King Harold’s Sister Gunhild (d. 1087), a Royal Exile in Flanders^*

Abstract

After the Norman Conquest of England, as the country gradually fell under the control of the newly crowned king, William the Conqueror, the English landed elite was in turmoil. Many of its men had died on the battlefield at Hastings and survivors were faced with dispossession. The Norman king handed out vast estates to his continental followers with the result that the indigenous elite lost their lands.¹ Much has been written about Englishmen’s loss of land and the allocation of vast tracts of land to William’s continental followers.² Much less, however, has been written about the fate of English women, including royal women, though this is a state of research that is fortunately changing.³ The present contribution focuses on the fate of one royal woman, Gunhild (d. 1087), sister of King Harold Godwineson (d. 1066).⁴ Despite this focus, it will become clear that the loss of the last English king’s life during the battle of Hastings had momentous consequences for other members of his family, who, as we shall see, scattered far and wide to England, Ireland, Scandinavia and Flanders. These men and women all reacted to the Norman Conquest differently and followed a wide variety of survival strategies in response to this political disaster. Most of their movements can be followed patchily in documentary or narrative sources, while Gunhild’s career, unusually, can be reconstructed on the basis of her Latin obituary, the longest inscribed on lead for a lay woman in the eleventh century. It has survived on a lead plaque that was buried with her in her tomb in the walls of the comital chapel of the canons of St Donatian at Bruges.⁵ Gunhild’s exile in Flanders raises questions about the opportunities available for English royal women after the Norman Conquest, either at home or abroad. Much depended on whether an elite English woman was married, widowed or single, and on how wealthy she was, in particular in terms of movable wealth; her life-cycle stage and the position of any protector were crucial in determining where she might end up. For Gunhild, an important question would be where she might find shelter, physical and moral, when, deprived of close kin, she ended up abroad. Was her life in exile as a woman religious a foregone conclusion? Who protected her, in what circumstances, and at what political and emotional cost? And who remembered her and in what way? These are the questions that this article addresses, seeking answers by putting Gunhild’s lead burial plaque and Latin obituary in its material and literary context, by analysing Gunhild’s life on the Continent as a case-study of an elite female survivor of the Norman Conquest, and by investigating her memorial tradition.

I

On 31 March 1786 a lead burial plaque was discovered in the church of St Donatian at Bruges with a long Latin obituary notice of Gunhild, sister of King Harold, who had died on 24 August 1087.⁶ Among the remains of a wooden coffin, still containing some bones, the lead plaque was found near where the head would have lain. The coffin itself was inside a cavity in the church’s wall on the north side at a height of about one metre from the floor. The coffin was opened in the presence of the bishop of Bruges and later sealed and reburied. The church was destroyed in 1799 and in 1804, during the upheavals of the Napoleonic occupation of the Low Countries, its ruins were sold. Only the lead plaque survived; it is now kept in the treasury of St Salvator’s cathedral in Bruges.⁷ The lead has badly corroded and the lettering faded but, because several transcriptions were made soon after its discovery, the text has survived more or less intact.⁸ Despite the fact that the plaque is reasonably well known, and is regularly referred to in scholarship on the Godwine family, its text has not been studied in its own right nor has it received the attention it deserves for its significance for the history of the Norman Conquest of England.

Sympathy for the Anglo-Saxon ‘doomed elite’ can be detected in the glowing terms in which Gunhild’s parents are described on Gunhild’s plaque.⁹ What strikes the reader most is the precise information about her English noble origins, the powerful position and wealth of her parents and the pinnacle of power reached by her brother Harold:

Gunhild, born of most noble parents, was English. Her father was Earl Godwine, who held a very large part of England under his military command, and her mother Gytha was born of famous Danish ancestry. As a young girl she took a vow of virginal chastity and wishing for a spiritual union she rejected marriage to several noble princes. She had already reached marriageable age when England was conquered by William, count of the Normans, who killed her brother Harold, king of the English.

Godwine’s status as an earl (which he held for more than three decades) and his military command (over Wessex, c.1018–53) is here recorded, as well as his wife Gytha’s noble origins in Denmark. She was the sister of Ulf Thorgilsson, who was married to Estrith, sister of King Cnut (1016–35). The marriage between Godwine and Gytha took place probably c.1022/3, five years after Godwine had been made earl.¹⁰ More politically explosive is the statement, without a reference to battle, that William, pointedly referred to as count of the Normans, who had conquered England, had killed her brother King Harold.¹¹ Assuming that her obituary was written shortly after her death in late August 1087, William was then still king (he died on 9 September) and Harold had been dead for more than two decades. The accusation that William had slain her brother, a legitimately chosen and crowned king, is an acknowledgement of the deep unease felt by Gunhild, many English, and indeed others in Europe about William’s invasion of England and his taking of its crown.¹² According to the Anglo-Saxon Chronicle, version D, in 1067 (for 1068), after the siege of Exeter, Harold’s mother Gytha in the company of ‘the wives of many good men’ (‘manegra godra manna wif’) fled from the island of Flatholme in the Bristol Channel to Saint-Omer in Flanders.¹³ That our Gunhild was almost certainly among them we only know from her burial plaque, which records—after the section quoted above—that ‘she fled her homeland and lived for some years in exile at Saint-Omer in Flanders’. This was not the first time the Godwine family had sought refuge there since the 1040s.¹⁴ In 1065, having been exiled by Edward the Confessor, Gunhild’s brother Earl Tostig of Northumbria and his Flemish wife Judith (widowed in 1066) had taken refuge in Flanders, where the couple were given a house and estate as well as the income of the town of Saint-Omer.¹⁵ It is therefore reasonable to assume that when in 1068 Gytha and Gunhild decided to leave England, Judith, as daughter-in-law and sister-in-law respectively, may have facilitated their stay at Saint-Omer.¹⁶

