Part III
Chapter 7
The Norman conquest of England was prepared and made possible by the growth of Norman power during the earlier half of the eleventh century, and by the consolidation of the duchy under the rule of Duke William II. To explain why that Conquest was undertaken, and to account for its success and its peculiar consequences it is necessary, however, to consider not only the character of the Norman duchy in his time but also the development of Norman policy during the same period. Not otherwise is it to be understood why, and how, the medieval destinies of England were during the third quarter of the eleventh century to be deflected from Scandinavia towards Latin Europe by a descendant of the Viking Rolf.
An intimate political relationship between Normandy and England was part of the inheritance of Duke William.1 The great Viking leaders of the ninth century had transferred their operations indiscriminately on both sides of the Channel, and the resulting settlements were associated both in the origin and in their character. In one sense the Danelaw can be regarded as the English Normandy, and Normandy the French Danelaw, inasmuch as in both cases similar problems arose as to their assimilation into the older political order in which they were planted. More direct political connexions inevitably ensued. Both Athelstan and William Longsword were concerned with the fate of Louis d'Outre-Mer,2 and after the intensification of the Viking onslaughts towards the end of the tenth century, the interrelations between the two dynasties became more pronounced. Ethelred II of England (who had to bear the brunt of these attacks) was naturally interested in what might be the policy of the Viking province of Gaul, and the Norman dukes were not always unwilling to turn the situation to their own advantage. The matter was clearly one of general importance to western Europe, and it is not altogether surprising that Pope John XV himself felt constrained to intervene. As a result a remarkable assembly met at Rouen in March 991 under the presidency of the papal envoy.3 It consisted (as it would seem) of Roger, bishop of Lisieux, and other Norman notables, together with Æthelsige, bishop of Sherborne, and two English thegns.4 There it was agreed that neither the duke nor the king should henceforth aid the enemies of the other, and the pact aptly symbolized both the special position which Normandy was attaining in western Christendom, and also the developing relationship between the ruling families of Normandy and England.
As the eleventh century advanced, the connexion became ever closer. The treaty of 991 was evidently only partially effective, for a Norman tradition which has some claims to credence refers to an unsuccessful English attack on the Cotentin in 1000, and if this in fact took place it was probably in the nature of a cutting-out operation designed to inflict punishment on a Viking fleet which after raiding England was refitting in the Norman harbours.5 Something more was, therefore, necessary to cement between the English king and the Norman duke an alliance which was becoming increasingly necessary to both. And this took place with the famous marriage in 1002 between Emma, sister of Duke Richard II, and Ethelred II of England.6 The full consequences of this momentous match were only to be revealed during the reign of the Conqueror himself, but even before his birth some of its implications became apparent. For when in 1013 Sweyn Forkbeard was successful in his last, and greatest, invasion of England, and the West Saxon royal family took refuge in flight, it was, as if inevitably, to Normandy that they turned. In the autumn of 1013 Emma arrived in the duchy with her two sons Edward and Alfred, and there they were joined in January 1014 by Ethelred himself.7 It was from Normandy, and with Norman backing, that in the next month Ethelred II returned to England to wage his last unavailing war against Sweyn's son, Cnut the Great.8
The fates of Normandy and England were in fact becoming intertwined. Thus in 1013, during the same months when Sweyn was conducting his attack upon England, Olaf and Lacman were ravaging Brittany, and after their depredations they were received as guests in Rouen by Duke Richard II.9 It is small wonder that the king of France was alarmed, and the assembly of Gaulish notables that he convoked at Coudres was a measure of his concern.10 The danger was averted when Richard, perhaps by bribery, divested himself of his pagan allies, and the changing character of Norman policy at this period was further illustrated when Olaf – later to be the patron saint of the Scandinavian world – received baptism from the archbishop of Rouen.11 Much of the future was here foreshadowed, and the situation was given further precision, when, after the death of Ethelred II, Emma joined her fortunes to those of his Scandinavian supplanter, and in July 1017 became the wife of Cnut the Great,12 who was now king of England and soon to be the lord of a wide Scandinavian empire. All these events concerned Normandy no less than England. Nor was it merely the ruling families in the two countries that were affected. The political filiations of England with Scandinavia, the changing position of Normandy within Gaul, were equally and reciprocally involved. And the consequences were to be far-reaching.
Duke William's relations with England were thus conditioned by an interconnexion between the duchy and the kingdom, which had been formed before his birth. But if he was here the heir to a tradition, he none the less transformed it. His achievements in this matter, the manner in which they were accomplished, and the personalities who were involved have, therefore, always to be regarded in relation to those larger political problems which were posed before his time, but which were to be given a special and permanent solution as a result of his acts. At the time of his birth the political pattern which was to result from the relation between England, France, and the Scandinavian world had not been determined; and it was Normandy, uniquely placed within that pattern, which was, through Duke William, to give it final shape. Duke William's policy towards England may thus be seen to possess, even in its complex details, a logical coherence. It progressively involved the chief powers of western Europe. And it culminated in one of the most dramatic episodes of history.
Even during his boyhood some of the many issues involved in this critical situation were being given more precise definition. It would seem that shortly after Cnut's establishment as king of England, the English athelings, Edward and Alfred, Emma's children by her first marriage, returned to Normandy as exiles. Their influence upon the formation of Norman policy at this time is not easy to determine, but it was not negligible. The remarriage of their mother and the friendship which prevailed between Cnut and Duke Richard II towards the end of the latter's reign must have kept them for some years in the background. After 1028, however, there is testimony to suggest that they may have been in fairly close attendance on Duke Robert I, so that William must often have seen them at his father's court. A charter given by Duke Robert for Fécamp about 1030 is, for instance, subscribed with their signs,13 as is also a gift to Saint-Wandrille made by Robert in 1033.14 Certain other texts also deserve consideration in this respect. Thus a charter of Duke William for Le Mont-Saint-Michel, which if genuine would fall in, or shortly before, 1042, contains in an early copy the attestation Haduardus rex.15 The authenticity of this deed has, however, been legitimately questioned, and partly on the assumption that Edward would not have used the title of king at that time.16 On the other hand, a charter given by Duke Robert to Fecamp between 1032 and 1035 commands greater confidence, and in both the early copies in which this deed survives there occurs the sign ofEdwardi regis.17 It is, of course, very probable that in this text (as in the suspect charter of Duke William for Le Mont-Saint-Michel) the royal title was given to Edward by a scribe writing after his accession to the English throne in 1042. The matter cannot, however, be regarded as finally settled, and whether Edward did, or did not, use the royal style at this early date, the Norman charters in which his name occurs give strong support to the statements of the Norman chroniclers that the English athelings in their exile kept alive at the Norman court their claims to the throne of England.
