Old Books On CD

We specialize in WARREN family History,

including "Memoirs of the earls of WARREN and Surrey"





on the

Parentage of Gundreda


Daughter of William Duke of Normandy


Wife of William de Warenne

[Warren Shield]



Sir George Duckett, Bart, FSA


Reprinted (with additions) from Vol III Cumberland and Westmorland Society's Translations.


[page 2- blank]

[page 3]

Observations on the parentage of Gundreda, the daughter of William, Duke of Normandy, and Wife of William de Warenne.

By Sir G. Duckett, Bart.

(Communicated at Whitehaven, December 11th, 1877.)

It is, doubtless, known to most members of the Cumberland and Westmorland Archaeological Society, that the second husband of Gundreda, Countess of Warwick, was William, surnamed of Lancaster, Baron of Kendal. She was the daughter of William de Warenne, the second Earl of Surrey, by Isabel de Vermandois, and consequently granddaughter of Gundreda, the fifth daughter of William, Duke of Normandy, the subject of the following notice.

This connection with Westmorland would alone entitle the paper, now submitted to the members of the Society, to figure in it's Transactions, but the additional fact of the Countess Gundreda's marriage with Roger de Glanvill, as her third husband, after the decease of W. de Lancaster, renders the much controverted point of the parentage of her grandmother, a matter of historical interest for the Northern counties, seeing that such husband was Sheriff of Northumberland from the 31 Henry II. To Richard I. The authorities for these authenticated facts are given below.1

1 Dugdale Bar. i, 73, 421 sq; John Rous of Warwick [Hist. MS. Bibl. Cotton]; Ordercus Vitalis [Hist. Eccles.]; William of Jumieges [Ed Ducesne 1619]; Milles' "Catalogue of Honor" [Ed 1610]; Mon i, 708a, n 60; Mag Rot. Pip. 16, H II., Lanc.; Coram Rege, i John, 3 m, 17 dorso; Madox's Exchequer ii, 236; 3d Report App. P. 321; William of Poictou [Ed Guizot].

[page 4]

In turning to page 149 of the "Yorkshire Archaeological Journal" (vol iv); to certain pages of "Notes and Queries" (5th S iv, pp 386, 476); and to Murray's "Handbook for Sussex" (under Lewes), one's attention is forcibly drawn to a question, which, it would have been imagined, ought long since to have been set at rest, viz., the parentage of Gundreda, as wife of William de Warenne (Guillaume de Varenne), 1st Earl of Surrey.

The writer in the otherwise ably written paper, in the periodical first named, in enumerating the several Yorkshire Tenants in Domesday, runs counter to the opinion usually received until of late years, and quoting Ordericus Vitalis, assumes Gundreda to have been the sister of one Gherbod, a Fleming. This alleged relationship seems first to have been assigned to her by Dugdale on the same authority, and after him, in more recent times, by Mr Stapleton,2 F.S.A., and a writer of considerable merit, Mr Freeman,3author of the "Norman Conquest."

It seems inconceivable that Dugdale, from whom so many have gleaned their information, should have overlooked the Conqueror's charter giving to the monks of St Pancreas the Manor of Walton, in Norfolk, and the Foundation Charter of Lewes Priory, still extant, which are incontestably clear, and of themselves sufficient to refute any doubt on this subject. As a question of history, it is a matter of regret, and to ourselves of supreme astonishment, that the arguments which of late years have been set forth by the late Mr Blaauw,4 to whom the merit is due of having so successfully controverted the attempt to cast a doubt on the parentage of Gundreda, as the Conqueror's daughter, should not long since have settled this disputed point. It only shews on what slight (or rather worthless) grounds a false position may be held, and it's evil consequences propagated.