The town was an important commercial centre with easy access to the North Sea.¹⁷ Its main church in the centre was that of Our Lady, an establishment of secular canons, while two miles further out was the well-known Benedictine abbey of Saint-Bertin.¹⁸ Women were forbidden entry to the Saint-Bertin complex and there is no evidence for communities of women religious attached to either house. Presumably Gytha, Gunhild and their servants lived either on the hospitality extended to them by Judith or whoever was their Flemish host, or independently on the considerable moveable wealth that they had been able to bring with them—an issue I will return to shortly. Gunhild’s age at that time is unknown, though her obituary tells us that ‘she had already reached marriageable age when England was conquered’. As she must have been born before her father Godwine’s death in 1053, this suggests that she was at least 13 in 1066 or 15 at the time of exile, but she could have been in her late teens or early twenties.¹⁹ At some stage, so the plaque records, Gunhild left Saint-Omer: ‘She went to Bruges where she lived for a few years and from there crossed to Denmark’. The move north from southern Flanders was probably prompted by the loss of the women close to her. Her mother died almost certainly at Saint-Omer, though in an unknown year.²⁰ In c.1070 her sister-in-law Judith left Flanders for Bavaria, where she married Welf IV as her second husband.²¹ Around the same time Flanders experienced a civil war.

Count Baldwin V (1035–67) had followed a policy of near ­neutrality when his son-in-law William (married to his daughter Matilda) prepared the invasion of England.²² He did not prevent his nobles from taking part, but neither did he encourage them to do so. The twelfth-century rumour, recorded by William of Malmesbury, that he provided his son-in-law with soldiers in return for a cash sum, may arise from backdating the series of twelfth-century Anglo-Flemish treaties to the Conquest.²³ As guardian of the under-age king of France, Philip I (1060–1108), Count Baldwin V’s primary responsibility was his own principality and the realm of his protégé.²⁴ On 1 September 1067 Baldwin V died and within three years his eldest son Baldwin VI too died, on 17 July 1070. Baldwin VI’s eldest son Arnulf III succeeded but his rule was challenged by his paternal uncle, Robert the Frisian. Despite the support of Philip I of France, most of the southern Flemish nobles and an army sent by King William, at the battle of Cassel in February 1071 Arnulf III was killed and Robert the Frisian emerged as victor.²⁵ Thereafter Count Robert never forgave his royal brother-in-law William for having supported Arnulf. Relations between the Anglo-Norman and Flemish rulers deteriorated further for three reasons. First, many southern Flemish nobles who had supported the losing side left for England to supplement the Flemish settlers who had helped win the Norman Conquest.²⁶ Secondly, in 1078–9 Robert the Frisian gave shelter to Robert Curthose (d. 1134), eldest son of King William and Matilda, who was at war with his father over his refusal to give him Normandy.²⁷ And thirdly, potentially even more threateningly for King William, the Flemish count allied himself with Denmark. In the late 1070s or early 1080s his daughter Adela had married Cnut IV. Two years after the death of his sister Queen Matilda in 1083, Robert the Frisian prepared a fleet to join Cnut’s invasion of England to oust William as king, a plan scuppered by Cnut’s murder in the cathedral of Odense.²⁸ These events in Flanders provide a context for Gunhild’s movements, from Saint-Omer to Bruges—perhaps seeking firmer protection from Robert the Frisian in his stronghold—and then, after some time, to Denmark, prompted, we may guess, by a wish to seek out her late mother’s Danish relatives.

Gunhild’s departure from Bruges for Denmark can plausibly be linked with Adela’s marriage to King Cnut IV, which is dated to the late 1070s or early 1080s, as it is conceivable that she joined Adela on her way to her new homeland.²⁹ Alternatively, Gunhild, with some of her treasure, may have left for Denmark in the mid-1080s as part of preparations for the Flemish–Danish campaign to invade England. King Cnut’s murder not only prevented the planned invasion of England but also caused his widow Adela to return to Flanders.³⁰ If at this time Gunhild was indeed in Denmark, she may have accompanied Adela on her return to Flanders together with Adela’s son Charles but leaving his two sisters behind. Without any more clues we do not know how long Gunhild stayed in Denmark, though her obituary suggests that it was not long before ‘she returned here [Bruges]’, where she spent the remaining years of her life.

That Robert the Frisian provided protection to King Harold’s sister Gunhild remains politically most significant as implicit support for the house of Godwine and for other Anglo-Saxon victims of the Norman occupation of England.³¹ His actions against William the Conqueror and by implication his sister Queen Matilda may well have had consequences for his mother Countess Adela (d. 1079), as we will see.³² Gunhild’s obituary ends with her death: ‘she died a virgin in the year 1087 on 24 August at the twenty second day of the full moon’. Gunhild’s obituary and the life it describes raise many issues. But ­before pursuing them, the artefact on which they are recorded and its context demand our attention, not least because Gunhild’s plaque was not the only one prepared for a member of the Godwine family.

