Читать книгу Robert Curthose, Duke of Normandy - Charles W. David - Страница 11

Table of Contents

[1] “Inter bellicas patris alas excrevit primaevo tirocinio, parenti morem in omnibus gerens.” G. R., ii, p. 459.

[2] Practically all the sources bear witness to Robert’s courage and special prowess in arms. E.g., Ordericus, ii, p. 295; iii, p. 262; William of Malmesbury, G. R., ii, pp. 459–460, 463; Interpolations de Robert de Torigny, in William of Jumièges, pp. 267, 284; Guibert of Nogent, in H. C. Oc., iv, p. 149. For the exaggerations to which this was carried in later tradition see infra, pp. 190–197.

[3] These qualities will become more evident in the sequel. Stenton characterizes Robert as “a gross anticipation of the chivalrous knight of later times.” William the Conqueror, p. 349.

[4] William of Malmesbury, G. R., ii, p. 459; Ordericus, ii, p. 295; iii, p. 262.

[5] The inimitable characterization of Ordericus Vitalis is worthy of reproduction in full. “Omnes ducem Rodbertum mollem esse desidemque cognoscebant … Erat quippe idem dux audax et validus, multaque laude dignus; eloquio facundus, sed in regimine sui suorumque inconsideratus, in erogando prodigus, in promittendo diffusus, ad mentiendum levis et incautus, misericors supplicibus, ad iustitiam super iniquo faciendam mollis et mansuetus, in definitione mutabilis, in conversatione omnibus nimis blandus et tractabilis, ideoque perversis et insipientibus despicabilis; corpore autem brevis et grossus, ideoque Brevis Ocrea a patre est cognominatus. Ipse cunctis placere studebat, cunctisque quod petebant aut dabat, aut promittebat, vel concedebat. Prodigus, dominium patrum suorum quotidie imminuebat, insipienter tribuens unicuique quod petebat, et ipse pauperescebat, unde alios contra se roborabat.” Ibid., iii, pp. 262–263. Cf. Ralph of Caen in H. C. Oc., iii, pp. 616, 642; William of Malmesbury, G. R., ii, pp. 459–463.

[6] Two charters dated 24 May 1096 at Bayeux, ‘xviiii. anno principatus domni Roberti Willelmi regis Anglorum filii ducis Normannie,’ the one by Robert himself and the other by Odo of Bayeux and attested by Robert. Haskins, pp. 66–67, nos. 3, 4, and n. 19. The style here employed of dating the reign from 1077–78 is unusual. It is ordinarily dated from Robert’s actual accession to the duchy upon the death of the Conqueror in 1087. Cf., e.g., Davis, Regesta, nos. 308, 310.

[7] Ordericus Vitalis makes Robert say: “Quid ergo faciam, vel quid meis clientibus tribuam? … Mercenarius tuus semper esse nolo. Aliquando rem familiarem volo habere, ut mihi famulantibus digna possim stipendia retribuere.” Ordericus, ii, p. 378. Cf. Achille Luchaire, La société française au temps de Philippe-Auguste (Paris, 1909), pp. 280–282, where it is pointed out that such demands and the quarrels and the open warfare which frequently resulted from them were perfectly characteristic of the feudal age.

[8] Ordericus, ii, pp. 294, 377 ff.; William of Malmesbury, G. R., ii, p. 459; Registers of Gregory VII, bk. vii, no. 27, in Bibliotheca Rerum Germanicarum, ed. Philipp Jaffé (Berlin, 1864–73), ii, pp. 420–421.

[9] The date at which the quarrel began is uncertain. It must have been after 13 September 1077, when Robert was present with his parents and William Rufus at the dedication of Saint-Étienne at Caen. Supra, p. 16. The siege of Gerberoy, which marks its termination, took place in December and January 1078–79. Infra, n. 38.

[10] William of Malmesbury, G. R., ii, pp. 316–317, 459–460; A.-S.C., a. 1079; Florence of Worcester, ii, p. 12; Chronicon Monasterii de Hyda, in Liber Monasterii de Hyda, ed. Edward Edwards (London, 1866), p. 297; Ordericus, ii, pp. 294–295, 377 ff.; Interpolations de Robert de Torigny, in William of Jumièges, p. 268; Registers of Gregory VII, bk. vii, no. 27, in Jaffé, Bibliotheca, ii, pp. 420–421.