A solitary passage, on the authority of this nearly contemporary historian, Ordericus Vitalis, extracted from

2 Archaeological Journal, Vol iii p.1-26

3 History of the Norman Conquest, vol iii, app p 645-658.

4 Archaeologia, xxxii p 108-125. See also some remarks by Lord Stanhope,- Proc Soc of Ant, vol v, p 138-139.


[page 5]

his so-called "Ecclesiastical History," is forcibly insisted on, and thus from time to time revives the attempt to disprove the marriage of William de Warenne with a daughter of the Conqueror. The passage alluded to has been construed as an inference that Gundreda was simply the stepdaughter of the Conqueror, namely, the daughter of Queen Matilda by a former husband, being therein recorded as the Sister of Gherbod, a Fleming to whom the Conqueror had first granted the City and Earldom of Chester, although the assertion is supported by no other chronicler, resting solely on the authority of the historian in question, and notwithstanding that the work is of considerable length, is nowhere again repeated by him.

After recapitulating the lands and counties which the Conqueror had portioned out to certain of his Norman followers, Ordericus, in passing to the county of Surrey, which was granted to William of Warenne, observes:- "et Guillelmo de Guarenna (Warenna) qui Gundredam sororem Gherbodi conjugem habebat, dedit Surregiam," having shortly before stated:- "Cestriam et comitatum ejus Gherbodo Flandrensi jamdudum rex dederat" (Ord. Vit. "Historia Eccles." Pars. Ii., Lib. Iv., c. vii.; Migne, "Patrologia," clxxxviii, pp. 271, 331, 515, 583). His reason for granting him the Earldom of Chester we believe to have made apparent in the sequel.

Now, with a view to refute the perversion of the real facts as regards Gundreda, we may observe that the question of her parentage is open to two entirely opposite considerations. First, as to the authority of Ordericus Vitalis as an authentic historian, if, as prima facie it would seem, he has used the word "soror" in its usually accepted sense, in which case, as will be evident, its employment is capable of entire refutation; and secondly, irrespective of any implied want of authenticity, whether Ordericus has not employed the word in the sense we are disposed to take it, which puts the question in a totally different light, as we will shew hereafter.

Mr Blaauw observes in his paper in the "Archaeologia" (xxxii., 108), in answer to Mr Stapleton (who had undertaken to disprove the marriage of William de


[page 6]

Warenne with a daughter of the Conqueror, on the ground of Gundreda being the Queen's daughter by a divorced husband, named Gherbod), that "on matters of the Royal pedigree his acknowledged errors are so numerous as to deprive him of much authority;" that he wrote his Chronicle "when a very old man, with a confused memory of the details;" that he has erred again "in making Matilda to survive the Conqueror," and again "in stating the grant of the Earldom of Surrey to have been conferred in that monarch's time, whereas it was bestowed in that of his son Rufus."

The foundation charter of the Priory of Lewes, dedicated to St Pancras, expressly states Gundreda to have been the Queen's daughter; the words of William de Warenne on the occasion of his founding that house, indubitably prove Queen Matilda to have been her mother, and can be taken in no other sense: the words are, "pro salute dominae meae Matildis Reginae matrix uxoris meae." It is therefore self-evident from this fact, that Gherbod the Fleming must equally have been Queen Matilda's son, but although sufficient opportunity is afforded Ordericus, he never once mentions him as her son, neither does he in any part of his "History" represent the Queen to have been united to a previous husband, in fact no trace of such an assertion can be found in any contemporary, or subsequent chronicler. As to the pretended marriage of (Queen) Matilda with Gherbod the Fleming, and her subsequent divorce, which Mr Stapleton endeavors to maintain, Mr Blaauw explains at some length how the confusion may have arisen [Archaeol. xxxii, 120], and we have elsewhere given additional, and we believe conclusive, reasons in disproof of this supposition. Not one of the Norman chroniclers, he observes, with any exception, "has dropped the smallest hint of any husband or child, or consequently any such divorce on the part of Matilda previous to her marriage with the King." All authorities in fact concur in proving the reverse; they all allude to Duke William's affianced bride as a young unmarried girl, pucelle (puella), and the only inference is that William of Normandy was Gundreda's father.

Sir H Ellis, in his "Introduction to Domesday" (i.

