Chapter 3
The period 1047–1060 is of cardinal importance in the history of Normandy. Often dismissed as comprising merely a welter of disconnected political disturbances, it none the less possessed its own cohesion, and without doubt it entailed momentous consequences for the future. It began with the revolt which in 1047 came near to annihilating the ducal power, and it reached its second crisis when in 1052–1054 the duke was forced to withstand not only a hostile confederation formed by his own magnates, but also a coalition of French fiefs led by the king of France. During these fourteen years the duke of Normandy was almost continually at war. Until after 1054 his survival was always in some doubt, and not until after 1060 could it at last be regarded as fully assured.
The unity of this period in the development of Normandy needs therefore to be stressed. The opening battle in these campaigns – Val-ès-Dunes – is properly to be considered as marking the end of the anarchic minority of the duke, and setting the seal on his authority. Yet the critical engagement on the banks of the Orne was but one episode in the duke's struggle for survival, and it was the beginning of a long period of uninterrupted warfare. The disturbances of these years are in short to be regarded not as a series of isolated revolts, but rather as embodying a prolonged crisis which lasted in the most acute form from 1046 to 1054; and the threat to the Norman future was no less formidable at Mortemer than it had been at Val-ès-Dunes. Only after 1054 can there be discerned a relaxation of the tension, and no subsequent menace to the integrity of Normandy was ever so severe during the Conqueror's reign as that which had been continuous between 1046 and 1054. Varaville in 1057 was an engagement of only minor importance, and the deaths of the count of Anjou and the king of France in 1060 did but give final assurance that the perils which had been faced during the previous fourteen years had at last been surmounted.
The condition of Normandy, and its place in the European order, was in fact so different in 1066 from what it had been in 1046 that it would be easy to underestimate the hazards which attended this transformation. It was not merely the survival of Duke William that was at stake. The events of this period assuredly offer the strongest testimony of the indomitable purpose of the young man who was the principal actor in this embattled drama, but these related campaigns decided much more than his own personal fate. The antagonism between Upper and Lower Normandy which was partially reflected in the conflict of 1047 had for political purposes been resolved by 1060; and the same period witnessed the beginning of the long struggle between Anjou and Normandy for the dominance of north-western Gaul. Finally, a yet more fundamental change was made at this time in the relations between the duke of Normandy and the king of France – a change which has been justly described as marking nothing less than a ‘turning-point of history’.1 The ultimate consequences of these developments were, indeed, to stretch into the far future, and were not in fact to be fulfilled until the establishment of Henry, count of Anjou, as King Henry II of England. The confused, but always interconnected, events of these years, when such large issues were still in question, thus deserve careful examination, for the outcome of the crisis which they reflected, and determined, was sensibly to affect the fortunes of Normandy and Anjou, of France and England, for more than a century and a half.
On the morrow of the battle of Val-ès-Dunes, Duke William was still not secure in his duchy. Nor did that battle mark the end of the warfare of which it formed a part. King Henry himself left Normandy shortly after his victory,2 but hostilities continued without interruption. It is in fact perhaps indicative of the duke's continuing insecurity at this time that most of the surviving leaders of the revolt were to escape somewhat lightly. Nothing is known of the immediate fate of Rannulf of the Avranchin, but he was not deprived of his vicomté and he was to transmit it to his son.3 Nigel of the Cotentin was in more desperate plight, but he too was treated with considerable leniency. He was forced to go into exile in Brittany, but before long he returned; by 1054 he was re-established in his vice-comital lands, and nearly all his forfeited possessions were restored to him.4 As for Guy of Burgundy, although wounded, he managed to escape from the battlefield with a considerable force, and he promptly fortified himself in his stronghold of Brionne.5
The reduction of Brionne with the minimum of delay was thus imposed upon the duke as an imperative necessity if his authority was to be restored, and it was a considerable check to his fortunes when he failed to take it by storm. He was thus compelled to invest it, and to undertake an operation which might prove to be disastrously protracted. He certainly did not minimize either the importance or the difficulty of his task. He erected large siege-works on both banks of the Risle and, in particular, wooden towers which might make the siege closer and at the same time protect his own investing troops against sorties by the garrison.6 Even these measures, however, for long proved ineffective, and it would seem that nearly three years elapsed before the castle surrendered.7The length of this operation, indeed, deserves some emphasis. For the delay was fraught with peril for the duke, and it postponed any reimposition of his authority over a united duchy. So long as Brionne remained untaken, Duke William could never proceed with any confidence much beyond the neighbourhood where his principal enemy, the potential leader of a new rising, was established. There is in fact no record of Duke William's presence in Upper Normandy at any time during the years 1047–1049, and it is possible that during these years Rouen itself passed out of his control. Excluded from the richest part of his duchy, and with a strong castle in central Normandy holding out against him, William remained throughout the period 1047–1049 in an equivocal position. It was probably not until the beginning of 1050, after Guy had surrendered and been banished from Normandy, that William was at last able once more to re-enter his own capital.8
During these years William's authority was in fact largely circumscribed by the extent of the power he had recently regained in Lower Normandy, and the fact was to entail considerable consequences. He found himself established in precisely that part of the duchy which had come most tardily under the control of the counts of Rouen, and it is possible that he realized that here might be provided an opportunity of mitigating that dichotomy between Upper and Lower Normandy which had always tended to impair the political unity of the duchy. At all events, it is significant that the growth of Caen to a position of greater importance begins about this time.9 By the end of the first quarter of the eleventh century a cluster of villages had been formed at the juncture of the Orne and the Odon, and the importance of this site was evidently recognized in 1047 when it was chosen as the meeting-place of the council which proclaimed the Truce of God. Thenceforward William, who doubtless appreciated the strategic and commercial advantages of this position, took positive action during the ensuing years to foster the growth of an urban agglomeration at this place, providing it with walls of stone and perhaps with a castle, and making it also one of his principal residences. The regard he had for it is displayed today in the two magnificent abbey-churches which remain the glory of the town, and it is significant that at the last William was to be buried, not like his ancestors in Rouen or Falaise, but at Caen. Although not an episcopal city, Caen rose during his lifetime to be the second town in Normandy, and its early growth owes more to William the Conqueror than to any other single man. The rise of Caen during the Conqueror's reign is thus symptomatic of his eventual success in integrating Upper and Lower Normandy finally into a single political unit.
It was, likewise, during the period immediately following the battle of Val-ès-Dunes that the rise of Anjou began to introduce a new factor into Norman politics. The relations between Normandy and Anjou, which were to colour so much of the history of the twelfth century and eventually to result in the formation of a great continental empire, entered on their first critical phase between 1047 and 1052, and it was then that there began to take shape a new political grouping in north-western France which was vitally to affect the future. So wide-reaching were to be the results of this development, and so fraught was it with immediate peril to the duke, that its obscure origins in Angevin policy need here to be watched in so far as these hazardously affected the subsequent course of William's career, and modified the fortunes of the duchy over which he ruled.
Hitherto, the expansion of Anjou had been directed southward. It had been achieved mainly at the expense of the counts of Blois,10 and so successfully that in 1044, a bare three years before Val-ès-Dunes, King Henry I had been constrained to give formal recognition to Angevin dominance over Touraine.11 Henceforth Anjou, holding in Tours the key to the Loire valley, could block the Roman road of Capetian governance which ran from Paris to Orléans and onward into Poitou.12 In this way was justified the political insight of earlier counts of Anjou who had realized that their power could best be based upon a control of the Loire, and it was Geoffrey Martel, who, succeeding as count in 1040, reaped the reward of their endeavour. Brutal, unprincipled, crude, and strong, Geoffrey lacked finesse in his ambitions and statesmanship in his policies. But he had in his character much of the hammer after which he was named, and his achievements are not to be minimized. From 1044 until his death in 1060 he was always a formidable menace to the duke of Normandy, and for most of that period he was a stronger force than Duke William in the affairs of northern France.