Nor is there any doubt that these claims were recognized and fostered by William's father. It was during this period, for example, that Goda, Edward's sister, who was with her brother in Normandy, was given in marriage to Dreux, count of the Vexin, who was Duke Robert's friend and ally,18 and beyond question the relations between Normandy and Cnut steadily deteriorated during Duke Robert's reign. It was later alleged, and it is widely believed, that the enmity between the two princes was caused by an intimate personal quarrel between them. Cnut, it is said, in order to propitiate the Norman duke gave Robert his sister Estrith, whom Robert married and subsequently repudiated. The evidence for this tale is, however, most unsatisfactory, and there is no need to have recourse to such a story to explain Robert's attitude towards England and its Danish ruler.19 Robert may (as his apologists suggest) have been genuinely moved by the plight of the athelings, but at all events he had every reason to be apprehensive at the rapid extension of Cnut's power in the Viking world. As a result he was drawn into hostility with Cnut and became implicated in English affairs on the side of the dispossessed West Saxon dynasty. A statement by an early Norman chronicler even asserts that he planned an invasion of England on behalf of Edward and Alfred, and collected ships for that purpose. The story, which is unconfirmed, should be treated with caution, but it cannot be summarily dismissed. According to the narrative, the fleet actually set sail for England, but being harassed by a storm was diverted to Brittany to aid the Norman troops who were there engaged against Count Alan III.20
Cnut and Robert both died in 1035, and Anglo-Norman relations thereupon entered a new phase. In both countries the succession involved grave problems. In England, Cnut's designated heir seems to have been Harthacnut, Cnut's son by Emma, but in 1035 this man was in Denmark, and his half-brother, Harold ‘Harefoot’, succeeded in getting himself recognized as joint king despite the opposition of Emma and of Godwine, the powerful earl of Wessex.21 In Normandy the troubles of William's minority were beginning. In these circumstances any Norman intervention in English affairs was clearly impossible, and the athelings who were still in the duchy were left to look after their own fortunes. And one year after the Conqueror's accession there occurred an event which was to have an enduring influence upon his subsequent policy. In 1036 Alfred the brother of Edward came to England from Normandy, ostensibly to visit his mother Emma. His advent was embarrassing to all parties in England and particularly to the supporters of Harold Harefoot with whom Earl Godwine was now joined. As a result, Godwine seized Alfred with his followers before they could reach Emma. He put many of the atheling's companions to death, and delivered Alfred himself to an escort of Harold Harefoot's men who took him on board ship, and there blinded him before bringing him to Ely, where he died from his mutilation.22 It was a crime which shocked the conscience of even that callous age, and it was to leave a legacy of suspicion and hatred. There is good reason to believe that Edward, later king of England, always held Godwine guilty of his brother's murder, and that he never forgave the earl.23 Moreover, it soon became clear that if in the future there was to be further Norman intervention in English affairs, the terrible fate of the Atheling Alfred might be cited as a justification.24
The conditions prevailing in Normandy and England at this time were such as to prevent the development of any settled policy between them. The death of Archbishop Robert of Rouen in 1037 precipitated increased disturbance in the duchy, and the English situation likewise became increasingly dominated by the rivalries of contending factions. Harold Harefoot was recognized as sole king from 1037 until his early death in June 1040, when Harthacnut became king in his turn.25 Already, however, some of the English magnates were looking across the Channel for a successor, and in 1041 the Atheling Edward was invited to come from Normandy to England. It must have demanded considerable courage for him to accept such an invitation in view of what had so recently happened to his brother. None the less he crossed the Channel, became a member of the household of Harthacnut, and (as it seems) was recognized, at least by one faction, as the successor to the English throne.26 On the other hand, there were Scandinavian princes whose claims could not be easily ignored. Sweyn Estrithson, of Denmark, though still a youth, might, as Cnut's nephew, be said to have an hereditary interest, whilst Magnus, king of Norway, had, apparently in 1038 or 1039, entered into a pact with Harthacnut that in the event of either of them dying without direct heirs the survivor should inherit his kingdom.27 A critical situation was thus created when on 8 June 1042 Harthacnut, then no more than twenty-three years old, died suddenly ‘as he stood at his drink’.28
It is not wholly certain whether the Atheling Edward was in Normandy or England at the time of Harthacnut's death, but he was at once acclaimed king by a strong party in England ‘as was his natural right’.29 For some months it would seem that his position may have been doubtful, but he was at length fully recognized, and crowned on Easter Day 1043.30 The event was clearly of major importance not only to England but also to Normandy. Edward, of course, owed his acceptance to the fact that he was the representative of the ancient and honoured West Saxon dynasty. But he was none the less in a special sense the protégé of Normandy, where he had spent so many years of his exile, and the Norman ducal house might well feel itself to some extent committed to his cause. Duke William himself was, it is true, only some fourteen years old at this time and his own position was precarious, but the connexion thus established was to prove of special consequence to the later development of his policy towards England.
