them. We hear on occasions of the Vikings taking English horses for their own use during their campaigns. But the Vikings used them as the English did, not for battle. The Normans needed their own specially bred horses, such as they took with them to Sicily, such as they transported across the Channel in 1066. The young Norman warrior learned to fight from horseback as a noble pursuit. The idea of a mounted élite was already forming in Norman and northern French minds in a way that was yet to happen in England. The heroes of English poetry were warriors who fought on foot. The warfare of Normandy, like most of north-western continental Europe, relied heavily on development from the Frankish Carolingian Empire. Fortified strongholds or castles, and warriors trained to fight on horseback had developed over a century or so. The Bretons, for example, ‘gave themselves to arms and the equestrian art’.46
There is still a question mark over the degree of continuity from Carolingian Neustria to Viking-ruled Normandy. There is little sign of those things we consider ‘feudal’ before 1100: castles, knights, mounted warfare. It may be significant that the great families of the Conquest era, such as the Beaumonts, Warennes and Montgomeries, do not appear much before 1100 either. This does not mean the families did not exist, but perhaps it signifies a change in attitudes to the importance of family and especially the concept of inheritance and lineage.
Both features, castles and mounted cavalry, were emerging as newly dominant in warfare in the mid-eleventh century. Western medieval cavalry methods developed very gradually, with the first evidence from the Merovingian age, and with less dominance in the Carolingian period than once thought. But when the Conqueror aided by the king of France fought at Val-ès-Dunes in 1047, there were probably cavalry attacks from both sides.47 The less reliable and later writer Wace, a Channel Islander, gave a graphic account: ‘horses were to be seen running loose over the plain, and the field of battle was covered with knights riding haphazard for their lives … the Orne swallowed soldiers and horses in great numbers’.48
There is one reference to English cavalry fighting in 1066, but it is made by Snorri Sturlusson. He has an English cavalry charge at Stamford Bridge, not confirmed in other evidence of that battle, and oddly similar in his account to the Norman charges at Hastings. It is likely that Snorri, writing an heroic saga at a much later date, used the only detailed evidence of a battle from the period (for Hastings) to fill out the account of the battle which mattered more to his work, in which occurred the death of his hero, Harold Hardrada, king of Norway. No evidence can be ignored, but Snorri’s is far less convincing than that from the period itself.
Perhaps the most telling piece of evidence in this debate is the one clear example of English fighting on horses close to this period. In 1055 Ralph, earl of Hereford, fought a skirmish against the Welsh. Ralph’s background was continental, he was the Confessor’s nephew by his sister’s marriage to the count of Mantes. Edward knew him during his own Norman exile, and brought him to England after he gained the throne. Ralph was then rewarded with his earldom. He was trained in the continental methods of war, albeit he was not a greatly renowned warrior, and indeed in England earned the epithet of ‘timid’. According to John of Worcester, against the Welsh, Ralph ordered his English troops ‘to fight on horseback, against their custom’.49 They were not successful. From this we may reasonably deduce that there were some continentals in England who were keen to develop cavalry fighting, but who found that at this time it was difficult. It would take time and the will to undertake drastic change. We may conclude that the English by 1066 had chosen to prefer their trusted infantry methods against the considerable changes that would be required to produce not simply cavalry forces but to make their élite troops into mounted warriors.