[11] Interpolations de Robert de Torigny, in William of Jumièges, p. 268; cf. Ordericus, ii, pp. 294–295, 389.

[12] Ordericus Vitalis is the only early writer who treats in detail of the quarrels between Robert and the Conqueror. He discusses them at length in two places (ii, pp. 294–298, 377–390), but unfortunately his accounts are confused and very difficult to disentangle. There clearly were two quarrels and two periods during which Robert was in exile. Ordericus himself (ii, p. 390) is specific with regard to this; and we know independently that the first quarrel—followed by a relatively short period of exile—ended in the reconciliation after the siege of Gerberoy (1079) and that Robert was again in exile at the time of the Conqueror’s death (1087). Pretty clearly the second exile was for a longer period than the first. But the two accounts of Ordericus do not deal each with one of these quarrels. Rather they both purport to relate to the earlier quarrel and to the banishment which followed it. Yet it is obvious that Ordericus, lacking contemporary knowledge of the events, has confused the two episodes and has related incidents of the latter as if they belonged to the former. For example (ii, p. 381), he represents Robert as wandering in exile for a period of five years. Clearly this was not after the first quarrel, to which he relates it, since that could have been followed by no such extended banishment. In the narrative detail which follows I have attempted to disentangle the accounts of Ordericus Vitalis conjecturally, striving to preserve something of the vivacity of style of the original, without supposing that I have been able to arrive at rigorous historical accuracy. Ordericus’s own narrative is obviously in a high degree a work of imagination.

[13] Ordericus, ii, pp. 294–295.

[14] Ibid., pp. 378–380.

[15] Ibid., pp. 294–295.

[16] Ordericus, ii, pp. 295–296.

[17] Ivo and Alberic of Grandmesnil are mentioned by name.

[18] Ordericus, ii, pp. 295–296.

[19] Ordericus, ii, p. 296.

[20] Ibid., pp. 296–298. Elsewhere Ordericus gives another list as follows: Robert of Bellême, William of Breteuil, Roger de Bienfaite, Robert Mowbray, William de Moulins, and William de Rupierre. Ibid., pp. 380–381. Robert of Bellême is the only one appearing in both lists, and it would be rash to assume that all the foregoing supported Robert Curthose against the king in his first rebellion. But if Ordericus Vitalis is to be trusted, they were all at one time or another associated in Robert’s treason.

[21] Ibid., p. 297.

[22] Ibid., p. 295; cf. p. 297: “cum Rotrone Mauritaniensi comite pacem fecit.”

[23] Ordericus, ii, pp. 297–298.

[24] Ibid., p. 298. Freeman’s interpretation of this passage regarding Aimeric de Villeray and the dapifer of the king of France, which differs greatly from that which I have given, appears to be based upon a careless and absolutely wrong reading of the Latin text. Norman Conquest, iv, pp. 639–640.

[25] This hypothesis would help to explain the vague statement of Ordericus Vitalis: “Galli et Britones, Cenomanni et Andegavenses, aliique populi fluctuabant, et quem merito sequi deberent ignorabant.” Ordericus, ii, p. 297.

[26] A.-S. C., a. 1077: “This year a peace was made between the king of France and William king of England, but it lasted only a little while.” Henry of Huntingdon, Historia Anglorum, ed. Thomas Arnold (London, 1879), p. 206; cf. Fliche, Philippe Iᵉʳ, p. 274.

[27] “Philippum … semper infidum habuit, quod scilicet ille tantam gloriam viro invideret quem et patris sui et suum hominem esse constaret.” William of Malmesbury, G. R., ii, p. 316.

[28] Ordericus, ii, p. 386.

[29] A. 1079.

[30] Ordericus, ii, p. 381. Bishop Odo died 11 November 1078. Ordericus is in error in saying that he was the brother of Robert the Frisian.

[31] Florence of Worcester, ii, p. 12: “Franciam adiit, et auxilio Philippi regis in Normannia magnam frequenter praedam agebat, villas comburebat, homines perimebat”; Chronicon, in Liber de Hyda, p. 297.

[32] Ordericus, ii, pp. 381–382.

[33] Supra, n. 12.

[34] William of Malmesbury, G. R., ii, p. 331: “aliquantula simultas inter eos innata extremis annis fuerit pro Roberto filio, cui mater militarem manum ex fisci redditibus sufficere dicebatur”; Ordericus (ii, pp. 382–383) is much more detailed.

