The predominating ethos of Dark Age societies was martial; the king functioned first and foremost as a war-leader and as the defender of his people, and the more effective he was in this capacity, the higher his standing with his own people and his enemies. It was King Alfred who, in his translation of Boethius’ Consolation of Philosophy, first defined what he called the tools of his kingship by separating them into what would be the traditional three classes, the praying men, the fighting men and the working men; but there was never any doubt about which of them formed the aristocracy. In England, as in Normandy, the ability to fight was the most important qualification for life, and the reputation of a renowned warrior the most eagerly sought.
It may seem rather contradictory, therefore, to make the point that major pitched battles, like Hastings, were on the whole avoided whenever possible, and the most successful rulers of the day were generally those who were most efficient in avoiding them. It has been pointed out that the only major battlefield on which William had appeared before Hastings was that of Val-ès-Dunes, when he was only nineteen and where the commander-in-chief was his overlord, the King of France. Over the next twenty years until Hastings, he contrived with considerable adroitness to achieve his objectives by more indirect methods such as siege-work, in which in his early years at least he appears to have been masterly. The battle of Mortemer, in which the Normans defeated the French under the French king and the Count of Anjou, was captained on the Norman side by William’s cousin, Robert of Eu, and the battle of Varaville against the same opponents, where William managed to catch the French army divided in two on either side of a ford, indicates patience and clever tactics but can hardly be compared with a battle of the scale of Hastings. Harold, in his warfare against the Welsh king Gruffydd ap Llewellyn in 1062, showed something of the same tendency. He pursued an extremely effective campaign of harassment against him, but the hands that eventually killed Gruffydd were Welsh, not English. On the whole he appears to have preferred negotiation to battle and to have resorted to force only when diplomacy failed. Edward, on the other hand, despite his saintly reputation, seems to have favoured warfare rather than diplomacy on the occasions for which we have evidence (for example, the exiling of the Godwin family in 1051 and the Northumbrian rebellion in 1065): a not uncommon example of the civilian who knew little at first hand of war contrasted with the soldier who knew all too much about it.
The reason is not difficult to find. Far too much must be hazarded on the outcome of a single pitched battle, and, unless the odds on one side were overwhelming (in which case the lesser side would if it could find a way to avoid battle, such as retreating), the eventual outcome was far too uncertain for the hazard to be worthwhile. The principle was summed up succinctly by Vegetius in his De Re Militari, the military bible of the Middle Ages, probably written about AD 390: battle should be the last resort, everything else should be tried first. ‘The main and principal point in war,’ he went on, ‘is to secure plenty of provisions for oneself and to destroy the enemy by famine. Famine is more terrible than the sword.’lii The tactic of harrying and devastating the enemy’s territory (as Harold did in Wales and as William did when he landed in England) had the double advantage of damaging the enemy’s prestige and economy, and maintaining the invading army at no cost to the invader. As has been pointed out, one man’s foraging is another man’s ravaging. It was also a procedure much more popular with the individual soldier. In ravaging (or foraging), he not only looked after his own commissariat, he also had the chance of plunder. In battle, he was much more likely to be killed. It has been suggested that the harrying of Harold’s lands in Wessex was the most effective stratagem used by William to provoke Harold into confronting him at the earliest possible opportunity. And it was, of course, what the Viking raiders did in the ninth and tenth centuries. Apart from the five pitched battles of Edmund Ironside against Cnut and, of course, the battle of Maldon, there had been few major battles against the Vikings since Alfred’s conclusive victory at Edington. It follows that over the century before Hastings, the English defence capability had been geared more to combat guerrilla Viking invasions than to battles on the Stamford Bridge or Hastings scale, the one notable exception being Athelstan’s great victory in 937 at the battle of Brunanburh over the combined forces of the kings of Scotland and Ireland.
The myth of an Anglo-Saxon army primarily made up of peasants fighting with sticks and stones was exploded many years ago but dies very hard. Such an army could not have held back the Normans for half an hour, let alone a full day. Another myth, strenuously promoted in some circles in recent years, is that the victory of the Normans was that of a highly disciplined feudal force, composed in large part of well-trained cavalry, over some kind of home guard fighting on foot, enthusiastic but poorly equipped and largely untrained, called together in haste from the shires to meet the threat of invasion but hampered by obsolete organization in the face of the sophisticated opposition. In part, this is due to the retrospective effect of the outcome: the English army was defeated by the Norman army, therefore it must, ipso facto, have been inferior. This argument does not take account of the circumstances in which the battle was fought.
