In the first two chapters we have reviewed, along with the earlier history of England, reasons why England might be invaded in 1066: in brief, the attraction of a wealthy land, together with the hope of success against a divided and unstable state. Now we need to consider, along with Normandy’s earlier history, how the ruler of that duchy was able to invade in 1066. This is an important consideration, for it is probably true that before 1066 such a venture would not have been viable.

Under the early rulers, from about 911 to 1026, Normandy grew into its recognised boundaries, and the interest of its rulers was inevitably upon this internal growth. Normandy then underwent a period of troubles, under Richard III (1026–7) and Robert I (1027–35), lasting through the minority of William the Conqueror. Only by about 1047 was the Conqueror really safe in his own duchy.

There then followed a period when his main task was to deal with enemies and rivals in France. He could not possibly have invaded England, and left Normandy open to invasion from the Capetian monarch or the count of Anjou. Only with the deaths of his main enemies on the continent, in 1060, was William relatively free to contemplate some broader project.

Even after 1060 for some time those broader projects were still nearer to home, in Maine and Brittany. That the death of Edward the Confessor occurred in 1066 and not earlier was in many ways a stroke of luck for the Conqueror. It came at just the moment when he could truly think about pursuing claims in England, with sufficient stability in the duchy and on its borders to leave it for some months, and with a degree of wealth and support which had not been available to him before the 1060s.

We shall note more closely than in the opening chapters the growing links between England and Normandy after ad 1000. The geographical proximity of the two was bound to bring some connection, but in the first half of the eleventh century there were new dimensions to the relationship: economic, social and political.

Both areas had strong Scandinavian settlements, and common interests from them. According to Dudo of St-Quentin, a clerk from Picardy who came to the Norman court, the links went back as far as Rollo or Rolf, the Viking leader and first ruler of Normandy. Dudo’s work, especially for the early years of Norman history, is now widely questioned. One historian has seen the Customs and Acts of the First Dukes of Normandy as ‘a mere farrago of distorted and altered fragments from the old annalists’.1 Dudo was a chaplain at the court of Duke Richard II, and becomes more trustworthy when dealing with his own lifetime, though never exactly reliable.2

The English kings, after a period of hostility, began to seek better relations with Normandy, and to make agreements for mutual benefit. In 991 Richard I of Normandy and Aethelred II of England made an agreement not to aid their respective enemies. Stemming from this, Emma, Richard I’s daughter, married Aethelred of the old West Saxon line in 1002. (After his death she married Cnut, the greatest of the Scandinavian kings of England in 1017.) She and Aethelred, with their two sons, took shelter in Normandy when Aethelred was in difficulties in 1013; and the sons, Edward and Alfred, were brought up at the Norman court.


The comital family of Normandy from 942. (Notes: William the Conqueror was William I, King of England, but William II, duke of Normandy; Robert I fathered two children by Herlève but did not marry her.

The Normans gave Emma’s sons assistance when they attempted to return to England. Robert I, who treated them ‘as brothers’, organised an invasion fleet at Fécamp in 1033, though a storm ruined its chances.3 Nevertheless, Norman aid for the exiled aethelings was a real threat to the Scandinavian kings of England. Edward, when he became king of England, brought Normans to the English court and made grants to those who had attached themselves to him during his Norman exile. William the Conqueror probably visited Edward in England in the 1050s, and was said to have received a promise of the English throne at Edward’s death. William of Poitiers claims that the throne was promised as ‘a lawful gift’ with the assent of the magnates.4

Normandy’s development in the period before Hastings was very different to that of England, but there were some similarities. Normandy had also been overrun by Viking invaders, and was developing into a powerful political unit. But Normandy was not a kingdom and acknowledged, however incompletely, the authority of the king of France. The most remarkable factor in Normandy’s position by 1066 was its readiness to expand. Not only England, but also Spain, several parts of the Mediterranean, and especially southern Italy were to receive often unwelcome Norman visitors. Perhaps some latter-day Viking will-to-voyage endured in the only part of France where a Viking settlement had taken root; other French principalities, though equally interested in expansion, did not go so far afield in their ambitions.

Part of the explanation of Normandy’s unique history lies in the kind of political unit that Normandy was. Several groups of Vikings settled on the western shores of continental Europe, and a few were able to establish some political authority, but Normandy was the only one to survive under a Viking dynasty. William the Conqueror was the direct descendant of a remarkable line of dukes descended from the first Viking ruler of Normandy, Rollo or Rolf – though the early leaders were not dukes or even perhaps counts.5

We need, in order to understand Normandy in 1066, to look back to the foundation of Normandy under the Viking leader Rolf. At the beginning of the tenth century, there were several similar leaders of war bands who had settled as best they could along the continental coast, mostly in regions where rivers entered the sea. There had been a series of raids against the Norman coast before any settlement occurred, for example in 841, 851, 855.

As in England, the first impact was frightening. One account tells of the consequences for the people attacked: ‘I was freed from the hand of the very cruel nation of the Normans. They took me, bound me as a wretched slave, [and] sold me to a foreign land.’ Having been dealt numerous blows, faced perils of the sea and storms, suffered extreme cold, nudity, atrocious hunger and a long voyage, the writer finally returned home.6

Rolf’s group had settled along the Seine. There is no doubt these were opportunistic groups, happy to take any wealth that might offer but also eager for land, and in competition with each other for it.

The ‘foundation’ of Normandy came out of a policy used by the increasingly desperate rulers of Western Francia. The Viking raids were only one of several serious problems disturbing Europe at the time. The troubles had led to the splitting of the old Frankish Carolingian Empire into several component parts in 888, one of which was West Francia, and from this nucleus emerged the kingdom of France. The kings of West Francia found it difficult to survive, and their kingdom was itself in danger of splitting into yet smaller independent units, such as counties.

