Gytha and the sons of the Godwin House I

Dublin Danes recruited by King Diarmuid of Leinster would sail wih Harold’s son Godwin, and his brothers Eadmund and Magnus. They were to secure Wessex against the Normans

King William, apparently somewhat overcome by his good fortune, took an extended triumphal leave in Normandy during 1067. Under new, white, sails he set out from Pevensey with booty and an entourage. With this new figurehead thus removed from the kingdom it is not surprising to see other opportunists now stepping forwards. In Herefordshire a thegn known as Eadric the Wild joined forces with a couple of Welsh petty-kings and raided the borders to acquire whatever the Norman mercenaries had not plundered, but with no claims to make on the English Crown. He was simply assuming that nature abhors a vacuum. The standing down of much of the invasion force made it impossible to apprehend him. Meanwhile in the troubled North, earl Copsig (who seems to have ingratiated himself with King William), who had been appointed as the ‘strong man’ required to pacify ever-troublesome Northumbria, proved incapable of the task: he was ambushed and killed. Having thus forfeited legality the area sank (not for the first time) into general lawlessness. Over the waters Eustace of Boulogne was secretly making his own preparations.

Where, in all this, were Odo of Bayeux and William fitzOsbern? Well their task was to secure the Channel coast and to administer the south but they both appear to have been busy enjoying conquests elsewhere for when Eustace slipped across the Channel with his own expeditionary force, Odo was nowhere near his caput of Dover. The panegyrists try to tell us that the people of Dover were so abused by Odo and his hearth-troop that they ‘invited’ Eustace to liberate them, a theme which has supplied so many historians with visions of English nationalism discarding the yoke and shackles of novel oppression, conveniently ignoring Eustace’s French identity! In point of fact when the Boulonnais forces arrived there was no evidence at all of collusion and the remaining Norman/French garrison of Dover put up a spirited defence, aided by Englishmen, then sortied, drove off the attackers and pursued them into the night. As most of the Boulonnais expedition had no knowledge of the area, they were, it seems, easily cornered; some were driven over cliffs and others swamped and drowned as they attempted to re-embark. Count Eustace and some of his hearth-troop escaped, but then he knew the territory: in 1051 he had fallen foul of the townsfolk of Dover, which probably helps explain their eagerness to help the Norman garrison in 1067.4 Here we see no clear ‘nationalist’ distinction between the forces.

It can only be speculation, of course, but my evaluation is that Eustace did not do his homework, he did not prepare. If he landed at Dover he cannot have found it easy to land his destriers and it may be that he was seriously lacking in heavy horse? He was certainly crossing against prevailing winds and currents, making for a very difficult passage, and his fleet cannot have been large or elaborate, otherwise advanced news of its preparation would have spread. Maybe he anticipated Odo’s assistance and a cosmetic siege rather than a battle but he clearly had not remembered the tactical formula which had benefitted Duke William the year before. As a ‘castle’ Dover would have had a strong contingent of crossbowmen as well as a heavy horse garrison. A deadly discharge of quarrels followed by a charge of the heavy horse would easily have broken besieging French infantry, especially if in the open. As to Odo, well he could take the credit either way, for preventing the coup (as happened) or for not opposing it, the garrison was his. As I have speculated elsewhere, it may be that he was in collusion with Eustace, William being out of the way? One possibility for the Bayeux Tapestry is that it was intended to celebrate a joint enterprise by Eustace and Odo. If this was it, failure would account for the Tapestry being quickly hidden away and then lost for some centuries. Odo would have needed to destroy such evidence. Maybe some of the garrison were William’s men, not Odo’s, and so he needed to act with circumspection, only committing himself when the case was certain? Maybe he reckoned without English loyalty to the Crown?

Later in this same year Harold’s mother, Gytha, appeared on the scene with sons of the Godwine house and an Irish mercenary fleet. She had massive estates around Dartmoor and probably felt safe in the West Country, which seems to have been settled by English incomers under a novel security system not unlike the Anglo-Irish ‘plantations’ of Elizabeth I’s reign. If so there would also have been a discontented Damnonian residual population susceptible to promises of loot and revenge against both English and Normans, so Gytha set up her caput at Exeter and there appear to have been rumours of a massacre (said to be of ‘Normans’, if there really were so many of them in residence) early in 1068. Intelligences had by then reached Normandy and in spite of the dangers of the season King William took ship early in December 1067 to be in London during Christmas.