It is not surprising that such a man finding himself now secure in the south should at once strive to push northward in his conquests. And the comté of Maine presented itself as the obvious field for his operations. For Maine at this time was in a state of great disorder. It was precisely during the earlier half of the eleventh century that a new feudal aristocracy arose in Maine. Such families as those of Mayenne, Château-Gonthier, Craon, Laval, and Vitré were being established in their possessions and in many cases there is charter evidence to show that the early eleventh-century lord was the original grantee.13 In face of this development the comital house of Maine found itself unable to exercise any effective control, and after the death of Herbert ‘Wake-Dog’ about 1035,14his successor Count Hugh IV was engaged in constant warfare with his vassals, many of whom had now fortified themselves in newly built castles.15 Among these, moreover, there was one feudal family which was to play a crucial part in the wider conflict which ensued. For at the exact point whereat Capetian, Norman and Angevin interests were to meet, there was established the house of Bellême.
Neither the French king, nor the count of Anjou, nor the young duke of Normandy, could ignore the family of Bellême, for it controlled a region that was vital to all of them – a wild stretch of hilly country on the border between Maine and Normandy which held the key to important lines of communications. At Bellême itself, six roads converged linking Maine with the Chartrain and with Normandy. Through Alençon passed the old Roman way from Le Mans to Falaise. And in the single gap in the ridge between Alençon and Domfront there ran yet another Roman road leading towards Vieux.16 It had for long been the ambition of the first house of Bellême17 to obtain control of this vital district and by 1040, that ambition had been virtually fulfilled. The family held Bellême itself nominally from the king of France, Domfront from the count of Maine, and Alençon from the duke of Normandy. But in reality it was possible for the lords of Bellême to act as virtually independent of their various overlords, and to play these off one against the other.18 Moreover, the power of the family had been yet further increased by the dominating position it had acquired in the church. Between 992 and 1055 three successive bishops of Le Mans – Siffroi, Avejot, and Gervais – were connexions of the family of Bellême,19 and in 1035 Yves, then head of the family, became bishop of Sées.20
Such was the situation in Maine when shortly after the battle of Val-ès-Dunes, the count of Anjou began to extend his activities northwards. In the extreme south of Maine, and near the Angevin border, was the fortress of Château-du-Loir whereat was established Gervais, bishop of Le Mans,21 and it was this stronghold that Geoffrey Martel now attacked.22 He failed to take the castle, which was partially burnt during the operation,23 but he had the good fortune to capture the bishop, whom he promptly threw into prison.24 A situation was clearly arising which the French king could not ignore, the more especially, when in 1050, Pope Leo IX, who had made unavailing protests against the imprisonment of Gervais,25 formally excommunicated the count of Anjou.26 Finally, on 26 March 1051, Count Hugh IV of Maine died,27 and the citizens of Le Mans immediately offered their town to Count Geoffrey. The count of Anjou at once seized his opportunity and occupied the capital of Maine.28
The crisis which ensued inexorably involved not only the French king but the Norman duke. Hugh's widow Bertha with her son Herbert and her daughter Margaret were expelled from Maine.29 At the same time, Gervais, who at last obtained his release by ceding Château-du-Loir to Count Geoffrey, repaired to the Norman court, and together with other exiles ceaselessly urged the duke to take action in Maine.30 Duke William must have realized that his own interests were now involved, and in any case he could hardly have refused assistance to his royal overlord on whose support he relied. Hostilities had in fact become inevitable and it is possible that about this time the duke joined the king in blockading the Angevin stronghold of Mouliherne near Baugé.31 Whether this engagement (which is only reported in William of Poitiers) ever took place, or whether in that case it should be referred (as is probable) to the spring of 1051 is doubtful. Soon, however, the war was to be extended up to the very border of Normandy. Once secure in Le Mans after the cession of the town in March 1051, Geoffrey moved north-eastward and occupied the fortresses of Domfront and Alençon. The Norman duchy was now directly threatened by the Angevin expansion and its duke was forced into action.32
In the late summer or early autumn of 1051,33 therefore, Duke William, with the approval of King Henry, entered the territory of Bellême to dispute with the count of Anjou the possession of the key fortresses of this region. His first advance was towards Domfront, and so immediate did the threat appear that Count Geoffrey at once hurried to the rescue.34 After considerable fighting the count was compelled to retire, and perhaps owing to a threat to Anjou by King Henry from Touraine, he subsequently left Maine.35Even so, William could not take Domfront by storm, and after constructing siege-works similar to those he had used at Brionne, he settled down to besiege it. But in the absence of his chief opponent he was not content merely with the prolonged blockade which certainly lasted for a considerable time during the months of winter.36 One night, therefore, leaving a section of his force in front of Domfront, he suddenly moved under cover of darkness to Alençon. He achieved surprise, stormed the town, and having inflicted horrible barbarities upon the defenders, he placed his own garrison within its walls. Then he moved back to Domfront, and so great was the fear excited by the atrocities he had sanctioned at Alençon that the inhabitants of Domfront determined to surrender in return for a promise of mercy and protection. Duke William thus found himself in possession of the two strongholds.37
The importance of this campaign was considerable. Alençon and Domfront were both in the future to stand in need of defence, but the duke had retained his overlordship over the one, and added the other to his dominion. The customs of Normandy thus soon came to prevail also in Domfront, whilst the surrounding district of the Passais was gradually to be absorbed in the duchy.38 A certain stabilization of this frontier had thus been achieved in the interests of Normandy, and the results were reflected in the changing position of the house of Bellême within the feudal structure of north-western France. The lords of Bellême were always to prove difficult subjects, but though the Capetian kings were ever ready to assert direct overlordship over them, the Norman dukes from this time forward were able to treat them more and more as vassals, and the new situation was forcibly expressed in one of the most important feudal alliances of this period. About this time Roger II of Montgomery, whose family had risen to power with the support of Duke Robert I, acquired for wife Mabel, daughter of William Talvas of Bellême.39 She was heiress of a large part of the Bellême lands, and her marriage was yet further to advance the fortunes of Montgomery. Henceforth the fortunes of Bellême were to be linked to those of Normandy, and with a family already closely associated with the Norman ducal house.
The control obtained by the duke over the debatable territory of Bellême was in the future to provide William with a base from which he might extend his authority westward. But this development was still remote in 1051, for Geoffrey was for long to remain the dominant force in Maine. Nor is it perhaps always realized how great a threat to the duke was latent in this war. Throughout, it would appear, he was surrounded by treachery,40 and any failure would undoubtedly have provoked a widespread revolt. In the event, however, his position was to be substantially strengthened by his success, and it is noteworthy that in this campaign William had been particularly assisted not only by Roger of Montgomery (who had a special interest in the Bellême inheritance) but also by William fitz Osbern, the son of Duke Robert's murdered steward.41 These young men were to be among the architects of the Norman conquest of England, and their presence along with others with William in front of Domfront indicates that the duke was already attracting to his support rising members of the new Norman nobility, who were now willing to stake their personal fortunes on his survival.
He was in truth almost immediately to stand in need of all the support he could inspire, for his fate was now to be perilously intertwined with a political movement which was to influence much of the future of France and England, and which after all but destroying the Norman power was to determine much of the subsequent character of Norman policy. Hitherto, the survival of the duke had depended in large measure upon the French king. During the minority Normandy had been treated almost as if it were part of the royal demesne of France. The prime feature of Val-ès-Dunes had been the action of King Henry; and in the subsequent wars the duke had discharged his feudal duty to the king. In 1052, however, the ancient connexion between the Norman dynasty and the house of Capet was disrupted. Val-ès-Dunes had been won by the king for the duke; Alençon had been stormed by William when fighting against the enemies of Henry; but when, shortly after the fall of Domfront, Duke William had to face a rebellion comparable in magnitude with that of 1047, he did so, not with Capetian assistance, but in opposition to the armed strength of the French king. There thus took place a transformation in the long-established relations between the Norman duchy and the French monarchy, and as a result a new epoch opened wherein Normandy was no longer to appear as the vassal and supporter of the Capetian monarchy, but henceforth, for a hundred and fifty years, as its most formidable opponent in Gaul.