[35] Ordericus, ii, pp. 382–383.

[36] Ibid., pp. 386–387.

[37] Ordericus, ii, pp. 386–387; cf. Florence of Worcester, ii, pp. 12–13.

[38] The Anglo-Saxon Chronicle seems to place the siege at the end of 1079, but this is an error. A.-S. C., a. 1079. The siege took place after Christmas 1078 and in the early weeks of 1079. Ordericus, ii, p. 387. This is made certain by a charter of Philip I in favor of Saint-Quentin of Beauvais, dated “in obsidione … circa Gerborredum, anno … millesimo septuagesimo viiiiⁿᵒ anno vero regni Philippi regis Francorum ixⁿᵒ xᵐᵒ.” Recueil des actes de Philippe Iᵉʳ, roi de France, ed. Maurice Prou (Paris, 1908), no. 94. Freeman, though having this charter in hand, still dates the siege in 1079–80. Norman Conquest, iv, pp. 642–643. But Prou has shown conclusively that Freeman is in error and that the correct date is unquestionably January 1079. Op. cit., p. 242, n. 1.

[39] Ordericus, ii, pp. 387–388.

[40] A.-S. C., a. 1079; Florence of Worcester, ii, p. 13.

[41] A.-S. C., a. 1079; Florence of Worcester, ii, p. 13; William of Malmesbury, G. R., ii, p. 317; Henry of Huntingdon, p. 206. According to the Chronicle the king was wounded in the hand, according to Florence in the arm. The Chronicon in Liber de Hyda, p. 279, is still different, stating that the king was wounded in the foot by an arrow.

[42] A.-S. C., a. 1079. Freeman with patriotic pride makes much of this exploit of Tokig the Englishman; but there appears to be no valid reason for accepting, as Freeman does, this version from the Chronicle and rejecting the different version of Florence of Worcester. Norman Conquest, iv, pp. 643–644; cf. pp. 850–852.

[43] Florence of Worcester, ii, p. 13.

[44] William of Malmesbury, G. R., ii, p. 317; A.-S. C., a. 1079; Florence of Worcester, ii, p. 13; Henry of Huntingdon, pp. 206–207.

[45] Florence of Worcester, ii, p. 13.

[46] G. R., ii, p. 317.

[47] Henry of Huntingdon, p. 207: “Maledixit autem rex Roberto filio suo”; Chronicon, in Liber de Hyda, p. 297: “Cumque sanguinem defluere cerneret, terribiliter imprecatus est ne unquam Robertus filius suus haereditatis suae iura perciperet”; Annales de Wintonia, in Annales Monastici, ii, p. 32; cf. William of Malmesbury, G. R., ii, p. 460.

[48] Prou, Actes de Philippe Iᵉʳ, no. 94.

[49] Friendly relations between the Conqueror and Philip are implied in the statement of Ordericus (ii, p. 390) that the king of France sent ambassadors to urge a reconciliation between William and Robert. Infra, p. 29.

[50] Ordericus, ii, p. 388.

[51] A.-S. C., a. 1079.

[52] Vita Beati Simonis Comitis Crespeiensis Auctore Synchrono, in Migne, clvi, col. 1219. We have here chronological data of some importance. St. Simon was present at Compiègne at the translation of the Holy Shroud from its ivory casket to the magnificent golden reliquary which Queen Matilda had presented to the church of Saint-Corneille; and on the next day (in crastino itaque solemnitate peracta) he proceeded to Normandy, where he acted as mediator between the Conqueror and his rebellious son. A charter by Philip I informs us that the translation of the Holy Shroud at Compiègne took place on the fourth Sunday of Lent. Prou, Actes de Philippe Iᵉʳ, no. 126. St. Simon, therefore, left Compiègne for Normandy on the Monday after Midlent. The year, however, remains in doubt. Presumably it was 1079 or 1080, probably the latter. Philip’s charter (dated 1092) refers to the translation only incidently and gives no information as to the year in which it occurred. Ordericus Vitalis (ii, p. 389) indicates that the peace negotiations were protracted: “Frequenti colloquio Normannici proceres regem allocuti sunt.” It cannot certainly be said that the reconciliation had been consummated earlier than Easter (12 April) 1080, on which date Robert joined with the king in the attestation of a charter. Davis, Regesta, no. 123. Gregory VII, writing on 8 May 1080, speaks of it as a recent event. Infra, n. 55. Émile Morel, editor of Cartulaire de l’abbaye de Saint-Corneille de Compiègne (Montdidier, 1904–09), i, p. 53, says that the translation of the relic took place on 3 April 1082, but he cites no authority, and I have been able to find none. Jean Pillet says: “Il est constant par des manuscrits qui parlent de cette translation, qu’elle a été faite … en 1081.” Histoire du château et de la ville de Gerberoy (Rouen, 1679), p. 85. But he does not indicate where these ‘manuscripts’ are to be found, and his method of dealing with chronological problems is so arbitrary as to inspire little confidence.