Normandy was thus concerned from the start with the fortunes of the new king of England, who had been placed in a situation of some hazard. Edward had succeeded despite the claims of Scandinavian princes; there remained a very strong Scandinavian element in his court;31 and the Danelaw districts in England had strong Scandinavian sympathies. It is not surprising therefore that the opening years of the reign of Edward the Confessor should have been coloured by an imminent threat from Scandinavia, and marked by the efforts which were made to withstand it. In 1043, for instance, Edward, supported by the earls of Wessex, Northumbria, and Mercia seized the person of Emma, who was apparently scheming on behalf of Magnus, and confiscated her property. In 1045 Magnus planned a large-scale invasion of England, and was only prevented from sailing because of his own war with Sweyn Estrithson of Denmark. Sweyn actually appealed to Edward for help, which was refused, and the menace to England became more acute when Magnus, having expelled him from Denmark, was once more ready to invade England. The sudden death of Magnus on 25 October 1047 may indeed have saved Edward from disaster, but the menace none the less remained. In 1048 the south-eastern shires were harried by a considerable Scandinavian force, and if the object on this occasion was plunder, it is clear that aspirations for the political reconquest of England were being fostered in the northern courts.32
Edward's hope of combating this threat depended upon the support he could elicit from his magnates, and here too he found himself in a position of difficulty. For the earldoms which had been created by Cnut as administrative provinces had now fallen into the hands of powerful families who between them exercised jurisdiction over the greater part of England. To implement any consistent policy therefore, Edward had in the first instance to reconcile the bitter rivalries of his great earls—Siward of Northumbria, Leofric of Mercia, and Godwine of Wessex—and if possible to use them to his own advantage. The struggles among them which ensued, and in particular the rise to dominance during these years of Godwine, earl of Wessex, was thus of major concern to the English king, and by implication a matter of moment to Normandy also. In 1045 Earl Godwine, presumably as the price of his allegiance, had forced the king to marry his daughter Edith,33 and from that time forward the increase of the power of Godwine though not undisputed was constant. He had, it is true, to overcome the opposition of the other earls, and this, as in 1049, sometimes resulted in violent disorder. But by 1050 the family of Godwine had come to dominate the English scene. The earl of Wessex himself held all southern England from Cornwall to Kent; his eldest son Sweyn, whose career had been ruthless and disreputable, held five shires in the south-western Midlands; whilst Harold, his second son, was established as earl in Essex, East Anglia, Cambridgeshire, and Huntingdonshire. It was a concentration of power in the hands of a single family that was in itself a menace to the royal authority, and it was especially repugnant to the king since it had been achieved by the man who was charged with his brother's murder.34
In these circumstances it was inevitable that the king should endeavour to form his own party, and it was natural that he should turn to the connexions he had made during his exile, and in particular to the Norman duchy which had for so long given him hospitality and protection.35 Thus Norman clerks began to appear in the royal household, and transferences of property in favour of Normans were made in the country. In Sussex, for instance, Steyning, then a port, was given to the abbey of Fécamp,36 and Osbern the brother of William fits Osbern, and himself a clerk, was established at Bosham which commanded the harbour of Chichester.37 In the west, a more significant development took place in connexion with the advancement of Ralph, nicknamed ‘the Timid’, the son of Dreux, count of the Vexin, who had married Edward's sister. This man had come to England with the king in 1041, and received extensive lands in Herefordshire, Worcestershire, and Gloucestershire. In due course he became earl of Herefordshire, and under him a Norman colony was established. Richard, son of Scrob (a Norman), was settled in Herefordshire before 1052, and about the same time another Norman, nicknamed ‘Pentecost’, acquired the important manors of Burghill and Hope. Before 1050, or shortly afterwards, both these men constructed castles: ‘Richard's castle’ near Hereford, and the more famous stronghold of Ewias Harold which dominated the Golden Valley.38 More important, however, was the king's success in introducing Norman prelates into the church. About 1044 Robert, abbot of Jumièges, was made bishop of London;39 in 1049 Ulf, another Norman, became bishop of Dorchester, the see which stretched across England to include Lincoln;40 and in 1051, after Robert of Jumièges had been promoted to Canterbury, William, a Norman clerk in the king's household, was appointed bishop of London.41
The infiltration of Normans into England during these years has attracted much comment, but it might be easy to exaggerate the originality of the king's policy in this respect. There is reason to believe that a group of men from Normandy had followed Edward's mother to England at the time of Emma's marriage to Ethelred II, and though these receded into the background after her marriage to Cnut, they can be regarded as the precursors of the men who responded to her son's invitation.42 Edward's Norman policy was in short only partially dictated by his personal predilections: it flowed naturally out of the previous relations between Normandy and England. Nor, except in the Church, did it at first entail any very widespread consequences. It is to be remembered that during these years Edward could expect no personal support from Duke William who was himself engaged in a struggle for survival, and the greater men of Normandy were likewise too fully occupied in establishing themselves in the duchy to pay much attention to England. Among the laymen who came to England from overseas during these years, few except Earl Ralph the Timid were of the first rank, and no Norman layman at this time seems to have been given possessions in England of very wide extent. None the less the tenacity with which the king pursued his Norman policy in opposition to Earl Godwine between 1042 and 1051 was bringing the affairs of the duchy and the kingdom into ever closer juxtaposition. And when in 1051 Edward found himself at last able to confront Earl Godwine on a major issue of policy, the relationship between Normandy and England was brought to a crisis.
The narratives which describe this crisis are mutually contradictory, and much controversy has taken place as to its immediate causes and consequences.43 What is certain is that, very early in 1051, Robert of Jumièges, then bishop of London, was translated to the metropolitan see of Canterbury, and shortly afterwards, in the same year, the citizens of Dover were involved in an affray with the retinue of Eustace of Boulogne, the king's brother-in-law, who was returning to France after a visit to King Edward. Edward called in Earl Godwine to punish the citizens, but the earl, being perhaps already outraged by the appointment of Robert of Jumièges, refused to do so, and forthwith collected levies in all the earldoms under the control of his family in order to oppose the king. On his part, the king appealed to the loyalty of his subjects, called for the assistance of the Norman party he had created, and managed also to obtain the support of Earls Siward and Leofric against the rebellious earl of Wessex. The two northern earls thus came with troops to the king at Gloucester where they were joined by Earl Ralph, and they sent to their earldoms for further reinforcements. It was a trial of strength, and Edward was victorious. The opposing armies were disbanded by mutual consent, but Godwine and his sons were ordered to appear before the royal council in London to answer for their misconduct. When they refused to do so, they were condemned to banishment as rebels and forced to fly the country. It was a notable triumph for the king. Indeed, as a contemporary remarked: ‘It would have seemed remarkable to everyone in England if anyone had told them that it could happen because he [Godwine] had been exalted so high even to the point of ruling the king and all England, and his sons were earls, and in the favour of the king, and his daughter married to the king.’44 Edward had been forced to wait long for his deliverance, but by the end of 1051 it seemed complete.