This transformation in political filiations was, if judged by its future consequences, one of the most important events in the Norman reign of Duke William, but the manner in which it was accomplished is difficult to ascertain. It would seem, however, that the change must have been initiated by the king rather than the duke. The warfare between Henry and Geoffrey had dragged on inconclusively, and it may have become clear to the French king that he had little to gain from its continuance. At all events, arapprochement between the count and the king took place during the first half of 1052, and a definite reconciliation between them was completed before 15 August of that year when the king and the count were together in amity at the royal court at Orléans.42William on his part was bound to be concerned at this development. It would seem that he hurried to the king for he is himself to be found on 20 September 1052 in the royal presence at Vitry-aux-Loges,43 where he doubtless strove to hinder the reconciliation between Henry and Count Geoffrey. He failed; and in fact this was the last time he was ever to be present as a friend at the court of the French king. The new alliance between Henry and Geoffrey at once began to take shape, and its critical consequences for Normandy were immediately to be disclosed. Anjou and Normandy had already been brought into direct collision, and now the former protector of the duke of Normandy had passed over to the duke's most dangerous opponent in northern Gaul. It only needed a formidable rebellion to break out within the duchy, and in connexion with the new alliance, to produce one of the most acute crises of Norman history. And such a rebellion was immediately foreshadowed when in the midst of the siege of Domfront, William, count of Arques, suddenly and without excuse, deserted from the ducal army, renounced his vassalage, and departed to his own lands in eastern Normandy.44
The establishment of this man as count of Arques had been one of the features of the minority, and in 1052 he and his full brother Mauger, now archbishop of Rouen, could between them exercise an overwhelming authority over the whole of Upper Normandy. In view, therefore, of the critical situation now developing in north-western France their support might almost have seemed essential to the survival of the young duke. From the first, however, William, count of Arques, seems always to have shown himself ill-disposed towards his nephew whom with some complacency he despised as illegitimate.45 He was in fact overwhelmingly ambitious, and having failed to become duke himself, he perhaps hoped to establish himself as an independent ruler east of the Seine.46Not without reason did William of Poitiers assert that his efforts to increase his own power and to diminish that of the duke were constant and of long duration.47 His importance as an opponent may, further, be gauged not only by the extent of his possessions, which were vast,48but also by the frequency with which his attestation was sought in order to fortify private charters. He appears about this time as witness to deeds relating to Jumièges and Saint-Ouen, to Saint-Wandrille, and Holy Trinity, Rouen.49 Nor was it only from his brother, the archbishop, that he might expect to receive support in his designs. He had already married a sister of Enguerrand II, count of Ponthieu, and his son Walter might expect to succeed him.50 Here in truth was a most formidable connexion based on Upper Normandy which by itself might have seemed capable of menacing the existence of the ducal power.
The formation within Normandy of this powerful coalition in opposition to the duke was, moreover, effected at the same time as the king of France and the count of Anjou were making their own alliance, and by 1052 the two developments, which were not wholly dissociated, had combined to threaten the ducal authority with extinction. It is in fact in the light of these events that the change in the traditional relationship between the duke of Normandy and the French monarchy is best to be appraised. Duke William had at this time nothing to gain by an assertion of independence from the French monarchy such as was the delight of later Norman chroniclers to acclaim, and for a long time he seems to have been very reluctant even to recognize the breach that had been made.51 Both Norman and later English writers thus insist that he showed himself averse from engaging in any personal conflict with his overlord. King Henry, on the other hand, having made peace with the count of Anjou, appears to have hoped that his dominance over Normandy could be sustained by means of a powerful faction within the duchy which would support the French king against the duke. Only in part therefore was the significance of the ensuing war to be explained by Ordericus Vitalis when he accurately reported that Count William of Talou rebelled with the counsel of Archbishop Mauger his brother, and that together these succeeded in bringing the king of France to their aid.52 King Henry after his negotiations with Anjou had a still more positive part to play in the drama. He assumed it: and the significance of his action was perhaps felt by a monk of Holy Trinity, Rouen, who saw fit to date one of his charters by reference to the event.53 The king and the duke were now at war, and a new era had opened in the relations between France and Normandy.
Certainly, the threat to the ducal dynasty was lethal; for if the full strength of this coalition from Talou and Rouen, from Paris, Anjou, and Ponthieu, could ever have been brought at one time in unison to attack the duke, it is very doubtful whether he could have survived. As it was, the hostilities opened on Norman soil where the insurgents had acquired a great accession of strength from the great fortress which the count of Talou had recently built on the heights overlooking Arques.54 This famous castle, whose erection marked an epoch in the development of Norman military architecture, was designed to be impregnable to direct assault. It was perhaps built of stone, and further strengthened by a deep surrounding fosse which can still be seen. The erection of such a castle by a count who was notoriously ill-disposed towards his nephew must have been a source of great disquiet to the duke, and according to William of Poitiers he early placed a garrison of his own within it. Whether he was in fact able to take this step against the magnate who had certainly built the castle for his own use is a little doubtful, as must also be the assertion that the ducal garrison afterwards betrayed the fortress to the count. In any case, at the opening of hostilities the castle of Arques was in full possession of the count of Talou, and his revolt was based upon it.
Strong in his great fortress, the count of Arques could begin to make himself supreme in the surrounding countryside, and at first he seems to have encountered little opposition except in one quarter where an interesting family group decided to withstand him. In what is now the small village of Hugleville,55 some twelve miles south of Arques, there was established a certain Richard who was related to the ducal house since his mother, Papia, who had married Gulbert, advocatus of Saint-Valery, was herself a daughter of Duke Richard III.56 This man who was in due course to build the little town of Auffay,57 had erected for himself a stronghold near Saint-Aubin between Hugleville and Arques. Here he stood against the count and brought to his assistance Geoffrey of Neufmarché58and Hugh of Morimont,59 the two sons60 of that Turchetil who had about 1040 been killed while acting as guardian of the ducal court. Geoffrey of Neufmarché had, moreover, previously married one of Richard's Ill's daughters,61 and so it was a close connexion of related magnates, all of whom had in different ways become associated with the ducal dynasty, that now determined to resist the count of Arques. At first their opposition was ineffective, and Hugh of Morimont with certain of his followers was slain after an encounter with the men from Arques at Esclavelles.62 The family connexion of Saint-Valery and Auffay was soon to play a great part in Norman history, but for the moment their opposition to the count of Arques was checked. The count could, in fact, now boast that he had on his side ‘almost all the inhabitants of the county of Talou’.63
The news of the count's overt rebellion came to Duke William when he was in the Cotentin.64 He may already have unsuccessfully summoned the count to appear before him, or it may have been a report of the betrayal of the castle which first reached him.65 At all events he acted with speed. He set out at once with a very few followers, and riding rapidly eastward, was joined on his way by a small body of men from Rouen who had already without success striven to prevent the provisioning of Arques. On his arrival at the castle he engaged in a skirmish with some of the count's men outside the walls and drove them back within the fortress. Then realizing that the castle was not to be taken by storm, he determined to besiege it, and following his earlier practice, he erected a large wooden tower by means of which he might threaten the defenders from outside. Having done this, he left Walter Giffard to conduct the siege and himself retired in order to prepare to meet any relieving force that might come from outside to the assistance of the beleaguered garrison.66