[53] Ordericus, ii, pp. 388–390.

[54] Supra, n. 52. It may also be noted that the raid of King Malcolm, though it occurred in 1079, did not cause the king to go to England until 1080. Infra, p. 31.

[55] Registers of Gregory VII, bk. vii, no. 27, in Jaffé, Bibliotheca, ii, pp. 420–421. The letter is of more than passing interest, since it throws much light upon the matters which had been in controversy and is strongly confirmatory of the narrative sources. “Insuper monemus et paterne precamur, ut menti tuae semper sit infixum, quam forti manu, quam divulgata gloria, quicquid pater tuus possideat, ab ore inimicorum extraxerit; sciens tamen, se non in perpetuum vivere, sed ad hoc tam viriliter insistere, ut eredi alicui sua dimitteret. Caveas ergo, fili dilectissime, admonemus, ne abhinc pravorum consiliis adquiescas, quibus patrem offendas et matrem contristeris … Pravorum consilia ex officio nostro praecipimus penitus dimittas, patris voluntati in omnibus adquiescas. Data Rome 8 idus Maii, indictione 3.”

It may also be noted that on the same day Gregory wrote letters of courtesy to William and Matilda. But in both he confined himself to generalities and said nothing of consequence, tactfully avoiding all reference to Robert or to the recent family discord. Ibid., nos. 25, 26.

[56] E.g., 1080, April 12, [Rouen?] (Davis, Regesta, no. 123); 1080, July 14, Caen (ibid., no. 125); 1080, [presumably in Normandy] (ibid., nos. 126, 127); 1081, February, [London] (ibid., no. 135); [1078–83, perhaps 1081], February 2, Salisbury (Historia et Cartularium Monasterii S. Petri Gloucestriae, ed. W. H. Hart, London, 1863–67, i, no. 411); 1081, Winchester (Davis, Regesta, no. 140); 1082, June 24, Oissel (ibid., nos. 145, 146); 1082, Downton (ibid., no. 147); 1082 (ibid., nos. 149, 150); [c. 1082] (ibid., no. 158); 1083, July 18 (ibid., no. 182); 1083 (Chartes de S.-Julien de Tours, no. 37); [1079–82] (Davis, Regesta, nos. 168–173); cf. ibid., 165, 175, 183a.

[57] A.-S.C., a. 1079; Florence of Worcester, ii, p. 13.

[58] Presumably they went over together, though we have no record of their actual crossing. They were still at Caen in Normandy 14 July 1080. Davis, Regesta, no. 125.

[59] Chronicon Monasterii de Abingdon, ed. Joseph Stevenson (London, 1858), ii, p. 9; Simeon, H. R., p. 211.

[60] Chronicon de Abingdon, ii, p. 9.

[61] Simeon, H. R., p. 211.

[62] “Proinde ut regno Angliae principatus Scotiae subactus foret, obsides tribuit.” Chronicon de Abingdon, pp. 9–10. Simeon of Durham says rather contemptuously that Robert returned from Eccles “nullo confecto negotio.” H. R., p. 211. But this statement is hardly inconsistent with the Abingdon account. A Durham writer, thirsting for vengeance, might very well use it in spite of the results accomplished by Robert’s peaceful negotiations. William of Malmesbury uses very similar language of the expedition of William Rufus eleven years later: “Statimque primo contra Walenses, post in Scottos expeditionem movens, nihil magnificentia sua dignum exhibuit.” G. R., ii, p. 365. The Abingdon account is circumstantial, and the presence of the abbot indicates a sure source of information, though perhaps a biassed one.

[63] Simeon, H. R., p. 211.

[64] Chronicon de Abingdon, ii, p. 10.