The struggle weakened the old Carolingian dynasty, and before the end of the tenth century it was to be replaced by the family of one of its dukes, the Capetians. However, in 911 there was still a Carolingian ruling West Francia, Charles III, known rather unfortunately as Charles the Simple (898–922). Translated more accurately, his name probably meant Charles the Honest. A chronicler explained: ‘during his life he was called simplex because of his good nature’.7

Charles III made the decision to try and save West Francia by allying with the Viking Rolf, the leader of one band of Vikings who had settled on the Seine. This was a bold move, and potentially dangerous for the West Franks, but it had great consequences. Charles probably had two major motives in making the agreement of 911. The first was to use Rolf as an obstacle against further Viking incursions into that part of Francia, now fighting for him rather than against him, and thus also dividing the Viking menace. He was seeking the alliance of one Viking leader not only against further Scandinavian incursions, but also against other Viking groups already settled in the region. The second purpose of Charles was to use Rolf as a buffer against the expansion of the Bretons, who had not come under the authority of the West Frankish kings, and who had been extending their power eastwards.

It is now generally accepted by historians that Charles III’s grant was a sensible if risky move which, it must be said, succeeded. It is also acknowledged that it was a narrower grant than the Normans would claim a century or so later. They made the claim to justify the expansion of comital power beyond the confines of the original grant. But Rolf in 911 was the leader of one band among many and was only recognised as an ally not as a count. Whatever territories were ‘granted’ would have to be won and held by the recipient of the grant.

Charles III certainly encouraged Rolf to take lands from the Bretons but, in the context of 911, this almost certainly meant land which was to become part of Normandy rather than Brittany proper.

It is very doubtful that Charles III envisaged any expansion of Norman power over the whole of Brittany. No more did he wish to see a Viking Normandy. Not that he had much right, and certainly no power in real terms over Brittany or even Normandy, to make such a broad grant. However, he would have been perfectly happy to see the Normans fight against the Bretons, who at the time held land which later became western Normandy. It is unlikely that the king was more generous than he needed to be.

Rolf had recently been defeated in battle by the Franks at Chartres. So the grant came at a time of Frankish strength rather than weakness. What Charles the Simple granted in 911 was almost certainly the right of Rolf to rule in the king’s name over the city of Rouen and a relatively restricted area around it by the Seine; at most it was Upper or eastern Normandy, probably the area where Rolf already held sway.

Rolf had emerged as the leader of a group settled in the area for some decades, who had recently taken over Rouen. He is thought to be of Norse rather than Danish origin, an idea supported by tradition held within the ducal family itself. It is believed that he was exiled from Scandinavia by Harold Fairhair, king of Norway (900–33), and had probably lived the life of a Viking, voyaging to Scotland, Northumbria and Ireland.8

Although from the evidence of a charter of 918 we can be sure than an agreement was made between King Charles and Rolf, we are uncertain about its details. As D.C. Douglas has suggested, ‘far less is known about pre-Conquest Normandy than about pre-Conquest England’. Something has been done about that by Douglas himself, and by David Bates and others since, but there is still a gap in the quantity of sources available.9

The problem over 911 is that the only details come from Dudo of St-Quentin, who was writing to bolster Norman ducal claims, and was doing so about a century after the event. Dudo was not himself a Norman, but had come to the court of Duke Richard I at the end of the tenth century. His Customs and Acts of the First Dukes of Normandy was written for the dukes, and finished before Dudo’s death in 1043. Like many such works it is less reliable for the early material, but among chroniclers Dudo was particularly prone to invention. Given that we must not believe his every word, it is still worth recording what he made of the meeting which he described as occurring at St-Clair-sur-Epte in 911.10

At the appointed time they came to the place that had been agreed, which is called St-Clair. The army of Rollo kept to one side of the River Epte, that of the king and Robert [the Strong, duke of France] to the other. Then Rollo sent the archbishop [of Rouen] to the king of the Franks, on a mission to say to him: ‘Rollo cannot conclude a peace with you, since the land you wish to grant him is not cultivated by the plough, and is altogether without herds and flocks and the presence of men. No one lives on it except by thieving and pillage. Give him rather a region from which comes nourishment, filled with riches. He will not negotiate with you unless you swear by a Christian oath, you and your archbishops, bishops, counts, abbots and all your realm, that he and his successors shall hold the land from the River Epte to the sea as a farm and an alod for ever’

Then Robert duke of the Franks and the counts, bishops and abbots who were there said to the king: ‘You will not win over such a strong leader unless you do what he wants. If you do not give him what he claims from you in return for his service, give it in return for Christianity, that at last such a numerous people should be won for Christ, then you will avoid diabolical error; and, finally, [give it] in order that the edifice of your realm and of the Church should not be destroyed by the attack of his army. You should employ his defence and protection in the name of Christ. You should act as a king and a firm defender of the Church’.

The king wanted to give him Flanders on which to support himself, but he did not wish to accept that because it was marshy. Also the king offered to give him Brittany, which was on the border of the promised territory. Then Robert and Archbishop Franco told all this to Rollo, and brought him to King Charles, after an exchange of hostages, under the protection of the Christian faith. The Franks, seeing Rollo, the aggressor against all Francia, said to each other: ‘This chief has great power, great courage, much wisdom and prudence, and even more energy, to have waged war in this way against the counts of the realm’.

So, placated by the words of the Franks, he placed his hands between the hands of the king, which his father, grandfather and great-grandfather had never done. So the king gave his daughter, Gisela by name, as wife to the leader, and certain lands as alods and in farm, from the River Epte as far as the sea, and all of Brittany, on which he should be able to live. The bishops, seeing that Rollo was unwilling to kiss the foot of the king, said: ‘He who receives such a gift ought to kiss the foot of the king’. He replied: ‘I shall never bend my knee to anyone, nor shall I kiss any foot’. But, compelled by the prayers of the Franks, he ordered a certain soldier to kiss the king’s foot. The latter, at once seizing the king’s foot, lifted it to his mouth and, having planted the kiss while he was standing, made the king fall down. So, much laughter arose, and a great disturbance among the men.11

The basic import of the agreement was that the king recognised Rolf as a chieftain (not a duke or even a count) over a territory centred upon Rouen, with the chief purpose of protecting the Frankish kingdom; ‘for the safety of the realm’ according to the 918 charter. In this charter Rolf and his companions are described as ‘the northmen of the Seine’.12 Flodoard, who is more to be trusted than Dudo on this, says that Charles granted Rouen and some coastal districts dependent upon that city.13 There seems little doubt that Rolf’s lands did not go beyond the River Orne to the west, perhaps not even so far; what he was given was, in essence, eastern Normandy. He also made concessions to the king. Dudo’s farcical account of one of Rolf’s men grasping the king’s leg and pulling him over is a way of belittling what was surely some act of submission or homage, with a promise of fidelity, and does not seem credible.