The duke's prime purpose was to invest the castle of Arques before the count of Talou could be joined by his allies from outside Normandy. In this he was so far successful that when in the autumn of 105367 King Henry in company of Count Enguerrand of Ponthieu entered Normandy by way of the Scie valley, he found Arques already cut off, and an opposing force lying between him and the stronghold.68 It therefore became the king's major objective to bring reinforcements and provisions to the garrison. Duke William, whose chief hope was to prevent such assistance reaching Arques, still seems to have been reluctant to oppose his overlord in person, but on 25 October69 some of his followers succeeded in ambushing a part of the French force near Saint-Aubin and in cutting it to pieces. The casualties were heavy, and Enguerrand himself was among the slain.70
This engagement which brought consternation to the defenders of Arques was a decided reverse for the enemies of the duke, and it is therefore little wonder that his panegyrists were concerned to emphasize its importance.71 King Henry was able to give some help both in men and material to the garrison, but he was compelled soon to withdraw. Thereafter the castle could be starved into surrender, and late in 1053 it yielded on the sole condition that the lives of the garrison should be spared. The fortress of Arques, so essential to the governance of Upper Normandy, thus passed into the hands of the Norman duke. Count William of Arques was granted terms of surprising leniency, but he was constrained to leave the duchy, and he fled to the court of Eustace, count of Boulogne. He troubled Normandy no more.72
Whilst the fall of Arques was of the first importance, the capture of the fortress did not itself decide the campaign. Already Count Geoffrey of Anjou was making his preparations, and on the western borders of Normandy there were men who were ready to support him. Thus even while Arques was being besieged, many magnates elsewhere in Normandy rebelled, and the men of Moulins which lay near the territory of Bellême actually gave up their town to Guy-William of Aquitaine, the brother-in-law and ally of the count of Anjou. The king of France in his turn was only delayed for a very short while by the defeat of his men on the Scie. It is not certain how long the reduction of Arques took after the action at Saint-Aubin on 25 October 1053, but several weeks were probably occupied in this task, and Duke William can hardly have been master of the fortress much, if at all, before December. Yet before the beginning of February 1054 the coalition arrayed against the duke was ready for joint action, and a large double invasion of Normandy then took place.73
The offensive planned by the French king was organized on a wide scale. The main body of the invaders assembled under the king at Mantes, and entered the comté of Évreux which was given over to pillage. In this body were troops drawn from all over north-western France, from Berri, for example, from Sens, Blois, and Tourraine and it would seem that there were also men from Anjou perhaps with Count Geoffrey at their head. The other body of invaders was recruited chiefly from the north-eastern fiefs of the French king and placed under the leadership of Odo, his brother, together with Rainald, count of Clermont, and Guy, count of Ponthieu, who was doubtless eager to avenge the death of his own brother who had fallen the previous year outside the walls of Arques. This force entered eastern Normandy and immediately began a widespread devastation. It was in all a formidable attack. Full allowance must be made for the exaggerations of later Norman chroniclers who were naturally prone to magnify the threat to the duchy. But their remarks receive some independent confirmation, and it would seem that a very considerable part of the feudal strength of the Capetian monarchy had been mobilized against Normandy.74
It was at this moment of peril that Duke William's previous capture of Arques saved him from destruction for he was able to collect a defending force from all over his duchy without the menace of a hostile fortress within his own borders. Wace gives a long list of the magnates who rallied to his support,75 and more reliable writers indicate that there was a wide response to his summons.76 This was in itself a notable achievement. For whilst it is very significant that many of the men particularly noted by William of Poitiers as having taken part in the campaign seems to have been as much concerned to preserve their own estates as to defend the duchy,77 none the less their interests had clearly become identified with those of their ruler. William's force was at any rate large enough for it to be divided into two contingents operating respectively to the west and east of the Seine. The duke himself with men from middle Normandy faced the invaders who were advancing under the French king through the Évreçin.78 On the other side of the river, Robert, count of Eu, with Hugh of Gournay, Walter Giffard, Roger of Mortemer, and the young William of Warenne, came out from their own lands to withstand the eastern incursion under Count Odo and Count Rainald.79
The French force under Odo and the counts seems to have been unprepared for this levy from eastern Normandy. Having entered the duchy by way of Neufchâtel-en-Bray, it advanced to the neighbourhood of Mortemer, and there gave itself up to unrestrained rape and pillage. Widely scattered and demoralized, it thus offered itself as an easy target for attack, and when the troops of the count of Eu rapidly advanced they achieved surprise, and fought with an initial advantage which was ultimately to prove decisive. The engagement lasted many hours and was fiercely contested, but no discipline could apparently be imposed on the French force, and the slaughter was considerable. It is unfortunate that no detailed account which merits credence has survived of this battle which was to be of decisive importance, but of the issue there could be no doubt. Odo and Rainald escaped with difficulty; Guy, count of Ponthieu, was captured; and their force was dispersed. So complete was the Norman victory that when the news of the battle reached the two opposing forces on the other side of the Seine, the king of France decided to withdraw. The duke of Normandy had been saved.80
The battle of Mortemer reflected a major crisis in Norman history, and never again was Duke William to be faced by so formidable a threat to the continued existence of his power. Within Normandy the results were immediately apparent as the strong coalition which had been arrayed against him began rapidly to break up. The king of France and the count of Anjou had departed, and William, count of Arques, who had already gone into exile, lost all hope of return. His comté was forfeit; his son, Walter, disinherited; and for the remainder of Norman history Talou was never to possess a count of its own, but was directly subject to the duke at Rouen. Even more noteworthy were the consequences to the archbishop of Rouen. Very soon after Mortemer, perhaps in 1055,81 but more probably in 1054,82 an ecclesiastical council, meeting at Lisieux in the presence of Hugh, bishop of that see, together with other bishops of the province and under the presidency of Ermenfrid, bishop of Sitten, the papal legate, solemnly deposed Mauger and appointed a reforming archbishop of a new type to succeed him.83 Duke William's power had been firmly re-established in Upper Normandy, and his victory might even seem to have had the blessing of the church.
It is therefore reasonable to conclude that at Mortemer a turning-point in the Norman reign of Duke William had been reached and passed, and it is interesting to consider how far he had himself been personally responsible for the achievement. Before 1046 he can hardly have had much influence on the conduct of affairs, and he clearly played a subordinate part in the campaign which culminated at Val-ès-Dunes. But his energy at that time was remarkable and his influence was not negligible, whilst in the succeeding years he moved rapidly into a position of greater prominence. His personal prowess during the warfare in Maine was noted with admiration,84 and he now began to show for the first time those qualities of efficient leadership which were later so signally to impress his generation. It is not known whether the planned methods of siege-work which he employed successively at Brionne, Domfront, and Arques were of Norman origin, but certainly the young duke used them with great effect, and his night march to Alençon, by which he achieved surprise, was a remarkable tactical feat. During the war in Maine he also exhibited for the first time that calculated combination of ferocity and leniency which was later to mark so much of his career. The horrible savagery with which he treated those who resisted him at Alençon was used as an example to the defenders of Domfront, who were offered pardon and protection in return for surrender. It was a device which he continually repeated and with notable results, both in Normandy and England. His successes in this warfare certainly caused him to be surrounded at this time with an ever-increasing regard, and the widespread support he received during the crisis of 1052–1054 was undoubtedly due to the prestige he had acquired by his character and through his acts.