The remnants of Harold’s family had been particularly affected by the defeat at Hastings. With the fall of the king and both his brothers, they had been deprived of their senior representatives. Harold’s sons and heirs were as young and untried as Atheling Edgar. Faced with a stark choice between these boys, the majority of the leading Englishmen had returned not unnaturally to their familiar loyalty to Alfred’s dynasty. Deprived of political influence and military support, Harold’s family were forced to retreat before William’s advance, and seek refuge in their lands in south-west England. All the bright hopes of the beginning of the year had been dashed before its end, but the family were Harold’s heirs and fully intended to make every attempt to restore their fortunes. They also had the resources to do this. Countess Gytha’s offer for the return of Harold’s body provides an indication of the wealth still available to the family.

The family began to construct a basis for their planned return to power in the south-west, beyond William’s reach, probably under the direction of the redoubtable Countess Gytha. They chose Exeter, the fourth largest town in England, as their base. It was a wise choice as a city with direct links by sea to sources of possible support in Ireland and Denmark. In addition, the family held wide lands in the surrounding counties of Devon, Somerset and Cornwall. There they began repairing the town fortifications and securing support among the thegns of the region. They were apparently supported by a remnant of Harold’s huscarls, perhaps those occupied on other duties during the Hastings campaign or survivors of the battle, and these men provided a nucleus of trained soldiers for a new army. The intention of the family was to make a bid for the throne based on the right of an atheling, or king’s son, to succeed his father on the throne. King Harold’s sons, Godwine, the eldest, Edmund and Magnus, were all eligible for the throne on this basis. It has been held that their mother Edith’s marriage more Danico to the then Earl Harold disqualified them from the kingship. However, the succession of Harold ‘Harefoot’, son of Cnut and Aelfgifu of Northampton, in 1035 shows that such descent was not necessarily a disqualification. Not surprisingly, they appear to have had no intention of supporting Atheling Edgar, who had already been put aside by Harold and who had now passed into Norman captivity.

The evidence for a revival by Harold’s family is scarce but not entirely lacking. Countess Gytha seems to have granted land at Warrington in Devon to Abbot Sihtric of Tavistock at this time, perhaps in return for his support. This appears to have been effective, as William of Malmesbury tells us that Sihtric later became a pirate, probably indicating that he joined the raiding fleet of King Harold’s sons. Abbot Ealdred of Abingdon was another supporter of Harold’s family, in breach of an oath of allegiance to King William, and he later travelled abroad with them. The fact that Ealdred supported Harold’s family is confirmed by the abbey’s own chronicle, which connects his opposition to William specifically with that of Countess Gytha. It is even possible that it was Gytha’s advice that the abbey’s thegns listened to when they went armed to join what is termed a gathering of William’s enemies. The Abingdon Chronicle places this latter event with Bishop Aethelwine’s rebellion in 1071, but there is no strict chronological sequence in this source and thus no reason to place it in 1071 rather than 1067 or 1068. Indeed, it is possible that the men of the abbey’s forces were on their way to Exeter when they were intercepted by Norman forces during William’s offensive of winter 1068. Another of the family’s supporters may have been Abbot Saewold of Bath, who fled to Flanders at the same time as they did, taking his valuable library with him. The family must also have rallied others to their cause and recruited support locally from their own lands and possibly former royal estates. It is possible that this scenario provided an occasion for the appropriation by Queen Mathilda of the wide estates of Brihtric, son of Aelfgar, which were situated in the region. If Brihtric had been among Gytha’s supporters at Exeter, William may have ordered the forfeiture of his estates and presented them as a gift to his wife when he met her at Winchester in Easter 1068 for her coronation, just after his campaign against Harold’s family. In addition, Orderic Vitalis records that envoys were sent from Exeter to urge other cities to join its stand and this may have been part of the family’s process of rallying support. They may also have sought aid from Swein of Denmark, but, if so, none was apparently forthcoming.

