by John France
The tactics we use and which seem the most efficacious against our enemy are these. The infantry with their antelope [hide] shields, lances and iron-tipped javelins are placed, kneeling in ranks. Their lances rest obliquely on their shoulders, the shaft touching the ground behind them, the point directed towards the enemy. Each one kneels on his left knee with his shield in the air. Behind the infantry are the picked archers who, with their arrows, can pierce coats of mail. Behind the archers are the cavalry. When the Christians charge, the infantry remains in position, kneeling as before. As soon as the enemy comes into range, the archers let loose a hail of arrows while the infantry throw their javelins and receive the charge on the points of their lances. Then infantry and archers open their ranks to right and left and through the gaps they create, the cavalry rushes the enemy and inflicts upon him what Allah wills.72
In recognising the limitations of cavalry and the value of infantry we need to bear in mind that the horses used at Hastings were comparatively small animals. Recent research suggests that in the late eleventh century a horse of twelve hands was quite large, and one of fourteen or more exceptional. To put this into perspective, a Shetland is ten hands, a twelve hand horse would now be classified as a pony, and fourteen a small hunter. These estimates are based on examining the representations of horses in the Bayeux Tapestry, particularly in relation to their riders.73 In the Tapestry all the horsemen arc riding ‘long’, that is with their legs at almost full-stretch and feet in stirrups fully extended, a configuration which gives stability. In all cases the rider’s legs project well below the body of the horse, suggesting a small animal. It is possible that this is an artistic convention but the story of Richard, son of Asclctin of Aversa, who liked to ride horses so small that his feet almost touched the ground is well-known. Moreover, similar representations are known in quite different contexts; an eleventh-century Spanish marble relief and the early twelfth-century Commentaries of Beatus (BM Add 11695) arc examples and many more could be cited. It is interesting that in the Aquileia mural of a crusader with spear couched pursuing and killing a Saracen, no difference in the size of horses is suggested, and this seems to be generally true of early twelfth-century pictures.74 William’s knights charging uphill against steady infantry must have needed good nerves and it is doubtful if they were aware of the ‘shock’ effect which later writers would ascribe to them. What happened along the crest of that hill where Battle Abbey now stands must have resembled the sixteenth-century ‘push at pike’, not the charge of some Hollywood Light Brigade. William exploited his good luck and, decisively, used the mobility of his cavalry with great skill. But the fact that cavalry was decisive does not mean that it was totally dominant, as later experience mentioned here shows. William was certainly careful to bring plenty of foot-soldiers with him.75 Battle was always chancy – William was able to rally his men against one early moment of panic which could have destroyed him. Once this crisis was over he held the initiative and could plan his attacks and he did so to great effect. Hastings was a decisive battle largely because the killing of Harold and his brothers, together with a large number of thegns whose deaths came on top of the butchery at Fulford and Stamford, deprived the Anglo-Saxon realm of much of its leadership. Harold himself paid the price for his folly in engaging too soon. Even so, the battle did not deliver the whole realm to William. He would soon be crowned, but it was only by terrible devastation in the north and covering the land with a network of castles that he was able to secure his hold. This process of conquest was greatly facilitated by the lack of castles in England. The English learned – Hereward built a castle at Ely in 1071 – but by then it was too late and William’s long war of attrition, which followed Hastings, was on the brink of success.76
The conquest of England is not isolated as an example of large scale and complex military effort in late eleventh century Europe. Only a few years later Robert Guiscard, the Norman conqueror of South Italy, launched a great expedition to capture the Eastern Roman Empire. This involved the raising of a fleet and a great army which was kept in the field for some four years from 1081–5. Guiscard had been seeking a Byzantine marriage for his family and when his efforts collapsed be took advantage of the internal weakness of the empire in the early years of Alexius I Comnenus (1081–1118). It was an extraordinarily bold act, for Robert’s brother, Roger, would not complete the conquest of Sicily until 1091, while he himself had promised to aid Pope Gregory VII (1073–85) against Henry IV of Germany. In these circumstances the Byzantines were able to create diplomatic difficulties by subsidising Henry IV, inflaming hostility amongst the many Norman leaders who had resented the Hauteville domination, some of whom were actually employed as mercenaries by Alexius, and by playing upon Venetian concern about a Norman dominion on both sides of the Adriatic. This diplomatic background severely hampered the Norman campaign.77 