[page 7]

507), observes; "Gundreda was really a daughter of the Conqueror. William de Warenne's second charter of foundation, granted to Lewes Priory in the reign of Rufus, states this fact distinctly:- Volo ergo quod sciant qui sunt et qui futuri sunt, quod ego Willielmus de Warenna Surreiae comes, donavi et confirmavi Deo et Sancto Pancratio, et monachis Cluniascensibus, quicumque in ipsa ecclesia Sancti Pancratii Deo servient in perpetuum; donavi pro salute animae meae, et animae Gundredae uxoris meae, et pro anima domini mei Willielmi regis, qui me in Anglecam terram adduxit, et per cujus licentiam monachos venire feci, et qui meam prioreum donationem confirmavit, et pro salute dominae meae Matildis reginae, matris uxoris meae, et pro salute domini mei Willielmi regis, filii sui, post cujus adventum in Anglicam terram hanc cartam feci, et qui me comitem Surregiae fecit." (Cott. MS. Vesp. F. XV; Lappenberg, p 216.)

Gundreda is also acknowledged by the Conqueror himself as his daughter. The charter, by which the King gave the manor of Walton, in Norfolk, to the same Priory, on its first foundation by W. de Warenne and his wife,5 distinctly styles her his daughter. He gives it, "pro anima domini et antecessoris mei Regis Edwardi ... et pro anima Gulielmi de Warenna, et uxoris suae Gundredae filiae meae et haeredum suorum." (Intro. Domesd. I. 507.)

Again, in the Ledger Book of Lewes are these words:- "Iste (William de Warenne), primo non vocabatur nisi solummodo, Willielmus de Warenna, postea vero processu temporis a Willielmo Rege et Conquestore Angeliae, cujus filiam desponsavit, plurium honoratus est," etc. (Watson's Memoirs, i. 36.)

Those who, relying on Ordericus Vitalis, seek to disprove this fact, insist that the words, "filiae meae" in the Conqueror's charter are an interpolation, but a minute inspection of the original MS. In the Cottonian Library (Vespas. F. iii. Fo. I), in no way warrants this belief; on the contrary, the words "filiae meae" are simply interlined in

5 The Queen, her mother, gave Carleton, on the same occasion, to their newly-founded Priory at Lewes.

[page 8]

explanation of words which were originally written, but which have disappeared from decay; indeed the whole charter has faded, and has been rubbed with gall. Mr Blaauw remarks on the words "filiae meae" thus: "A close and repeated examination of this MS. By Weston Styleman Walford, Esq., and myself, has furnished very sufficient proof that the words of the original should be read thus:- "Pro anima Gulielmi de Warenna et uxoris sue Gondrade filie mee et heredum suorum," not "pro me et heredibus meis," as substituted by Mr Stapleton (Archaeol. Jour., iii. 2), for words "filie mee et heredum suorum" (Archaeologia, xxxii. 117).

Moreover, among the original benefactors of the Abbey of Lewes, Gundreda is named conjointlywith her brother Henry I. of England:- "In Norfolcia (among other possessions) Karletuna quam dedit Matilidis regina, mater Henrici Regis et Gundredae Comitissae;"6whilst the Chartulary of the same Abbey (at fo. 106) further certifies to her affinity to the Conqueror:- "Iste (William de Warenne) jacet in capitulo Lewensi juxta Dominam Gundredam, comitissam suam et filiam praedicti Regis Conquestoris;" and again (at fo. 103) - "Domina Gundreda, Comitissa Surreyae, filia Conquestoris, quae sepulta est in capitulo Lewensi cum conjuge suo." [These last facts the excavations in the Chapter house of the above Priory in 1845 have thoroughly established.]

Irrespective of the undeniable evidence of these charters and chartulary references, until a recent date, with the sole exception of Dugdale, who, as we have observed, was entirely misled by Ordericus Vitalis, in the way in which he has applied the word "suror" in his "Ecclesiastical History," Gundreda had always been handed down and acknowledged as the Conqueror's daughter. Leland, for example, in the reign of Hen VIII. gives us in his "Collectanea" the following extract:- "And these thinges that folow I translatid owte of an olde French Historie yn Rime of the Actes of the Guanines onto the death of Fulco the II."