The development in Normandy may have been later than once supposed, but it had happened. At Hastings, his men were astonished to see William fighting on foot at all.50 The Norman nobility began to see mounted warfare as a part of being the warrior élite. When the Conqueror himself was knighted, sword, shield and helmet were part of the ceremony, as were lance and the reins of his horse. So also were promises to protect the Church and the weak, to act justly. We see the beginnings of chivalric ideals.51
The Normans now used well-armed mounted cavalry as a regular feature of their warfare. In 1041, during the course of their rise to power in southern Italy, a Norman force beat the Byzantine Varangian Guard at Monte Maggiore by a mounted charge. At Civitate in 1053, Richard of Aversa led a successful charge against the opposing papal force, and had the ability to regroup and return to the fray, thus winning the battle. In Normandy, we hear of William commanding forces of three hundred knights; and it was the speed of the charge at Varaville, cutting off the enemy when only half were across the River Dives, which gave William victory in 1057.52
They had also, by 1066, developed some tactical achievements. It used to be questioned whether medieval cavalry was capable of the feigned flight, and some historians wrote off this feature at Hastings as coming from the imagination of writers. It is now widely accepted that feigned flights were possible, and had indeed been employed by various Frankish armies and others over a period of time, for example, by the Alans, Huns, Byzantines and Magyars. The Bretons employed the feigned flight against Fulk IV, count of Anjou. The Normans used the feigned flight on several occasions before Hastings, for example, under Walter Giffard at St-Aubin in 1053: ‘the Normans succeeded in drawing away a considerable part of the army and, as if in flight, they led the French into a trap. For suddenly the Normans who seemed to be fleeing, turned round and began violently to cut down the French.’ Walter Giffard, incidentally, probably fought for William at Hastings.53 The Normans also made a feigned flight to draw the enemy away from the walls of Messina in Sicily in 1060.
Later Norman armies were organised in small tactical groups, sometimes called conroys, and there is little reason to dispute the likelihood that these had a long history. Wace suggested that the conroys were based on men from the same geographical area. The conroy is unlikely to have numbered much more than about twenty men.
There was a third major difference between English and Norman warfare, but its significance is less vital so far as Hastings is concerned. The Normans built and used fortifications which we call castles, the English on the whole did not. There were fortifications in England, but their nature was different. From Alfred onwards there had developed a network of over thirty fortified strongholds, virtually towns – the burhs. Some of these were fortified with stone walls, as Towcester, while some towns already possessing walls had them repaired.54 But these were on a national basis, and were large enough to contain urban populations and offer shelter to those living in the neighbourhood. Nevertheless, it may be significant that our word for ‘castle’ derives from the word the English used for their own fortifications.55
There was at least one private fortification, found at Goltho in Lincolnshire, which suggests that the English were moving along the same sort of path as the Normans. It has been called ‘a pre-Conquest castle’, and its function like that of a castle was as a ‘defended residence’, though Goltho’s defences were not as massive as those of a typical castle, and at present it is a unique site. When it did become a castle in about 1080, it was a particularly small one, but was still more strongly defended than its English predecessor.56
The continentals brought to England under Edward the Confessor had, it is true, built a handful of such buildings, including Pentecost’s Castle and Robert’s Castle, so the English were not entirely unfamiliar with this kind of fortification. It is interesting to find Harold Godwinson making earthwork defences with a ditch around Hereford in 1055. Given time and without the Conquest, castles would almost certainly have developed in England, but the rate would have been slower.57
Equally truly there do not seem to have been so many castles in Normandy, nor were they built so early, as was once believed. Normandy was certainly not the base or centre of castle-building, and seems to have adopted the practice which grew up probably first in the Loire region. In England there were no more than half a dozen castles at the most before the Norman Conquest.
In Normandy, castles were at this time mostly what we might call citadels, fortifications built within towns. But there were some separate structures and a few at least were earthwork and timber (i.e. motte and bailey). There is mention of large ditches cut to defend both Arques and Domfront. When the Conqueror besieged Brionne for three years from 1047, he constructed earthwork fortifications on the banks of the River Risle; similarly at the siege of his uncle’s new castle at Arques, William built a mound for the protection of his own men, and again no less than four mounds at Domfront. At both Domfront and Alençon, previous dukes had permitted the erection or rebuilding of castles.58 There were pre-1066 towers at Ivry and Brionne, as well as ‘The Tower’ at Rouen. J. Yver believed that the length of some sieges meant that the castles concerned were probably constructed of stone.
Breton castles are shown in the Bayeux Tapestry’s account of the Conqueror’s campaign there probably in 1064: at Dol, Dinan and Rennes. The Tapestry also shows the castle of Bishop Odo, the Conqueror’s half-brother, at Bayeux; and at Beaurain where Harold was held prisoner; and probably William’s own fortification at Rouen. There is reference to the Conqueror building a new castle at St-James-de-Beuvron. The use of castles may seem less relevant to Hastings than the possession of archers and cavalry, but it is not entirely without significance, when one considers how William used Pevensey and constructed a castle at Hastings in the period immediately preceding the great battle, and indeed how the Conquest was carried through in the years after Hastings.59Castles of course developed quickly in England after the Conquest partly because the Norman lords were ruling a hostile land and needed to protect themselves.