[65] Davis, Regesta, nos. 135, 140; cf. Hist. et Cart. S. Petri Gloucestriae, i, no. 411, a charter of 1078–83, perhaps of 1081.

[66] Supra, p. 14.

[67] On the date (September-October 1076) see Halphen, Anjou, p. 182; Prou, Actes de Philippe Iᵉʳ, nos. 83, 84; Annales dites de Renaud, in Recueil d’annales angevines et vendômoises, ed. Louis Halphen (Paris, 1903), p. 88.

[68] Ibid. On the Norman siege of Dol in general see Fliche, Philippe Iᵉʳ, pp. 271–272.

[69] Ordericus, ii, p. 256.

[70] “Turbulentis tempestatibus, quas a Cenomannensibus et Normannis permotas esse diximus, fomes (ut ferunt) et causa fuit Rodbertus regis filius.” Ibid., p. 294; cf. p. 297.

[71] Halphen, relying upon the Annales de Saint-Aubin, has assigned Fulk’s first attack upon La Flèche to 1076, suggesting that Fulk launched it while the Conqueror was engaged in the north at the siege of Dol. Anjou, pp. 182–183. These conclusions, however, seem too dogmatic. There is no evidence which indicates a connection between the attack upon La Flèche and the king’s Breton enterprise; and it seems hardly likely that Fulk would have entered upon an undertaking against La Flèche which proved beyond his powers, while he was also operating against the Conqueror in Brittany. Further, the date 1076 from the Annales de Saint-Aubin (Halphen, Annales, p. 5) is not to be relied upon: because (1) the numeral “mlxxvi” is entered twice in the MS., the entry concerning La Flèche being the second of the two, and no such repetition appears elsewhere in these annals. We are, therefore, forewarned of a scribal error. And (2) the probability of such an error is made stronger by the fact that MSS. C, A, and B all read “mlxxvii,” while the Annales de Saint-Florent (ibid., p. 119) read “mlxxviii.” Having no other chronological data than are furnished by these meagre and uncertain annals, it is impossible to fix the date of the first attack upon La Flèche. It may have taken place in 1076, 1077, or 1078. On the whole, one of the later dates seems more probable than 1076, in view of the vague indications of some connection with Robert’s rebellion (supra, n. 70), and in view of the fact that Fulk was involved in Breton affairs in 1076.

[72] Ordericus, ii, p. 256. Ordericus says that Fulk had the support of Hoël, duke of Brittany; but his narrative is confused—he apparently puts together the first and second sieges of La Flèche and treats them as one—and it is impossible to say whether Breton aid was given during Fulk’s first or second expedition.

[73] “Blessé grièvement à la jambe, à la suite d’un accident de cheval, et quittant le siège de la Flèche pour se faire transporter par eau à Angers.” Halphen, Anjou, p. 311, no. 233—from an eighteenth century copy of an undated notice in the cartulary of Saint-Nicolas of Angers.

[74] “Eo tempore quo Willelmus rex Anglorum cum Fulcone Andegavensi comite iuxta castellum Vallium treviam accepit.” Cartulaire de Saint-Vincent, no. 99. The document is undated, but it is witnessed by Abbot William of Saint-Vincent, who was appointed bishop of Durham 5 November 1080 and consecrated 3 January 1081. The ‘trevia’ of this document, therefore, cannot refer to the treaty of La Bruère (1081) and it seems probable that it refers to a truce concluded after the failure of the first attack upon La Flèche.

[75] “MLXXXI … Fulcho Rechim castrum Fisse cepit et succendit.” Annales de Saint-Aubin, in Halphen, Annales p. 5. “MLXXXI. In hoc anno … comes Andecavorum Fulcho iunior obsedit castrum quoddam quod Fissa Iohannis dicitur atque cepit necnon succendit.” Annales dites de Renaud, ibid., p. 88. Ordericus Vitalis does not admit that La Flèche was taken, doubtless because of the confusion which he makes between the two sieges. Ordericus, ii, p. 256.

[76] On the exaggeration of numbers by mediaeval chroniclers, see J. H. Ramsay, “Chroniclers’ Estimates of Numbers and Official Records,” in E. H. R., xviii (1903), pp. 625–629; and cf. the same, “The Strength of English Armies in the Middle Ages,” ibid., xxix (1914), pp. 221–227.