6 Ex libro Computorum, olim Prioratui de Lewes spectante (Dugdale).

[page 9]

"The Genealogie of the Countes of Guarine, alias Surreie:"

"William the firste Counte Guarine married Gndreda, the Doughter of King William Conquerour."

Benolte, also, who was Clarenceux from 1516 to 1534, in his Visitation, in remarking on the tombs existing in his day," in the Chapter House of Lewes Priory, observes:- "Willyam the firste Erle Waryne and Surrey, furste founder of the howse of Saynt pancrase, assituate within the towne of Lewys, in the countye of Sussex, wiche Willyam and Gondrede, his wyffe, lieth buryede in the Chapytre of the same howse, wich Gondrede was dawghter unto Willyam the Conquerour." (MS. D. 13. Col. Arm.) (Suss. Arch. Coll., iii, 187.)

Again, dating back to an early part of the same reign, we find in one of the Harleian MSS. some remarkable portraits or sketches,7 though somewhat rude, of William of Normandy, his Queen, and three sons, with three only of his six daughters, to the exclusion, namely, of Cecilia (the abbess of Caen); Adeliza (a nun); and Agatha (who died, betrothed to the King of Galicia). Inasmuch as the three therein depicted Princesses are partly robed in the Royal purple, and partly display the heraldic ensigns of their respective husbands, embroidered on their mantles, that of Gundreda having the Checky or. and az. Coat of Warenne,8 the inference is clear, that the excluded sisters, being dead to the world, needed no portraying or worldly vestments, according to the artist's view, and that the remaining daughters, of whom the wife of William de Warenne was one, were well-known and acknowledged daughters of the Conqueror at that date. In addition to these instances, Ralph Brooke, York Herald, [in his "Catalogue of the Earls of Surrey," Ed. 1619]; and Vincent, Rouge Croix [in his "Discoverie of Brooke's Errours," Ed. 1622],

7 Also noticed by Mrs Everett Green, in her "Lives of English Princesses."

8 The Warenne Checky coat, (or that of Vermandois), was first assumed in the next generation by Gundreda's son, the 2nd Earl of Surrey, as that of his wife, Isabel de Vermandois; but this fact is not material; it proves that one of the Conqueror's daughters was admitted to have allied herself to the House of Warenne.

[page 10]

equally agree in respect of Gundreda's degree of affinity to the Conqueror. The former goes so far as to say that she had, on her marriage with W. de Warenne, "all Chirkland, Bromfield, and Yale," but this is disallowed by the latter.

These different extracts establish Gundreda as the Conqueror's daughter beyond any reasonable doubt, and of themselves prove the entire fallacy of the hypothesis of Mr. Stapleton and his followers,9 which has already received, as may be said, its coup de grace at the hands of the late Mr Blaauw,10 indeed, an impartial study of them can lead to no other conclusion.

There remain still, however, many other forcible reasons which may be cited, conclusive of the illogical nature of this doctrine, relying as its main, or sole, argument upon a previous marriage of Queen Matilda.

The very wording of the epitaph11 on Gundreda's tomb at Lewes is conclusive of her affinity to the Conqueror- "Stirps Gundrada ducum"- and clearly points to her paternal descent from the Dukes of Normandy; whilst on her mother's side, that on Matilda's tomb at Caen, "Germen regale Mathildem," indicates royalnot ducal descent, her mother having been the daughter of Robert, King of France, the son and nephew of kings, from other

9 Archaeological Journal iii, pg 20

10 Archaeologia xxxii, 108

11 These well known epitaphs scarcely need repetition, but we give them to save the trouble of reference:--

"Stirps Gundrada ducum decus evi nobile germen

Intulit ecelesiis Anglorum balsama morum

Martir - - - - - - - - - - - - - - - - - - - - - - - - - - -

- - - - - - - -uit miseris suit ex pietate Maria

Pars obiit Marthe superest pars magna Marie

O pie Pancrati ter - - - - - - - - - - - - - et equi

Te facit heredem tu clemens suscipe matrem

Sexta kalendarum Junii lux obvia carnis

Ifregit alabastrum ..."