Even Dudo accepts that Rolf and his men agreed to become Christian as part of the deal. Rolf and his ‘companions and soldiers’ were immediately baptised.14 The slowness of Christianity to recover in Normandy suggests that the Normans were being forced into an act for which they had little enthusiasm. It was reported that, when dying, Rolf made gifts to churches, but also arranged for human sacrifices in the pagan manner. It is probable that Rolf married a Christian wife, though it may not have been Popa or Gisela, and that his children were brought up as Christians.15


The sources for the period from 911 to 1026 are not very full. We have to rely a good deal on Dudo of St-Quentin, though some other Frankish chronicles give assistance. We do not, for example, even know the date of the death of the first Viking ruler of Normandy, Rolf.

Rolf’s original territory was a fertile land. Eastern Normandy contained much open country and provided valuable produce in grain and fruit. A charter of Charles III, dated to 905, shows that normal administration was operating within Rolf’s territories then, so we may believe that Rolf was keeping order within his own region in eastern Normandy, which can only have improved after the 911 agreement.16

Rolf himself began the expansion of his lands westwards, which would culminate in forming Normandy as we know it. He probably moved his authority beyond Eu, and beyond the Orne as far as the River Vire. Rolf made the first major extension of his original grant, adding what we think of as middle Normandy, especially the Bessin. By the time of his death, Rolf had become essentially a Frankish count with his capital at Rouen.

Rolf’s son, William, succeeded him in about 924. He is known as William Longsword (c. 924–42). Among the early rulers, the descendants of Rolf, there was no weak link. Each one in turn added something to what he acquired, and increased the strength of Normandy. In William I’s case it was the second major extension of Norman authority.

To the west there was trouble between the Bretons and Scandinavian settlers along the Loire, and the Normans intervened for their own profit. By 933 William had gained western Normandy as far as the Couesnon, which it seems he had recovered from the Bretons, adding the Cotentin and the Avranchin, though comital power in these areas remained weak through the next century.17 Normandy as we recognise it had been created, or perhaps recreated, since it responded closely to the old ecclesiastical province of Rouen and the even more ancient boundaries of the Roman province of Lugdunensis Secunda; it also had a rough correspondence to the Frankish region of Neustria, which it was still sometimes called. William Longsword established a family connection with Fécamp, where a palace was constructed.

William Longsword was christian and encouraged Christianity. He married a christian noblewoman, Liégarde, daughter of the count of Vermandois, though his successor was born to a Breton mistress.18 But Christianity’s revival in Normandy proved slow and uncertain. Five successive bishops appointed to Coutances were unable to reside in their see. Bishops appointed outside Rouen also found themselves unable to live in their own sees.

However, William I was known as a friend of monastic restoration, and was especially associated with the great house of Jumièges. In 942, the year of his death, William welcomed King Louis IV of France (936–54) to Rouen, which suggests that he recognised the king’s authority over Normandy. Neither the emphasis on Christianity, nor the friendship with the West Frankish monarchy, seems to have been favoured by William’s subjects, and may have been the cause of his assassination in 942. His death, though treated by some as martyrdom, led to a pagan revival in Normandy, and a period of disorder.

William Longsword’s son succeeded as Richard I (942–96). He grew into ‘a tall man, handsome and strongly built, with a long beard and grey hair’. But in 942 he was only ten years old, and as usual a minority meant disorder and difficulty.19 Scandinavian raids were still occurring, and were a cause of disturbances within Normandy. The king of France and the duke of the Franks established themselves in Norman territory, and won a victory against the Viking leader Sihtric. They looked for the overthrow of the Viking county rather than the defence of Richard. But in 945 Louis IV was himself defeated by Harold, a Viking leader probably based in Bayeux.20


Capetian Kings of France: Hugh Capet to Philip I.

Gradually, over the years, Richard I emerged as a man of strength and determination. He took as his wife, though perhaps not by a Christian ceremony, a woman of Danish descent called Gunnor, from a family settled in the pays de Caux. By her he had several children, including his eventual successor. Most members of the Norman nobility of the Conqueror’s time claimed some sort of relationship either with Richard I or with Gunnor, which brought a coherence to the ruling group that in turn added strength to their combined efforts at expansion.21 Given the circumstances of the minority, it is hardly surprising that Richard I continued to keep links with Scandinavia, but he also made an agreement with the new king, Lothar, at Gisors in 965. For a long period after this the ruler of Normandy kept on good terms with the king of the West Franks, of importance to them both.

But Richard I did not continue his support for the old Carolingian family. He had already been closely associated with the duke of the Franks, and took as his ‘official’ wife, Emma, daughter of Hugh the Great, duke of the Franks (d. 956). The Normans were among the firmest supporters of this family. In 968 Richard I recognised Hugh the Great’s son, Hugh Capet, as his overlord, and when Hugh became the first Capetian king of France (987–96) the Normans were among his earliest adherents. Richard also sought to restore Christianity, and from this time on paganism in Normandy waned. One of his most enduring acts was to aid the revival of the monastery at Mont-St-Michel. Richard I’s reign also saw the beginnings of an important monastic revival in Normandy.

Richard II (996–1026) succeeded his father in a year marked by a peasant revolt in Normandy. The peasants called assemblies, and made ‘laws of their own’, but the movement was brutally suppressed by the nobility.22 When the count of Ivry was approached by rebels to put their case, he cut off their hands and feet. But the new reign was a period of significant economic progress for Normandy. Despite being ‘highly skilled in warfare’, Richard II kept out of the conflicts which raged around him in north-west Europe, though he did push Norman interests beyond his own boundaries. He had ‘decidedly pacific tendencies’, and brought a period of significant stability to the duchy.23

Richard II married the sister of the count of Rennes, the ‘fair of form’ Judith.24 He also had contacts with the Scandinavian world: Vikings could still be welcomed at Rouen in 1014, and a Norse poet was received at court in 1025. To Franks outside Normandy the rulers still seemed Vikings, and Richer of Reims continually referred to Richard as ‘duke of the pirates’. The name given to the territory itself, ‘Normandy’, came from the same attitude to its inhabitants, meaning the land of the northmen or Vikings.