The events of 1051 are, however, chiefly important in this context because they brought to a logical conclusion the earlier development of Anglo-Norman relations which we have been concerned to watch. Edward was childless; a feature of his triumph in 1051 had been his repudiation of his wife who was Godwine's daughter;45 and there can be no reasonable doubt that before the end of 1051 he had nominated William of Normandy as his heir. It is, moreover, to 1051 itself that in all probability this grant should be assigned.46 One authority even seems to suggest that in 1051 the duke came over to England to receive the grant in person.47 But this, although very generally believed, is most unlikely48 – if only for the fact that William was desperately concerned with affairs in Normandy throughout that year.49 It is more probable that, as another narrative states,50 Robert of Jumièges was sent to acquaint the duke of the bequest, and that he did so between mid-Lent and 21 June 1051, when on his way to Rome to seek his pallium as archbishop of Canterbury. The rebellion of the earl of Wessex may even have been caused by knowledge of this transaction, and the affair at Dover would in that case have to be regarded as only a secondary cause of the upheaval which followed. At all events, by the end of 1051, Godwine and his sons had been banished from England; the king who had for long received Norman support was master of his English kingdom; and Duke William of Normandy was his designated heir.
If the conditions prevailing in England at the close of 1051 had been allowed to continue, it is even possible that the political union of Normandy and England under the royal rule of a Norman duke might have been peacefully achieved. Events on both sides of the Channel were, however, at once to modify this situation. In England the year 1052 saw the re-establishment of Godwine and his sons by force of arms. The great earl had taken refuge in Flanders whilst his sons Harold and Leofwine fled to Ireland. From these countries a co-ordinated and brilliantly organized attack upon England was made by sea. It was overwhelmingly successful, and the king was forced to submit. He was compelled to readmit Godwine and his sons to their English dignities, and to receive back to favour Godwine's daughter Edith as his wife. Whilst moreover some of the king's continental advisers, such as Ralph the Timid and William, bishop of London, were allowed to remain in England, most of the members of the Norman party were sent ignominiously into exile, including Robert the archbishop of Canterbury. A counter-revolution had in fact taken place, and its consequences were to be profound.51
The expulsion of Robert of Jumièges raised, for instance, the question of the metropolitan see he held, and a new element was introduced into Anglo-Norman politics when his place was promptly filled by Stigand, bishop of Winchester.52 This man, whose reputation as a churchman was not untarnished,53 had been a strong supporter of Godwine, and his promotion was clearly due to the triumphant earl. But the substitution of such a prelate, by such means, for an archbishop who had not been canonically deposed offered a challenge to the movement of ecclesiastical reform which was now being sponsored by the papacy. Stigand was in fact to be excommunicated and declared deposed by no less than five successive popes, and even in England his position was held to be so equivocal that prelates hesitated to be consecrated by him.54 From 1052 onwards, therefore, the family of Godwine was to find itself out of favour with the reforming party in the Church, and for this reason it incurred the constant opposition of the papacy. William in Normandy could, by contrast, take considerable advantage from this situation, and he was to exploit it. The support which in 1066 he was to receive from the papacy was in fact partly inspired not only by the reformed state of the province of Rouen but also by the ecclesiastical situation which had been created in England twelve years earlier. And one of the inevitable consequences of the Norman conquest when it came was to be the final deposition of Stigand.
In 1052, however, the victory lay with the family of Godwine. The royal authority in England had been challenged and defeated, and the Norman policy of the king had been broken. Nor could the ultimate consequences of this to the future development of Anglo-Norman relations be other than catastrophic. The events of 1052 (as has been shrewdly observed) ‘established the house of Godwine so firmly in power that neither the king nor any rival family could ever dislodge it. It reduced the Normans in England to political insignificance, and thereby decided that if the duke of Normandy was ever to become king of England it could only be through war.’55
Whether the duke of Normandy would ever be strong enough forcibly to assert his claims on England was, however, in 1052 still very doubtful. Between 1052 and 1054 the duke was facing one of the great crises of his Norman reign, and this was not to be resolved until after the battle of Mortemer. Not before 1060 was he to be wholly free from peril. Every stage in the very confused history of England during these years was thus to be of vital concern to his future, and he must have watched with intense interest the manner in which death was changing the pattern of English politics at this time. In 1053 Earl Godwine died,56 and since Sweyn, his eldest son had died on pilgrimage in 1052, the leadership of the family passed to his second son Harold. Two years later occurred the death of Earl Siward, and the Northumbrian earldom was given to Harold's brother Tosti, thus further increasing the power of the family.57 Finally in 1057 two other deaths were to take place which were of great consequence to the future. In that year Earl Leofric of Mercia passed from the scene, leaving his earldom to his son Ælfgar, and on 21 December 1057 there died Earl Ralph the Timid58 who, as a grandson of Ethelred II through his mother Goda, might have been considered as one of the possible claimants to the English throne. He passed on his claims to his brother Walter, count of the Vexin.
It was against this rapidly changing background that between 1053 and 1057 there was formed a plan to deflect the English succession away from Normandy and to substitute for Duke William a member of the West Saxon royal house. This was Edward, son of Edmund Ironsides, who since 1016 had been in exile in Hungary. He was quite unknown in England, but negotiations were opened for his return,59 and eventually in 1057 he arrived in England accompanied, as it would seem, by his wife Agatha and his three children – Margaret, Edgar, and Christina.60 It was an important occasion, for he came in state, with the support of the emperor, a great noble with much treasure. As was the case with Alfred in 1036, therefore, his advent was recognized as having deep political significance, and again, as in the case of Alfred, his coming was the prelude to tragedy. He died in mysterious circumstances before he could reach the royal court. ‘We do not know,’ exclaimed a contemporary, ‘for what reason it was brought about that he was not allowed to visit his kinsman King Edward. Alas, it was a miserable fate and grievous to all the people that he so speedily ended his life after he came to England.’61 The words, it is true, do not in themselves warrant any specific accusation of foul play, but the phrases seem almost designed to invite suspicion, and there were many powerful men in England to whom Edward's arrival, like that of Alfred fifteen years earlier, must have been unwelcome. Nor was his removal without advantages to some of them. At all events, from this time forward Harold, earl of Wessex, seems to have begun to think of the succession for himself.