The epitaph of Queen Matilda, given by Vitalis (Ed. Prevost, iii. 192-3), is :--

"Egrgie pulchri tegit haec structura sepulchri

Moribus insignem, germen regale, Mathildem;

Dux Flandritae pater, huic exstitit Hadala mater,

Francorum gentia Roberta filia regis

Et soror Henrici regali sede potiti,

Regi magnifico Willelmo juncta marito, &c., &c.

[page 11]

kings descended.12 "Regali ex genere descendentem nomine Mathildem," are the words also of William of Jumieges (lib. 7, c. xxi.); and William of Poictou (the Conqueror's chaplain), corroborates this fact.

A very material point in the present enquiry are the dates of Duke William's birth, his marriage, and his death; and still more so those of Gundreda.

Without adducing other proof, if reference be made to William de Jumieges, and Thierry ("Histoire de la Conquete de l'Angeleterre par les Normans"), it is plain that the birth of William of Normandy may be fixed in the year 1027, or the commencement of 1028. The former states that he was in his 60th year ("fere sexagenarius), at the time of his death, and that he succeeded his father in 1036, being then a boy of 8 years old.13 Matilda was married to him, if we are to take the Roman de Rou of Wace chronologically, after he had fortified Ambrieres, near Mayenne14; an event, however, of certainly later occurrence, for both William of Poictou and William de Jumieges (contemporary chroniclers) record how, when only just a young man, having received at the Chateau d'Eu (Ponthieu), his youthful bride the French King's niece, at the hands of her father, Baldwin, Count of Flanders, Duke William made his public entry with her into the city of Rouen.15

Pluet [Ed. Of Wace's "Roman de Rou"], observes-

12 Baldwin, Count of Flanders, (Queen Matilda's father), was known as Baldwin de l'Isle, because he rebuilt the city of l'Isle (Lille). He married the daughter of Robert Capat, King of France, and by her had two sons, Baldwin (surnamed de Mons), and Robert le Frison, and one daughter, the above-mentioned Matilda. (Chronique de Flandres).

13 According to William of Malmsbury, the Conqueror died in the fifty ninth year of his age, and his death occurred in 1087.

14 Par cunseil de sa barunie,

Prist une fame de haut lin,a

En Henri li rei France,

Fille soe fille b Cunstance;

A maint noble home fit parente,

Mahelt c out nom, mult bele e gente.

["Roman de Rou", Ed. Pluquet, ii, 58.]

a Lineage b Fille de sa fille. c Matilda

15 Gemmet, lib. Vii, c. 21.


[page 12]

"No Norman historian had thought proper to fix the date of this marriage, and we are constrained to seek it in the "Chronicle of Tours." Here we are told that the marriage was celebrated in the 1053, the same year in which must be placed the revolt of William of Arques against his nephew, William of Normandy (the Conqueror in question). It seems that this projected marriage between the Duke and Matilda had been an affair of long standing inasmuch as it was prohibited by the Council of Rheims in 1049; the Count of Flanders being enjoined not to bestow his daughter on the Duke, or the latter to take her to wife."

The reason which the French editor of the "Roman de Rou" assigns for the incorrectness of the date (viz 1053), although adduced in respect of Agatha, another daughter of the Conqueror, applies with equal force to Gundreda.

We know from Wace's Chronicle that the Conqueror called for, and mounted before the Battle of Hastings, his Spanish war horse;--

Sun boen cheval fist demander,

Ne poeit l'en meillor trover;

D'Espaingne li ont enveie

Un Reis par mult grant amistie.

["Roman de Rou", ii. 193. Ed. Pluqet.]

as to which Pluquet makes this significant remark;- "It is possible that this horse was sent to the Duke by (Alphonso) the King of Galacia, to whom was betrothed (against her will) Agatha (or Adelaide), the same daughter who had been previously affianced to the Anglo-Saxon King Harold. Constant to her first love, this young Princess implored the Almighty to take her to Himself rather than that she should become the wife of the Spanish monarch, and as we know, the prayer was so far heard, that she died previous to the consummantion of the event. This touching anecdote (he continues), recorded by Ordericus Vitalis (l. V. p 579), leads one to conclude that the "Chronicle of Tours" has assigned too late a date to the marriage of William and Matilda, in placing

[page 13]

it in the year 1053. It is highly improbable that Harold would have inspired so violent a passion in a young girl of eleven years, and we have already observed the more probable date of this marriage would have been in 1049."