But Scandinavian influence was decreasing in Normandy. Place-name studies suggest that the original Scandinavian settlement did not extend evenly throughout Normandy. The names cluster along the coast and the rivers. It is clear in any case that the settlers began to integrate with the existing population through intermarriage. Some Scandinavian attitudes and customs continued but, as is so often the case, the surviving population from the old world recovered its strength, if only in influencing language and a way of life. By the tenth century French was taking over as the main language in Normandy, if it had ever been overtaken. According to David Douglas, by the eleventh century Normandy was ‘French in its speech, in its culture, and in its political ideas’.25

The administrative system which developed in Normandy was largely Frankish, and similar to that in surrounding counties. We have a nice picture of Richard II at Rouen, in ‘the city tower, engaged in public affairs’. We are told that those in attendance feared to break in upon him unless summoned by his chamberlains or doorkeepers: ‘but if you wish to see him, you can watch him at the usual time, just after dinner, at the upper window of the tower, where he is in the habit of looking down over the city walls, the fields and the river’.26

Richard II continued the family’s reputation for defending the Church, and was responsible for inviting to Normandy the reformer William of Volpiano. By this time the episcopal organisation of Normandy had developed, and the bishops were able to function normally within their sees. Under Richard II a new social structure of Normandy emerged. It is clear now that this was not the emergence of new families, but of old families in a new guise: as castellans, with stress on primogeniture and lineage. The families were not new, but their way of looking at themselves and their ancestry was.

There is, for example, no mention of the Montgomerys (one member of whom considered himself ‘a Norman of the Northmen’) before a charter dated to 1027 at the earliest, after Richard’s death. Montgomery itself was not fortified until after 1030. The use of toponyms to define an individual and his family did not become common until about 1040.27 It was about this time that the residence at Le Plessis-Grimoult was turned into a castle. During the period of political instability old families began to see themselves as lineages, to build castles, to latch on to offices at the ducal court, to become vicomtes in ducal administration of the duchy, indeed to threaten ducal power itself.28

Richard II married twice, to the Breton Judith, whose sons, Richard and Robert, succeeded him, and to the Norman Papia, by whom he had two further sons, William of Arques and Mauger, the later Archbishop of Rouen. Richard II used members of his family to rule over divisions of his territory on his behalf: at Mortain, Ivry, Eu, Évreux and Exmes. It was the acknowledgement of the rights of this second family which caused many of the problems of the subsequent period.

With local magnates called counts came the transfer of the ruler’s title from count to duke, marking his superiority. Those appointed to rule over the new Norman counties, mostly in sensitive areas on or near the frontier, were members of the ducal family. Ducal government also developed, and we begin to hear of vicomtes, who were not deputies for the counts, but were all direct representatives of the count of Rouen himself, that is of the duke. During the period 1020 to 1035 some twenty vicomtes have been identified, and they represent a growing structure for comital government throughout Normandy.


After the death of Richard II in 1026, Normandy underwent a long period of difficulty. The next duke, Richard III, survived only a year, until 1027. There was rumour that he had been poisoned, possibly by his successor.29 Robert I (1027–35) was the only member of the family of Rolf who proved something of a failure, despite being known as Robert the Magnificent or sometimes the Liberal, and reputed to be ‘mild and kind to his supporters’, with an ‘honest face and handsome appearance’, and of a ‘fine physique’.30Others, it must be said, called him Robert the Devil.

External relations deteriorated, and Normandy faced a period of severe internal disorder. Yet the duchy retained vestiges of its earlier position. When King Henry I of France (1031–60) found himself in desperate trouble in the year of his accession, it was to Normandy that he fled for refuge. Surviving gratitude for this help accounts for his aid to the young William the Conqueror during the latter’s minority, the years of his greatest vulnerability.

Robert’s decision to go on pilgrimage to the Holy Land is something of a puzzle. Perhaps he was overcome with piety, though his life to that date shows little sign of it. Perhaps he was overcome by remorse, for which he no doubt had good cause. However, for his duchy it was a perilous moment to depart on such a distant adventure, from which, as might have been feared, he was never to return, dying unexpectedly at Nicaea during his return journey.

One of Robert I’s sins was a liaison with Herlève, variously said to be the daughter of a tanner or perhaps an undertaker of Falaise called Fulbert.31 In any case the duke, as dukes will, had his way with her, made her pregnant without any thoughts of marriage, and thus fathered William the Bastard, perhaps Robert’s chief contribution to his duchy.

When you look down nowadays from the walls of the great stone castle at Falaise (not in that state when Duke Robert lived), you are told that you are standing (presumably approximately) where Robert was when he espied the fair Herlève beside the pond below, outside the castle wall. Another story is that he had ‘accidentally beheld her beauty as she was dancing’. The twelfth-century writer described William’s birth, on rushes laid out on the floor, and said that Herlève had a dream about her new son: she saw her intestines spread out over Normandy and England which forecast William’s ‘future glory’!32

William the Conqueror (William II, duke of Normandy, 1035–87) thus came to rule the duchy in unpromising circumstances. His father had died when William was aged about nine, possibly even younger. The duchy had passed through decades of instability, which had included a peasants’ revolt and divisions among the aristocracy, while ‘many Normans built earthworks in many places, and erected fortified strongholds for their own purposes’.33

Added to that, William was not the legitimate son of the old duke. Bastardy was not the stain it was about to become in terms of moral attitude or right to inherit, but it was, nevertheless, a drawback, as one can see from the very fact that he was called ‘the Bastard’, and from the way the citizens of Alençon and others later would taunt him with his bastardy. William’s reaction to this insult at Alençon shows how much it smarted: he ordered the hands and feet of thirty-two mockers to be cut off. A chronicler considered that ‘as a bastard he was despised by the native nobility’.34 The taint of bastardy added to the dissatisfaction of the nobles at having a minor succeed to the duchy.