He now reaped the reward of his endeavours. But if his position in Normandy after 1054 was stronger than it had ever been before, there was still to be much fighting before there could be formal peace between him and the French king. Yet the duke was able to take the offensive, and it was probably at this time that William fitz Osbern was charged with the fortification of Breteuil over against Tillières.85 Meanwhile negotiations between the duke and the king proceeded. The details of these have been lost, and even the date and terms of the ensuing agreement are uncertain. It would seem, however, that by the end of 1055 Duke William had become formally reconciled to his overlord, and on terms which were not disadvantageous to himself. The French king, it would appear, obtained the release of some of his vassals who had been captured at Mortemer, and in return he is said to have confirmed William in possession of all the lands which the duke had taken from Count Geoffrey.86
The count of Anjou, however, could hardly be expected to acquiesce in these arrangements, and it is probably to the years 1054 and 105587 that should be referred a renewal of hostilities between Normandy and Anjou. These were concentrated once again in the border country around Domfront, where Duke William had garrisoned the strongholds of Mont Barbet and Ambrières.88 Once again, also, a great border family became involved in these disputes. In 1054 Mayenne, which is some seven miles from Ambrières, was held by a certain Geoffrey, the son, as it seems, of Haimo ‘de Medano’, who was established as early as 1014.89 This Geoffrey of Mayenne held land not only in Maine but also in the diocese of Chartres, and he was later found witnessing charters in favour both of Marmoutier and Le Mont-Saint-Michel.90 His position was in fact (on a smaller scale) analogous to that occupied by the lords of Bellême, and it was in his interests likewise to prevent too close a definition as to who were his immediate overlords: the counts of Maine or Anjou, the king of France, or the duke of Normandy. Faced by the Norman advance, he now appealed to Geoffrey of Anjou, and the count immediately responded. He called to his assistance Guy-William of Aquitaine and Count Eudo of Brittany, and together with Geoffrey of Mayenne they moved against Ambrières. Duke William thereupon came to rescue the stronghold, and he compelled the besiegers to withdraw. Geoffrey of Mayenne was himself captured, and carried off to Normandy, where he was compelled to do homage. Thus the vassalage of another of the great border families was passing to Normandy.91
It would be easy, none the less, to overemphasize the duke's success on this occasion. Whatever he may have achieved at Ambrières, his exploits at this time do not seem sensibly to have affected the Angevin dominance over Maine. In August 1055, when Bishop Gervais was translated to Rheims, the count of Anjou was able without difficulty to secure the appointment to the vacant see of Le Mans of his own nominee Vougrin, who had formerly been abbot of Saint Sergius at Angers,92 and throughout these years there can be little doubt that Geoffrey Martel continued to be master of Maine. As such, he remained a perpetual menace to the Norman duke, and the natural centre of any coalition against Normandy which might be formed. Thus it was that King Henry, anxious to avenge the defeat of Mortemer, again turned to the count of Anjou, and charters indicate that he was associated very closely with him during the early months of 1057.93 The same alliance which had dominated the crisis of 1054 thus once more came into being, and immediately precipitated a new attack on Normandy. It was in August 1057 that there thus occurred a combined invasion of the duchy by the king of France and the count of Anjou.94
On this occasion the king and the count entered Normandy by way of the Hiémois with the object of laying waste the whole of that district, and pushing their destruction northward towards Bayeux and Caen. Duke William, however, still seems to have been reluctant to oppose his French overlord in person, and he contented himself with massing a considerable force in the neighbourhood of Falaise, where by means of spies he kept himself informed of the movements of the enemy. In due course his opportunity came. Glutted with pillage, the invaders reached the Dives at the ford near Varaville, and proceeded to cross the river. When part of the force was across, the incoming tide made it impossible for the remainder to follow. William thereupon launched a savage attack on those who had not effected the crossing, and according to his panegyrists inflicted something like a massacre upon them. So heavy were the losses sustained by the French that, as we understand from the Norman writers, the king of France felt he had no option but to beat a hasty retreat. Never again was he to invade Normandy at the head of a hostile army.95
The so-called battle of Varaville does not possess the same critical significance in the history of Normandy as does that of Mortemer, and its importance has been overestimated. Unlike Mortemer, it is scarcely noticed by the annalists who record the earlier battle. William of Poitiers and William of Jumièges are almost the only early writers to describe it; William of Malmesbury very briefly repeats their story; but most of the chroniclers, including Ordericus himself, pass by this campaign with little or no notice;96 and it is not until the time of Wace that the full Norman tradition of a great victory is developed.97 It might therefore be wise to suspect some exaggeration here in a Norman tradition which is so little corroborated. Nevertheless, in the years 1057–1060 there seems to have been a significant extension of the Norman influence in Maine. It will be recalled that after the death of Count Hugh of Maine in 1051, Herbert his son had been driven into exile by Geoffrey Martel. He now turned with some confidence towards the Norman duke for support against his Angevin supplanter. William, on his side, was quick to see the advantage that might be gained by promoting Herbert's claims under his own direction. As a result, and some time after 1055, a notable pact was made between them. By it, Herbert promised to marry a daughter of the Norman duke, and engaged his sister Margaret to marry William's son, Robert. Moreover, it was agreed that in the event of Herbert's dying without children the comté of Maine should pass to the duke of Normandy.98From this time forward Herbert could be regarded as William's protégé, if not his vassal, and it was in this capacity that during these years he began to recover some authority in Maine.99 Duke William, on his part, might see opening up before him the distant prospect of adding Maine eventually to his own dominion.
In the meantime, however, it was with King Henry that the Norman duke was most closely engaged. After Varaville he could take the initiative, and soon he is to be found on the south-western frontier of his duchy. Here there was a district which was to provide the occasion for long disputes between the dukes of Normandy and the kings of France. In Carolingian times the whole region between the Andelle and the Oise had formed a single pagus – the Vexin – but after the establishment of the Viking dynasty, the northern part of this had been made an effective part of Normandy, whilst the section south of the Epte containing the towns of Mantes and Pontoise passed under the control of local counts.100 In the time of William's father, one of these, named Dreux, came under the overlordship of Duke Robert, and after Dreux's death his eldest son, Walter, continued in this vassalage.101 The French Vexin was, however, always liable to cause disturbance, and its counts were in fact to play an important part in the relations between Normandy, France, and England during the central decades of the eleventh century. For Dreux had married a sister of Edward the Confessor, and his son, Ralph, was to have a notable career in England,102 whilst Walter, on his part, by subsequently claiming thecomté of Maine, was to precipitate one of the most important of William's continental wars.103
In 1058, however, the Norman overlordship in the French Vexin was not immediately threatened, and it was slightly to the west of this district – that is to say on the northern borders of the Chartrain – that William now began hostilities against King Henry. According to one account it was about this time that he regained Tillières which he had lost during the minority,104 and certainly he now succeeded in capturing the French king's stronghold of Thimert, some twelve miles from Dreux.105 Between 29 June and 15 August 1058106 King Henry therefore came to lay siege to this fortress, and there opened the last episode in the war between him and Duke William. The siege of Thimert was in fact to last almost as long as the siege of Brionne. Several royal charters are dated by reference to it in 1058 and 1059, and it was still continuing when Philip I was consecrated on 23 May of the latter year.107 The contest was, however, losing its significance, and there were already tentative negotiations for a truce. According to a Norman chronicler, the bishops of Paris and Amiens were sent to Normandy by Henry on an embassy designed to bring about a peace,108 and the fact that about this time William and many of his magnates are to be found near Dreux109 suggests that a personal interview may have taken place between the king and the duke. None the less the war dragged on. It is unlikely that the siege of Thimert was over in August 1060, and it is probable that King Henry died before peace was concluded with Duke William.110
The negotiations of Duke William with Herbert of Maine and with the French king, together with the desultory fighting which accompanied them, form a fitting epilogue to the long period of almost uninterrupted warfare which occupied the Norman duke from 1047 to 1060. During these years the duke of Normandy was engaged in a continuous struggle for survival, and it falsifies the character of this prolonged crisis in the Norman fortunes to treat its various episodes in isolation. The siege of Brionne began within a few weeks of Val-ès-Dunes, and Brionne was scarcely taken before the opening of the war in Maine. The campaign round Domfront and Alençon was in turn directly linked to the revolt of Count William of Arques, and this merged into the invasion of Normandy by King Henry which was repelled at Mortemer. Thus while it is true that the battle of Val-ès-Dunes marks the beginning of the effective reign of Duke William, it is equally true that his authority was never undisputed or secure during the seven years which followed the victory on the banks of the Orne, and during the last two years of that period, the revolutionary change which turned the king of France from a friend into an enemy provided a new and most formidable element of danger. In retrospect it may even appear a matter of wonder that the ducal power was enabled in this period to survive the onslaught which was directed against it from so many quarters, and its preservation reflects authentically not only the strong tradition of ducal authority which was never allowed to disappear but also the indomitable character of the future conqueror of England.
Between 1054 and 1060 the tension was relaxed, but it remained acute. The joint invasion of the duchy by King Henry and Count Geoffrey in 1057 gave some indication of the danger which was still latent. Even after Varaville, Maine continued to be an Angevin rather than a Norman dependency, and until his death Geoffrey Martel was a stronger force in north-western Gaul than was the young Duke William. On his other frontier, it is true, the duke was now able, albeit with some reluctance, to take the offensive against his overlord, though the Capetian monarchy had large reserves of strength which might again be turned against Normandy. But the tide had evidently turned in favour of the duke before two events completed the deliverance which had been so hardly earned. On 4 August 1060 King Henry I died, leaving the French realm in the hands of his son Philip, who passed under the guardianship of William's father-in-law, Baldwin V, count of Flanders.111 And on 14 November 1060 there occurred the death of Geoffrey Martel, which removed William's greatest rival in the west, and plunged both Anjou and Maine into a civil war in which the Norman duke might take his profit.112 Thus the stage was cleared for a new act in the personal drama of Duke William, and there opened for the architect of Norman greatness a new era of Norman opportunity.
