^{page 58 note 1} David, , ‘A tract’, 383–7Google Scholar; id., RC 212–5Google Scholar; Offler, Tractate', 322–3.

^{page 58 note 2} Gransden, Antonia, Historical Writing in England c. 550 to c. 1347 (London, 1974), 122 and n. 113Google Scholar, while claiming that DIV was ‘a remarkable piece of contemporary reportage’, denied that the vita was a later addition to the libellus. These positions are difficult to reconcile. Since the vita is later than April 1109, if it and the libellus were composed at the same time, then all DIV must be later than that date. This seems incompatible with Dr Gransden's view that the treatise was a contemporary record of events in 1088.

^{page 60 note 3} See e.g. the narratives in Freeman, WR i, 95–120Google Scholar and Barlow, , WR 85–9.Google Scholar

^{page 61 note 4} As David Knowles supposed, The Monastic Order in England (Cambridge, 1949), 169, n.1.Google Scholar

^{page 61 note 5} Offler, , ‘Tractate’, 321–41.Google Scholar

^{page 61 note 6} Hoffmann, , ‘Zur Echtheit’, 438–40.Google Scholar

^{page 61 note 7} See the note 56 to DIV.

^{page 62 note 8} Southern, Anselm (1963), 148 and n. 1.Google Scholar

^{page 62 note 9} Hoffmann, , ‘Zur Echtheit’, 439Google Scholar; Symeon, , HDE 132Google Scholar. See note 98 to DIV.

^{page 62 note 10} HR 214–7.Google Scholar

^{page 62 note 11} Offler, , ‘Tractate’, 341.Google Scholar

^{page 62 note 12} Gibson, , Lanfranc, 220–1.Google Scholar

^{page 63 note 13} Barlow, , EC 281, n. 46Google Scholar; id., The Norman Conquest and Beyond (London, 1983), 234, n. 7.Google Scholar

^{page 63 note 14} Barlow, , WR 85–9.Google Scholar

^{page 63 note 15} EC 281, n. 46.Google Scholar

^{page 64 note 16} See note 45 to DIV.

^{page 64 note 17} Barlow, , WR 77, n. 111, 89, 92, 168Google Scholar; see DIV lines 581–2 and note.

^{page 64 note 18} See note 28 to DIV.

^{page 64 note 19} Barlow, , WR 85, n. 160.Google Scholar

^{page 64 note 20} Orderic iv, pp. xxvii–xxx; Offler, ‘Tractate’, 340–1. See note 57 to DIV.

^{page 65 note 21} RAN i nos. 315, 319.

^{page 65 note 22} C & S I.ii, 635, n. 2; 855, n. 1.

^{page 66 note 23} DCL ms. B.iv.40, f. 21v = Catt.vett., p. 55 entry P.Google Scholar

^{page 66 note 24} [Doyle, ‘Claxton’, 338–9, 344.]

^{page 66 note 25} Chaplais, Pierre, ‘William of St Calais and the Domesday Survey’, in Holt, J. C. (ed.), Domesday Studies (Woodbridge, 1987), 65–76Google Scholar; on the script, cf. Alexander R. Rumble, ib., 82–5.

^{page 67 note 26} See note 98 to DIV.

^{page 67 note 27} Printed by Rud, Thomas, Codd. mss. eccksiae cathedralis Dunelmensis Catalogas (Durham, 1825), 111Google Scholar; cf. Mynors, , DCM no. 32, p. 35.Google Scholar

^{page 67 note 28} Mynors, , DCM no. 47, p. 41Google Scholar. [For further examples of his work, see pp. 68–9 in Gullick, M., ‘The scribe of the Carilef Bible: a new look at some late-eleventh-century Durham Cathedral manuscripts’, in Medieval Book Production: Assessing the Evidence, ed. Brownrigg, L. L., (Los Altos Hills, 1990), pp. 61–83.]Google Scholar

^{page 67 note 29} Bishop & Ghaplais, ERW, pl.viii(a) and note.

^{page 67 note 30} Mynors, DCM no. 31 and pl. 20.

^{page 67 note 31} Offler, , Medieval Historians, 9.Google Scholar

^{page 68 note 32} Gullick, M., ‘The scribes of the Durham cantor's book (DCL ms B.iv.24) and the Durham Martyrology scribe’, in Anglo-Norman Durham 1093–1193, ed. Rollasen, D., Harvey, M. and Prestwich, M., (Woodbridge, 1994), 93–109.Google Scholar

^{page 68 note 33} See Offler, , ‘Tractate’, 323–4.Google Scholar

^{page 68 note 34} Described in Summary Cat.of Western MSS. in the Bodleian Library, ii, pt. ii, no. 3886, pp. 773–5.Google Scholar

^{page 68 note 35} Offler, , Medieval Historians of Durham, Durham, 1958, 23, n. 39Google Scholar; repr. as I in North of the Tees.