William’s own relatives were among those who stirred up trouble during his minority, suggesting the unwise nature of Richard II’s acknowledgement of families by two wives. The period was marked by a series of internal rebellions and external threats. At times, William’s security, and even his life, was at risk. At Valognes, he was once roused from sleep to be warned that conspirators were about to kill him; William got away half-dressed on a horse. He was protected by a few loyal retainers and given some support from the Church and by the king of France, but he often escaped by the skin of his teeth. Among those around him who were killed were his guardian Gilbert de Brionne, his tutor Turold, and his steward Osbern.35

The worst period of anxiety ended when Henry I of France came to his aid against the Norman rebels, enabling the young William to win his first major engagement at the battle of Val-ès-Dunes in 1047. The duke’s enemies gathered in the west of the duchy and advanced to the Orne, where their way was blocked by the duke’s supporters. The rebels broke, and many drowned in the river. If Wace is to be trusted, horses were seen running loose on the plain, while mounted men rode haphazardly in their efforts to escape. William of Poitiers confirms that riders drove their mounts into the Orne trying to get away, till the river was full of soldiers and horses.36

The Conqueror’s main enemy and rival at this time had been Guy de Brionne, but the victory at Val-ès-Dunes crushed his ambitions. However, William showed little gratitude to the French king. Once freed from his greatest fears, he began to flex the muscles of his Norman war machine, and to attack neighbouring powers in a way that his predecessors had avoided. This caused growing resentment and hostility from those neighbours, and from the king.

We do not know the precise reasons, but it was in this context that the king of France joined the enemies of Normandy from 1052, and turned to attacking the duke he had previously defended. Possibly it was because of Norman participation in a rebellion against the king in the Ile-de-France. Whatever the reason, the king’s hostility added considerably to Normandy’s dangers in the mid-eleventh century.37

William was equal to the new threat, and in the 1050s transformed Normandy into a greater military power. A serious problem was posed by the building of private castles during the worst of the disturbances. Now William had to spend much of his time besieging, destroying, or taking over these strongholds. Any rebel of standing could shelter behind the walls of his own castle. In the early part of the decade his own uncles, Count William of Arques and Mauger, Archbishop of Rouen, remained the greatest internal threats, and they could now look to assistance from France and the growing rival of Normandy, the county of Anjou.

Count William of Arques’ opposition turned into rebellion against his nephew by 1053. He had never readily accepted the succession of his brother’s illegitimate child. In 1053 Henry I of France tried to relieve Arques, but was beaten in a conflict at St-Aubin-sur-Scie by some of the Conqueror’s men, using a feigned flight. The surrender of Arques and the submission of Count William symbolise the triumph of the Conqueror over the rebels. His uncle was treated leniently and allowed to go into exile.38

In 1054 the enemies of William combined in rebellion and invasion, but he thwarted their attack by a great victory at Mortemer. Here, according to William of Poitiers, the invading army was decimated. At midnight, William ordered a herald from the top of a tree to cry the details of the victory to the defeated king, who then fled.39 One of the duke’s enemies in the field was the neighbouring Count Guy of Ponthieu. Guy was captured during the battle and submitted to the Conqueror, transferring allegiance to him. A few years later this move would have important consequences.

At this time, William was building a close group of familiars and friends from the Norman nobility, who would form a strong support to his activities throughout his life, men such as William fitz Osbern and Roger Montgomery, William de Warenne and Roger de Beaumont, together with his own half-brothers Robert, count of Mortain, and Odo, Bishop of Bayeux (sons of William’s mother, Herlève, by the husband Duke Robert had found for her). William II was beginning to take a grip on his duchy, and the great families and the lesser lords swung in behind his lead. He was also building a sound administration, revived after the period of troubles.

But William’s difficulties were far from over, and in 1057 he faced a new invasion from France and Anjou. The Angevin counts had expanded their territories in a manner even more remarkable than the successes to date of the Norman dukes. The Angevins had started from smaller beginnings, had no obvious frontiers to work towards, and were surrounded by hostile powers. At this time, Anjou was ruled by one of its greatest counts, Geoffrey Martel (1040–60). Normandy and Anjou were almost inevitably rivals since between them, and of interest to both, was the county of Maine, while both hoped to intervene also in Brittany.

William responded with energy to the new invasion and again defeated his enemies, this time at Varaville in 1057. Here William caught the invaders attempting to cross a ford on the River Dives, and attacked the rear section, when the change of the tide caused the river to rise. About half the enemy army had crossed and could not return. According to Wace, the Normans used archers and knights with lances to annihilate the men at their mercy.40 Because of the tide, William was unable to pursue those on the far side, but Henry I was forced to flee from the duchy.

Even this victory did not ensure William’s triumph. Both the king of France and the count of Anjou had escaped and continued to oppose him with some success. It is sometimes overlooked that although William made claims upon Brittany and Maine, while Geoffrey Martel lived the latter was more successful.


Counts of Anjou, 987–1109.

For William the year which brought great change and transformed his position and his hopes was 1060. His two greatest enemies died: Henry I of France, leaving an eight-year-old son, Philip I (1060–1108); and Geoffrey III Martel of Anjou, whose death resulted in a conflict between his nephews, Geoffrey IV the Bearded (1060–7, d. 1096) and Fulk IV le Réchin (1067–1109), to control the principality.

It was at this point that William could seriously undertake a programme of expansion beyond Normandy. However, even in 1060 his first concern was not with England, where any success must have still seemed a fairly distant likelihood. His first action was against Maine, situated on Normandy’s southern border. William captured the stronghold of his opponent, Geoffrey de Mayenne, by ‘throwing fire inside its walls’, and for a time from 1063 Maine fell under the power of Normandy.41

Without Geoffrey Martel, Anjou went through a period of internal troubles from which William took advantage. In the following year he moved into the second area where Norman ambitions had been thwarted by Anjou and Brittany. This was the campaign in which Harold Godwinson took part.


Counts of Flanders, 988–1111.