Certainly his success in surmounting the difficulties which faced him between 1046 and 1060 had been due in large measure to his own personality. Consequently it is noteworthy that the politics of these years, and their hazards, were throughout this period further complicated by a series of events which were in a special sense personal to himself. Before 1049, that is to say shortly after the battle of Val-ès-Dunes, plans had been made for a marriage between the duke and Matilda, daughter of Baldwin V, count of Flanders, by Adela, daughter of Robert II, king of France.113 The projected marriage was, however, forbidden by Leo IX at the council of Rheims in October 1049, and though, in the report of the council, no specific reason for the prohibition is given, it is generally assumed that the ground for the objection was that William and Matilda were within the prohibited degrees of relationship.114 The marriage none the less took place. Perhaps in 1050, probably in 1051, and at all events not later than 1052,115 Baldwin V brought his daughter to Eu, where the marriage was celebrated, and the duke forthwith conducted his bride with fitting pomp to Rouen. Not until 1059, however, was the papal sanction to the union obtained from Pope Nicholas II at the second Lateran Council.116
The marriage of Duke William, and the circumstances in which it took place, were sensibly to affect the position of the duke in his duchy, and indeed the place he was to occupy in the political structure of western Europe. And no event in his career has given rise to more controversial discussion.117 Much speculation has, for instance, taken place as to the ecclesiastical objections to the match, and the nature of the consanguinity (if such existed) between William and Matilda. At one time it was held that when William sought her hand, Matilda was already the wife of a certain Gerbod, by whom she had a daughter, Gundrada, who later became the wife of William of Warenne, the first earl of Surrey. This, however, has now been finally disproved, and it is in the highest degree improbable that Matilda was married to anyone before the Conqueror. Some other explanation has therefore to be sought for the ecclesiastical ban on the marriage. It has thus been suggested that both William and Matilda were cousins in the fifth degree, being both directly descended from Rolf the Viking. It has also been suggested that the ground for the probition was a marriage, which is alleged to have been contracted (though it was certainly not consummated) between Duke Richard III of Normandy and Adela (Matilda's mother) before the latter was married to Baldwin V. Finally it has been suggested (perhaps with greater probability) that the prohibition was based on the fact that after the death of Baldwin V's mother, Ogiva, his father, Baldwin IV, had married a daughter of Duke Richard II of Normandy. All these theories have difficulties to overcome, and the matter may well therefore be left in some suspense.
It is thus much more profitable to consider William's marriage, the opposition it excited, and the political consequences it entailed, in relation to the politics of the age. Viewed against the background of the duke's war for survival, it is not difficult to conjecture the motives which impelled him to seek the match. After Val-ès-Dunes he was still strictly dependent upon the loyalty of a few trusted magnates, and on the support of his overlord the king of France. Consequently, it is noteworthy that the duke was, as it seems, urged to the marriage by his followers, and it is equally significant that Matilda was the niece of the French king.118 The marriage must furthermore have in itself seemed highly advantageous to a young man labouring under the stigma of bastardy, and as yet only partially in possession of his inheritance. Moreover, the rising power of Flanders under Baldwin V was involved,119 and might well have appeared to offer to the Norman duke the prospect of a useful alliance. In any circumstances, therefore, such a marriage might be expected to increase the influence of Normandy in Gaul, and its results were to be more profitable than even the astute duke could have foreseen in 1049. Much was to happen in the ensuing decade, and William was, in particular, to lose the support of his royal overlord. But after the deaths of Count Geoffrey and King Henry I in 1060 when Baldwin V had become guardian to Philip I, the earlier marriage between William and Matilda of Flanders thus began in some measure to condition the pattern of power in north-western Europe that formed the essential background to the Norman conquest of England.
It is, perhaps, harder to understand why the count of Flanders should have welcomed the project, but such was apparently the case,120 and here too the explanation may be sought in the political situation then prevailing in Europe. Baldwin V was already engaged in turning Flemish policy in the direction of France and away from the empire.121 His own marriage to Adela the daughter of the French king had been of great moment to him,122 and it was to be the corner-stone of Franco-Flemish relations for the ensuing forty years.123 Moreover, in 1049, his affairs were approaching a crisis, for in that year both he and his ally, Duke Godfrey of Upper Lorraine, were being hard pressed by Emperor Henry III,124 and the repercussions of this struggle were felt even in England, where Edward the Confessor collected a fleet to serve if necessary against the count of Flanders.125 Finally, the pope – Leo IX – was also involved, being still himself committed to the imperial cause.126 The two opposing interests at the council of Rheims were thus neatly defined. Baldwin V might feel constrained to receive favourably the project of marriage between his daughter and a loyal vassal of the French king who had recently been rescued by his overlord at Val-ès-Dunes. The pope, by contrast, must have viewed with some alarm the prospect of the confederation against the emperor being so substantially strengthened.
The prohibition thus conformed broadly to the grouping of political forces in western Europe in 1049, and in the event the marriage was postponed. When, about 1052–1053 it did take place, the situation had changed in so far as Duke William and the French king were no longer allies but enemies. But the war in Germany still dragged on, and Baldwin V cannot have wished to renounce the alliance, the more especially as he had become more directly involved at this time in the turbulent politics of England. He had evidently not forgotten the action of the Confessor in 1049. In or before 1051, he had given his half-sister Judith in marriage to Tosti, son of Earl Godwine, when the latter was still one of Edward's opponents, and in 1052 he sponsored the armed return of Godwine to England in the king's despite.127 In 1053, therefore, the count of Flanders was still himself in need of allies, so that he had little reason to do other than welcome the marriage of William and Matilda. And the marriage itself was notably to affect the course not only of Flemish but also of French and Anglo-Norman history.
The general political implications of this important marriage were, however, only slowly to be disclosed, and at the beginning of the project, the ecclesiastical opposition to the match must have sensibly increased the difficulties of Duke William within his own duchy. It is significant that Norman writers of the period seem to have been very reluctant to discuss the ban, or the reasons for its imposition, and their silence was later to be broken only by legends which, however picturesque, are historically valueless.128 The matter was evidently one of great delicacy. A strong tradition supports the view that the marriage was, for whatever reason, a matter of deep concern to the Norman church, and it has even been suggested that the very strong contingent of Norman bishops at the council of Rheims in 1049 had been sent there by the duke in the hope that they might prevent the ban being promulgated. After the prohibition had been pronounced it would seem that an influential party in the province of Rouen was actively hostile to the marriage,129 and it was later asserted that Archbishop Mauger's denunciation of the match was one of the causes of his eventual banishment.130 Disloyalty could certainly, in this matter, cloak itself in the garments of righteousness, and jealous enemies of the duke might find themselves in company with ecclesiastical reformers. According to writers from Le Bec, Normandy was actually placed under an interdict at some time during this period.131
The situation was certainly dangerous for a young ruler whose own position was precarious, and whose own title was weakened by illegitimacy. It is not surprising therefore that after the improvement in his fortunes in the years following Mortemer, the duke took active steps to try to effect a reconciliation with the papacy, and it is further significant that the long controversy about the marriage which then took place should have played its part in developing the personal relationship between Duke William and Lanfranc, then prior of Le Bec, whose later co-operation was to be the basis of the ecclesiastical policy of the Anglo-Norman kingdom. Since, however, the duke was never at any time prepared to recognize any bar to his marriage, negotiations inevitably moved slowly. Not until 1059 did the reconciliation eventually take place, and then apparently only on conditions. Pope Nicholas II removed the ban but (as is alleged) only in return for a promise that the duke and his wife should each build and endow a monastic house at Caen.132Certainly, these two magnificent churches can be regarded as symbolizing an important feature of the growth of eleventh-century Normandy. A close interconnexion between temporal and ecclesiastical authority within the duchy was to prove one of the chief sources of Duke William's Norman power, and the long controversy over his marriage was mainly important in so far as it impeded or retarded the realization of that harmony, which today at Caen is still so strikingly represented in stone.