^{page 68 note 36} See p. 13.

^{page 69 note 37} DCL ms. B.iv.46, f. 21v = Catt.vett., p. 55.Google Scholar

^{page 69 note 38} Catt.vett., 55, 81.Google Scholar

^{page 69 note 39} Coxe, H. C., Catalogue of the Laudian mss.Google Scholar, corrected reprint, Oxford, 1975, 580; cf. A. G. Watson (ed.), Supplement to Ker, N. R., Medieval Libraries of Great Britain, 2nd. edn. (London, 1987), 31, n. 3.Google Scholar

^{page 69 note 40} Craster, , ‘Red Book’Google Scholar; cf. Dobson, R. B., Durham Priory 1400–1450 (Cambridge, 1973), 379–81.CrossRefGoogle Scholar

^{page 69 note 41} Craster, , ‘Red Book’, 518, n. 1.Google Scholar

^{page 70 note 42} ‘Red Book’, 513.Google Scholar

^{page 70 note 43} Described by Colgrave, B., Two Lines of St Cuthbert (Cambridge, 1940), 28–9Google Scholar; Pantin, W. A., ‘Some medieval English treatises on the origins of monasticism’, in Medieval Studies presented to Rose Graham (Oxford, 1950), 201–2.Google Scholar

^{page 70 note 44} [DCD, Bursar's Book L ff. 146v–147v, a list of monks serving chantries; they took turns in order of seniority in the community, making it possible to identify one of them as the older William Tode.]

^{page 70 note 45} Emden, A. B., A Biographical Register of the University of Oxford A.D. 1501 to 1540 (Oxford, 1974), p. 570Google Scholar, [with the description of H misplaced under Todde, alias ? Smerthwaite, John, p. 569].

^{page 70 note 46} Mynors, , DCM no. 88, pp. 60–1Google Scholar. Dr Doyle kindly allowed me to consult the detailed description he has prepared for his forthcoming catalogue of Cosin's mss.

^{page 70 note 47} [Doyle, , ‘Claxton’, 344]Google Scholar

^{page 71 note 48} Preface, p. vii.

^{page 73 note 1} Lines 3–28 are mostly quotation or paraphrase from Symeon, HDE 119–24, 127–8Google Scholar. But David, , ‘A tract’, 385Google Scholar erred in supposing that the author had no other source, since HDE does not give the place of St Calais's consecration nor the date of William II's. The latter could easily be found: cf. ASC 1086 [1087], ‘Florence’, ii, 20Google Scholar and HR 214Google Scholar. Only HR 211Google Scholar gives St Calais's consecration at Gloucester, though with the wrong date, 2 January.

^{page 73 note 2} Walcher was killed on 14 May 1080: Symeon, HDE 117.Google Scholar

^{page 73 note 3} sex menses et decem dies: i.e. (6 × 28) + 10 = 178 days.

^{page 73 note 4} William was chosen bishop (ab ipso rege electus) on 9 November 1080, Symeon, HDE 119Google Scholar. For his consecration by Thomas of Bayeux, archbishop of York, assisted by suffragans of Canterbury, at the Council of Gloucester on 3 January 1081, see C & S I.ii, 629 and n. 2.Google Scholar

^{page 73 note 5} William's early career is discussed by Offler, , ‘William of St Calais, first Norman bishop of Durham’, Trans. Architect. & Archaeolog. Soc. Durham & Northumberland 10 (1950), 260–5Google Scholar; reprinted as V, with unchanged pagination, in North of the Tees. He had become abbot of St Vincent-des-Prés at Le Mans by 1078.

^{page 73 note 6} St Calais's gift copy of Bede's Historia ecclesiastica to the priory is still at Durham: DCL ms. B.II. 35, ff. 36–150; cf, Mynors, , DCM no. 47, p. 41Google Scholar. A version of Bede's prose life of St Cuthbert written in Durham at this time is now in Oxford, Univ. Coll. ms. 165.

^{page 73 note 7} The date of the bishop's journey to Rome is uncertain, though it must have been between 1081 and early 1083. He witnesses with the king in Normandy on 5 September 1082, RAN i no. 146a, and in the same year at Downton in Wiltshire, RAN i no. 147. From May 1081 onwards Pope Gregory at Rome was under heavy pressure from hostile imperialist forces.