War opened late in 1080 when Bohemond landed at Avlona with the vanguard of an army 15,000 strong whose core was a purely Norman force of 1,300 knights.78 By 17 June 1081, after seizing Corfu, Robert and Bohemond were before Dyrrachium, the western terminus of the Via Egnetia, the great road to Constantinople, held for Alexius by George Paleologus. A close siege was established around Dyrrachium with the construction of a great leather-covered siege-tower. Against it, Paleologus built a tower on the wall equipped with wooden beams to hold off the Norman attack, and as the two towers engaged, his troops sallied out and burned the siege-tower.79 In July 1081 the Venetians largely destroyed the Norman fleet, and Guiscard was now faced with a strong Greek army under Alexius which by 15 October was close to Dyrrachium. Guiscard’s situation was now extremely difficult, his communications were cut and an enemy force was in the field. Alexius debated whether to attack, or to establish a counter-blockade which would starve the Normans. There was much to commend either course of action. The problem with blockade was that it would take time and Alexius had problems elsewhere, and it was probably because of this that he advanced to battle on 18 October 1081. Guiscard burned the remnant of his fleet, forcing his troops to fight. He seems to have surprised Alexius by leaving his camp early in the morning, so that it was captured by the garrison of Dyrrachium and other forces sent by Alexius. As the Greek army deployed, the Varangian guard, numbering in its ranks many Anglo-Saxons, prepared for action.80 Then they charged, contrary to Alexius’s orders and though they pushed back the horse and infantry under the count of Bari, they were overextended and defeated by an infantry charge in the flank. Many of Alexius’s compound force, including the Turks and the large Slav force under their ruler Bodin, then fled making no effort to intervene as the Normans fell upon Alexius in the centre.81 Guiscard’s victory opened the way for the fall of Dyrrachium in February 1082 enabling the Normans to advance via Deabolis to Kastoria in the spring of 1082. At this point Guiscard was forced to return to Italy by revolt in his own lands, fanned by Byzantine money and by Henry IV’s assault on Rome which Alexius had encouraged, leaving Bohemond to conduct a campaign whose immediate purpose was probably to secure a firm base for further advance. Although a number of cities fell and Bohemond twice defeated Alexius’s efforts to relieve Joannina the Norman expedition was now in difficulties. Bohemond failed to seize Ochrida and Berroea, while the fort at Moglena fell to a Byzantine counterattack. Skopia, Pelagonia and Trikala, amongst others, fell, but the siege of Larissa was undertaken late in 1082 at a time when there had been desertions and treachery in the Norman force.82 These symptoms of exhaustion prepared the way for Alexius to challenge Bohemond in the open field. His earlier experience had not been good. Anna tells us that after the defeat at Dyrrachium Alexius had decided that: ‘the first charge of the Keltic cavalry was irresistible’. In his attempts to relieve the siege of Joannina he used strategies to counter this. In his first effort he strengthened his centre with wagons mounted with poles, whose presence was intended to break up enemy cavalry assault. However, Bohemond was forewarned and attacked on the flanks. It was not a decisive defeat and the emperor returned, th
is time protecting his centre with coltrops, iron barbs scattered on the ground – but Bohemond again attacked on the flank. At Larissa in the spring of 1083, however, Alexius lured much of Bohemond’s force away from his camp which the Byzantines captured, thus forcing the Normans to raise the siege, although the victory left the Norman army intact.83 Bohemond was now faced with retreat and a discontented army which had not been paid and this forced him to return to Italy, while Alexius mopped up his garrisons. In the summer of 1083 a Venetian fleet took Dyrrachium and with the fall of Kastoria to Greek forces in November it seemed that the campaign was over. In the autumn of 1084 Robert Guiscard raised another army and a fleet of 150 ships. He defeated the Venetian fleet before Corfu, which he again seized, but his army was decimated by illness on the mainland and it dissolved totally when he died in July 1085.84
The Norman war against Byzantium was a long affair. It was almost certainly prompted by the weakness of the empire at this juncture, but Guiscard had underestimated his own problems and the range of his enemies, whose various attacks sapped his army. It became a war of attrition in which both sides were desperately short of resources. After his defeat at Dyrrachium Alexius had to resort to seizure of church wealth to raise another army. Bohemond, left in charge by his father, prosecuted a skilful campaign. The Normans continued to be a strong fighting force, but their two victories over Alexius were inconclusive, as was his sole victory over them. In the end, shortages of money and men were more acute on the Norman side than on the Greek, but it was a close-run affair. It is remarkable that the Normans of South Italy could sustain such an effort at all in the circumstances. Certainly the campaign made Bohemond’s name as a soldier.