1. Carmen, eds Morton and Muntz, p. 29. This work has been the subject of much debate over its dating: R.H.C. Davis, ‘The Carmen de Hastingae Proelio’, EHR, xciii, 1978, pp. 241–61, proposing a later date than had previously been accepted; and historians taking sides over the issue since. In the present writer’s view Davis was probably correct to consider there was a problem, and the work may date from about 1100. It would still have interest as a source on armies of the eleventh century.
2. John of Worcester, eds Darlington and McGurk, p. 480.
3. R.P. Abels, Lordship and Military Obligation in Anglo-Saxon England, London, 1988, is an excellent recent work which reinforces this modern trend in thinking about the composition of English forces. See e.g. p. 37: the army was ‘aristocratic in its basis’; also pp. 32, 160, 168, 175. The key work on housecarls is N. Hooper, ‘The housecarls in England in the eleventh century’, ANS, vii, 1985, pp. 161–76, which should be taken with the additional thoughts in N. Hooper, ‘Military developments in the reign of Cnut’ in A. Humble (ed.), The Reign of Cnut, London, 1994, pp. 89–100.
4. S. Pollington, The English Warrior from Earliest Times to 1066, Hockwold-cum-Wilton, 1996, p. 144: the Benty Grange helmet.
5. Pollington, English Warrior, p. 145: the Coppergate helmet.
6. William of Jumièges, ed. van Houts, ii, p. 24.
7. William of Poitiers, ed. Foreville, p. 185.
8. William of Poitiers, ed. Foreville, p. 169. The author, like many others, has been allowed to don a hauberk made by the historian of arms and armour, Ian Peirce.
9. Bayeux Tapestry, pl. 69; William of Poitiers, ed. Foreville, p. 183.
10. Pollington, English Warrior, p. 136.
11. William of Poitiers, ed. Foreville, p. 207.
12. Pollington, English Warrior, pp. 83, 97, 125, 244. The poem about Maldon is useful, but was probably written about thirty years after the event.
13. M. Strickland, paper to Battle Conference, to be published in ANS, xix. See N. Higham, The Kingdom of Northumbria, Stroud, 1993.
14. John of Worcester, eds Darlington and McGurk, p. 486.
15. Pollington, English Warrior, p. 127.
16. Henry of Huntingdon, Historia Anglorum, ed. T. Arnold, RS no. 74, London, 1965, p. 200.
17. J. Bradbury, The Medieval Archer, Woodbridge, 1985, pp. 22–40.
18. Anna Comnena, The Alexiad, Harmondsworth, 1969, pp. 163–5, 416.
19. Bayeux Tapestry, pl. 40–4; Carmen, eds Morton and Muntz, p. 8.
20. Anna Comnena, pp. 56–7; R. Glover, ‘English warfare in 1066’, EHR, xvii, 1952, pp. 1–18, p. 14.
21. See Hooper, ‘Military developments’.
22. Abels, Lordship, pp. 115: ‘caballum in exercitu’; 110. John of Worcester, eds Darlington and McGurk, pp. 389, 934: ‘equestri exercitu non modico’.
23. Abels, Lordship, p. 13.
24. Abels, Lordship, p. 45; A. Williams, The English and the Norman Conquest, Woodbridge, 1995, pp. 191–2.
25. Abels, Lordship, p. 146.
26. Carmen, eds Morton and Muntz, p. 17.
27. William of Poitiers, ed. Foreville, p. 37.
28. Whitelock et al. (eds), Anglo-Saxon Chronicle, p. 142.
29. Bayeux Tapestry, pl. 36–9; William of Poitiers, ed. Foreville, p. 151.
30. John of Worcester, eds Darlington and McGurk, pp. 460–2.
31. Barlow (ed.), Vita, p. 20.
32. John of Worcester, eds Darlington and McGurk, p. 530.
33. Pollington, English Warrior, p. 151.
34. John of Worcester, eds Darlington and McGurk, p. 356; Pollington, English Warrior, pp. 236, 242; 244, ll. 267–71; ‘bogan waeron bysige’.