The reason why Harold went to Normandy is not clear. Edward the Confessor seems to have sent him, and at least one of his aims was to try and help two relatives who were hostages in Normandy. The Durham chronicler, perhaps rightly, claims that the trip was made at Harold’s initiative and against the king’s advice: he ‘begged the king’s permission to go to Normandy and liberate his brother and nephew, who were detained there as hostages, and to bring them back with him in freedom’.42 William of Poitiers has William the Conqueror later in England claiming: ‘the king [Edward] gave me Godwin’s son and grandson as hostages. What is more, he sent Harold himself to Normandy, so that he might swear in person in my presence what his father and the others whom I have mentioned had sworn … he confirmed in writing that the kingship of England should without question be mine.’43

There is a puzzle over this matter of the hostages. From the Norman sources they were handed over to guarantee Edward’s promise of the throne to William of Normandy, and it is difficult to think of an alternative reason. That then raises the question of why the hostages should be Harold’s younger brother, Wulfnoth, and his nephew, Hakon. The apparent answer would be to guarantee the Godwin family’s support for William. This in turn raises the question of the Godwin family’s attitude. It would surely have been impossible for Edward and William to arrange for such hostages without Harold’s consent. This would suggest that Harold favoured or at least accepted the idea of William’s succession.

If in 1064 Harold was seeking the release of the hostages, he could hardly obtain it without convincing William that he could trust in his support even without the hostages. This is conjectural, but it at least explains the nature of the oath. The probable explanation is that the Godwin interest in the throne through most of Edward’s reign was not in seeking it for themselves, but in ensuring that, whoever came to the throne, the Godwin position would be secure. They were therefore not especially opposed to either Edgar the Aetheling or William, if their own family position was guaranteed.

The Tapestry shows Harold setting off in a leisurely manner, perhaps hunting on the way. He rested at his own manor of Bosham, where he feasted before boarding ship in Chichester Harbour. It was probably a storm which blew him to the shores of Ponthieu where he was arrested by Count Guy and taken to his castle at Beaurain. What Guy hoped to gain is uncertain, perhaps to use Harold as a bargaining counter with William.

The Conqueror was informed of the event, and ordered Harold’s release. Count Guy was no great friend of the Norman duke, but he had been forced into recognising his overlordship after being among the defeated at Varaville. At any rate, Guy decided not to oppose William and escorted the captive to the duke, to whom he was handed. The act of obtaining his release gave William an advantage over Harold, whose ability to act freely in Normandy is uncertain.

William received Harold in the palace at Rouen. The Tapestry refers to some now forgotten scandal there between a woman with an English name, Aelfgyva, and a cleric, and then shows William setting off with Harold on the Breton campaign.44 They passed by the great coastal monastery of Mont-St-Michel. Crossing the River Couesnon some of the Norman soldiers got into trouble in the quicksand, and were saved by the heroic action of Harold.

Their first objective was the castle at Dol, which the Tapestry shows as a wooden keep on a mound, a motte. It also shows Conan II, count of Brittany, escaping down a rope, though chronicle sources tell us that he had gone before the Normans arrived. They took Rennes and moved on to Dinan, which resisted. These two castles are also portrayed as wooden towers on mottes. Dinan was fired with torches, and the Bretons handed over the keys in surrender.

The Breton campaign had been successful, though its effects were soon to be reversed. William recognised the English earl’s contribution, and ‘gave arms to Harold’.45 The Tapestry version is surely the portrayal of a knighting ceremony. It probably means that Harold recognised William as his lord, and must be taken along with the oath in defining how the Normans viewed Harold’s subsequent actions.

So the victorious Norman army, having temporarily imposed its authority on eastern Brittany, rode back to Bayeux, where William’s half-brother, Odo, was bishop. It was here, according to the Tapestry, that Harold took the famous oath to William.46 Chronicle reports say it took place earlier and elsewhere, and Bonneville-sur-Touques, which is named for the event by William of Poitiers, is the most likely location. The Tapestry, in naming Bayeux, may have been trying to puff up the role of Bishop Odo, for whom the English artist was probably working.

There can be little doubt that an oath was made. What its exact content was we shall never know. Nor can we be quite certain if Harold was forced or tricked into swearing. The Norman interpretation was that Harold had made a promise to support William, perhaps as his man. William of Jumièges says that Edward sent Harold to ‘swear fealty to the duke concerning his crown and, according to the Christian custom, pledge it with oaths’. William of Poitiers confirms this, saying that Harold promised to do all in his power to ensure William’s succession to the English throne. He goes on to say that Harold promised to hand Dover to William with various other strongholds in England. We cannot take the Norman view without retaining some doubts about its accuracy on the detail, but it is impossible to discount the oath altogether, and William’s actions throughout point to his belief that in 1066 Harold betrayed his trust.47

The theme of the Tapestry is that Harold had made a sacred oath, he is shown swearing on a reliquary; by taking the English crown he therefore broke his promise. The implication is that Harold had promised to aid William in getting the English crown, though the oath may not have been so specific; but it surely at the least promised Harold’s fidelity to William. In any event Harold was able to return to England. One of the hostages, his nephew Hakon, was released, but Harold had to leave his brother Wulfnoth in Norman hands: this smacks of a compromise.

The Tapestry’s portrayal of Earl Harold reporting back to the Confessor suggests that his trip was seen as a failure and a humiliation. Had he been sent merely to inform William of a promise, this would not be the case. Perhaps it means the partial failure to get a good deal on the hostages. Perhaps Edward no longer favoured William’s succession; his attempts to bring over Edward from Hungary suggest this. It is also possible that Edward returned to the idea of William for the succession, as the Normans have it.

The other possibility is that Edward was toying with the thought of Harold succeeding. If so he may have been disappointed at the dilemma Harold had created for himself in Normandy. The comment of the author of the Vita Aedwardi that Harold was ‘rather too generous with oaths (alas)’ further suggests that Harold had promised more than was thought good, which would not be the case if Edward had sent him expressly to promise support to William for the English crown. Edward’s general reluctance to make public promises belies the idea that Edward arranged for Harold to take a solemn vow in public about the English succession. In short the Tapestry suggests that what had happened in Normandy was not at the wish of the king, and hints that support for William as king may not have been welcome in England.48

It seems that William had, on an earlier occasion, visited Edward the Confessor in England, and been given some promise of the succession. William was related to Edward, whose mother we recall was Emma of Normandy. Given that Edward had no children, William’s claim by relationship was as good as anyone else’s. Harold’s was, at best through his sister, Edith, who was married to the king. William’s interest in England was opportunistic. Had the chance offered at another time, he may well have had to ignore it. But in 1066 he could contemplate a military venture.

