by Marc Morris
Meanwhile Odo, bishop of Bayeux, and William fitz Osbern were administering their prefectures in the kingdom … They burned with a common desire to keep the Christian people in peace, and deferred readily to each other’s advice. They paid the greatest respect to justice, as the king had admonished, so that fierce men and enemies might be corrected and brought into friendship. The lesser officials were equally zealous in the castles where each had been placed.
Orderic, confronted by this passage, replaced it with his own:
Meanwhile, the English were groaning under the Norman yoke, and suffering oppressions from the proud lords who ignored the king’s injunctions. The petty lords who were guarding the castles oppressed all the native inhabitants of high and low degree, and heaped shameful burdens on them. For Bishop Odo and William fitz Osbern, the king’s vicegerents, were so swollen with pride that they would not deign to hear the reasonable plea of the English or give them impartial judgement. When their men at arms were guilty of plunder and rape they protected them by force, and wreaked their wrath all the more violently upon those who complained of the cruel wrongs they suffered.5
Clearly, not much common ground. The only point on which Poitiers and Orderic seem to agree is that the newcomers were based in castles. The Normans, as we’ve seen, had started digging in from the moment of their arrival. We hear of new fortifications being established at Pevensey, Hastings, Dover, London and Winchester, and we can make a strong circumstantial case for supposing they were also begun at other places, such as Canterbury, Wallingford and Berkhamsted. What form these castles took is a matter of debate. As every schoolchild knows, the common-or-garden model in Norman England was the so-called ‘motte and bailey’ – a giant mound of earth to support a wooden tower, paired with a shallower but more extensive enclosure to house and protect the castle’s other buildings; around three-quarters of all known sites conform to this type. It could be that the Normans were establishing such castles from the first – the Bayeux Tapestry shows men erecting a motte at Hastings, and a motte can still be seen on the site today. But, then again, the Tapestry depicts every castle in this way, so it could be that drawing a motte was simply a convenient artistic shorthand. While there are surviving mottes at Canterbury, Wallingford and Berkhamsted, excavation has shown that the one at Winchester was added a few years after the Conquest, and at Pevensey, Dover and London there were evidently never any mottes at all. Construction on this scale would have required weeks running into months, so it may be that the very earliest Norman castles were simply enclosures (or ‘ringworks’ as they are often termed): the one on Castle Hill near Folkestone is a good example.6
Whatever form they took, the important point – quite obvious from the fact that the Normans felt obliged to build so many from the moment of their arrival – is that castles in England were a new phenomenon. England, for all the woes that it had suffered in the eleventh century, remained a strong, united kingdom. It had not experienced anything like the political fragmentation that had engulfed the principalities of western Europe. Its coinage, its courts and its laws were all part of a long-established royal monopoly, and so too were its fortifications – the burhs established by the conquering kings of Wessex in the tenth century. The private dwellings known as burhgeats, associated with thegnly status and claimed by some historians to be castles in all but name, were clearly nowhere near as strong and defensible as the true castles we find on the Continent.7 In England we do not read of rebellions based on fortresses, or find the king besieging his greater subjects in their own homes. When English magnates fall from grace they flee into foreign exile, and resist if they can by raising fleets.