II

Lead plaques inscribed with basic biographical information in Latin, such as a name and year of death, can be found in early medieval Europe north of the Alps and Pyrenees from the late Merovingian period onwards.³³ They were produced as cheaper alternatives to inscribed stone, which was a legacy of the Roman period. There are some late eighth-century Carolingian stones with incised lettering where the letters were filled with lead.³⁴ The Roman heritage of inscriptions in stone inspired the later northern European practice of recording the dead person’s biographical details in lead.³⁵ It seems to have been a Carolingian practice to bury men and women with a plaque placed on or near their body. Originally lead plaques were probably nothing more than what we might call identification tags, made of durable material that, when still soft, was easy to write on with a stylus.³⁶ Before burial they were placed either near the head of the body, sometimes under the head, or, most often, on the body’s breast. Most lead plaques were in the shape of rectangles or squares but a good number were in the form of a cross, bearing witness to the liturgical rite for death and burial. An important clue to the penitential aspect of the lead plaques and crosses for monks is given by the monastic rule of Lanfranc (d. 1089), William the Conqueror’s first appointee to the see of Canterbury.³⁷ After death the body is sewn into a cloth and his fellow monks ‘lay the body out carefully in the grave, and put upon its breast a written absolution that has been read out by the brethren’.³⁸ Lanfranc, however, does not specify the material on which the ‘written absolution’ should be inscribed, but it is worth noting that—whether made of parchment or of lead—it is put on the shroud and not actually on the body. Penitential lead plaques and crosses were left with the bodies of the dead as evidence that the deceased had confessed their sins, or at least that God had been asked for forgiveness. Increasingly, however, the penitentiary purpose of the cross or plaque that was prevalent in England gave way to a more elaborate biographical account, especially in northern France and Flanders.³⁹ A good example is the Flemish ‘titulus’ of John of Warneton, bishop of Thérouanne (d. 1130). The author of his Life, Archdeacon Walter, explained that he had written the ‘titulus’ for Bishop John on lead and parchment (sealed in a glass container) and that both were buried with him.⁴⁰ Although neither object has survived, the text of the obituary notice was copied into Bishop John’s Life. When parchment and lead tituli were produced for the same person and buried with them, we may wonder whether the text on the lead plaque and on the parchment may have followed the same layout. This question is important because in Gunhild’s case her obituary text was written out on lead as if on parchment.

Gunhild’s plaque measures 24.5 cm x 19.5–20.5 cm, which is about the average size for a lead plaque of a high-status person. It was prepared with lines for a text consisting of twenty-four lines written in Carolingian minuscule script with a layout that resembles that of a parchment sheet. This layout explains some last-minute adjustments about three-quarters of the way down, halfway through line eighteen, starting with the words gustando sed uix, when the writing becomes squeezed as if the scribe realised that he might be running out of space. Among the lead plaques that have survived, this parchment layout is highly unusual. Moreover, the lettering on Gunhild’s plaque is not the uncial script used for lead inscriptions, which tended mostly, but not exclusively, to imitate inscriptions in capital letters on stone.⁴¹ It is in Carolingian minuscule with only some capital letters, and clearly imitates a statement on parchment. Lead was chosen for preservation purposes but the layout was perhaps meant to convey extra authority or intended as a tribute to a person’s high status. Conceivably, the imitation of parchment might indicate that a parchment copy had also been produced, as in the case of Bishop John of Thérouanne discussed earlier.⁴²

Crucially, lead burial plaques with penitential prayers or biographical details, or both, were buried with elite persons, prompting the question: for whose benefit? Hidden in tombs, nobody but the dead person was in contact with them. It has been suggested that the evidence would help the identification of bodies at the Last Judgement.⁴³ But would God not have known who had been buried where? Alternatively, and in my view more probably, it was well enough known (often through direct experience) that graves could be disturbed either on purpose for reburial, or transfer, or accidentally as a result of building work to a church or damage. Any help with identification of bodies, whether at the Last Judgement or sooner, was clearly deemed useful for future generations.⁴⁴ Some biographical obituaries were highly political in their statements, suggesting that the dead, or their families, friends or colleagues, wished to make sure that the circumstances of their death and any significant events in their lives were recorded for posterity. Gunhild’s lead plaque falls into this category; but it was not unique in the Godwine family, as another one has survived, albeit from several (short) generations later, and not in Flanders but in Lincoln.

The so-called d’Aincourt lead plaque from Lincoln was made for William d’Aincourt, a young boy who died in the 1090s.⁴⁵ It was discovered in 1675 in the grounds of Lincoln Cathedral and measures 34 cm x 22 cm. It is thus another fairly large plaque with text covering only the top two-thirds of its length, suggesting that it was meant to be placed underneath the boy’s head. The text is as follows:

Here lies William, son of Walter d’Aincourt who was a kinsman of Remigius, bishop of Lincoln, who built this church. The said William, born of royal stock, died on 30 October, while living in fosterage at the court of William, son of King William the elder who conquered England.⁴⁶

William’s identification by Richard Sharpe as a great-grandson of King Harold put an end to the commonly held interpretation that the reference to ‘royal stock’ (stirps regia) was to kinship with the new Norman king William the Conqueror rather than the English king Harold. The English blood-relationship can be explained in the female line via the boy William’s mother Matilda, married to Walter d’Aincourt (d. before 1116), who was a daughter of Count Alan Rufus (d. 1093) and Gunhild, the daughter of King Harold and thus the niece of the Gunhild buried at Bruges. A genealogical schema provides a visual image of William’s ancestry:⁴⁷

King Harold (d. 1066) = Eadgifu Swanneck (d. after 1066)

|

Gunhild (c.1054–d. after 1094) = Alan Rufus, earl of Richmond (d. 1093)

|

Matilda (c.1073–d. after c.1090) = Walter d’Aincourt (d. by 1116)

|

William d’Aincourt (c.1090–d. by 1100)