When Harold took the throne, William began to make plans to invade England. It is impossible to know exactly what happened in all the behind-the-scenes negotiating about the succession; almost certainly Edward gave out conflicting signals. Although we rely on Norman sources which have a bias, it is most likely that their version is close to the truth: that Edward promised William the succession, and that Harold took some oath to support William.

The Norman interest in expanding beyond the duchy was becoming a reality by the 1060s. Roger de Tosny and Robert Crispin led forces against the Muslims in Spain, helping in the recapture of Tarragona and Gerona. Probably the most interesting of the projects, apart from England, was the Norman venture into southern Italy. A Norman principality was established at Aversa by 1030. In time they would conquer the southern mainland of Italy and by 1061 were ready to begin the invasion of the island of Sicily, where they established a new Norman kingdom in the twelfth century.

An Italian chronicler recognised the adaptability of these conquerors: ‘the Normans are a cunning and vengeful people … they can endure with incredible patience the inclemency of every climate’. It has recently been argued that at least some of the adventurers were not of Norman origin, including some of the more important such as Roger de Tosny in Spain.49 It may be that part of the impetus was too tight a ducal control at home rather than Viking spirit. But there is still no doubt that men from Normandy played a vital role in expansion from north-western Europe and especially in opposition to the power of Islam. Norman efforts in the early crusades and in the eastern Mediterranean underline this ability to utilise their military abilities in varying circumstances. The point is the desire to leave Normandy for distant lands. But until 1066 such ventures had not been led by the duke with a ducal army.

The invasion of England was made possible by a combination of factors. William had been freed from many of his continental anxieties by the 1060s, but there were other considerations. He needed support. First he must be secure in Normandy. In 1055 a monk at Marmoutier could write that William was ‘ruler of his whole land, something which is scarcely found anywhere else’.50 By the 1060s practically every major family had accepted his authority, even ones on the fringes of the duchy found it advisable to have representatives at his court.

Officials in Normandy such as steward, butler, constable and chamberlain were not new, but the duke’s authority over them was stronger. The Peace of God, introduced into Normandy in 1047, was repeated in 1064, guaranteeing peace from violence in the duchy from Wednesday evening till Monday morning: only the ducal forces were allowed to use arms during that period. With William II power reverted to the duke, and his government became dominant in the duchy. He dispensed justice at his court, and could even afford to indulge in informal acts, as when he made a grant while sitting on a carpet outside a house at Bernouville.51

He also needed to be sure that powerful neighbours would not take advantage of his absence overseas. His most important move was to marry Matilda, daughter of Baldwin V, count of Flanders (1035–67). She was thought ‘a very beautiful and noble girl of royal stock’, and is believed to have been just over 4 feet tall.52 The marriage was forbidden by Pope Leo IX (1048–54), probably because of too close a blood relationship. But William went ahead with the ceremony in either 1050 or 1051 at Eu. This made Flanders an ally rather than a threat, and indeed a number of Flemings came with William on his invasion.

The Conqueror also desired the support of the Church in his venture. This was threatened by that selfsame marriage, since William had ignored the Church ban in order to marry. He was condemned for not awaiting papal approval, but he and his wife managed to placate the Pope, in part by building two great religious houses at Caen: St-Étienne for men, and La Trinité for women. In 1059 the Church gave formal approval of the marriage. By 1066 the Church was prepared to sanction the English venture, and a papal banner was given to the duke and proudly displayed for propaganda purposes when the troops embarked.53

William had prepared the ground well. By 1066 he was safe at home with firm authority over the ducal administration. After the death of the king of France and the count of Anjou he was free from major concern about neighbouring powers. He had made a marriage alliance with the most important of these, Flanders. He had also overcome difficulty with the papacy regarding his marriage and won not only acceptance but support for his venture in England. If ever the time was ripe to cross the sea and seek his fortune across the Channel, that time was in 1066.


  1.  H.H. Howorth, ‘A criticism of the life of Rollo as told by Dudo of St-Quentin’, Archaeologia, xlv, 1880, pp. 235–50, p. 250.

  2.  Dudo of St-Quentin, De Moribus et Actibus Primorum Ducum Normanniae, ed. J. Lair, Caen, 1865, p. 159.

  3.  William of Jumièges (and Orderic Vitalis and Robert of Torigni), Gesta Normannorum Ducum, ed. E.M.C. van Houts, 2 vols, Oxford, 1992, 1995, ii, p. 76.

  4.  William of Poitiers, ed. Foreville, p. 31.

  5.  ‘Rollo of Normandy’ in D.C. Douglas, Time and the Hour, London, 1977. pp. 121–40, pp. 121–4. Rolf or Rorik, or Hrolfr, seems a likely original for Rollo or Rou, which are clearly Latin and French versions of the name. However, to accept the evidence of Snorri Sturlusson, writing in the thirteenth century, for Rolf and with Norse origins is even more perilous than accepting Dudo of St-Quentin’s apparent belief in Danish origins. A lost charter of 913, in M. Fauroux (ed.), Recueil des Actes des Ducs de Normandie de 911 à 1066, Caen, 1961, pp. 19–20, and n. 4, suggests that Rolf was also christened Robert, which explains the popularity of that name among his descendants: p. 19: ‘Igitur Rollo, qui et Robertus nomine sacri baptismatis’. See E. Searle, Predatory Kinship and the Creation of Norman Power, 840–1066, Berkeley Ca, 1988, on ‘counts’.

  6.  E. Privat (ed.), Documents de l’Histoire de la Normandie, Toulouse, 1972, p. 70, from Adelhelm, Bishop of Sées.

  7.  E. James, The Origins of France, London, 1982, p. 181. The quote is from the Chronicle of St Benignus of Dijon.

  8.  Douglas, ‘Rollo’, p. 126.

  9.  ‘The rise of Normandy’ in Douglas, Time and the Hour, pp. 95–119; see also in the same volume, ‘Rollo’, pp. 121–40, p. 127; and D. Bates, Normandy Before the Norman Conquest, Harlow, 1982.