[77] Ordericus (ii, pp. 256–257) has given a spirited account; but he manifestly wrote without any clear conception of the geographical or topographical setting of the proposed engagement, and all efforts to render his account intelligible have proved in vain. For a discussion of the problems involved and of the conjectures which have been made, see Halphen, Anjou, p. 184.

[78] Freeman conjectures that this is the “ubiquitous Hubert,” cardinal legate of Gregory VII. Norman Conquest, iv, p. 562.

[79] Ordericus, ii, pp. 257–258.

[80] “Qui et ipse a Fulcone bello lacessitus, obsidibus pacis pro fide datis fratre suo, consule videlicet Mauritanie, et filio suo et multis aliis, recessit.” Halphen, Annales, p. 88.

[81] “Haec nimirum pax, quae inter regem et praefatum comitem in loco, qui vulgo Blancalanda vel Brueria dicitur, facta est, omni vita regis ad profectum utriusque provinciae permansit.” Ordericus, ii, p. 258.

[82] Halphen, Anjou, pp. 185–186; Latouche, Maine, p. 79.

[83] Halphen, Anjou, p. 186; Latouche, Maine, p. 39.

[84] Davis, Regesta, nos. 145, 146; cf. nos. 149, 150, 158.

[85] Ibid., no. 147.

[86] Ibid., no. 182. He also attests with the king, queen, and William Rufus, in 1083, a charter in favor of Saint-Julien of Tours. Chartes de S.-Julien de Tours, no. 37.

Davis cites a “confirmation by William I” in favor of the abbey of Lessay, which is attested by Robert, along with King William, Bishop Odo of Bayeux, Henry “the king’s son,” and others, and which he assigns to 1084, remarking, “The appearance of Bishop Odo is strange, considering that he was at this time in captivity.” Regesta, no. 199. It cannot, of course, be supposed that the Conqueror really gave a confirmation in company with Odo of Bayeux while he was holding the latter in close confinement as a most bitter and dangerous enemy; and some other explanation of the apparent inconsistency must be found. A glance at the document as printed in full in Gallia Christiana (xi, instr., cols. 228–229) makes it clear that we have to do here not with a single diploma of known date, but rather with a list of notices of gifts. At the head of the list stands the record of a grant by Roger d’Aubigny, dated 1084, and accompanied by a list of witnesses. Then follow no less than six separate notices of grants, each with its own witnesses; and finally come the attestations of King William, Bishop Odo, Henry the king’s son, Count Robert, and others. There is no reason to suppose that these attestations are of the year 1084—a date which applies certainly only to the first grant in the list—and they are evidently of a later period, perhaps of the year 1091, when the abbey of Lessay might naturally seek a confirmation from the three brothers after the pacification which followed the siege of Mont-Saint-Michel. The king in question, therefore, is probably William Rufus rather than the Conqueror. The style of Henry “filii regis” is certainly surprising, but it can be matched in another document, also probably of the year 1091. Davis, Regesta, no. 320; cf. The New Palaeographical Society, Facsimiles of Ancient Manuscripts, etc. (London, 1903-), 1st series, pt. 2, plate 45a and text.

[87] Interpolations de Robert de Torigny, in William of Jumièges, pp. 265, 267–268; William of Malmesbury, G. R., ii, p. 332; Ordericus, iii, p. 268.

[88] “Serenitas pacis diu quaesitae inter regem et filium eius celeriter obnubilata est. Protervus enim iuvenis patrem sequi, vel ei obedire dedignatus est. Animosus vero princeps ob ignaviam eius crebris eum redargutionibus et conviciis palam iniuriatus est. Unde denuo post aliquod tempus, paucis sodalibus fretus, a patre recessit, nec postea rediit; donec pater moriens Albericum comitem, ut ducatum Neustriae reciperet, in Galliam ad eum direxit.” Ordericus, ii, p. 390.

[89] Ibid., p. 380.

[90] Robert appears in no reliable charter between the queen’s death and his own accession to the duchy.

[91] Because of the extended period during which he is not to be found in the charters, and because Ordericus (ii, p. 381) speaks of his being in exile “ferme quinque annis.” Cf. supra, n. 12.

[92] Ordericus, ii, p. 381.