At the time the cathedral at Lincoln, although an old institution, was a relatively new complex, the episcopal seat having been transferred from Dorchester-on-Thames in 1072.⁴⁸ There is no doubt that technical and literary expertise was available for the production of lead tablets and the composition of the texts.⁴⁹ William’s obituary is shorter and less literary than that of his great-aunt Gunhild. The obituary of the boy William is especially significant for its unique revelation that as a boy of recent, albeit short-lived, English royal stock he was brought up at the court of the Conqueror’s son William Rufus (1087–1100).⁵⁰ It is noteworthy that the remark on William’s (English) royal stock is made not in connection with Lincoln Cathedral so much as in connection with his upbringing at the Norman royal court in England. A decade or so after the Conqueror himself had sanctioned Harold’s damnatio memoriae treatment in Domesday Book—Harold is never referred to as king—it is striking how the plaque reminds us, in an oblique way, of William’s condemned great-grandfather.⁵¹ It illustrates how Harold’s daughter Gunhild and in turn her daughter Matilda kept the memory of Harold’s kingship alive, albeit on a tablet buried with William, named after the Conqueror, at Lincoln.⁵² Since he was only a relatively young child the commission for the plaque and its text presumably came from his parents, Matilda and Walter d’Aincourt. In the case of Gunhild’s plaque, before we can come to a conclusion about the authorship and audience of Gunhild’s obituary with any confidence, we need to explore the friendship she established with the canons of St Donatian and her life as a religious woman in their vicinity.

III

St Donatian’s church in Bruges was a house of canons founded either by Count Baldwin I (c.863–79) or by his grandson Count Arnulf I of Flanders (918–65).⁵³ It was an imposing church, built in imitation of Charlemagne’s chapel at Aachen with a massive west end and polygonal core, where forty years later in 1127 Count Charles the Good was murdered.⁵⁴ Badly damaged on that occasion, the church was rebuilt as a Romanesque basilica and then as a Gothic church. Significantly for our purpose, St Donatian was the chapel of the counts of Flanders; it had a provost who from time to time acted as the counts’ chancellor, and a chapter of canons some of whom have been identified as scribes of the Flemish counts’ charters.⁵⁵ Why did the canons of St Donatian support Gunhild by offering her a place of burial? The answer lies, partly, in her wealth. Evidence for her wealth comes from Domesday Book in England and from St Donatian itself. Domesday Book, compiled in 1086, is the only other eleventh-century evidence concerning Gunhild apart from her plaque.⁵⁶ It reports that in 1066 Gunhild held the income of four manors, one in Sussex and three in Somerset, comprising approximately 30 hides, giving her an annual income of just under £30.⁵⁷ By 1086, and probably much sooner after the Conquest, this income had been confiscated and reallocated to Norman holders. Contemporary historians mention her mother Gytha’s vast wealth. William of Poitiers, biographer of William the Conqueror, recorded that after the battle of Hastings the victor refused Gytha’s offer of the weight of Harold’s body in gold.⁵⁸ Probably basing his account on the now lost section of this biography, Orderic Vitalis mentions that when Gytha went into exile she carried with her ‘a great store of treasure’.⁵⁹ His Anglo-Norman contemporary William of Malmesbury added the rumour that Gytha’s wealth could in part be explained as a result of her activity as a slave trader with Denmark.⁶⁰ Presumably some if not all of Gytha’s wealth after her death went to Gunhild. We can get some idea of its size from the archive of St Donatian, which in the fourteenth century kept records of her bequest of gold and silver to the canons.

In 1389 the bishop of Tournai approved the sale of certa jocalia (‘certain jewels’ or ‘precious objects’) given by Lady Gunhild for the repair of the church’s roof, vault and choir.⁶¹ More detail is provided under the same year in the annals published in 1561 by Jacobus Meyerus, a Bruges chronicler, which specified that Gunhild’s gifts consisted of ‘gold and silver’.⁶² Meyerus also refers to a then extant Latin psalter known as Gunhild’s psalter, which contained ‘writings [glosses?] in Old English’ (‘enarrationes linguae Saxonicae’), ‘which here nobody can understand’.⁶³ Meyerus’s information was picked up in 1631 by Johannes Isacius Pontanus in his Rerum Danicarum Historia, where he supplemented it with details from Antonius Schoonhoven, a canon of St Donatian, who had consulted the archives of his house, presumably the paperwork from the sale in 1389. He reported that among the gifts Gunhild had given to the canons in her lifetime and at her death ‘were among others an imperial crown (corona augustalis) and many works of the best authors richly ornamented, of which the Psalter “in the Danish language” only remained’.⁶⁴ The most striking detail from Schoonhoven’s archival work for us is the reference to a crown. Might this have been Harold’s crown, which after his death was recovered by his mother and sister?   ⁶⁵ If so, it must have been a prestigious possession for the canons of St Donatian in 1087. Equally, when the jocalia—perhaps including a crown—were sold to fund the substantial repairs to the church (roof, vault and choir), they must have been of enormous value. The transaction hints at the huge value of the portable wealth that the Godwine women possessed at the time of the Conquest.