^{page 74 note 8} 28 May 1083. This is indeed in the Conqueror's seventeenth regnal year, though Symeon, HDE 122 says the eighteenth.

^{page 74 note 9} sequens libellus: cf. HR 216–7: cujus ordinem causae libellus in hoc descriptus aperte ostendit. Arnold's note is misconceived.

^{page 74 note 10} The immediate effects of this royal proclamation of the bishop's disseisin on 12 March are difficult to estimate. Most vulnerable would have been his comparatively small estates in Bedfordshire, Northamptonshire and Essex (DB i, 210b and 220; ii, 15bGoogle Scholar). While his lands in Lincolnshire and Yorkshire were clearly affected, this seems much less certain in respect of his possessions north of Tees; see below, lines 565–7, 609–10. Assuming that the bishop had left the king's court just before the proclamation, he would have been back in Durham by or shortly after the middle of March. These early dates raise problems, as Freeman, WR i, 29Google Scholar remarked. He concluded that there must have been seditious movements in south-east England before the open revolt broke out after Easter (16 April), as ASC 1087 [1088] and ‘Florence’, ii, 22Google Scholar report; cf. Barlow, , WR 75–7Google Scholar. This is perhaps implied by the libellus itself, lines 487–91 below.

^{page 74 note 11} If authentic, this letter is to be dated in late March or in April 1088. While noting its somewhat insolent tone, Barlow, , WR 83Google Scholar, regards it as ‘written in impeccable chancery style’.

^{page 74 note 12} terras meas: before the Conquest the bishops of Durham held 7 manors totalling 81 carucates in the North Riding of Yorkshire: DB i, 304bGoogle Scholar. To them William I had added two great areas in the East Riding: King Edward's manor of Howden and Morcar's of Welton, LV, f. 50v. By 1086 Durham's Yorkshire lands amounted to more than 243 carucates, though much was uncultivated. For Durham's lands in Lincolnshire, see Craster, ‘Red Book’, 529; Foster, C. W. (ed.), The Lincolnshire Domesday and the Lindsey Survey (Lincs. Rec. Soc. 19, 1924), 30–7.Google Scholar

^{page 75 note 13} Ralph Paynel: probably from the family which held Les Moutiers-Hubert (Calvados, arr. Lisieux, cant. Livart) and Hambaye (Manche, arr. Coutances, cant. Gavray): Loyd, Origins, 77Google Scholar. For his career in England, see Clay, C. T., ETC vi, 2–5, and 56–65Google Scholar for the Paynel fee. Following William Farrer, EHR 30 (1915), 282–4Google Scholar, Clay accepted the evidence of the libellus that Ralph was sheriff of Yorkshire in 1088. But it should be noted that ‘R. Painel’ witnesses a charter of Duke Robert in Normandy during 1088: RAN i no. 299. If this was our Ralph, it was more likely to have been before than after the rebellion in England; cf. Barlow, , WR 72.Google Scholar

^{page 75 note 14} For Guy (Wido), abbot of St Augustine's, Canterbury from 1087 to 1093, see HRHEW 36Google Scholar. St Calais witnesses royal writs in favour of Wido, 1087 × 1093: RAN i nos. 371,372. Wido appears as witness to a spurious charter of William I for Durham, RAN i no. *286, and to a spurious confirmation by Bishop William of a grant to the prior and convent of Durham, RAN i no. 318; cf. DEC no. *6 and pp. 48–53.

^{page 75 note 15} If this letter is authentic, its date can hardly be before early June.

^{page 76 note 16} For Howden and Welton, see note 12.

^{page 76 note 17} Odo (Eudes III, nephew of Thibaud I, count of Blois-Champagne), disinherited count of Champagne, was the third husband of Adelaide, sister or half-sister of William the Conqueror, and father of Stephen of Aumale; on him, see Orderic, ii, 264, n. 3 and iv, 182, n. 2; English, Barbara, The Lords of Holderness 1086–1260 (Oxford, 1979), 9–13Google Scholar; Bur, M., ‘Les comtes de Champagne et la “Normanitas”’, Battle 3 (1980), 29Google Scholar. Odo appears to have acquired his lordship of Holderness in Yorks ER in 1086/7: GEC i, 351–2; EYC iii, p. 26Google Scholar. His only appearances in RAN i are as witness to royal charters issued overseas, nos. 30, 323. Complicity in Robert Mowbray's conspiracy in 1095 cost him his lands in Holderness, which were granted to Arnulf of Montgomery. Mason, ‘Roger of Montgomery’, 17.