35. Bradbury, Medieval Archer, pp. 17–22; I. Gollancz (ed.), The Exeter Book, 2 vols, London, 1895, 1934, ii, p. 112, no. 23: ‘Agof is min noma’; Aldhelm, Prose Works, eds M. Lapidge and M. Herren, Cambridge, 1979, p. 163; Aldhelm, Opera, ed. R. Ehwald, MGH Auctorum Antiquissimorum, xv, Berlin, 1919, p. 230.
36. Snorri Sturlusson, King Harald’s Saga, eds M. Magnusson and H. Palsson, Harmondsworth, 1966, p. 152.
37. Pollington, English Warrior, p. 155; Whitelock et al. (eds), Anglo-Saxon Chronicle, p. 145.
38. Carmen, (eds) Monroe and Muntz, p. 25.
39. Pollington, English Warrior, p. 152.
40. John of Worcester, eds Darlington and McGurk, p. 481: ‘cum multo equitatu’.
41. John of Worcester, eds Darlington and McGurk, p. 592: ‘equitatu’, ‘equestri … exercitu’; a saying quoted by Pollington, English Warrior, p. 188: ‘eorl sceal on eos boge’; Williams, Norman Conquest, pp. 196–7.
42. Pollington, English Warrior, pp. 236, 242, 244; Carmen, eds Morton and Muntz, p. 25; John of Worcester, eds Darlington and McGurk, p. 487; William of Poitiers, ed. Foreville, p. 187, he also has the English abandoning the use of horses at this point.
43. Whitelock et al. (eds), Anglo-Saxon Chronicle, p. 130; John of Worcester, eds Darlington and McGurk, p. 576.
44. T. Cain. seminar paper at Institute of Historical Research, 1997, makes several interesting points about the early English use of cavalry at least in the north, as illustrated on northern carvings, though the evidence is rather of riding than fighting on horseback. I am grateful to Tom for passing me a copy of this paper more recently, entitled ‘A hoary old question reconsidered: a case for Anglo-Saxon cavalry’.
45. Bayeux Tapestry, pl. 67; Abels, Lordship, p. 26; one notes that ‘boys’ (pueri) also appear in English households.
46. William of Poitiers, ed. Foreville, p. 109.
47. William of Jumièges, ed. van Houts, p. 120: ‘impetus’ and ‘concursu’ suggest charges, but is only cavalry by inference.
48. Wace, ed. Holden, ii, pp. 39–41, ll. 4091–156.
49. John of Worcester, eds Darlington and McGurk, p. 576: ‘contra morem in equis pugnare’.
50. William of Poitiers, ed. Foreville, p. 199.
51. William of Poitiers, ed. Foreville, p. 13.
52. William of Poitiers, ed. Foreville, pp. 25, 81–3.
53. William of Jumièges, ed. van Houts, p. 104; Giffard may be the ‘Gilfardus’ in Carmen, eds Morton and Muntz, p. 34, l. 539.
54. John of Worcester, eds Darlington and McGurk, p. 376, on Towcester, and Colchester which was repaired by Edward the Elder, who ‘restored the wall’.
55. John of Worcester, eds Darlington and McGurk, e.g. about 885, p. 318, calls the fortification built against Rochester in the siege, therefore at least performing the function of a siege castle, a ‘castellum’; and p. 341 seems to distinguish smaller fortifications as ‘castella’ from towns.
56. Abels, Lordship, p. 92; G. Beresford, ‘Goltho Manor, Lincolnshire: the buildings and their surrounding defences c.850–1150’, ANS, iv, 1981, pp. 13–36, pp. 18, 31, 34.
57. Whitelock et al. (eds), Anglo-Saxon Chronicle, 1052, p. 125; 1055, p. 131.
58. William of Jumièges, ed. van Houts, pp. 102, 122, 124; William of Poitiers, ed. Foreville, pp. 19, 37, 43, 55.
59. J. Yver, ‘Les châteux forts en Normandie jusqu’au milieu du XIIe siècle’, Bulletin de la Société des Antiquaires de Normandie, liii, 1955–6, pp. 28–115, pp. 47, 49; William of Jumièges, ed. van Houts, p. 208; William of Poitiers, ed. Foreville, p. 106.