The exception that proves the rule is a tiny handful of castles constructed in the years immediately prior to 1051 by some of the French friends of Edward the Confessor. One was apparently built at Clavering in Essex, and three others were built in Herefordshire as part of an attempt (quite unsuccessful) to keep the Welsh at bay. Their novelty is suitably underlined by the fact the Anglo-Saxon Chronicle, describing the situation in Herefordshire in 1051, employs the earliest recorded example of the word ‘castle’ in English. The word, like the thing itself, was a foreign import. Beyond this there was nothing in England that could be meaningfully regarded as a castle. As Orderic Vitalis later put it in a justly famous passage, ‘the fortifications that the Normans called castles were scarcely known in the English provinces, and so the English – in spite of their courage and love of fighting – could put up only a weak resistance to their enemies’.8
Consequently, we sense the shock in the English sources when castles suddenly start appearing in large numbers. As in France, so too in England, their introduction went hand in hand with oppression. The foreigners who built the castle in Hereford, says the Chronicle, ‘inflicted all the injuries and insults they possibly could upon the king’s men in that region’. This may have been because the local population were forced to build them, or because their homes were demolished to make way for them, or simply because they served as bases for soldiers and knights who would ride out each day to cow the surrounding countryside into submission, indulging in the acts of plunder, rape and violence that Orderic Vitalis describes. Orderic’s account of the activities of the Conqueror’s regents during his absence in 1067 accords far better with our English sources than the panegyric of William of Poitiers. ‘Bishop Odo and Earl William were left behind here,’ groans the Anglo-Saxon Chronicle, ‘and they built castles far and wide throughout the land, oppressing the unhappy people, and things went ever from bad to worse. When God wills may the end be good!’9
Small wonder, then, that the English, as Orderic puts it, ‘plotted ceaselessly to find some way of shaking off a yoke that was so intolerable and unaccustomed’.10 During the year 1067 we hear of several local risings by the English against their new castle-building overlords. In Herefordshire, one of the most powerful English thegns, the aptly named Eadric the Wild, fought back with some success against the new Norman garrisons installed in those original pre-Conquest castles. According to John of Worcester, they frequently devastated his lands, but whenever they attacked him they lost many of their knights and soldiers. At length, in mid-August, Eadric joined forces with two Welsh kings, Bleddyn and Rhiwallon, with whom he ravaged Herefordshire ‘up to the bridge of the River Lugg’, and ‘brought back great spoil.’11
Around the same time, or perhaps a little later, a potentially more serious rebellion took place in the south-east. The men of Kent, either ‘because they hated the Normans’ (William of Poitiers) or ‘goaded by Norman oppression’ (Orderic Vitalis), sent emissaries across the Channel in an effort to persuade Eustace, count of Boulogne, to help them seize Dover Castle. This might seem a surprising appeal, given their relations in the not too distant past – it was Eustace’s attack on Dover in 1051 that had sparked the great crisis of that year, and the count had also fought for the Normans at Hastings: he is, indeed, one of the four individuals credited by the Carmen with the killing of King Harold. Yet his relations with Normandy were not as close or cordial as this might suggest: in 1053 Eustace had joined the rebellion that had tried to topple William and, when it failed, had been forced to surrender his son as a hostage. At some point in the first half of 1067 the two men had fallen out again, for reasons unknown; Orderic refers simply to jealousy between them, while Poitiers says only that, were he to go into details, he would easily convince us that William was in the right. Modern historians tend to assume, in lieu of any better explanation, that the count had been disappointed by the amount of land he had received as reward.
Whatever the reason for the rift, it predisposed Eustace to accept the Kentish offer: he assembled an invasion force and sailed across the Channel at night, intending to take Dover by surprise at dawn. His intelligence was seemingly good, because both Odo of Bayeux and the castle’s commander, Hugh de Montfort, were far away at the time, on the other side of the Thames, and had taken with them most of their troops. The men of Kent wer
e already up in arms, says Poitiers, and would have been joined by rebels from other regions, had the siege lasted as long as two days; but Eustace and his English allies found Dover’s defenders to be more doughty than expected. Rather than wait for more attackers to assemble, the Norman garrison sallied out of the gates and put their foes to flight. Eustace himself, familiar with the terrain, managed to find his way back to a boat and the safety of Boulogne, but many of his men were pursued to the cliffs and plunged to their deaths. The English, meanwhile, scattered in all directions, their plan of replacing one foreign lord with another having come to nothing.12
Lastly there had been trouble in Northumbria. Notionally the north of England remained under the command of Earl Morcar, elected by the northerners themselves in the wake of the successful rebellion against Tostig Godwineson in 1065 (the new earl was at least still in possession of his title in the period immediately after the Conquest). From the very first, however, Morcar had been obliged to share power with others. As we have seen, the 1065 rebellion had been triggered by the murder of Gospatric, head of the house of Bamburgh, and Morcar, mindful of this, had immediately ceded authority above the Tyne to Gospatric’s nephew, Oswulf.