We have, however, additional and even stronger reason for assigning the year 1049 as the marriage of Matilda and Duke William, the very year, namely, in which the Council of Rheims had prohibited it, as just observed by Pluquet; indeed, the recently published MS of an ancient Flemish chronicle places the marriage as early as 1047.16According to William of Jumieges, (who was not only contemporary with the events he relates, but dedicated his work to the Conqueror, so that his authority on this point is unquestionable), it was not until after the marriage that the fact of their near relationship was brought to the cognizance of the Pope. Seeing that if he pronounced a divorce between them (as Jumieges observes), a probable war would ensue between Flanders and Normandy, the Pope wisely determined that Duke William and his consort should atone for the crime, and from which he absolves them, on their agreement to found two separate monasteries. The result of this decision was that the monastery of the Holy Trinity was founded by Matilda at Caen, whilst that of St Etienne was founded by the Duke at the same place, [Gemmet, lib. vii., c. 26]

A similarity of reasoning applies to Gundreda. We are told that in 1070 she and her husband went on a pilgrimage to Rome, as detailed by W. de Warenne himself in his second charter of foundation of Lewes Priory [MS Cott. Vesp. F. xv.; Mon. v., I sq], and that their progress was arrested on reaching Burgundy by the contest then going on between Henry IV of Germany and Pope Gregory VII (Hildebrand). The year 1070, however, has manifestly been wrongly fixed as the date of this occurrence. The period at which the war ensued between the Emperor and the Pope, was between the years 1073 and 1077, so that it must necessarily be referred to some year between those dates, for Pope

16 "Corpus Chronicorum Flanderiae," i., 552.

[page 14]

Gregory did not succeed to St Peter's chair until the death of Pope Alexander in 1073. Assuming, therefore, the Conqueror and Matilda to have been married in 1053 (as in the case of Agatha), Gundreda would have been scarcely marriageable unless she had been the first-born child. The death of Gundreda, may, from her epitaph, be safely placed in 1085, so that although it has been said she died in her thirty-fifth year, she might thus have been the offspring of a marriage in 1049 (or 1047), though scarcely at the date of 1053.

These dates equally tend to show the fallacious and absurd reasoning of thse who maintain a previous marriage of her mother, Queen Matilda.

That William of Normandy was Matilda's first and only husband is plain from the following facts. From Domesday [B vol. i., fo. 100] we know that Queen Matilda had conferred upon her the lands of the Saxon noble Brictric (Brihtrik or Bric'trich Mau), the son of Earl Algar. [Intro. Domesd. i. 452.] Thierry mentions her name as the first inscribed on the partition roll of the territory of England, receiving as her portion this same Saxon's lands. It is recorded of her, that being in love with the young noble in question, when a representative at her father's court from King Edward (the Confessor), her advances to him were not reciprocated17 [Dug. Mon.,

17 Malde de Flandres fu nee

Meis de Escoce fu appelee,

Pur sa mere ke fu espuse

Al roi de Escoce ki l' out rove;

Laquelle jadis, quant fu pucele,

Ama un conte d'Engleterre,

Bric'trich Mau le oi nomer,

Apres le rois ki fu riche her;

A lui la pucele envela messager

Pur sa amur a lui procurer;

Meis Bictrich Maude refusa,

Dunt ele mult se coruca,

Hastiyement mer passa

E a William Bastard se maria.

[MS. Cott. Vitellius A. x., fo. 129]

The continuator of Wace, who wrote temp. H. III., and is the authority for this fact, has, however been guilty of more than one anachronism, in confounding two subsequent Matildas, viz., the daughter of Malcolm, King of Scotland, and the daughter of Henry I of England, married to the emperor of Germany.