[93] “Robertus, patre adhuc vivente, Normanniam sibi negari aegre ferens, in Italiam obstinatus abiit, ut, filia Bonifacii marchionis sumpta, patri partibus illis adiutus adversaretur: sed, petitionis huiusce cassus, Philippum Francorum regem contra patriam excitavit.” William of Malmesbury, G. R., ii, p. 332.

[94] “Porro ille, quae ab amicis liberalibus ad subsidium sui accipiebat, histrionibus et parasitis ac meretricibus insipienter distribuebat; quibus improvide distractis, egestate gravi compressus mendicabat, et aes alienum ab externis foeneratoribus exul egenus quaeritabat.” Ordericus, ii, p. 382. Ordericus reserves his worst criticisms for Robert’s later life, but doubtless the moral decay set in early. Cf. ibid., iv, pp. 105–106.

[95] Ibid., iv, pp. 81–82. The author embellishes his account with a further tale of how the boys were brought up in obscurity by their mother, who in later years took them to Robert, then become duke, and proved their parentage by undergoing the ordeal of hot iron.

[96] Ordericus, iv, p. 82; Florence of Worcester, ii, p. 45; William of Malmesbury, G. R., ii, p. 333.

[97] Ordericus, ii, p. 82.

[98] Ibid., iii, p. 320.

[99] Ibid., ii, p. 390; iii, p. 228; William of Malmesbury, G. R., ii, p. 338.

[100] It is clear that the war grew out of the inevitable antagonism between the interests of the two monarchs, and particularly out of the determination on King William’s part to reassert the Norman claim to the Vexin. Ordericus, iii, pp. 222–225. As to the immediate provocation, Ordericus explains that the Conqueror’s attack upon Mantes was in retaliation for predatory incursions which certain lawless inhabitants of the city had been making across the border into Normandy (ibid., p. 222); William of Malmesbury attributes it to an insulting jest which Philip had made about William’s obesity (G. R., ii, p. 336); while Robert of Torigny ascribes it to the aid which Philip had been giving Robert Curthose against his father (Interpolations de Robert de Torigny, in William of Jumièges, p. 265).

[101] Ordericus, iii, pp. 222–226; William of Malmesbury, G. R., ii, p. 336; A.-S. C., a. 1086; Florence of Worcester, ii, p. 20; Interpolations de Robert de Torigny, in William of Jumièges, p. 265.

[102] William of Malmesbury, G. R., ii, pp. 332, 338; Ordericus, iii, p. 228; cf. Chronicon in Liber de Hyda, p. 298. Robert of Torigny is more specific: “Cum igitur in Pontivo apud Abbatisvillam, cum sui similibus iuvenibus, filiis scilicet satraparum Normanniae, qui ei, quasi suo domino futuro, specie tenus obsequebantur, re autem vera novarum rerum cupiditate illecti, moraretur et ducatum Normanniae, maxime in margine, excursionibus et rapinis demoliretur.” Interpolations de Robert de Torigny, in William of Jumièges, p. 268.

[103] This is the plain inference from Ordericus, iii, p. 242; William of Malmesbury, G. R., ii, pp. 332,337; De Obitu Willelmi, in William of Jumièges, pp. 146–147.

[104] That is, (1) before the Conquest (supra, p. 12), (2) after the Conquest on the occasion of the king’s illness at Bonneville (supra, p. 15), (3) at the reconciliation after the siege of Gerberoy (supra, p. 29). Cf. also the charter of Stigand de Mézidon, 1063, in Le Prévost, Eure, i, p. 562.

[105] Ordericus, iii, pp. 239, 242–243.

[106] Ibid., p. 242.

[107] De Obitu Willelmi, in William of Jumièges, pp. 146–147.

[108] That Maine was included is clear from the fact that Robert’s right to rule there was not questioned. Wace, too, is specific:

E quant Guilleme trespassa,

Al duc Robert le Mans laissa.

Roman de Rou, ed. Hugo Andresen (Heilbronn, 1877–79), ii, p. 416. The Annales de Wintonia are clearly wrong in stating that the Conqueror left Maine to Henry. Annales Monastici, ii, p. 35.

[109] Ordericus, ii, p. 390: “pater moriens Albericum comitem, ut ducatum Neustriae reciperet, in Galliam ad eum direxit”; Interpolations de Robert de Torigny, in William of Jumièges, p. 268.

[110] William of Malmesbury, G. R., ii, p. 338.

[111] Interpolations de Robert de Torigny, in William of Jumièges, p. 265.