APPENDIX A
William the Conqueror was the bastard son of Robert I, duke of Normandy. He was born at Falaise, and his mother was Herleve, a girl of that town.
If, however, these facts respecting William's birth may be thus baldly stated, it must be added that all the circumstances surrounding the event are obscure, and that the evidence relating to it is tenuous and contradictory. Thus, William of Poitiers has here no precise information to impart; William of Jumièges merely records Robert's paternity; and William of Malmesbury, who gives a graphic account of Robert's first encounter with William's mother, does not mention her name or the place where the meeting took place.1 Herleve's name seems to have been supplied for the first time by Ordericus Vitalis;2 and that William was born at Falaise appears first to have been categorically asserted by writers of the twelfth century.3 There was, indeed, a later legend that Herleve was of Flemish stock, and that William was born at Rouen.4 But such stories have nothing to commend them.5 The tradition that the Conqueror was born at Falaise is very strong, and can (as will be seen) be supported by circumstantial evidence. It may be accepted without undue misgiving.
The origins of Herleve were humble. Contemporary writers are discreetly silent about her father, but Ordericus Vitalis gives his name as Fulbert, and also tells that when at a later date William was besieging Alençon, the defenders waved hides and skins from the walls to taunt the duke with the fact that his mother's relatives were polinctores.6 The very firm tradition that Herleve's father was a tanner is thus supported, and the tanneries of Falaise were famous. It must, however, be noted that the word polinctor could more readily be translated as embalmer.7 The conclusion would seem to be that Herleve's father was very probably named Fulbert, and that Fulbert was very probably a tanner, but perhaps a man who prepared corpses for burial.
The date of William's birth has been exhaustively discussed. I have accepted the conclusions of M. Henri Prentout, as reinforced by the testimony examined by Mr L. C. Loyd and Mr G. H. White in the Complete Peerage.8 The evidence is, however, not as conclusive as might be wished. In the death-bed speech, which Ordericus Vitalis puts into the mouth of William the Conqueror, it is stated that at that time (i.e. 9 September 1087) William was sixty-four years of age, and this would put his birth in or about 1023.9 In view of other testimony, this may be confidently set aside. William of Malmesbury remarks that when Robert I departed on pilgrimage, that is to say in January 1035 or at the very earliest at the end of 1034, the Conqueror was seven years old,10 and Ordericus says that he was then a boy of eight years.11 These remarks would place William's birth in 1027 or early in 1028. Similarly, William is said to have been eight years old when Robert died in July 1035.12
It will appear how difficult it is to give precision to these remarks, many of which might be translated with almost equal propriety as ‘in his seventh (or eighth) year’, or ‘seven (or eight) years old’. Ordericus, and indeed Wace, are, moreover, self-contradictory. An early narrative De Obitu Willelmi13 written by a monk of Caen before the end of the eleventh century states, however, that in September 1087 William was in the fifty-ninth year of his life.14 This statement carries higher authority than the others which have been cited. If taken literally, it would place William's birth between 9 September 1028 and 9 September 1029. In view, however, of the other testimony which inclines to an earlier date, I am myself disposed to place the Conqueror's birth early within these limits. A date during the autumn of 1028 has therefore here been adopted, but the matter cannot be regarded as finally settled.
It is probable, though not certain, that Herleve bore to Robert another child – a girl named Adelaide who in due course married (i) Enguerrand, count of Ponthieu; (ii) Lambert of Lens; and (iii) Odo, count of Champagne.15 It is certain that the Conqueror had a sister, or half-sister, of this name, and with this career; it is also certain that she was not the daughter of Herleve's husband, Herluin, vicomte of Conteville. She may, however, have been the daughter of Robert by some mistress other than Herleve, but it is perhaps more probable that she was the Conqueror's sister of the whole blood.
Herleve's career advanced the fortunes of her kinsfolk. It seems that the obscure Fulbert became in due course a cubicularius in the ducal household,16 and something is known of Herleve's brothers. Charter evidence gives their names as Osbern and Walter.17Walter, moreover, is stated to have watched over the future Conqueror during his perilous childhood, and on one occasion to have saved his life by snatching the lad from his cot and carrying him for safety into the ‘dwellings of the poor’.18 This Walter had at least two daughters. One named Clara became a nun at Montivilliers.19 The other, Matilda, married Ralph Tesson.20 The Tessons were a very considerable family in Middle Normandy, and this marriage further illustrates the advancement of Herleve's relatives.