[page 15]

i. 154; Wace's "Brut d'Angleterre," i. 73; Thierry's "Conquest of England," i. 428 (Hazlitt)], and that she thereupon resolved on marrying William of Normandy.

It thus becomes clear, that Matilda of Flanders could not have been married at the time of this occurrence, neither were her affections set on William of Normandy, until after the Saxon's slight, in fact one was a consequence of the other; so that had her marriage with him been indefinitely postponed, on the ground of its being within the forbidden degrees of consanguinity, or the prohibition by the Council of Rheims, and thus not consummated for some considerable time, (which some maintain, although the very reverse is stated by that contemporary historian as we have shown, William de Jumieges) (lib. vii. c. 26), it is contrary to all likelihood that she, in the interim, would have contracted marriage with another person (as alleged by Mr Stapleton), or have been in addition the mother of three small children, the improbability of which is otherwise sufficiently apparent. What are we in such case to understand by these words:- Laquelle jadis quant fu "pucelle;" what by the same word "puella," twice occurring in the "Chronicle of Tours," with reference to Matilda and her marriage with William of Normandy; and again by the following in Benoit's "Chronicle of the Dukes of Normandy?"--

"Cist out une fille trop bele

Maheut out nom jeune "pucele."

[Harl. MS 1717]

On the other hand, with regard to the alleged brother Gherbod; how could Ordericus Vitalis, whether his errors were many or few, who says of Matilda of Flanders that she derived her descent from the Kings of France and Emperors of Germany, have reasonably intended to imply that she was the mother of this Gherbod the Fleming? We know that some of his assertions are unfounded; they have been corrected by some of Mr Blaauw in his already quoted paper in the "Archaeologia:" by Ellis, "Introduction to Domesday," i. 506, 429, 502, 364; by Masseres, p 217; they have been pointed out by Lappenberg; and Watson ("Memoirs of Warren," i); but we are disposed in this case to


[page 16]

consider that he meant otherwise, as we shall proceed to shew, or else how comes it to pass that throughout his whole work, Ordericus never hints at any connection or supposed marriage with any one on the part of Matilda? Gherbod is never once mentioned as Queen Matilda's son; sooner or later in his narrative of events it would not have escaped his notice had such been the undoubted fact. He would certainly have intimated as much when relating how the Conqueror had given him the Earldom of Chester. Instead of saying- "Cestriam et comitatum ejus Gherbodo," &c (as in the context), his words would have been somewhat to this effect: Cestriam et comitatum ejus Gherbodo Flandrensi, filiovidelictet Mathildis postea Reginae Angelorum, jamdudum rex dederat.

If there is any foundation whatever on the part of Ordericus in naming her "sister" to Gherbod, we would suggest that she was simply his foster-sister, for we see improbability stamped on the face of any other supposition.18 We arrived at this solution of the point in question, that "sororGherbodi" has been used in the sense of "seur de lait," for these reasons. In old (and Norman) French "seurer" signifies to wean from suck; (seuree, weaned from suck; qui est seure that is weaned). We know apart from this, that "soror" is often used figuratively, but inasmuch as the old Norman French of Vitalis's time would readily suggest "soror," so are we convinced that the term is used by him without respect to consanguinity. A very singular application of the word in the same sense (which we give over leaf), may be found in the "Collection of Latin Inscriptions," by

18 Assuming (for the sake of argument) the Conqueror and William de Warenne to have been nearly of equal age, seeing that they died within a year of each other, the first, namely, in 1087, and the latter in 1088, the possibility might certainly exist of a prior marriage on the part of W. de Warenne. The sister of Gherbod might have been the first wife, and his marriage with the Princess Gundreda a subsequent event. Against this reasoning, however, it is highly improbable that Ordericus Vitalis, who wrote in the succeeding century, would, in such case, have confined his mention to one wife only; he would assuredly have named the wife of the greatest rank and importance, had he even omitted (which is equally improbable), all mention of another. We are forced, therefore, to conclude that Vitalis alludes to Gundreda alone, as the "Soror Gherbodi."