The opposition to William developed and fostered in this way by Harold’s family during this period was not a national movement. At this stage, the prospect of Godwine, Harold’s son, as king of England did not receive much support outside Harold’s former earldom. Godwine was young, unproven and little known in comparison with his father. The family could probably rely on old loyalties in Wessex to provide them with support, but could not count on such support beyond its borders. Nevertheless, the rebellion in the south-west inspired by Harold’s family was one of the most significant of those which occurred in William’s absence.

The situation in England was very tense and confused by the end of 1067, with widespread unrest but no recognized central authority among the English. King William’s control was limited largely to the south-east, and his soldiers were probably little in evidence outside an area from Kent to Hampshire and from the Channel coast to East Anglia. He himself had been absent since March 1067 and his forces had been left without his direct oversight. This situation allowed many separate rebellions to develop under their own leaders and with differing aims. By the time William returned to England in December 1067, many areas of the country were in open revolt, including Dover, the Welsh Borders and much of Northumbria, as well as the south-west. In most of these areas the uprisings were locally inspired and led by fairly minor local figures and until the return of the captive English leaders from Normandy, it was the rebellion in the south-west which provided the principal rallying point.

The significance of Harold’s family’s stand is made clear by the fact that, despite the wide extent of these rebellions, William on his return chose to strike first against Exeter. He did so almost immediately, undertaking a difficult winter campaign early in 1068. This haste was undoubtedly because William considered this to be the greatest threat to his position. It is likely that William reached this conclusion not simply because it was indeed the most substantial threat he faced, but also because it involved the sons of his dead rival, King Harold. Although Godwine, Edmund and Magnus were still young, they were old enough to lead military forces later in 1068 and therefore already represented a potential menace to William’s still insecure throne. They had refused to submit to William in 1066, unlike Atheling Edgar, and were obviously dangerous threats to the legitimacy of his kingship. Taking these things into account, the reason that William directed his entire army against Exeter in the depth of winter immediately on his return is readily understandable.

Correspondingly, the strength of Exeter’s resistance to William surely indicates something more than Orderic’s statement that it arose from a desire to preserve the laws and customs of the town. This purpose would surely be much better served by a speedy submission and a request for a writ from the new king securing these customs, as in the case of London. The real reason may have been loyalty to King Harold’s family; Countess Gytha was certainly present, if not also his sons, in the town during the siege. The attempts of the citizens to recruit wider support certainly speak of more serious reasons. Whatever the reason, William spent some eighteen days laying siege to the town, and a large part of his army perished in the process. The eighteen-day defence put up by the citizens demonstrates that the English burh was still a very effective defensive structure when properly garrisoned, despite the fashion among some for considering the Norman castle superior. Nevertheless, William’s persistence paid off and the city was forced to submit, according to the Chronicle, ‘because the thegns had betrayed them’. This may indicate that the citizens expected relief or aid from the local thegns which did not come. Alternatively, it may mean that some of the thegns who made up the garrison deserted. John of Worcester portrays the garrison as consisting of the citizens and only some thegns. In either case, this failure probably reflects the political and military inexperience of those directing the defence, possibly Harold’s sons.

Whether King Harold’s sons were actually present in Exeter during the siege, escaping before the surrender as his mother did, or whether they were perhaps among those who failed to bring relief is unknown. John of Worcester mentions Gytha’s escape with ‘many others’, and perhaps this included Harold’s sons and other family, and Abbot Sihtric of Tavistock. Escape by ship down the river Exe seems most likely given the siege conditions on land and Gytha’s subsequent move to Flatholme. William’s swift and decisive action against Exeter had severely undermined the family’s position in the south-west, and his subsequent pursuit into Cornwall compelled them to take flight overseas. Countess Gytha and other ladies, probably including Harold’s daughter, Gytha, and his sister, Gunnhild, went out to the island of Flatholme in the Bristol Channel where they were relatively safe but prepared for a quick return. Despite the set-back at Exeter, King Harold’s sons were not yet ready to give up and, perhaps recalling stories of their father’s previous exile, they sailed to Dublin with their huscarls to seek aid from King Diarmait. He still ruled Dublin and commanded its mercenaries, and he welcomed them as he had their father. He and ‘his princes’, perhaps a reference to his son, Murchad, provided them with support, possibly in memory of their father but more likely in return for some of the family treasure. Indeed, part of this treasure, ‘the battle standard of the King of the Saxons’, was presented by King Diarmait to his ally Toirdelbach, King of Munster, that same year. This may have been the standard of the deceased King Edward, perhaps the Dragon of Wessex shown in the Bayeux Tapestry, as King Harold’s own personal banner undoubtedly fell into Norman hands after Hastings.