After 1057 the pre-eminence of Earl Harold in England rapidly became more marked. No royal atheling remained in the country to overshadow his prestige, and the deaths of Earls Leofric and Ralph enabled him once more to increase the territorial possessions of his house. He himself annexed Herefordshire; East Anglia passed into the hands of his brother Gyrth; whilst Leofwine, another brother, was given an earldom stretching from Buckinghamshire to Kent.62 As a result of these arrangements Harold and his brothers, Tosti, Gyrth, and Leofwine, controlled the whole of England under the king with the exception of the Mercian earldom under Ælfgar which now had been diminished in size. It is little wonder that Ælfgar felt himself menaced. He was constantly in rebellion during these years, seeking support from Griffith, king of North Wales, and even from Scandinavian raiders. Sometimes in exile, and sometimes in precarious possession of Mercia, he survived until after 1062.63 After his death his earldom passed to his young son Edwin, who could offer no effective opposition to the great earl of Wessex. By 1064, therefore, Harold had reached the apogee of his power, and it is little wonder that an annalist could refer to him as ‘under-king’ (sub-regulus).64 Nor could it be any longer doubted that he was hoping eventually to acquire the royal dignity itself.
A new factor was thus intruded into the impending question of the English succession, and it was not only the duke of Normandy whose interests were challenged. Walter, count of the Vexin, for instance, might feel himself concerned as the grandson of Ethelred II, as might also Eustace, count of Boulogne, who as the second husband of Goda was the Confessor's brother-in-law.65 More formidable reactions to Harold's rise might, however, be expected from Scandinavia, for after the death of Magnus the kingdom of Norway had fallen to Harold Hardraada, half-brother to St Olaf, a man whose adventures were already legendary, and whose ambitions were boundless. He certainly considered that the pact between Harthacnut and Magnus had given him a claim to the English throne, and he was ready to seize any opportunity to enforce it. The developing ambitions of the earl of Wessex thus presented to Harold Hardraada a challenge which he could hardly ignore. It is significant, therefore, that in 1058 his son attacked England on his father's behalf with a large fleet collected from the Hebrides and from Dublin,66 and though the attempt was not successful, it clearly foreshadowed the larger invasion which Harold Hardraada was himself to make in 1066.
None the less, it was Duke William of Normandy who was most directly concerned with the developments which were taking place in England. Few political relationships in northern Europe were closer than that which had grown up between Normandy and England during the earlier half of the eleventh century, and this connexion had been fortified by the circumstances of Edward's accession, and more particularly by the promise of the English succession to the Norman duke. William was, moreover, just at this time himself attaining a position from which he could effectively assert the rights which he considered to have devolved upon him. The feudal and ecclesiastical consolidation of Normandy under its duke was now far advanced, and William's situation in France was much improved. The battle of Mortemer in 1054 was the last of the great crises of William's Norman reign, and the deaths of Geoffrey of Anjou and of Henry, king of France, in 1060 removed two most formidable rivals from his path. The same years that witnessed the rise of Earl Harold in England thus also saw the advancement of Duke William to a pre-eminent position in northern France, and in 1062 he was given the opportunity of turning that pre-eminence to practical advantage. The war of Maine which ensued was in fact to be of cardinal importance in producing the conditions that were necessary to the success of his English enterprise four years later.
It will be recalled that after the occupation of Le Mans by Geoffrey Martel in 1051, Duke William had sponsored the cause of the exiled Count Herbert II of Maine, and that a pact had been made between them by which it was agreed that if Herbert died without children, Maine should pass to the Norman duke.67 At the same time Herbert engaged himself to marry a daughter of the duke, whilst Robert, the duke's son, was betrothed to Margaret the infant sister of Herbert. So long as Geoffrey Martel lived, such arrangements were of little consequence, but after 1060 they began to be significant, and on Herbert's death on 9 March 106268 they precipitated a crisis. Duke William immediately claimed Maine on behalf of his son, whilst a strong party in Maine led by the border lord, Geoffrey of Mayenne, determined to resist him by putting forward as their candidate for the succession Walter, count of the Vexin, who had married Herbert's aunt Biota.69 The situation was of particular interest since two of the most disputed fiefs of France were called in question by a single challenge: Maine which had for so long been a battle-ground between Normandy and Anjou, and the Vexin which had been similarly debatable between the duke of Normandy and the French king. William's reply to this challenge was thus understandably vigorous. Norman troops began to harry the Vexin while the duke himself invaded Maine. The war was prolonged, but before the end of 1063 William, who had captured Le Mans, had already begun to consolidate his conquest of the comté. The fortifications of Le Mans were strengthened; Mayenne was taken and sacked; and probably about this time the duke reconstructed his castles at Mont Barbet and Ambrières. By the beginning of 1064 he had made himself the effective master of Maine.70
The Norman acquisition of Maine altered the balance of power in Gaul in a manner that was substantially to affect the course of events during the next critical decade. It helped to ensure that the duke need fear no interference from northern France in any enterprise he might undertake overseas. Count Walter and his wife Biota, who after the fall of Le Mans had been taken into custody, died shortly afterwards in suspicious circumstances,71 and the comital house of the Vexin, robbed of its chief members, passed under the leadership of a collateral branch. Walter was succeeded by his cousin Ralph of Crépi, or of Valois, whose policy towards Normandy had still to be disclosed. Similarly, the successful campaigns of 1062–1063 had freed William at last from the Angevin menace. After the death of Geoffrey Martel in 1060 a struggle for the Angevin succession developed, and for the next decade no Angevin count was to be able as heretofore to use Maine as an effective base for operations against Normandy. Finally, William's position in respect of the French royal house had also been strengthened. Since 1060 the young King Philip I, himself a minor, was the ward of the count of Flanders,72 and able to exercise as yet no personal control over the conduct of affairs, and now just at the time when the Capetian monarchy was in eclipse the duke of Normandy had enhanced his prestige, and multiplied his resources by the conquest of Maine.
In 1064, therefore, Duke William might feel that his chance of eventually winning the long promised realm of England had been substantially improved, but he must also have realized that the rising fortunes of the earl of Wessex had erected a formidable obstacle across his path. The new security of the duke in Gaul had been won during the same years as the earl had so signally prospered in England, and now, as the life of Edward the Confessor ebbed to its close, the two men confronted each other across the Channel as possible rivals for the succession to the childless king. It was a situation which called for dramatic treatment, and it was soon to receive worthy commemoration in the famous stitchwork of the Bayeux Tapestry. The story there displayed was of course only one factor in the developing crisis which was overtaking northern Europe, but it was none the less an essential feature of that crisis. The connexion between England and Normandy which had been developing inexorably since 1035 was already in 1064 becoming crystallized into an individual opposition between two of the most remarkable personalities of eleventh-century Europe.