Yet there is one problem with this information in that, by the time of the sale in 1389, the canons (and therefore, later, Jacobus Meyerus) had forgotten who our Gunhild was. They confused her with another royal woman: Gunhild (d. 1038), daughter of King Cnut (1016–35) and Queen Emma (d. 1052) and, briefly, wife of Henry, later Emperor Henry III (1039–56).⁶⁶ When this confusion originated cannot now be established. However, this misidentification in itself is strong evidence that the canons had no access to the information contained on Gunhild’s plaque, which is unambiguous about her identity as the daughter of Earl Godwine and sister of King Harold. If a copy of Gunhild’s obituary had been kept by the canons, perhaps as part of her liturgical commemoration, it got lost. In its absence, the scale and value of the bequest of Lady Gunhild was such that it was associated with a better-known imperial woman for whose wealth there would have been a more persuasive rationale than for the sister of a man who was king of England for only ten months.

Gunhild’s movable wealth may have been one reason for her move to Bruges, where the comital castle next to the church of St Donatian might have guaranteed her safety.⁶⁷ In Saint-Omer the monks of Saint-Bertin provided ‘banking’ services that might have allowed Gytha and Gunhild to store their possessions in their safekeeping. It is well known that monastic houses acted as banks providing loans and mortgages from the eleventh century.⁶⁸ From the mid-twelfth century there is evidence that all Peter’s Pence collections from England were sent to the abbot of Saint-Bertin at Saint-Omer, who would organise its secure transport to Rome.⁶⁹ In return, the monks of Saint-Bertin at Saint-Omer and the canons of St Donatian at Bruges would have expected to receive charitable donations from the English noblewomen. The canons in Bruges certainly did.⁷⁰

IV

Despite her wealth, Gunhild, having opted to become a woman religious, did not live a life of luxury in Bruges. As we have seen, her obituary records that, when still in England, ‘as a young girl she took a vow of virginal chastity and wishing for a spiritual union she rejected marriage to several noble princes’. If we take this information at face value, Gunhild had already chosen a life of religion before she went into exile. While in England she may have lived with her mother, or in a small community of like-minded religious women, but she never took the veil.⁷¹ Entry into a nunnery in Flanders, whether earlier on in the exile or after her return from Denmark, seems extremely unlikely. There was no nunnery in either Saint-Omer or Bruges, and in fact the only nunnery available in the county of Flanders would have been the recently founded one at Messines (Mesen) near Ypres.⁷² The most prominent veiled woman there between 1067 (the year of her husband Count Baldwin V’s death) and her own death in August 1079 was Adela, dowager countess of Flanders, who was the mother of Queen Matilda, wife of William the Conqueror. There is no evidence that Adela made any effort to support Gytha or Gunhild when they arrived in 1068. This may be surprising in view of the fact that that Gytha and Adela had met when in 1051–2 the Godwine family spent a winter in Bruges.⁷³

What, then, do we know about Gunhild’s life as a woman religious at Bruges? This is what her obituary tells us:

She piously loved Christ in her heart and she always honoured him righteously in her deeds. To those serving her she was cheerful and modest. She was benevolent and just towards strangers and generous to the poor. She was very restrained when it came to her body—what should I say—so restrained that she abstained from all pleasures and many years before the day she died she did not eat meat. She did not eat anything that seemed sweet to her but instead took only enough to stay alive. She wore a hair shirt and did not even show it to certain members of her household. In battling the vices she triumphed in the virtues.

The text bears resemblance to hagiographical portrayals of women in female religious communities from Lotharingia (not Flanders) from the late tenth century onwards.⁷⁴ The abstinence of luxurious food and sweets is mentioned for Ansoaldis of Maubeuge (1012–50), as depicted in the anonymous Life of her brother Theoderic of St Hubert written c.1050.⁷⁵ The hidden hair shirt reminds us of John of Gorze’s vivid portrait of Geisa, a young woman whose hair shirt he felt as he touched her breast at the nunnery of Saint-Pierre-aux-Nonnains at Metz c.950.⁷⁶ Both portrayals, it should be said, are of veiled women living in nunneries. Nevertheless, literary portraits like these, inspired by the Lotharingian reform movement of the late tenth and eleventh century, remind us also of those of women religious in England. In Lives written by the Flemish hagiographer Goscelin, formerly a monk of Saint-Bertin at Saint-Omer (d. after 1108) and active in England since the late 1060s, we learn about the nun Eve at Wilton, who c.1080 became a recluse in exile at Angers, and Seitha, a resident recluse at Bury St Edmund’s in the 1080s and 1090s.⁷⁷ Gunhild’s obituary, although much shorter than the other accounts, fits the same tradition. Whereas the longer lives of women religious were recorded ‘publicly’ and for posterity in hagiographical tracts preserved in monastic libraries, Gunhild’s survives on her private lead plaque only. What these women religious in England and Lotharingia, and Gunhild in Flanders, shared was a relatively wealthy background, a single life (having rejected suitors) of extreme abstinence from food and of penitence (wearing hair shirts), and of intense religiosity in prayers (often from their own psalters) while living under the protection of male religious (which some paid for in kind or with wealth).⁷⁸ While they all shared a life of penance, for Gunhild it is tempting to suggest a specific additional purpose.