William's marriage, and his eventual reconciliation with the papacy which had taken place by 1060, form a fitting counterpart to the duke's war for survival and its successful conclusion. After 1054, and more particularly between 1060 and 1066, the duke strengthened his duchy so that he could undertake a successful foreign invasion. It must, however, be emphasized that only twelve years and seven months separated Mortemer from Hastings, and less than seven years elapsed between the death of King Henry and William's own coronation at Westminster. The interval was too short for even the ablest ruler to attain such power had he not been able to rely upon earlier institutions of government which had escaped destruction during the period of confusion, and had he not been able, also, to link to his purpose the social and ecclesiastical movements which in his time were sweeping with exceptional force through an exceptional province. In particular, his political power was to derive from the rise of a remarkable secular aristocracy, from a notable revival in the Norman church, and from the duke's control and co-ordination of both these movements. These developments have now therefore to be considered, for much of the future history both of Normandy and England was to depend upon them. The culminating achievement of William the Conqueror was to be securely based upon the developing strength of Normandy during the decades immediately preceding the Norman conquest of England, and to the position then attained by the duke in his duchy.
1 F. Lot, Fidèles ou Vassaux?, p. 198: ‘un tournant d'histoire’.
2 Will. Jum., p. 123.
3 Below, pp. 92, 93.
4 Delisle, Saint Sauveur, pp. 20, 21.
5 Will. Poit., p. 18; Ord. Vit., vol. III, p. 232.
6 Will. Poit., p. 21.
7 Ord. Vit., vol. III, p. 342; vol. IV, p. 335.
8 Will. Poit., p. 18. He says that after William re-entered Rouen, he punished those who had recently revolted against him in the city.
9 De Bouard, Guillaume le Conquérant, pp. 58–61.
10 Norgate, England under Angevin Kings, vol. I, pp. 143–185.
11 Halphen, Comté d'Anjou (1906), p. 48.
12 Powicke, Loss of Normandy (1913), vol. II, p. 14.
13 B. de Broussillon, Maison de Craon (1893), vol. I, pp. 18 et sqq.; Maison de Laval (1895), chap. I; R. Latouche, Comté du Maine (1910), pp. 60–62, 116–127.
14 Latouche, op. cit., pp. 22–31.
15 Ibid., pp. 60–62.
16 J. Boussard, in Mélanges – Halphen (1951), pp. 43, 44.
17 On this family, see G. H. White, R. Hist. Soc., Transactions, series 4, vol. XXXI (1940), pp. 67–68.
18 Lemarignier, Hommage en Marche (1945), pp. 65, 66.
19 Latouche, op. cit., pp. 132–136. Siffroi (d. c. 1000) was brother-in-law to Yves of Bellême (d. after 1005), and Avejot (d. c. 1032) was the son of that Yves. Gervais, bishop of Le Mans (1038–1055), was nephew of William of Bellême who died in 1027.
20 Gall. Christ., vol. XI, cols. 680–682.
21 The original grantee was Haimo, the father of Gervais: he was already established there by 1007 (Latouche, op. cit., p. 62).
22 Ibid., pp. 27–28.
23 Cart. Château-du-Loir (ed. E. Vallé (1906)), no. 27.
24 Actus pontificum Cenommanis (Rec. Hist. Franc., vol. XI, p. 136).
25 Already at the opening of the council of Rheims (3–6 October 1049) Leo IX was concerned with the imprisonment of Gervais.
26 Act. pont. Cenomm. (Rec. Hist. Franc., vol. XI, p. 138).
27 Necrologie de la Cathédrale du Mans (ed. Busson and Ledru), p. 72 ; Latouche, op. cit., p. 29; Halphen, Comté d'Anjou, p. 73.
28 Act. pont. Cenomm. (Rec. Hist. Franc., vol. XI, p. 138).
29 Latouche, op. cit., p. 29.
30 Act. pont. Cenomm., loc. cit.
31 Everything connected with the episode at Mouliherne, including its date, is obscure. It may have taken place in 1048 (Halphen, op. cit., p. 72), though in that case it must have been before October when the king was at Carignan, near Sedan. The action could, however, be better referred to the spring of 1051 (Prentout, Due de Normandie, p. 146). There is, however, nothing in Will. Poit. (the sole authority) to connect it with the operations round Domfront.
32 Will. Poit., pp. 23, 34–36; Will. Jum., pp. 124–126.
33 The traditional date of 1048 for this campaign is only to be rejected with hesitation. Accepted by Halphen (op. cit.) when that notable scholar was a very young man, his conclusion has been very generally followed. Henri Prentout (Duc de Normandie, pp. 140–143) has, however, shown good reason why it is unacceptable, and in this he has been followed by de Bouard, though without further comment (op. cit., p. 41). Some of the evidence is set out below (Appendix B), and I have few doubts in assigning the operation round Domfront and Alençon to the autumn of 1051 and the early spring of 1052.
34 Will. Poit., p. 40.
35 Ibid., p. 42.
36 Ibid., p. 38.
37 Will. Jum., p. 126.
38 Lemarignier, Hommage en Marche, pp. 19, 20, 35.
39 The Complete Peerage (vol. XI, p. 686) places the marriage ‘probably between 1050 and 1054’. The inference taken from the date of the birth of Roger's son, Robert of Bellême (ibid., p. 689, note j), is valueless, since the date of Robert's birth is in fact unknown, and since it is uncertain whether any or all of the daughters of Roger II of Montgomery by Mabel of Bellême were born before Robert, who was himself the second son of that marriage. The date suggested by Jean Marx (Will. Jum., p. 125) of ‘1048–1049’ is evidently based on the assumption that the Domfront campaign took place in those years.
40 Will. Poit., p. 36.
41 Ibid., p. 38.
42 Charter printed in Rec. Hist. Franc., vol. XI, p. 590; cf. Soehnée, Actes d'Henri I, no. 91.
43 Charter printed in Rec. Hist. Franc., vol. XI, p. 588. For date, see Soehnée, op. cit., no. 92.
44 Will. Poit., p. 52.
45 Ord. Vit., vol. III, p. 232.
46 Will. Poit., p. 52.
47 Ibid.
48 He had lands as far west as the forest of Brotonne (R.A.D.N., no. 100).
49 R.A.D.N., nos. 113 (Jumièges), 103 (Sigy); Lot, Saint-Wandrille, nos. 22, 29 (R.A.D.N., nos. 108, 125); Chevreux et Vernier, Archives de Normandie, plate IV; Cart. S. Trin. Roth., p. 447, no. 50.
50 Lot, Saint-Wandrille, no. 15.
51 Will. Poit., p. 26; Will. Malms., p. 289.
52 Ord. Vit., vol. I, p. 184.
53 Cart. S. Trin. Roth., no. VII (R.A.D.N., no. 130).
54 Will. Jum., p. 119; Will. Poit., p. 54; A. Deville, Château d'Arques.
55 Seine-Maritime; cant. Longueville.
56 Ord. Vit., vol. III, pp. 41–42 and 483–484. Robert de Torigni (ed. Delisle), vol. I, pp. 33–34.
57 Ord. Vit., vol. III, p. 42.
58 Seine-Maritime; cant. Gournai.
59 Morimont is a hamlet of Esclavelles near Neufchâtel.
60 Ord. Vit., vol. III, pp. 42, 43; R.A.D.N., no. 104.
61 The full brother of this girl was Gulbert of Auffay, who was present at the battle of Hastings, founded the abbey of Auffay in 1079, and died 14 or 15 August 1087.
62 Ord. Vit. (vol. III, p. 45) places this fighting later – at the time of the action at Saint-Aubin in October 1053.
63 Ord. Vit., vol. III, p. 42.
64 Will. Poit., p. 54.
65 Will. Jum., p. 119.
66 The sequence of events here followed is that given by Will. Poit., which is not wholly consistent with that supplied by Will. Jum. or by Wace (Roman de Rou, vol. II, pp. 167 et sqq.).