A good deal of research has been done on the composition of the two armies that met at Hastings, but in essentials there are several unknowable facts, the most important of which is our ignorance of the size of the two forces. Many efforts have been made to compute these, on the one side from the numbers of men and horses whose heads are shown above the gunwales in the Norman ships in the Bayeux Tapestry (and the belief that one can extrapolate from this a calculation based on the number of ships believed to have sailed), on the other on the assumed length of the English position and the probable depth of Harold’s deployment along it. Neither hypothesis can provide a reliable result. The depictions on the Tapestry are symbolic, not naturalistic, and we have no detailed knowledge of the types or sizes of the ships William built; and the topography of Harold’s original position has been changed so much by time and building work that it cannot support any reliable calculation. There is also the unreliability of the contemporary evidence. On the English side, the Anglo-Saxon Chronicle, corroborated by some later chroniclers, states that Harold fought before many of his troops had come up – contradicted by a different assertion, that in fact he had too many men for the position he occupied. On the Norman side, there are wildly unrealistic estimates of the size of the English army, designed in part, probably, to enhance the duke’s prestige in having beaten so colossal a force. As for the Norman army, William announced before the battle that if he had only 10,000 men rather than the 60,000 he had brought he would still fight Harold, but this was undoubtedly a rhetorical figure and 60,000 is not credible. The possibility of armies of 20,000 or more on each side has been suggested, but is unlikely. The best guesses of the best authorities, based on calculations of the size of the battlefield, the number of men it could accommodate and the space between them that would be necessary for them to fight effectively, are that the two armies were fairly evenly matched, at between 6,000 and 8,000 men on each side, and this is borne out by the length of the battle since one would not expect a battle where one side was demonstrably superior to the other to last so long. But they are no more than guesses.
Ironically, of the two forces we know far more about the obsolete English than the professional Norman, and what can be deduced from the written evidence available does not support the idea of an out-of-date amateur force, crushed by a highly trained professional army organized along feudal lines. This is not surprising, given the length of time over which the English system had evolved and the various Viking invasions to which it had been forced to react. There must always, from the earliest days and on both sides of the Channel, have been some arrangements, of varying degrees of formality, linking the defence of a country to the holding of land in that country, so that all free landholders had a responsibility to the king, or to some intermediary lord, to give armed service when required. Vestiges of such a system over the previous centuries can be found in many parts of Europe. England, by coincidence, provides more evidence of its development than Normandy.
Long before the advent of the Vikings, the rulers of the individual kingdoms of the Heptarchy had had occasion to call on their subjects for fighting men. It was as a result of the ninth-century raids that Alfred made the most far-reaching changes yet to the organization of that requirement. It was he who initiated the systematized construction of fortified towns or burhs (or boroughs, as they became) throughout his kingdom, to be permanently maintained in a state of readiness and defence; in theory no one was more than twenty miles away from a place of refuge in the event of Viking attack. The germ of this idea had been found earlier in Mercia, but it was Alfred who saw its relevance to the kind of hit and run raids to which so many parts of England had been subjected, though it was left to his son, Edward the Elder, to complete the scheme. Edward also produced the complicated Burghal Hidage document, which provided for the maintenance and defence of the burhs and has been described as a watershed in the history of Anglo-Saxon governance.liii The lack of castles in England has been seen as a sign of the general backwardness of the English in military matters, in comparison with the achievements of the castle-building Normans, and Orderic Vitalis ascribes the speed with which William was able to subdue the country after Hastings to the absence of English castles. But the virtue of castles lay chiefly in the part they could play in defending border territory (most of the few English castles that existed before 1066 were on the Welsh marches), or in holding down rebellious or conquered territory (the reason why William built so many of them) or in providing a focus for insurrection. In the centuries after the conquest, the many English castles held by rebellious nobles were to prove a mixed blessing to William’s successors. Alfred’s system of fortified burhs whose administration and upkeep were in the hands of their inhabitants, not of individual nobles, and into which country-dwellers could retreat for safety in time of danger represented a much nobler vision and proved an effective deterrent to Viking raiders.
It was Alfred, too, who solved the problem of raiders who could attack and disperse before the English shire levies could be called out; he arranged a rota, as described in the Anglo-Saxon Chronicle, so that half the fighting force would always be on military duty while the other half remained at home for purposes of immediate local defence and maintaining the general affairs of the country. This produced what was, in effect, the first English standing army, and it worked. But in order for it to work, he had to tighten up the legislation that laid the duty on the land to produce the fighting men he needed. It is also Alfred who is credited with laying the foundations of the English navy when in 897 he commissioned the building of a fleet of longships to his own design, though in action these proved less successful than his other innovations, being too deep in the draught for the river work in which the shallower Viking ships excelled. None the less, his was the first recorded attempt to construct an official naval force for the defence of the nation, and his successors were to build on his achievements.