An important trope of the English tradition of memorialising post-Conquest defeat was that of divine punishment for the sins of the English.⁷⁹ Gunhild, as a survivor of 1066, embodies quite literally that sense of English guilt. And although her obituary contains detailed biographical information, as we have seen, it is combined with her own strong sense of penance. The text starts with two abridged prayers, the Pater Noster and the Apostles’ Creed, which are among the most common prayers Christians would know.⁸⁰ The purpose of inserting them here was to testify to Gunhild’s belief in God and to give evidence of her knowledge of them in the hope of absolution for her sins. The reference to her struggle with vice and overcoming it because of her virtue is a theme from Prudentius’s Psychomachia that recurs elsewhere in writings about the Godwine family. As Tom Licence has shown, the Life of King Edward written for Queen Edith by a Flemish monk, probably Fulcard of Saint-Bertin at Saint-Omer, uses this theme to show that King Edward, his wife Edith (Gunhild’s older sister) and Earl Godwine (Gunhild’s father) triumphed in their struggle with vices in England, but that Harold and Tostig (Gunhild’s brothers) did not.⁸¹ Gunhild’s obituary bears testimony to the belief that she too was one of the Godwines who was virtuous, albeit in exile in Flanders.⁸²

Who then was the author of Gunhild’s obituary and who was or were his informers? It seems to me likely that a Flemish, rather than an English, clerk composed Gunhild’s touching memorial portrait. Naturally, the suggestion of Flemish authorship does not preclude the possibility that the author was informed by Gunhild’s servants. Her obituary twice refers to those who attended on her to illustrate how cheerful and kind she was to them but also how she concealed her hair shirt from some of them. Her household probably consisted of English women who had accompanied her mother and herself into exile in 1068, and perhaps some Flemish girls as well as male clergy. Among the latter, one name springs to mind as a possible informant and that is Blacmann.⁸³ Originally a rich priest, he left England to accompany Gytha in 1068 as her chaplain. On his departure abroad he granted the island of Andersey in the river Thames, an important strategic site, to the monastery of Abingdon. King William blocked the move and kept it for himself, before later returning it to the monks. After Gytha’s death, probably at Saint-Omer, Blacmann may have stayed on to provide Gunhild and the rest of her household with pastoral care. The obituary with its hagiographical tropes was probably composed by someone closely connected to the canons of St Donatian who wished to portray Gunhild as an exemplary woman religious who should be remembered in their community. If Gunhild’s penitential lifestyle was chosen in part to atone for the sins of her family in England, the form her burial took could be taken as a reinforcement of her commitment in life and in death.

Gunhild’s burial place, immured in the wall of St Donatian church, suggests very strongly that she wished to be commemorated as a vowess, if not a recluse. This elite woman religious, who lost most of her family and all her lands in England and who had lived a life in exile, arranged with the canons of St Donatian that she be buried in their church.⁸⁴ Undoubtedly, on this occasion she handed them such movable wealth as she still possessed, which, as we have seen, was of substantial value and would finance her commemoration.⁸⁵ The earliest evidence for the site of her tomb comes from the 1389 record of the sale of her jocalia, which locates the tomb ‘in the cloister’ as part of the canons’ confirmation to the bishop that her liturgical commemoration would continue. They included the directions ‘that commendations be sung for her tomb in the cloister and that in that place the procession will stop’.⁸⁶ This record and the sixteenth-century evidence of Jacobus Meyerus are in line with the report of the discovery of the grave of Gunhild by G.F. Beltz in 1833. According to Meyerus in 1561 her tomb was situated ‘next to the doorway on the north side of the church, her monument now obscured by a picture of the Virgin Mary’.⁸⁷ Apparently a plaque on blue stone was known to have been placed on the wall outside her tomb giving her name as Gunhild but with the wrong identification, as Gunhild, daughter of King Cnut and Queen Emma.⁸⁸ From Beltz’s report we learn that her coffin was indeed found buried in a very thick wall next to the doorway that led to the cloister of St Donatian with its graveyard. The position suggests that Gunhild in death was literally entombed in the church in a manner that reminds us of medieval recluses who lived and were buried in the churches where they led penitential lives, or in their graveyards.⁸⁹

Was the mural burial a posthumous extension of Gunhild’s penance in life for her own sins and those of her family? I would tentatively suggest that, upon her return from Denmark, Gunhild went to live with her servants somewhere in the grounds of St Donatian as a woman religious, perhaps a vowess or a recluse. Her obituary, however, does not describe her using the vocabulary of inclusa/reclusa, unlike the obituaries on lead of Olardis (d. 1078) and Emma (d. 1124), two recluses at the monastery of Saint-Amand (Elnone) (now in the French Départemente du Nord, part of medieval Flanders).⁹⁰ The precise sites of their tombs are lost and therefore we do not know whether their tombs, like Gunhild’s, were immured. No other evidence for immured burials in eleventh-century Flanders has survived. At Bruges the only testimony to a hermit or recluse is that of Everelmus, who died in 1060 in a small church he built south of the burgh (the fortress), near the count’s fishpond.⁹¹ The combination of her bequest to the canons, its associated wish to be commemorated, and her actual burial site strongly suggests that Gunhild herself was involved in plans for her burial as well as the text for her lead plaque obituary.

V

Finally, what does the evidence for the career of Gunhild, sister of King Harold, as we have been able to reconstruct it, contribute to the knowledge we have about royal English women? Several of them were married off. If the career of Gunhild’s niece Gunhild, Harold’s daughter and grandmother of William d’Aincourt, is anything to go by, an initial stay in a religious community before or after the Norman Conquest would not necessarily have precluded later marriage. After 1066 the younger Gunhild spent a few years at the nunnery of Wilton. Whether her entry there was out of choice or out of ‘fear for the French’, a scenario discussed by Lanfranc in relation to elite English women who had sought refuge in monasteries, we do not know.⁹² She left Wilton and married c.1075 the Breton noble Alan Rufus, lord of Richmond, who died in 1093; she was then approached by his brother Alan the Black (d. 1096 x 1098).⁹³ After the Conquest (inter-) marriage was used as a tool on a modest scale by William the Conqueror to marry off English widows and heiresses to his followers, not least to legitimate their holding of English lands.⁹⁴ In the younger Gunhild’s case the Conqueror allowed her to inherit lands from her mother Eadgifu Swanneck (d. after 1066), a rich Anglo-Saxon heiress from East Anglia and King Harold’s first wife.⁹⁵ Some of her lands then formally passed on to the Aincourts through inheritances sanctioned by the Norman kings. These arrangements included protection for little William d’Aincourt, who was offered an education at the court of William Rufus. Perhaps the fosterage was something of a goodwill gesture towards the descendants of his father’s adversary King Harold, or a hostage situation to prevent William’s father Walter from causing trouble on behalf of his wife’s royal ancestor. Marriage was also used for Margaret (d. 1092), a great-granddaughter of King Aethelred (d. 1016), who in c.1068 married King Malcolm of Scotland in an alliance that had been initiated by Edward the Confessor.⁹⁶