67 Below, Appendix B.
68 Will. Poit., pp. 58–60.
69 The date is fixed by the obituary of Enguerrand (C. Brunel, Actes des Comtes de Ponthieu, p. iv).
70 Will. Poit., p. 58; Will. Jum., p. 120.
71 Will. Poit., p. 50. According to a later chronicler (Rec. Hist. Franc., vol. XI, p. 330) the dying cries of Enguerrand were heard by his sister on the walls of Arques. It is a good story which might even be true, for less than a mile separates the castle from Saint-Aubin-sur-Scie, and the exact site of the engagement is unknown.
72 Will. Poit., p. 60; Will. Jum., p. 120.
73 Will. Poit., pp. 62, 65; Will. Malms., Gesta Regum, vol. II, p. 290; Ord. Vit., vol. III, pp. 234–238.
74 Will, Poit., pp. 68–72. Freeman (Norman Conquest, vol. III, pp. 144, 164) denies that Geoffrey was present, but Will. Jum. (p. 129) mentions him.
75 Roman de Rou (ed. Andresen), vol. II, pp. 224, 225.
76 Will. Poit., p. 72.
77 Many of the magnates such as the count of Eu, Hugh of Gournai, William of Warenne, and Walter Giffard mentioned as having taken part in the battle of Mortemer came from that neighbourhood.
78 For example, Ralph of Tosny, who was with the duke west of the Seine (Ord, Vit., vol. III, p. 238), was a great landowner in Central Normandy.
79 Will. Poit., p. 70.
80 Ibid., p. 72; Will. Jum., p. 130; Ord. Vit., vol. III, pp. 237, 238.
81 The Norman annals give this date which has been generally accepted, but it is based on doubtful evidence. The charter which might support it has been printed in R.A.D.N. (no. 133), and in Cart. Îles normandes (p. 185, no. 116). It has also been calendared in Cal. Doc. France (no. 710). All three editors assign this instrument to 25 December 1054, and since it is subscribed by Mauger as archbishop this would imply that he had not been deposed at that time. The charter could, however, in my opinion, be dated with equal plausibility (according to modern reckoning) at 25 December 1053, for Christmas was frequently taken to mark the beginning of the new year.
82 The Acta of the archbishops of Rouen (Rec. Hist. Franc., vol. XI, p. 70) speak of Mauger as having been deposed by the authority of Leo IX, who died 19 April 1054. This by itself is not wholly conclusive. But a charter for Le Mont-Saint-Michel (R.A.D.N., no. 132), which all the elements of a complicated date seem to place in 1054, is witnessed by Mauger's successor Maurilius as archbishop. The matter is not removed from doubt, but I have only slight hesitation in assigning Mauger's deposition to 1054.
83 Bessin, Concilia, pp. 46, 47.
84 Will. Jum., p. 127; Will. Poit., p. 44.
85 Ord. Vit. (Will. Jum., p. 180); Benoit (ed. Michel), vol. III, p. 132.
86 Will. Poit., p. 74.
87 Will. Poit. (p. 74) states that there was much fighting between the battle of Mortemer (February 1054) and the truce between William and King Henry (? 1055). In particular, he assigns to this period operations round Ambrières. Will. Jum. (p. 127) places an attack on Ambrières at the time of the Domfront campaign, and this may well have taken place. But the same writer alludes to the seizure of ‘two townships’ in Maine after Mortemer, and Ordericus, interpolating this passage (p. 184), says this action was in the neighbourhood of Ambrières.
88 Ord. Vit., interp. Will. Jum., p. 184.
89 Cart. S. Vincent du Mans, no. IV
90 Cart. S. Père Chartres, pp. 149, 184, 193, 211, 403; Round, Cal. Doc. France, no. 1168; Cart. S. Michel de I'Abbayette, no. 5.
91 Will. Poit., pp. 66–78.
92 Latouche, op. cit., pp. 32, 79.
93 The king and the count were together at Tours on 19 January 1057, and at Angers on 1 March 1057 (Rec. Hist. Franc., vol. XI, pp. 592, 593; Soehnée, op. cit., nos. 106, 107).
94 Not 1058, as has been very generally supposed. The correct date has been established by J. Dhondt (Normannia, vol. XII (1939), pp. 465–486; also Revue Belge de Philologie et d'histoire, vol. XXV (1946), pp. 87–109). That the expedition took place in August (of 1057) is indicated by the remark of Wace (Roman de Rou, vol. II, pp. 238) that the crops were still on the ground and ready for harvest.
95 Will. Jum., p. 131; Will. Poit., pp. 80–82.
96 Gesta Regum, vol. II, p. 291. The Norman annals are all silent on this campaign, and the name Varaville is not mentioned by either Will. Poit. or Will. Jum.
97 Roman de Rou, vol. II, pp. 236 et sqq.; Benoit (ed. Michel), vol. II, pp. 14 et sqq.
98 Will. Poit., p. 88; Ord. Vit., vol. II, pp. 102, 252; C. W. David, Robert Curthose, pp. 7, 8.
99 A charter given by Herbert of La Milesse between August 1055 and 14 November 1060 is assented to by his lords, Geoffrey, count of Anjou, and Herbert of Maine (Cart. S. Vincent du Mans, no. 303).
100 Le Prévost, Soc. Antiq. Norm., Mémoires, vol. XI (1840), pp. 20–25. Cf. Stenton, Anglo-Saxon England, p. 611.
101 Ord. Vit., vol. III, p. 224.
102 Round, Studies in Peerage and Family History, pp. 147, 149.
103 Below, pp. 173, 174.
104 Ord. Vit., interp. Will. Jum. (ed. Marx), p. 184.
105 On what follows, see R. Merlet, in Moyen Age, vol. XVI, 1903.
106 Dhondt, Normannia, vol. XII, p. 484.
107 Rec. Hist. Franc., vol. XI, pp. 431, 598; Soehnée, op. cit., no. 116.
108 Chronicle of Fécamp (Rec. Hist. Franc., vol. XI, p. 364).
109 Cart. S. Père Chartres, vol. I, p. 153 (R.A.D.N., no. 147).
110 Merlet, op. cit., vol. XVI (1903), p. 208.
111 Rec. Actes – Philippe I (ed. Prou), p. xxviii.
112 Halphen, op. cit., p. 12.
113 Below, Appendix B.
114 Hefele-Leclerc, Histoire des Conciles, vol. IV, p. 1018.
115 Below, Appendix B.
116 Will. Jum., pp. 127, 128.
117 Below, Appendix B.
118 Will. Jum., p. 127.
119 Grierson, in R. Hist. Soc., Transactions, series 4, vol. XXIII (1941), pp. 95 et seq.
120 Will. Poit., pp. 52–54.
121 Lot, Fidèles ou Vassaux?, p. 13.
122 Will. Jum., pp. 103–104.
123 Flach, Origines de l'ancienne France, vol. IV, p. 71.
124 Grierson, op. cit., p. 98.
125 AS. Chron., ‘C’, s.a. 1049.
126 Grierson, lot. cit.
127 Below, pp. 169, 170.
128 The most picturesque of these derives from the Chronicle of Tours (Rec. Hist. Franc., vol. XI, p. 348). It tells that the duke, who had been sustained by Baldwin V, asked for the count's daughter in marriage. The girl declared that she would never marry a bastard. Whereupon, Duke William went secretly to Bruges where she was living and forced his way into her bedroom, where he beat and kicked her. The girl then took to her bed, but was so impressed with the treatment she had received that she declared that she would never marry anyone else. The tale may be regarded as of more interest to the student of psychology than to the student of history.
129 Milo Crispin, Vita Lanfranci (Opera (ed. Giles), vol. I, p. 286); Ord. Vit., interp. Will. Jum. (ed. Marx), pp. 181–182.
130 Will. Malms., Gesta Regum, vol. II, p. 327.
131 Milo Crispin, loc. cit.; Chronique du Bec (ed. Porée), p. 190.
132 Milo Crispin, loc. cit.; Chronique du Bec (ed. Porée), p. 190.