But at the start of 1067, just before his return to Normandy, William had upset this arrangement by granting the earldom of Northumbria, or at the very least its northern half, to a Yorkshire thegn called Copsig. It was an astonishing appointment, for Copsig had previously been Tostig’s lieutenant, and was no less hated in the north than his former master. He had, however, submitted to the Conqueror in a way that Oswulf evidently had not, and by some unknown magic had convinced the king that, when it came to controlling the north, he was the best man for the job. (‘He was’, says William of Poitiers, ‘entirely favourable to the king and supported his cause.’) As it was, the confidence of both parties turned out to be seriously misplaced: within just a few weeks of his arrival in Northumbria, Copsig was ambushed and killed by Oswulf, who personally hacked off his rival’s head. Oswulf might therefore have expected some future day of reckoning with the Conqueror or his regents, but at some point in the autumn of 1067 he too came to an untimely end, run through by the lance of a robber.13
It was none of these risings or rebellions, however, that caused William to return to England, but reports of some wider conspiracy. The details are sketchy, and have to be reconstructed forensically from the often allusive comments of the chroniclers. The general conclusion, however, seems clear: in the last weeks of 1067, the Conqueror learned of a conspiracy against him, organized by the surviving members of the Godwine family.14
Orderic Vitalis gives us the general context. During his stay in Normandy, he says, the king was disquieted by intelligence which intimated that the troops he had left behind in England were about to be massacred as part of an English plot. William of Jumièges provides the more specific but uncorroborated information that the plotters’ intention was to attack the Normans on Ash Wednesday, the beginning of Lent, when they would be walking barefoot to church. In 1068 this day fell on 6 February.
William, therefore, hurried to cross the Channel before the end of 1067, sailing from Dieppe on 6 December despite the rough sea and wintry weather, and arriving safely at Winchelsea the following morning. From there he made straight for London, where he celebrated Christmas and tried to sniff out the conspirators. According to Orderic, he was very gracious to the English lords and bishops who attended him, granting them favours and offering them the kiss of peace. Such behaviour, explains the chronicler, often brings back to the fold persons whose loyalty is doubtful, but at the same time William warned his Norman followers, behind the backs of the English, not to relax their guard for a moment.
Early in the new year, William received the certain intelligence he had been seeking: the conspiracy was based in the south-west of England, in the city of Exeter. The king’s suspicions must already have lain in that direction. Before leaving Normandy he had apparently sent some of his knights across the Channel to investigate the rumours, and those who had gone to Exeter had been ‘ill-treated’. Confirmation came in 1068 when Exeter sent messages to other cities, urging them to join the rebellion – messages which William intercepted. The plot uncovered, the king sent a message of his own, demanding that Exeter’s citizens swear fealty to him.
Orderic Vitalis, presumably following William of Poitiers, says nothing about the identity of the plotters. It is only from the terse accounts of this episode in our English sources that we can see that the ringleaders were the surviving members of the Godwine family, and, in particular, Harold’s mother, Gytha. Last reported bargaining with the Conqueror over Harold’s body, Gytha had evidently gone west after the battle to lick her wounds and plan her revenge. We can well imagine the intensity of her hatred: besides Harold, she had lost three other sons in 1066 (Tostig, Leofwine and Gyrth), and her sole surviving son, Wulfnoth, still languished in a Norman prison.