The Breton count Alan Rufus, lord of Richmond and founder of St Mary's, York, was in the first flight of great post-Conquest landowners: Mason, J. F. A., ‘The “Honour of Richmond” in 1086’, EHR 76 (1963), 703–4CrossRefGoogle Scholar. After the imprisonment of Morcar in 1071 William I had granted the western half of the North Riding to count Alan, who must be esteemed ‘the greatest man in the north of England’: Southern, Anselm (1963), 184.Google Scholar

^{page 76 note 18} partem terrarum mearum: thus the libellus does not make bishop William claim to have been disseized of all his lands.

^{page 77 note 19} Quod si dissaisitus: foreshadowing the appeal to the principle spoliatus ante omnia restituendus on which according to the libellus St Calais consistently based his resistance to a feudal judgement against him in 1088: Nam nee convocari ad causam nee diiudicari potest expoliatus vel expulsus, quia non est privilegium, quo expoliari possit iam nudatus (Pseudo-Isidore, Eusebius, Decret, c. xii, ed. Hinschius, , 237Google Scholar; cf. Actio quintae Synodi sub Simacho, Hinschius, 676Google Scholar; Sixtus I, Decret, c. vi, Hinschius, 109Google Scholar; Sixtus II, Decret, c. vi, Hinschius, 192Google Scholar; Julius, Decret, c. xii, Hinschius, 468Google Scholar). On the extracts from Pseudo-Isidore concerning the exceptio spolii in Lanfranc's collection of canon law, see F. Joüon des Longrais, ‘Les réformes d'Henry II en matière de saisine’, Rev. hist, de droit fiançais et étranger, 4th ser., 15 (1936), 548Google Scholar. Though William will not submit to trial before his possessions are restored, he offers to clear himself of guilt in the king's court by canonical purgation.

^{page 77 note 20} There is no other evidence for this meeting between Rufus and St Calais. If it took place, it was presumably in south-east England at some time during the summer, since from mid-April until early July Rufus was preoccupied in besieging Tonbridge, Pevensey and Rochester. Barlow, WR 83, n. 150 does not specify the ‘several reasons’ for supposing a date after the fall of Rochester in July.

^{page 77 note 21} Thomas of Bayeux, archbishop of York 1070–1100.

^{page 77 note 22} Lanfranc, archbishop of Canterbury and primate 1070–1089.

^{page 78 note 23} If Bishop William had returned to Durham during June, the letter which follows would have to be dated soon after that. But on Barlow's reckoning (note 20) it cannot be dated before July.

^{page 78 note 24} I take the sense of this convoluted passage (lines 138–42) to be: If Rufus insists that the bishop shall clear himself in lay fashion (i.e. from a charge of treason in the curia regis), the bishop is prepared to submit to a preliminary lawful finding (i.e. by ecclesiastics) on this issue, provided that, if he considers this ecclesiastical judgement unjust, he may, while remaining within the protection of the king's peace, appeal against it, as canon law demands, in that place where a canonical judgement is to be had (that is, at the papal curia; cf. lines 163–4 below). In effect, with subtlety and impudence the bishop offers to toss a coin with Rufus on the terms: ‘Heads you lose, tails I win’. The ubi canonice iudicatum fuerit in line 142 should be compared with Anselm at Rockingham: me paratum inveniet ei sicut debeo, et ubi debeo, respondere, Eadmer, HN 61.Google Scholar

^{page 78 note 25} The bishop renews his offer to clear himself in the curia régis by canonical purgation (recto iudicio); cf. lines 107–08, 131–5.

^{page 78 note 26} Talis enim dominus: that intended further generosity by William the Conqueror towards Durham was frustrated by his death is suggested by Symeon, , HDE 124.Google Scholar

^{page 78 note 27} exercitum suum misit: presumably after Rufus had overcome his troubles in southern England by mid-July. A date in August or early September seems indicated; see below, lines 156, 245–6, 649–50.