Naturally, there had been many changes in the detailed organization of the national defence between Alfred’s death in 899 and the mid-eleventh century. During the more peaceful years of the tenth century, the upkeep and manning of the burhs was not maintained with the same rigour as had originally been intended, and it was no longer necessary to keep a standing army in the field. As time went on, though the service due from landowners was strictly maintained, there was a move to allowing it to be commuted for a cash payment with which paid troops could be hired. The Danegeld or heregeld, in the reign of Æthelred, was not used exclusively for paying the Danes to go away; it often paid one lot of Danish troops to fight off another, as Æthelred paid the famous Viking Thorkell the Tall for many years. After Cnut’s introduction of the housecarls in 1018, and their development into the front-line troops of the English army, cash was needed to pay them. But the theoretical obligation of all free men to give military service remained, and could be and was called on. There were clearly enormous variations in detail in different parts of the country, but in general there seem to have been two different types of service: first, the system by which the king could call on a force of warriors for a particular purpose, based on the provision of a man for a certain unit of land; second, the responsibility of all free men to defend the country in an emergency such as an invasion. In the first case, the man was provided, armed, paid and provisioned by the lord, the abbey, the village, the hundred from which he was due, and was expected to serve anywhere needed, at home or abroad, for a certain period, usually sixty days. These would be the men who were normally referred to as the shire levies or the select fyrd. In the second, when all free men were expected to turn out for a national emergency, they were not normally expected to go beyond their own locality and had to be able to return to their own homes at night (there were exceptions on the Scottish and Welsh marches); this has been described as the general fyrd. When Byrhtnoth called together his force to meet the Danes at Maldon, he would have called out the shire levies and his own retainers, but he would almost certainly have regarded this as the sort of emergency in which all the local free men should come to the national defence. This might account for the presence of Dunhere, the ‘unorne ceorl’, the simple peasant, though he might just as easily have been there as part of the shire levies.
The provision of men for the shire levies was the most important part of the defensive system and was related to the tenure of land. The general basis (allowing for regional differences, for example in the Danelaw where the land was assessed in carucates, not hides) is set out fairly clearly in the account of Wallingford in the Domesday Book, and it was probably in accordance with this system that the summons would have been sent out in 1066:
If the king sent out an army anywhere only 1 thegn went out from 5 hides, and for his sustenance or pay 4s. for 2 months was given him from each hide. This money, however, was not sent to the king but given to the thegns. If anyone summoned on military service did not go he forfeited all his land to the king.liv
If the landowner went himself, he presumably paid himself from the profits of his land; if a man who was not a landholder (or held a unit of less than five hides) went, he would collect his pay or the balance of it from those who did own the land that he was representing. But there are indications throughout the Domesday Book that it was usually the same man who answered the call on any individual territory unless he was unavoidably prevented (perhaps by increasing age or by sickness), and that he went well armed and equipped, suggesting that he was a reasonably experienced fighter. The evidence of the Domesday Book indicates that in 1066 the summons was to the general fyrd, since it gives instances where more than one man had gone to the army for less than five hides; and in this case they cannot all have been able to return to their homes at night.
There is also evidence that in general, representatives from the same shire tended to fight together, and thus would have become accustomed to operating as a trained unit. Throughout the Chronicle there are reports that the men of this shire turned out on this occasion or the men of that shire repelled a landing on another or that an ealdorman opposed an enemy with the levies of such and such shires. Legend has it, for example, that the men of Kent traditionally had the honour of forming the vanguard in any campaign in which they fought and striking the first blow in any battle; the men of London traditionally fought around the king. It is tempting to see in these early arrangements the origins of the county regiments that mostly survived until the late twentieth century and were such a distinguished part of English military history. It was, on the whole, a remarkably efficient and sophisticated system. It could be abused, though there is little evidence that it was in pre-conquest days; the worst example is that of William Rufus who in 1094 summoned an army of 20,000 men and marched them to the coast where he demanded that each man gave him 10s. they were carrying out of the 20s. that they were due (they would have received the balance when they returned home), and then sent them home again, thus raising the enormous sum of £10,000. This is recorded by several chroniclers. The story does nothing for William Rufus’s reputation but it does indicate that if that number of men had set out each with the same sum of money, there must have been a fairly uniform system in operation and that it had outlasted the conquest. There is no reason to see why it should have produced a less efficient or well-trained force than any other. It has been well described by C. Warren Hollister:
The personnel of the select fyrd was heterogeneous because the obligation was based upon units of land rather than social rank. Throughout much of England, each five-hide unit was obliged to produce a warrior-representative. The miles who was produced was normally a thegn, but if no thegns were available he might be a man of lower status. He might be a member of one of the intermediate groups – a cniht, a radmannus, a sokeman. And he might, if necessary, be a well-armed and well-supported member of the ordinary peasantry. The important thing was that he represented an appreciable territorial unit which was obliged to give him generous financial support. As such, he belonged to an exclusive military group which can, in a sense, be considered a class in itself. And he may well have taken considerable pride in his connection with the select territorial army of Saxon England.lv