[page 17]

Jo. Gaspar Orellius, published at Zurich, in 1828. The derivative word "sororiae" [maiden's paps], may be also taken as indicating analogy, in connection with the above words, seur, [soror,] seurer. There is something in the very wording of the passage which implies this sense:- "Sororem Gherbodi conjegem habebat." Gherbod must have been her foster-brother. In the early days of chivalry foster-brotherhood was one of its peculiarities, and the foster-sister or brother was socially more than the brother or sister by blood. That the family of Gherbod was one of more or less pretension, may be inferred from the instances of the name which can be quoted; and bearing this in mind, the relative position of Gundreda to Gherbod becomes quite explainable. In the case of the former, the rank of the wife of the avoue (or protector) of St Bertin, would be quite in unison with that of a mother called upon to foster a child of Royal birth; whilst the subsequent elevation of Gherbod to the Earldom of Chester is in accordance altogether with the custom of those ages, and the spirit of foster-brotherhood; his promotion to honour, as Gundreda's foster-brother, being a natural consequence of such position.

This simple view of the matter goes far also to explain the otherwise unintelligible and far-fetched story of a former "divorced husband," the burden of Mr Stapleton's song,19 of which we confess, apart from the question of Royal and Ducal descent, we never could see the exact bearing, either as applicable to Matilda or the disproof of her daughter's affinity to Duke William, as under our hypothesis the "divorced" husband would have been simply a "foster-father."

A few instances of the use of the word "foster", will rather elucidate the peculiar applicability of the word, taken from the older writers:--

Chaucer says of some one:- "She was fostered in a nunnerie."

[The Reve's Tale]

Again, of another:--

"Fostered she was

With milk of Irish breast; her sire an Earl, her dame of Prince's blood."


19 Archaeological Journal, iii, p 20, seq.

[page 18]

Again, Bacon:- "The Duke of Bretaigne having been a host or a kind of parent or foster-fatherto the king, in his tenderness of age and weakness of fortune, did look for aid this time from King Henry." [Bacon]

In Ireland Foster children do love and are beloved by their foster-fathers, and their sept, more than of their own natural parents and kindred. [Davies]

In the opinion of the Irish, fostering has always been a stronger alliance than blood.

[Davies on Ireland]


"My father was your father's client, I

His son's scarce less than foster-brother."

[Byron's "Doge of Venice"]

There still remains in the Islands, though it is fast passing away, the custom of fosterage, &c. [Johnson, "A Journey in the Western Islands.]

That the word "soror" may be interpreted in its strictly classical sense, foster-sister, as well as real sister, irrespective of any employment as a term of mediaeval Latin, or the greater latitude habitual to it, the following inscription (Orellius, 3007 [Zurich, 1828]) offers a very strong presumption, and in support of our hypothesis we lay considerable stress upon it:--




Here we have unmistakably combined mistress and sister; the one owing allegiance to the other, her superior in blood, though equal on the score of fosterage. We do not find either in Ducange or elsewhere, any word, classical or mediaeval, which would have suggested itself to Ordericus so applicable to represent seur (de lait) as the one he has used, unless he had said "collactea" or (collactanea), an expression corrupt even to him as a mediaeval writer, and which, with French uppermost in his mind, we can quite understand his rejecting for the purer word "soror."

In the foregoing we have thus adduced different unanswerable arguments in favour of the Royal parentage of Gundreda. That the occasional errors attributed to Ordericus should be able to be brought home to him is not

[page 19]

a matter of much surprise, but in the main his authority, as a nearly contemporary historian, is trustworthy, his assertions generally supported by other chroniclers and historians, and on this one particular point, the parentage of Gundreda, he has not erred, if the above is the sense in which he intended the passage to be understood, and that he did so is evident, seeing that on the score of "consanguinity" all evidence goes otherwise to prove the expression totally unfounded, and (one may add) impossible.

Finally, should others concur in the same view with ourselves, it will happily tend to reconcile the opposite opinions of some modern writers on this disputed subject.