The campaigns of William the Conqueror and Robert Guiscard were, however, somewhat unusual for the ferocity with which they were fought and the readiness of both sides to resort to battle. When the Conqueror died in 1087 he divided his land between his sons. Robert Curthose held Normandy and William II ‘Rufus’ became king of England. The third son, Henry, was given money which he used to found a lordship in the Cotentin. These dispositions were soon challenged by the brothers, each of whom hoped to gain the whole inheritance of his father. When Rufus died in a hunting accident in 1099 the youngest brother, Henry, took up the challenge with ultimate success, for he seized the English throne and then Normandy with the victory of Tinchebrai in 1106. In nearly twenty years of war Tinchebrai was the only major battle. In the first stage of the conflict, Odo of Bayeux conspired with many of the nobility of England against the king, and Robert Curthose sent Robert of Bellême and Eustace of Boulogne who seized Pevensey and Rochester. However, he failed to raise an expedition to support them and the plot fizzled out. In the next phase, William, with his far greater resources, set about seducing the duke’s vassals and thereby securing castles as bases. It was in eastern Normandy north of the Seine that William concentrated his efforts from 1089 onwards, building a strong position. Robert’s counter-offensive was supported by King Philip of France who, however, allowed himself to be bought off by William. In November 1090, the English king was able to take advantage of factional struggles in Rouen and all but seized the city. It was not until 1091 that William came in person to the scene of this desultory fighting and raiding, which were brought to an end in February 1091 by a peace between the warring brothers. This gave William a strong position in Normandy, in part at the expense of Henry’s lands in the Cotentin and inaugurated a period of rapprochement during which the two brothers tried to impose order in Normandy.85 By 1093, however, the two brothers were again at war and the following year William led a strong army into Normandy. This time Robert waged quite a successful campaign against William and his allies, seizing important castles and threatening his long-established hold on eastern Normandy, until Philip of France was once again bought off with English bribes. It was probably in anticipation of this campaign that in 1093 William met Robert II of Flanders and concluded a treaty under which the count of Flanders undertook to supply mercenaries to the English king. In the end, the English campaign came to a halt when Robert Curthose took the cross. Abbot Jarento of St Bénigne, the papal legate, then negotiated an arrangement whereby Robert pawned the duchy to William for three years for the sum of 10,000 marks. This freed Robert Curthose to join the crusade and provided finances for him.
The men who went on the First Crusade were clearly familiar with the eternal verities of war – above all the need to seize food and deprive your enemies of it. In the context of eleventh-century society, that meant a war of position against castles which guarded lands and lordships, in which battle was to be avoided in favour of attrition. The strength of castles could be challenged by siege, and specialised equipment for this purpose was devised, but more often it was circumvented by raiding which brought the knight, whose mobility and heavy equipment prepared him for every eventuality to the fore. The manorial system, which underpinned this warring world, enabled its masters to sustain and organise warfare over long periods of time. Even at the most trivial levels we have noted, fighting often went on for years. Kings and great lords had the economic capacity to make very considerable military efforts, like the conquest of England or the German expeditions to Italy, which must excite our admiration. They could draw on large reserves of trained manpower from within and without their borders – a wide and rather ill-defined range of people were able to raise their sons in the skills of war and for them almost any war could exert enormous attractions. It was as lances for sale that Normans had emigrated to South Italy at the very beginning of the eleventh century; knights were attracted to war for the pay and the profits and, one is tempted to add, the pleasure. For plunder and ravaging were not merely a way of war, they were a way of life against which the church’s injunctions made little headway. The expedition to the east proclaimed by Pope Urban in 1095 offered an outlet for this drive, reinforced by a sense of righteousness. Some of its leaders were experienced in command. Robert Curthose had fought a long and bitter war, while Godfrey had participated in major expeditions and battles. By contrast Stephen of Blois, Hugh of Vermandois and Robert of Flanders were relatively untried while the count of Toulouse had risen to power in different circumstances. Bohemond alone had commanded a large army and had knowledge of eastern methods of fighting. The real challenge for all of them, however, was working together, for of that none of them had any real experience. Urban II provided an able Legate, Adhémar of Le Puy, who had military experience, but he never appointed a single commander. That was the problem which the leaders had to resolve.