With Diarmait’s assistance, King Harold’s sons returned unexpectedly from Ireland in the summer of 1068. Their naval force, recruited in Dublin, landed at the mouth of the river Avon and ravaged the surrounding district. They then attempted to take the town of Bristol, perhaps to replace their lost base at Exeter with one closer to their new supporters in Ireland. The citizens of Bristol proved unsympathetic and they were forced to attempt to take it by storm, which suggests that they had a considerable force under their command. The citizens resisted them fiercely, perhaps fearing the same fate as Exeter when William retaliated or simply distrustful of the brothers’ Hiberno-Norse mercenaries. Their assault unsuccessful, the brothers were forced to take what booty they had gathered and move down the coast to Somerset, perhaps near the mouth of the river Parret. There, Godwine held lands at nearby Nettlecombe and Langford Budville, and the Taunton mint could be raided. Again the brothers met with resistance, this time led by Eadnoth the Staller, who had previously served their father but had now submitted to William. In the battle which resulted, many fell on both sides, including Eadnoth and probably one of the king’s sons as only two appear to have returned a year later. The brothers were apparently victorious in this battle but it did not provide the swift breakthrough or accession of support they hoped. William’s decisive action at Exeter had been too effective in cowing English resistance. The surviving brothers, Godwine and perhaps Edmund, though the latter’s name is nowhere recorded, returned to Ireland with their remaining forces and a considerable amount of loot. The reference to spoils in John of Worcester may reflect these events, or perhaps those of the following year as he mentions raids in Devon and Cornwall and Harold’s sons were to raid that area in 1069.

King Harold’s sons were still not prepared to give up their attempts at the restoration of their fortunes in England. In the next year, 1069, at midsummer they secured another large fleet from Dublin, consisting of over sixty ships this time. Orderic says ‘they landed first at Exeter’, and if this is correct and not a confused reference to their possible earlier sojourn there this probably represents an attempt to revive their earlier success. However, the new Norman castle in the town effectively prevented any further rebellion on the part of the citizens. In addition, William’s relatively lenient treatment of their rebellion in 1068 encouraged them not to risk these terms by further insurrection. The brothers then appear to have turned to raid the south coasts of Devon and Cornwall, perhaps in frustration at their failure to rouse Exeter. Domesday Book records lands laid waste there by the Irishmen of their fleet between Kingsbridge Estuary and Bigbury Bay. Similarly, waste recorded in the Lizard peninsula may also be attributable to their activities.

The brothers then rounded Lands End and came ashore in the mouth of the river Taw in north Devon. They laid waste the countryside around Barnstable and moved inland, perhaps again heading for Godwine’s estates at Nettlecombe and Langford Budville near Milverton in Somerset. However, the lack of any opposition so far on this trip appears to have made the brothers incautious, and they were caught out by a large Norman force under Count Brian. In the battle or series of encounters that followed, most of the brothers’ best men were slain, as many as 1,700 according to William of Jumieges and including a number of thegns, and only a small force escaped at nightfall to return to Ireland.

This disaster finally put an end to the immediate attempts by King Harold’s sons to reclaim the English throne. They were unable to recruit further mercenary forces, perhaps because their resources of treasure were running out, but principally because of their lack of success to date, and especially in 1069. It was probably at this time that Countess Gytha and her ladies finally abandoned their refuge on Flatholme and sought refuge at St Omer in Flanders. King Harold’s dynasty had abandoned its hopes of regaining the throne directly and, as circumstances were to prove, finally.