Some time after the Conqueror's birth, Herleve was married to Herluin, vicomte of Conteville, and to him she was to bear two famous sons, Odo, bishop of Bayeux, and Robert, count of Mortain, and at least one daughter who married William, lord of La Ferté-Macé.21 Ordericus states that Herluin's marriage to Herleve took place after 1035,22 but though the statement has won some credence, it is open to grave objections. Odo became bishop of Bayeux between October 1049 and 23 April 1050,23 and if Ordericus's assertion were accepted, Odo would then have been barely fourteen. Such an appointment would hardly have passed without notice by Odo's hostile critics, who would certainly have called attention to such a scandal at the beginning of his career. Much interest therefore attaches to the remark of William of Malmesbury that Herleve was married off to Herluin before the death of Robert.24 And I have myself very little doubt that the marriage took place very soon after the birth of the Conqueror; and that Odo, being born about 1030, was some nineteen years old when he became a bishop.
Besides Herleve, Herluin married a certain Fredesendis, and there seems little doubt that she was his second wife.25 From this union there were at least two children, Ralph and John, the former of whom may possibly be the Radulfus de Contivilla who held lands in Somerset and Devon in 1086.26
Some inferences are possible respecting the date of Herleve's death. Towards the end of his life, Herluin founded the abbey of Grestain,27 and Robert of Torigny asserted that both he and Herleve were buried there.28 It is, however, most improbable that Herleve was buried at Grestain,29 and the absence of her name from the list of benefactors to that abbey, and the presence therein of the name of Fredesendis,30 suggests very strongly that Herleve died before Herluin founded the monastery.31 The foundation of Grestain is, moreover, usually placed in or shortly after 1050.32
If Herleve, in fact, died, as is here suggested, about 1050, she can hardly have been more than forty years of age at the time of her death. Yet she had accomplished much. In view of her humble origins, her career and connexions33 may even now challenge some attention. She was a remarkable girl.
1 Will. Jum., p. 115; Will. Malms., Gesta Regum, p. 285.
2 Interp. Will. Jum., p. 157.
3 Wace, Roman de Rou (ed. Andresen), vol. II, p. 204; Benoit (ed. Michel), vol. II, p. 555.
4 Cf. J. Depoin Congrès millenaire normand, vol. I, pp. 305–309.
5 H. Prentout, Guillaume le Conquérant: légende et histoire (Caen, 1927), pp. 20–23.
6 Interp. Will. Jum., p. 171.
7 Complete Peerage, vol. XII (1), Appendix K, p. 30. The matter is further confused by the fact that Wace (Roman de Rou, vol. II, p. 204) calls Fulbert parmentier which might be rendered ‘tailor’.
8 Prentout, ‘De la Naissance de Guillaume le Conquérant’ (Études sur Quelques Points d'Histoire de Normandie (Caen, 1927), pp. 73–89); Complete Peerage, vol. XII (1), Appendix K, where all the testimony is assembled.
9 Ord. Vit., vol. III, p. 228.
10 Will. Malms., Gesta Regum, vol. II, p. 285: habebat tunc filium septennem.
11 Ord. Vit., vol. III, p. 229: puer utpote octo annorum.
12 Ord. Vit., vol. II, p. 11: tunc octo annorum erat.
13 Will. Jum., pp. 145–149.
14 anno vitae suae quinquagesimo none.
15 Rot. Scacc. Norm., vol. II, p. xxxi; Complete Peerage, loc. cit.
16 Ord. Vit., interp. Will. Jum., p. 157.
17 R.A.D.N., no. 134 (Signum Walteri avunculi comites). This may legitimately be placed beside the list of witnesses given in Lot, Saint-Wandrille, no. 17 (Osbernus avunculus comitis. Walterius frater eius).
18 Ord. Vit., vol. III, p. 229.
19 Gall. Christ., vol. XI; Instrumenta, col. 329.
20 Ibid., col. 65a.
21 L. C. Denis, Chartes de Saint-Julien de Tours, nos. 24, 29; Douglas, Domesday Monachorum, pp. 35, 36.
22 Interp. Will. Jum., p. 157.
23 Gall. Christ., vol. XI, col. 353.
24 Gesta Regum, p. 333.
25 Douglas, Domesdav Monachorum, p. 34.
26 Bréard, L'abbaye Notre-Dame de Grestain, nos. I, II; Douglas, loc. cit. But there are several Contevilles.
27 Bréard, loc. cit.
28 Ed. Delisle, vol. II, p. 202. There is – or was recently – an inscription among the ruins to this effect.
29 Gall. Christ., vol. XI, col. 83.
30 Bréard, op. cit., nos. I, II.
31 Herluin was alive in 1059 (Chevreux et Vernier, Archives de Normandie, plate V), but since none of his gifts to Grestain were in England it is probable that he died before, or very shortly after, the Conquest.
32 Bréard, op. cit., p. 20.
33 Below, Table 6.