10.  R. McKitterick, The Frankish Kingdoms under the Carolingians, 751–987, Harlow, 1983, p. 237; Dudo, ed. Lair, pp. 168–9: p. 168: ‘locum qui dicitur ad Sanctum Clerum’. Douglas, ‘Rollo’, p. 129, discusses the possibility that Dudo invented the occasion using a meeting between Duke Richard I and Lothar at St-Clair as his inspiration.

11.  Dudo, ed. Lair, pp. 168–9; reprinted in Privat (ed.), Normandie, with French translation, pp. 74–5; compare William of Jumièges, ed. van Houts, i, pp. 64–6, who closely follows Dudo. Dudo, ed. Lair, p. 168: ‘ipsam terram ab Eptae fluviolo ad mare usque, quasi fundum et alodum, in sempiternum’; ‘Tunc Flandrensem terram, ut ex ea viveret, voluit rex ei dare’; ‘ei Britanniam dare, quae erat in confinio promissae terrae’; p. 169: ‘manus suas misit inter manus regis’; ‘Dedit itaque rex filiam suam, Gislam nomine, uxorem illi duci, terramque determinatam in alodo et in fundo, a flumine Eptae usque ad mare, totam Britanniam de qua posset vivere’; ‘Rolloni pedem regis nolenti osculari … jussit cuidam militi pedem regis osculari’.

12.  P. Lauer (ed.), Recueil des Actes de Charles III le Simple, roi de France, 893–923, Paris, 1940, i, no. 92, p. 209–12, p. 209: ‘pro tutela regni’; Privat (ed.), Normandie, p. 75.

13.  Douglas, William, p. 129; MGH Script, xiii, p. 577.

14.  Dudo, ed. Lair, p. 170: baptised ‘comites suos et milites omnemque manum exercitus sui’.

15.  Douglas, ‘Rollo’, p. 133; Bates, Normandy, pp. 8,13; Dudo gives ‘Gisla’ or Gisela as the wife’s name. The name Popa is suspiciously similar to Papia, wife of Richard II.

16.  Privat (ed.), Normandie, p. 72.

17.  Bates, Normandy, p. 9.

18.  Bates, Normandy, p. 13.

19.  William of Jumièges, ed. van Houts, i, p. 132.

20.  Bates, Normandy, p. 14.

21.  Searle, Predatory Kinship, e.g. pp. 131–42.

22.  William of Jumièges, ed. van Houts, ii, p. 8.

23.  William of Jumièges, ed. van Houts, ii, p. 6; Bates, Normandy, p. 73.

24.  William of Jumièges, ed. van Houts, ii, p. 28.

25.  Douglas.

26.  William of Jumièges, ed. van Houts, ii, p. 30.

27.  Bates, Normandy, p. 113.

28.  K. Thompson, ‘The Norman aristocracy before 1066: the example of the Montgomerys’, Historical Research, lx, 1987, pp. 251–63, pp. 251–2, 255: Roger II Montgomery called himself ‘ex northmannis northmannus’.

29.  William of Jumièges, ed. van Houts, ii, p. 46.

30.  William of Jumièges, ed. van Houts, ii, pp. 48, 60, 82.

31.  Douglas, William, p. 379. His mother’s relatives are referred to as ‘pollinctores’ which means embalmers: Orderic Vitalis in William of Jumièges, ed. van Houts, p. 124.

32.  William of Malmesbury, Chronicle of the Kings of England, ed. J.A. Giles, London, 1895, p. 259.

33.  William of Jumièges, ed. van Houts, ii, p. 92; compare William of Poitiers,ed. Foreville, p. 22: castles built in ‘seditious zeal’.

34.  William of Jumièges, ed. van Houts, ii, p. 96.

35.  William of Jumièges, ed. van Houts, ii, p. 92.

36.  Douglas, William, p. 50; William of Poitiers, ed. Foreville, pp. 12–18; Wace, Le Roman de Rou, ed. A.J. Holden, 3 vols, Paris, 1971; for a translation see E. Taylor (ed.), Master Wace, his Chronicle of the Norman Conquest from the Roman de Rou, London, 1837: pp. 18–27; William of Poitiers, ed. Foreville, p. 19.

37.  Bates, Normandy, p. 74.

38.  William of Jumièges, ed. van Houts, ii, p. 104.

39.  William of Poitiers, ed. Foreville, pp. 73–5.

40.  Wace, ed. Holden, ii, p. 80, l. 5205–6: ‘de lances fierent chevaliers/e od les ars traient archiers’; Wace, ed. Taylor, p. 60.

41.  William of Jumièges, ed. van Houts, ii, p. 150; William of Poitiers, ed. Foreville, p. 99.

42.  Symeon of Durham, ‘History of the Kings’ in Stevenson, viii, pt II, 1855, p. 547.

43.  L. Thorpe (ed.), The Bayeux Tapestry and the Norman Invasion, London, 1973, p. 46.

44.  J.B. McNulty, ‘The Lady Aelfgyva in the Bayeux Tapestry’, Speculum, lv, 1980, pp. 659–68; M.W. Campbell, ‘Aelfgyva: the mysterious lady of the Bayeux Tapestry’, Annales de Normandie, xxxiv, 1984, pp. 127–45; Bayeux Tapestry, pl. 17, 18.

45.  Bayeux Tapestry, pl. 19–24; Bates, Normandy, p. 83.

46.  Bayeux Tapestry, pl. 25–6.

47.  William of Jumièges, ed. van Houts, ii, p. 160; William of Poitiers, ed. Foreville, p. 103.

48.  Barlow (ed.), Vita, p. 53.

49.  Bates, Normandy, 241–3.

50.  Bates, Normandy, p. 57.

51.  K. Thompson, ‘Family and influence to the south of Normandy in the eleventh century: the lordship of Bellême’, JMH, xi, 1985, pp. 215–26; Bates, Normandy, p. 152.

52.  William of Jumièges, ed. van Houts, ii, p. 128; Douglas, William, pp. 369–70: her remains were re-examined in 1961, but had previously been disturbed, so that there must remain a doubt about the bones being hers.

53.  This granting of the banner has been questioned, e.g. by Bates, Normandy, p. 189, but it is difficult to see why William of Poitiers, ed. Foreville, p. 155, should not be accepted on this.

If you find an error please notify us in the comments. Thank you!