The single life was an alternative scenario for some royal women, perhaps depending on their stage in the life cycle and the reputation of their protector. After the death of her husband King Edward, Queen Edith (d. 1075) surrendered to King William, who allowed her to keep her dower lands as the widow of a king whom William recognised. She retired to Winchester but regularly spent time at the nunnery of Wilton.⁹⁷ A hoard of coins bearing the late King Harold’s name, containing a large number minted at Wilton in the autumn of 1066, has prompted the suggestion that Edith was involved in the design of the dies and issue of the coins, an action that could be construed as against the interests of the new king.⁹⁸ As a gesture to King Edward’s memory, the Conqueror did not interfere with the lands that Edward had assigned to his sister Goda (d. after 1066), widow of Count Drogo of Vexin (d. 1035) and divorced from Eustace II of Boulogne since the early 1050s. Goda spent her final years after the Conquest as a single woman at Lambeth near London.⁹⁹ Another of Edward’s paternal cousins, Christina, sister of Queen Margaret of Scotland, remained in England and was attached to the royal nunneries of Romsey and Wilton.¹⁰⁰ Thus, whether they were related to his acknowledged predecessor or his defeated predecessor, royal women received support from William the Conqueror as long as they toed the line.

The fates of a group of women associated with Harold, of whom Gunhild was one, merit special attention. When Harold died at Hastings, his second wife Ealdgyth of Mercia, widow of King Gruffydd ap Llywelyn (d. 1064), was in London, from where her brothers, earls Eadwin and Morcar, collected her and took her to relative safety in Chester.¹⁰¹ This is the last we hear of her; we do not know what happened to her after the fall of Chester and that of her brothers in 1070.¹⁰² With Queen Edith having been powerless to protect her mother Gytha and sister Gunhild, both held out with other noble women against the Conqueror for two years. Unfortunately the identities of these noble women are unknown. As we have seen, by 1068 their position in England had become untenable and remaining defiant in exile was their only option.¹⁰³ Did they expect that Flanders might again be a springboard for a return home, just as it had been in 1051–2? If they had Harold’s crown among their possessions, what might they have hoped for? A coronation for a son of Harold?

These questions raise the issue of the fate of elite Englishmen and in particular Harold’s sons. As experienced fighters, when capable, they were inclined to revenge their father. The sons who could not do so were Wulf, in captivity in Normandy but released in 1087, and Ealdgyth’s unnamed son (perhaps named Harold), born in 1066.¹⁰⁴ The sons who could had fled to Ireland, another safe haven for the Godwines from the recent past.¹⁰⁵ From there in 1068 Godwine, Edmund and Magnus attacked first Bristol and then Somerset, but were defeated in both places.¹⁰⁶ The next year two unnamed sons landed on the Dorset coast at the mouth of the Taw but were again pushed back.¹⁰⁷ A son called Harold found refuge at the royal court in Norway and in 1098 accompanied King Magnus III on a raid on Anglesey.¹⁰⁸ Thereafter Harold’s sons disappear from view.

Among royal men there was also King Edward’s kinsman Edgar the aetheling, who in 1066 was for a mere few weeks recognised as his successor.¹⁰⁹ Initially he was protected by the Conqueror and kept under close watch at his court, for similar reasons to those that kept little William d’Aincourt at the court of William Rufus. But then Edgar allied himself variously with Anglo-Saxon resistance fighters, with his brother-in-law King Malcolm of Scotland, and others. Ultimately, he disappeared from view only to re-emerge during the First Crusade, from which he returned to retire to Somerset, where he died in the early twelfth century.

The decision by Gytha and Gunhild to go into exile, and to remain on the Continent, was an act of defiance against the new Norman king of England. It was in line with the exile of Harold’s sons and with previous Godwine family experiences of seeking refuge overseas. In terms of the fates of royal women in England, including Queen Edith, the flight of the widow and her single daughter to Flanders was exceptional. In the first instance they wished to safeguard their movable wealth (perhaps including Harold’s crown) in order to finance a future return to England. Exile also may have safeguarded Gunhild’s choice to live a life as a religious woman, a status which, had she remained in England, might have been compromised by the Normans. In Flanders, living under the protection of its counts, and with Gytha dead, Gunhild’s temporary move abroad became a permanent one. On her own death Gunhild donated the family wealth to the canons of St Donatian at Bruges in return for the concession that she be entombed in the wall of their church as an act of penance. It is at this stage of her life that she emerges from obscurity in her Latin obituary on lead. This records, apart from her penance, her pride in her English elite ancestry. The unusual, near-hagiographical portrayal of Gunhild is the longest eleventh-century text of its kind on lead for a lay woman in western Europe. As such it provides a unique female testimony to an English royal woman’s life abroad in the aftermath of the Norman Conquest.