The relations of Duke William and Earl Harold, and their modification during the last two years of the Confessor's reign, thus possess a general importance as well as the personal interest which attaches to one of the most picturesque and controverted episodes of history. In 106473 (as it would seem) Earl Harold set sail from Bosham in Sussex on a mission to Europe. Almost every detail of his ensuing adventures, and their purpose, has been made the subject of controversy and no finality can be claimed for any single interpretation which may be put upon them.74 Following the three earliest accounts75 of these events which have survived, it may, however, seem reasonable to suggest that the earl of Wessex on this occasion had been commanded by the Confessor to proceed to Normandy in order formally to confirm in the presence of William the grant of the succession to the English throne which had previously been made by the king to the duke. Earl Harold, doubtless, had little liking for this task, but he may well have felt it unwise to disobey the king's order, and he may also have hoped to derive some personal advantages from its execution.76 At all events he set out, but encountering a stiff wind he was blown out of his course and compelled to make a forced landing near the mouth of the Somme on the coast of the comté of Ponthieu.77 There, ‘according to the barbarous custom of the country,’ he was seized by Guy, the reigning count, ‘as if he were a shipwrecked mariner,’ and thrown in prison within the castle of Beaurain situated some ten miles from Montreuil.78
The situation presented an immediate opportunity to Duke William which he was quick to seize. Perhaps the Norman duke was aware that Harold's journey was connected with the promise which had previously been given by Edward, and he appreciated at once the advantages he might obtain from personal contact with the earl of Wessex in circumstances highly favourable to himself. He therefore lost no time in demanding the person of Harold from Count Guy (who since 1054 could be regarded as in some sense a vassal of Normandy),79 and he perhaps agreed to pay the count some ransom. Count Guy on his part felt it prudent or profitable to accede at once to the duke's request. He brought Harold to Eu, where the duke with a troop of armed horsemen came to receive him. The earl was thereupon conducted with honour to Rouen.80 And either at Rouen,81 or perhaps at Bayeux,82 but more probably in the presence of an assembly of magnates held at Bonneville-sur-Touques,83 Earl Harold was brought to swear his famous oath of fealty to the Norman duke, with particular reference to the impending question of the English succession. William of Jumièges states baldly that the earl ‘swore fealty regarding the kingdom with many oaths’,84 whilst the Bayeux Tapestry indicates the solemn character of the transaction, and emphasizes the relics on which the oath was taken.85 William of Poitiers, however, records the terms of the undertaking which was so solemnly given.86 The earl swore to act as the duke's representative (vicarius) at the Confessor's court; he engaged himself to do everything in his power to secure the duke's succession in England after the Confessor's death; and in the meantime he promised to maintain garrisons in certain strongholds, and particularly at Dover. About the same time, and as part of these arrangements, Duke William vested the earl as his vassal with the arms of Norman knighthood, and (as it seems) also entered into an undertaking whereby the earl became pledged to marry one of the duke's daughters.
Such are the only facts given in the earliest accounts of this famous transaction, though legend was soon to add many embellishments to the story. Whether, as Eadmer suggested,87 Harold acted under duress on this occasion, or was the subject of trickery,88 must remain in doubt. Certainly, the position of the earl in Normandy was a difficult one. He may well have hesitated to defy the orders of his king, and the forceful desires of the duke who at this time was both his host and his protector. On the other hand, it is not impossible that he willingly consented to what took place, and perhaps even (as William of Malmesbury asserted) he acted here on his own initiative.89 The perils which surrounded his own designs on the English throne were very obvious, for such an attempt could only succeed if he could obtain sufficient support at home to override the rights of surviving members of the English royal house and the claims of strong Scandinavian princes. The earl may thus perhaps have thought to safeguard his future position in the event of his own failure or the duke's success. And he may have felt that in any case he could later repudiate the oath, or plead that it had been taken under compulsion.
Whatever may have been Harold's motives on this occasion, there can be no doubt that William's policy had been clearly conceived, and the newly established relationship between the two men was demonstrated before Harold returned to England,90 when the duke associated the earl with him as his vassal in a venture which was yet further to strengthen William's power on the eve of the English crisis. The acquisition of Maine which had been accomplished by 1064 had stabilized the western frontier of William's duchv. There remained, however, Brittany as a potential menace that needed to be removed. Since the death of Alan III in 1040 the nominal ruler of Brittany had been his young son, Conan II, whose interests were protected, albeit with great difficulty, by his mother Bertha, and opposed by his uncle Eudo of Penthièvre.91 The result was prolonged disturbance, and it was not until 1057 that Conan was able to assert his own authority as sole ruler. Even so, he was dependent upon the precarious allegiance of the new feudal nobility which had arisen in Brittany.92 Here was therefore another opportunity for Duke William. In 1064 Conan was engaged in operations against the Breton rebels near the stronghold of Saint-James-de-Beuvron which William had erected near the Norman border.93 Some of the rebels, notably Riwallon of Dol, thereupon appealed to the duke, who forthwith invaded Brittany with the earl of Wessex in his force. He crossed the estuary of the Couesnon with some difficulty, and advanced to the relief of Riwallon, who was besieged in Dol. The town fell to the Norman assault, and Conan retreated towards Rennes. William thereupon moved to Dinant, which he captured. He then retired, leaving Conan to take his revenge in due course upon Riwallon, who was sent into exile.94
The details of William's Breton war of 1064 are highly obscure, but the results of this inconclusive fighting were considerable. A powerful Brittany, active and hostile on his western frontier, might have been a grave menace to Duke William during 1066, and in 1064 this danger was by no means to be ignored. The declaration alleged to have been made by Conan that he would resist any expedition made by William against England may safely be regarded as apocryphal,95 but it represented what might very well have proved a disturbing factor in the development of Norman policy at this time. It had therefore been an astute move on William's part to foster opposition to Conan among the Breton magnates, and his raids into Brittany in 1064 also served to turn Conan's attention elsewhere than Normandy during the critical years that followed. In 1065 Conan is to be found at Blois seeking an alliance against Anjou, and all through 1066 he was engaged in operations in Angevin territory until he died suddenly in December of that year while besieging Château-Gonthier.96 His death was a further stroke of good fortune for William, but even more important was the fact that the duke had already created a powerful party of his own in Brittany. No less than four of the sons of Count Eudo of Penthièvre were to follow Duke William across the Channel, and together with many other Bretons to receive in due course large estates in England.