Gytha evidently had her hopes pinned on a new generation of Godwine men. Harold, of course, had been married to his queen, Ealdgyth, for only a short time before his death; had this match produced any children – one later chronicler claims it produced a son – then they would still have been babies at this point. But the dead king had been married earlier, according to Danish custom, to Edith Swan-Neck, by whom he had multiple offspring – no fewer than five children, at least three of whom were boys.15 These young men, probably in their late teens or early twenties in 1066, had also fled west in the wake of their father’s death, crossing the sea to Ireland. Their plan, along with Gytha, was evidently to restage the successful Godwine comeback of 1052: a mercenary fleet from Ireland, a fifth column in England – and perhaps also an invasion from Scandinavia. William of Poitiers says that the plotters had ‘repeatedly sent envoys to the Danes or some other people from whom they might hope for help’, while Orderic says the plot was ‘supported by the Danes and other barbarous peoples’.16
William was evidently determined to stop this plan before it started. When Exeter refused his demand for fealty, he raised an army and began to march westwards. Orderic notes that in doing so the king for the first time demanded military service from his English subjects. Naturally William still had many Norman troops by his side and stationed elsewhere around the country. But summoning Englishmen to fight by his side had wider implications. It made the rebellion in the West Country a litmus test for loyalty: those who did not aid the king in crushing it would themselves be counted as rebels.
At first it seemed that there would be no struggle at all. As the royal army drew near to Exeter, a delegation of leading citizens rode out to sue for peace, much as other urban leaders had done in 1066. They promised to open their gates to William and to obey his commands, guaranteeing their good faith by handing over hostages. Yet on their return to the city, says Orderic, these men ‘continued their hostile preparations, encouraging each other to fight for many reasons’.17
What were the reasons for this eccentric behaviour? One possibility is that the delegates were simply playing for time, and hoping to hold out until the arrival of their overseas allies. The other scenario, perhaps more likely, is that differences of opinion existed within the rebel ranks. Gytha and the other ringleaders had evidently been able to attract widespread popular support during William’s absence, and perhaps also for a time thereafter. One of the king’s first actions on his return, says John of Worcester, had been the imposition of ‘an unbearable tax’, and this was clearly a key factor for many in Exeter. According to Orderic, even as they defied William, the citizens indicated that they would be willing to pay tax at the customary rate. Some rebels, in short, were merely hoping for better treatment from the Conqueror; others remained determined to see him toppled. For the time being, it seems, the will of the diehards had carried the day.18
Inevitably, therefore, the matter was decided by violence. Exeter was a walled city and on his arrival William found the rebels manning the whole circuit of its ramparts. In a final
attempt to induce a surrender he ordered one of the hostages to be blinded in view of the walls, but, says Orderic, this merely strengthened the determination of the defenders. Indeed, according to William of Malmesbury, one of them staged something of a counter-demonstration by dropping his trousers and farting loudly in the king’s general direction. The siege that followed was evidently hard-fought. Orderic says that for many days William attempted to storm the city and undermine its walls; ‘a large part of his army perished’, adds the Anglo-Saxon Chronicle.
At length, after eighteen days, the citizens agreed to surrender. Orderic, no doubt toeing the line established by William of Poitiers, says that they were compelled to do so because of the Normans’ relentless assault. William of Malmesbury claims that the king’s forces gained entry after a section of wall collapsed – a break which he attributes to divine intervention after the fashion of Jericho rather than any Norman mining operations. The English chroniclers, by contrast, suggest that the surrender came about because of the desertion of the Godwine faction. Gytha, says John of Worcester, ‘escaped with many in flight from the city’, while the D Chronicle, apparently describing the same incident, says that the citizens surrendered ‘because the thegns had betrayed them’. Accompanied by ‘many other distinguished men’s wives’, Gytha sailed into the Bristol Channel and took refuge on the tiny island of Flat Holm. Presumably these pro-Godwine women remained hopeful that their husbands and grandsons would soon be crossing from Ireland.19
With the Godwine party gone, there was nothing to stop the more moderate of Exeter’s citizens from seeking terms of surrender. According to Orderic and Poitiers, these were very favourable: William refrained from seizing their goods and guarded the gates in order to prevent any post-siege plunder. The Chronicle, predictably, offers a more acid assessment: the king ‘made fair promises to them, and fulfilled them badly’. The truth probably lies somewhere in the middle, but that the surrender was negotiated rather than dictated seems clear. On one point, at least, the citizens seem to have got their way: the Domesday Book shows that in 1086 Exeter paid the same tax as it had done ‘in the time of King Edward’.