* * *
1 The accounts of William of Jumièges p. 123 and WP, pp. 16–18 say almost nothing about Val-ès-Dunes. Wace’s account has very graphic incidents, some of which may reflect actual events, but tells us nothing of the course of the battle. Guy of Burgundy was the younger son of the marriage between Raynald count of Burgundy and Adelaide, daughter of Richard II of Normandy (996–1026), who therefore had a claim to Normandy against the illegitimate William: Glaber, p. 106 n. 2; for the history of the duchy in these years see D. C. Douglas, William the Conqueror (London, 1964).
2 Douglas, William the Conqueror, pp. 55–9. William’s son Robert later was betrothed to Margaret and, even though she died, assumed the title count of Maine: David, Robert Curthose, pp. 7–10; WP, p. 38.
3 This account of events is drawn from Gillingham, ‘William the Bastard’, pp. 149–51 whose careful study establishes the generalship of William, on which see also J. Beeler, ‘Towards a re-evaluation of medieval English generalship’, Journal of British Studies, 3 (1963), 1–10.
4 See above p. 42. On the succession dispute in Anjou on the death of Geoffrey Martel in 1060 see O. Guillot, Le Comte d’Anjou et son entourage au xie siècle (Paris, 1972), 2 vols., i. 102–116; L. Halphen, Le Comté d’Anjou au xie siècle (Paris, 1906, Geneva, 1974), pp. 133–51.
5 Glaber, pp. 138–9, testifies frequently to the good relations between the Capetian kings and the Norman dukes, for example in th
e matter of the Orléans heresy. It was precisely because of these good relations that he tells us so little about the duchy and its politics: J. France, ‘Rodulfus Glaber and French politics in the early eleventh century’, Francia, 16 (1989), 111. Duke Robert of Normandy gave Henry I (1031–60) support when his vassals rose against him: J. Dhondt, ‘Une crise du pouvoir capétien’, in D. P. Blok et al. eds., Miscellanea Medievalia in Memoriam J. F. Niermeyer 137–48.
6 France, ‘Glaber and French politics’, 109.
7 J. Gillingham, ‘William the Bastard’, pp. 151–53 describes these events very clearly; for Wace, see above p. 36, n. 29.
8 OV, 2. 282–3; William of Jumièges, p. 25; Gislebert of Mons, Chronique, ed. L. Vanderkinadere (Brussels, 1904), p. 7; Annales Egmundani MGH SS 16. 447; Flandria Generosa, MGH SS 9. 322; Lambert of Hersfeld, p. 124.
9 C. Verlinden, Robert I Le Frison, Comte de Flandre (Antwerp/Paris/’S Gravenhage, 1935) pp. 46–70.
10 It is not certain who held the castle of Cassel. Verlinden, Robert le Frison, p. 66, suggests Arnulf, though its castellan Boniface was in Robert’s force. A. Fliche, Le règne de Phillippe I roi de France 1060–1108 (Paris, 1912), pp. 258–9 thinks, therefore, that Robert held it.
11 Verlinden, Robert le Frison, pp. 70–71, 80–6; on family relationships see Bur, Comté de Champagne, pp. 128–9, 286–7. For Baldwin of Hainault’s death see AA. pp. 434–5; Guibert of Nogent, Gesta Dei per Francos, RHC Oc. 4. [hereafter cited as GN], p. 208.