It is impossible not to admire the high competence of Duke William's policy in 1063–1064, or the manner in which it was steadfastly directed towards the eventual fulfilment of his English purpose. Full advantage had been taken of the weakness of the Capetian and Angevin dynasties; Maine had been added to his resources; and a strong faction in Brittany had become favourable to his cause. Finally, one of the most formidable of his potential opponents had publicly given an undertaking, recognized as sacrosanct by the feudal and ecclesiastical opinion of Europe, to support the duke's claim upon England, or at all events not to oppose it. During this same year, also, this same English magnate had been publicly displayed as the duke's vassal in a campaign which was indirectly to facilitate the later deployment of Norman arms in England. When before the end of 1064 Earl Harold returned to England, laden with gifts from Rouen,97 his position as a possible rival to Duke William for the English throne had, in the public opinion of Europe, become irreparably compromised.
The advantage gained over the earl of Wessex by Duke William in 1064 was soon to be increased by events within England itself. In the autumn of 1065 a rebellion occurred in Northumbria against Earl Harold's brother, Tosti, who had been earl since 1055.98The revolt rapidly spread and the rebels, having massacred many of the earl's supporters in the north, took it upon themselves to proclaim Tosti an outlaw. They then offered his earldom to Morcar, the brother of Earl Edwin of Mercia, and in order to compel the king to confirm their acts, they marched southward in force to Northampton. Earl Harold tried to arrange some compromise in favour of his brother, but he failed to do so, and King Edward was forced to recognize Morcar as earl of Northumbria. Tosti and his wife Judith fled from England to take refuge with Judith's half-brother, Count Baldwin V of Flanders.99 Earl Harold's position in England was thus seriously weakened by these events which removed his brother from an important earldom and substituted in his place, and against the wishes of the earl of Wessex, a member of the Mercian house of Leofric who had no reason to feel friendly to any representative of the family of Godwine.
The immediate advantage which thus accrued to Duke William is obvious. But there was a further implication of the Northumbrian revolt which was equally significant. It is wholly remarkable that despite the bitterness of the dispute of 1065 there seems to have been no desire on the part of the rebels to establish a separate kingdom north of the Humber.100 King Edward had been compelled to accept an earl of whom he disapproved, but his status as the sole legitimate king over all England had been preserved for his successor. England, albeit comprising two ecclesiastical provinces and three great earldoms, was evidently conceived as a single kingdom whose political identity must override the differences inherent in the individual traditions of its several parts. At the same time as the resources of the earl of Wessex had been diminished, the unity of the inheritance which Edward the Confessor would leave to his successor had been sustained and indeed re-emphasized. The prize was soon to be disputed: its integrity remained unimpaired.
The long developing relationship between Normandy and England had thus at last produced a situation which involved the medieval destiny of a large part of northern Europe. The aged childless king, revered and qualifying for sainthood, was approaching his final problems. Earl Harold of Wessex, weakened by the events of 1064–1065, was still strong, and might yet acquire sufficient support in England to hold the crown he coveted. In the northern world, Harold Hardraada, representing a very strong tradition that England's true links were with the Baltic lands, was reaching the climax of his power as king of Norway, and must soon put to the test his long expressed claim to the English throne. These were no mean opponents for the duke of Normandy who, having welded a unique province into a political unity, was now, at the beginning of 1066, brought to the crisis of his fate.
1 Above, chap. I.
2 Freeman, Norman Conquest, vol. I, pp. 197—205.
3 See the papal letter given in Will. Malms., Gesta Regum, vol. I, pp. 191–193. The letter is not wholly satisfactory in its present state, but in substance it is probably genuine (Stenton, Anglo-Saxon England, p. 371).
4 Will. Malms. refers to Bishop Roger. The bishop of Lisieux was the only Norman bishop at this date with that name, and he appears in a charter of 990 (R.A.D.N., no. 4). It is not inconceivable, however, that Will. Malms, made a mistake over the name and that Robert, archbishop of Rouen, is intended.
5 The story is only found in Will. Jum. (p. 76), and must be treated with caution. It may be noted, however, that a Viking fleet in 1000 on leaving England went to Normandy, and that in the next year the coast of England opposite Normandy was ravaged, so perhaps there is some confirmation (AS. Chron., ‘C’, s.a. 1000, 1001).
6 AS. Chron., ‘C’, s.a. 1002.
7 Ibid., s.a 1013, 1014.
8 Campbell, Encomium Emmae, vol. XLIV.
9 Above, p. 21.
10 Pfister, Robert le Pieux, pp. 214, 215.
11 Acta Sanctorum (Palmé), July, vol. VIII, p. 125.
12 AS. Chron., ‘C’, s.a. 1017.
13 R.A.D.N., no. 70.
14 Ibid., no. 69: accepted as genuine by Lot (Saint-Wandrille, no. 13) and by Round (Cal. Doc. France, no. 1422).
15 R.A.D.N., no. III; Cart. #aflles Norm., plate I.
16 Haskins, Norman Institutions, p. 261.
17 R.A.D.N., no. 85; Haskins, op. cit., plates IV and V
18 Above, pp. 73, 74.
19 Douglas, in Eng. Hist. Rev., vol. LXV (1950), pp. 292–295.
20 Will. Jum., p. 110.
21 Stenton, op. cit., p. 414.
22 AS. Chron., ‘C’, s.a. 1036; Campbell, op. cit., p. lxiii.
23 Will. Malms., Gesta Regum, vol. I, p. 240; and below, Appendix F.
24 Will. Poit., pp. 11–13.
25 AS. Chron., ‘C’, s.a. 1040.
26 Ibid., s.a. 1041.
27 Stenton, op. cit., p. 415.
28 AS. Chron., ‘C’, s.a. 1042.
29 Ibid., s.a. 1042.
30 Ibid., s.a. 1043.
31 Oleson, Witenagemot in the Reign of Edward the Confessor, Appendix B.
32 AS. Chron., ‘C’, s.a. 1043; ‘D’, s.a. 1045, 1047 (equals 1046, 1048); ‘E’, s.a. 1043 (equals 1045).
33 AS. Chron., ‘E’, loc. cit.
34 Stenton, op. cit., p. 553.
35 Round, Feudal England, pp. 317–341.
36 Edward's declaration to this effect (Harmer, Anglo-Saxon Writs, p. 16) is not above suspicion. Harold had retaken Steyning in 1065, but Fécamp possessed it in 1086 (D.B., vol. I, fol. 17). That a grant was in fact made before 1066 is indicated by two confirmations by William (Regesta, vol. I, no. 1; Chevreux et Vernier, Archives, plates VIII). The matter is further discussed by D. Matthew, Norman Monasteries and their English Possessions, pp. 38–41.