Many have seen the family’s hopes as foolish, and the actions of King Harold’s sons as irrelevant. However, England during the period 1067–9 was in chaos and the whole country seethed with rebellion. During this time, several claimants to the throne competed for support, including Atheling Edgar, Earls Edwin and Morcar, using Harold’s son by Alditha, Swein of Denmark, and William himself. Therefore, it was not unreasonable for King Harold’s sons also to enter this contest, and to feel that they had as good a chance as any of success. The failure even to come near to achieving their aim was due in the main to William’s decisive reaction to the very real threat they posed. His campaign in the south-west decisively nipped their schemes in the bud. Subsequently, they were forced to base themselves abroad and use mercenary troops, making it difficult for them to win any real support in England. Another significant factor was their failure to win the support of their cousin, Swein of Denmark, who was clearly intent on pursuing his own claim to the throne rather than supporting that of his cousins. The fact that many of Harold’s key supporters had fallen at Hastings and that the Normans controlled a large part of the family lands in the south-east severely handicapped them. The brothers inexperience in warfare was also a contributory factor to their failure, although the long defence of Exeter and their victory over Eadnoth suggest this was not decisive. Perhaps if either Gyrth or Leofwine, with their greater authority and experience, had survived the battle at Hastings things might have been different.

The brothers’ bid for the throne was over, but this was not the end of the story. A considerable amount is known about the fate of the remnants of King Harold’s family after their final withdrawal from England in 1069. The elderly Countess Gytha, with Harold’s sister, Gunnhild, probably settled in quiet retirement at St Omer in Flanders, where Count Baldwin VI apparently received them charitably as relatives of his aunt Judith and in spite of their rivalry with his brother-in-law, William of Normandy. Countess Gytha’s remaining treasure may have helped to persuade Baldwin to provide them with refuge. Thereafter, the royal ladies performed good works, and the death of the king’s sister, Gunnhild, was recorded at Bruges in 1087. She bequeathed a psalter with Anglo-Saxon glosses to St Donation’s in Bruges and this book, known as ‘Gunnhild’s Psalter’, was still there in the sixteenth century. She also donated a collection of religious relics to St Donation’s, most notably the mantle of St Bridget. A copy of Aelfric’s works donated to St Bertin’s may, perhaps, have been a legacy of Countess Gytha.

It seems likely that King Harold’s sons escorted these ladies to Flanders, as it would have been rather risky for them to navigate the Norman controlled Channel alone. Although Baldwin’s hospitality was undoubtedly extended to the ladies of Harold’s family, it might seem unlikely that he would also provide refuge for Harold’s sons. After all, he had close ties with King William, and Harold had also been responsible for the death of his aunt Judith’s husband. It is possible that Baldwin intended to use them as a form of insurance should his alliance with William fail. It should be recalled that the Flemish counts had a long history of hostile relations with England, and William, of course, was now King of England. If this assumption is accurate, then the presence of these exiles must have caused King William considerable unease. Indeed, the arrival of this group in Flanders, perhaps in late 1069 or early 1070, may have prompted William to depose Bishop Aethelric of Sussex on 24 May 1070, in case he became a fifth column in support of their return. He was, after all, a relative of the family and based in the ancient family heartland just across the Channel from them. Therefore, it became imperative to remove him for political reasons. It seems likely that this was the reason for Papal concern about this particular deposition, as expressed in a number of later Papal letters. On 16 July 1070 Baldwin VI died and a succession dispute broke out between his infant sons, supported by King William, and his brother, Robert the Frisian. This dispute ended on 22 February 1071 at the battle of Cassel, the victory falling to William’s enemy, Robert, who became the uncontested Count of Flanders. The threat of a descent by Harold’s sons on Sussex from a hostile Flanders may have contributed to the unusual organization of the Norman castellanries in the Sussex rapes.


At first the surviving citizens of Exeter sent a delegation to humbly offer surrender but on arrival William found the walls manned and defiant, explicable if Damnonian elements had united with the Irish invaders. The siege which followed was apparently bitter and bloody but when the Godwine faction circumspectly deserted it came to an end. To their surprise the now helpless citizens were reasonably well treated (it had hardly been their fault) and the King then moved on into Cornwall, a further indication of the ethnic and regional nature of the support obtained by Gytha and her Irish mercenaries. So that is how the English and Norman colonists presumably regained control of their estates in Devon and Cornwall. Disbanding his army, William celebrated Easter at Winchester, distributing those lands now forfeited by Gytha and her followers and also selling Northumbria to a petitioner (Gospatric) for a handsome sum. Already the Crown was in need of ready money having taken so much loot to Normandy in 1067. At Whitsun Mathilde, William’s consort, was crowned Queen in England, a further indication of settled conditions.