Footnotes

Appendix

Gunhild’s plaque is registered with the Royal Institute for Cultural Heritage Belgium, KIK-IRPA, Brussels, object no. 144772, and cliché no. Z011814. Measurements are 24.5 cm x 19.5–20.5 cm. The plate is kept in the treasury of St Salvator’s cathedral in Brussels.

My edition is based on the photograph of the plaque as published by Luc Devliegher. As for editorial conventions, I have silently expanded words with abbreviation marks and indicated with square brackets where letters have been added by me. I have added capital letters and the interpunction is modern. I list substantial alternatives by previous editors but have not signalled previous reproductions of e as ae or v for u, or any capitalisations which are different from mine. I note that Gailliard did not expand any abbreviations in the transcript he published.

Previous editions and translations:

1. Beltz, ‘Observations’, pp. 400–401 (Latin text) and 411 (engraving by J. Basire).

2. [P.J. (i.e. Pierre Jacques) Scourion], ‘Notice historique et critique au sujet d’une inscription gravée sur une plaque de plomb trouvée dans le tombeau de Gunilde, princesse anglo-saxonne, qui, après que son frère Harold II eu été tué à la bataille de Hastings, gagnée par Guillaume-le-Conquérant, en 1066, se retira à Bruges où elle mourut en 1087’, Messager des Sciences et des Arts de la Belgique, i (1833), pp. 425–41, at 426–7 (Latin text and French translation).

3. A General Introduction to Domesday Book accompanied by Indices of the Tenants-in-Chief and Undertenants at the Time of the Survey, ed. Henry Ellis (London, 1833), pp. 136–7 (Latin text made available to Ellis, via John Bidwell of the Foreign Office, by M. Moke in Bruges).

4. Jean Gailliard, Inscriptions funéraires et monumentales de la Flandre occidentale … Arrondissement de Bruges (1 vol in 3 pts, Bruges, 1861), pp. 206–7 (Latin text), plate xliv with an engraving of the plaque by A. Ancot.

5. Freeman, History of the Norman Conquest, iv, pp. 754–5 (Note M) (Latin text made available to him by Sir Thomas Hardy from a text published in Belgium in 1833, presumably no. 2 above).

6. Devliegher, De Sint-Salvatorkathedraal te Brugge, p. 97 for Latin text and Dutch translation, and plate 163.

I am most grateful to Moreed Arbabzadah and Tom Licence for their help with the Latin text and the translation.

Text

Pater noster. Credo in Deum patrem,/ et cetera que in simbolo apostolorum sunt scripta./ Gunildis nobilissimis orta parentibus, genere Angla, pa/tre Goduuino comite, sub cuius dominio maxima pars/ militabat Anglie, matre Githa, illustri prosapia Dacorum oriunda./ Hec dum uoueret adhuc puella uirginalem castitatem, desiderans/ spirituale coniugium, spreuit connubia nonnullorum nobilium principum./ Hecque dum iam ad nubilem etatem peruenisset, Anglia deuicta/ a Willelmo^a Normannorum comite et ab eodem interfecto/ fratre suo rege Anglorum Haroldo, relicta patria, apud/ sanctum Audomarum aliquot annos exulans in Flandria, Christum/ [qu]em pie amabat, in pectore sancte^b semper colebat in opere/ [ci]rca sibi famulantes hilaris et modesta, erga extra/neos beniuola et iusta, pauperibus larga, suo cor/pori admodum parca. Quid dicam? Adeo ut/ omnibus illecebris se abstinendo, per multos annos/ ante sui diem obitus non uesceretur^c carnibus, neque/ quicquam quod sibi dulce uisum est gustando, sed uix ne/cessaria uite capiendo, cilicio induta, ut nec etiam quibusdam pateret fami/liaribus, conflictando cum uiciis uicit in uirtutibus. Dehinc/ transiens Bruggas, et ibi^d transuolutis quibusdam annis et^e inde/ pertransiens^f in Dacia^g, huc reuersa, uirgo transmigrauit/ in Domino, anno incarnationis Domini millesimo lxxxvii, nono kalendas Septembris, luna xxii.

^a Guillelmo Scourion;^b sce: scilicet Scourion, Freeman. Ellis notes that his reading of Scc’ is conjectural;^c vexeretur Freeman;^domitted Scourion;^e ac Gailliard;^fEllis notes that his reading of pertransiens is conjectural;^g Daciam Beltz

Translation

Our Father; I believe in God the Father, and the other things which are written as the Apostles’ Creed. Gunhild, born of most noble parents was English. Her father was Earl Godwin who held a very large part of England under his military command and her mother Gytha was born of famous Danish ancestry. As a young girl she took a vow of virginal chastity and wishing for a spiritual union she rejected marriage to several noble princes. She had already reached marriageable age when England was conquered by William, count of the Normans, who slew her brother Harold, king of the English. She fled her homeland and lived for some years in exile at Saint-Omer in Flanders. She piously loved Christ in her heart and she always honoured him righteously in her deeds. To those serving her she was cheerful and modest. She was benevolent and just towards strangers and generous to the poor. She was very restrained when it came to her body—what should I say—so restrained that she abstained from all pleasures and many years before the day she died she did not eat meat. She did not eat anything that seemed sweet to her but instead took only enough to stay alive. She wore a hair shirt and did not even show it to certain members of her household. In battling the vices she triumphed in the virtues. She then went to Bruges where she lived for a few years and from there crossed to Denmark. She returned here and died a virgin in the year 1087 on the 24th of August on the twenty-second day of the new moon.