37 Will. Malms., Gesta Pontificum, pp. 201, 202; D.B., vol. I, fol. 17. He became bishop of Exeter in 1072.
38 Stenton, op. cit., p. 554.
39 Freeman, op. cit., vol. II, p. 69.
40 AS. Chron., ‘C’, s.a. 1049.
41 Ibid., ‘D’, s.a. 1052 (equals 1051); ‘E’, s.a. 1048 (equals 1051).
42 R. L. G. Ritchie, Normans in England before Edward the Confessor (Exeter, 1948).
43 B. Wilkinson, ‘Freeman and the Crisis of 1051’ (Bull. John Rylands Library, 1938).
44 AS. Chron., ‘D’, s.a. 1052 (equals 1051).
45 Ibid.
46 Stenton, op. cit., p. 553.
47 AS. Chron., ‘D’, s.a. 1052.
48 The whole matter is discussed by me in Eng. Hist. Rev., vol. LXVIII (1953), pp. 526–534; but see also Oleson, in Eng. Hist. Rev., vol. LXII (1957), pp. 221–228.
49 Above, pp. 55–69, and below, Appendix B.
50 ‘Will. Jum., pp. 132, 133.
51 AS. Chron., s.a. 1052.
52 AS. Chron., ‘E’, s.a. 1053.
53 He continued to hold Winchester with Canterbury in plurality.
54 Stenton, op. cit., p. 460.
55 Ibid., pp. 558, 559.
56 AS. Chron., ‘D’, s.a. 1053; below, Appendix F.
57 Ibid., s.a. 1055.
58 Ibid., s.a. 1057.
59 Stenton, op. cit., p. 563.
60 Ritchie, Normans in Scotland, p. 8.
61 AS. Chron., ‘D’, s.a. 1057.
62 Stenton, op. cit., p. 566.
63 Oleson, Witenagemot, p. 117.
64 Flor. Worc, vol. I, p. 224.
65 Round, Studies in Peerage and Family History, pp. 147–155.
66 Stenton, op. cit., p. 597.
67 Above, pp. 72, 73.
68 Latouche, Comté du Maine, p. 33, note 3.
69 Ibid., pp. 113–115.
70 Will. Poit., pp. 91–93; Will. Jum., pp. 130, 184; Prentout, Guillaume le Conquérant (1936), pp. 149–153.
71 Below, Appendix F.
72 Fliche, Philippe I, chap. I.
73 It is highly probable though not absolutely certain that the visit took place in that year.
74 Douglas, Eng. Hist. Rev., vol. LXVIII (1953), pp. 535–545; Oleson, ibid., vol. LXXII (1957), pp. 221–228.
75 Will. Jum., pp. 132, 133; Will. Poit., pp. 100–106; Bayeux Tapestry.
76 He may have wished to safeguard his eventual position in the event of Duke William's success, and it is very probable that members of his family were at the Norman court as hostages to safeguard the duke's succession, and Harold may have wished to obtain their release.
77 Bayeux Tapestry, plates VI, VII. There is no suggestion of a shipwreck.
78 Will. Poit., pp. 100–102; Bayeux Tapestry, plates IX, X – ‘Belrem’ (E.H.D., vol. II, p. 242).
79 Ord. Vit., vol. II, pp. 237, 238.
80 Will. Poit., p. 102.
81 Ord. Vit., vol. II, p. 217.
82 Bayeux Tapestry, plates XXVIII, XXIX – ‘Bagias’.
83 Will. Poit., p. 102.
84 Will. Jum., p. 133: ‘Facta fidelitate de regno plurimis sacramentis.’
85 Plates XXVIII, XXIX (E.H.D., vol. II, p. 251).
86 Will. Poit., pp. 104–106, 230.
87 Historia Novorum (ed. Rule), p. 8.
88 The story that the oath was taken on concealed relics is of later date. It appears in Wace, Roman de Rou (ed. Andresen), vol. II, p. 258. Plate XXIX, Bayeux Tapestry, may be studied in this connexion.
89 Gesta Regum, vol. I, p. 279.
90 Will. Poit. and the Bayeux Tapestry differ as to the sequence of events, the latter placing the Breton campaign before the taking of the oath.
91 Lobineau, Histoire de Bretagne, vol. I, pp. 93–98; La Borderie, Histoire de Bretagne, vol. III, pp. 14–23; Durtelle de Saint Sauveur, Histoire de Bretagne, vol. I, p. 118. The chaotic condition of Brittany under Bertha is well illustrated in the remarkable charter printed by Morice, Histoire de Bretagne: Preuves, vol. I, col. 393.
92 The character of that nobility in Brittany at this time is revealed in a charter of Conan (Cart. de Redon, p. 23); Morice, op. cit., Preuves, vol. I, col. 408; see also La Borderie, Les Neufs Barons de Bretagne (1905), pp. xiii, xiv.
93 Will. Poit., p. 106.
94 Will. Poit., pp. 106–112; Durtelle de Saint Sauveur, op. cit., p. 119; Bayeux Tapestry, plates XXI–XXIV (E.H.D., vol. II, pp. 248, 249).
95 Ord. Vit., interp. Will. Jum., pp. 193, 194; and below, Appendix F.
96 La Borderie, Histoire de Bretagne, vol. III, p. 20; Le Baud, Histoire de Bretagne, p. 157. He apparently captured Pouancé and Segré before his death.
97 Will. Jum., p. 133.
98 B. Wilkinson, ‘Northumbrian Separatism in 1065 and 1066’ (Bull. John Rylands Library, vol. XXXII (1939), pp. 504–526).
99 AS. Chron., ‘C’ ‘D’, s.a. 1065.
100 Wilkinson, loc. cit.