^{page 79 note 28} For Count Alan and Count Odo see note 17. On Roger le Poitevin, third son of Roger of Montgomery, earl of Shrewsbury, cf. Mason, , ‘Roger of Montgomery’, 1–28Google Scholar; Chandler, Victoria, ‘The last of the Montgomerys: Roger le Poitevin and Arnulf’, Historical Research 62 (1989), 1–14CrossRefGoogle Scholar. By 1086 Roger le Poitevin had forfeited his lands in England, as Galbraith, V. H., The Making of Domesday Book (Oxford, 1961), pp. 187–8Google Scholar, pointed out. It is uncertain whether he was restored by the Conqueror, or by Rufus before 1088, surviving his family's participation in the revolt against the king, or by Rufus immediately after and despite the revolt: Mason, 16. Whether he was one of the two unnamed sons of Earl Roger who according to ASC 1087 [1088] aided their eldest brother Robert of Bellême against Rufus at the siege of Rochester is perhaps uncertain; it has been denied: Mason, , ‘Roger of Montgomery’, 16Google Scholar; Lewis, C. P., ‘The King and Eye: a study in Anglo—Norman polities’, EHR 104 (1989), 572, 575–6Google Scholar. If he did dabble in rebellion in 1088, he can hardly have been reconciled with Rufus before the surrender of Rochester, which is probably to be dated in early July; cf. Dr Chibnall's note 1 to Orderic, iv, 134. About Roger le Poitevin's title the libellus is inconsistent. Though Barlow, EC, 283 refers to ‘three earls making the pact of 8 September’, in fact the libellus here avoids calling Roger comes. This seems correct, for Roger acquired the tide of count only in 1091, on the death of his wife's brother, Boso III, count of La Marche: Mason, 17; Chandler, 3. It is simply as Roger le Poitevin that the libellus has him as a party to the pact of 8 September and accepting the bishop's counter-assurances (line 181). Later on the libellus becomes confused and perhaps anachronistic about Roger's title. Like his fellow guarantor, Count Alan, he is called comes at line 268, and like Count Odo at line 443. While the Rogerus comes at lines 282–3 may possibly be his father, the earl of Shrewsbury (as Mason, 16, n. 4), it seems more likely that Roger le Poitevin was intended (cf. Barlow, , WR 85Google Scholar, n. 160 ). At lines 420 and 622 Roger seems to be included among the comites responsible for the pact on 8 September. Lines 637–8 refer to the guarantors explicitly as comites Alanum et Rogerum et Odonem, and here Roger le Poitevin must surely be intended. From these contradictory instances it might be argued: (a) that the libellus must have been written at a date after Roger acquired the title of count in 1091; or (b) that the author of the libellus was not particularly well-informed about Roger's status and title in 1088; or (c) that to an author in 1088/9 Roger's extensive holdings in England seemed to justify ranking him as a count though he had not yet achieved that status officially. None of these inferences carries complete conviction; the least likely seems the last.

^{page 79 note 29} These are extraordinarily favourable terms (lines 156–80) to have been granted to a suspect and defeated traitor. Bishop William is indeed to appear before the curia regis. But (a) he is to be allowed trial by churchmen, or be escorted safely back to Durham; (b) if the judgement given in such a trial is considered by the bishop to be unjust, and either the king or the (ecclesiastical) judges hinder his appeal to the pope, then the bishop is to be returned to his castle at Durham; (c) if the bishop is unwilling or unable to accept the judgement, he is to have a safe-conduct overseas with men and goods. The date of this agreement was 8 September.

^{page 80 note 30} ubi contenciosa pontificum iudicia juste debent terminari: a reference to the papal curia, as in line 142. See Gregory, Pope VII, Reg. viii. 21Google Scholar, ed. E. Caspar, MGH Epp. sel. ii,549: omnes maiores res et precipua negotia necnon omnium ȩcclesiarum iudicia ad earn [Romanam ȩcclesiam] quasi ad matrem et caput debere referri, ab ea nusquam appellari, iudicia eius a nemine retractari aut refelli debere vel posse. Cf. Reg. ii.55a (Dictatus pape), nos. 21 and 18, p. 206.

^{page 80 note 31} 29 September 1088.

^{page 80 note 32} septem: presumably seven of the bishop's major tenants, men of the kind later to be known as ‘barons of the bishopric’. Are they the same as the seven knights referred to in lines 286 and 347–8?

^{page 81 note 33} 2 November 1088.

^{page 81 note 34} The court was probably held in the castle (or just possibly the cathedral) at Old Salisbury: Barlow, EC 283, WR 85.

^{page 81 note 35} Urse, royal constable, sheriff of Worcester, despoiler of monks, active administrator and judge on Rufus's behalf, came from Abbetot, Seine-Maritime, arr. Le Havre, cant. Saint Romain. Loyd, Origins 1–2Google Scholar; see Round, J. H., DNB xx, 52Google Scholar; Mason, Emma, ‘Magnates, Curiales and the Wheel of Fortune’, Battle 2 (1979), 135–8Google Scholar; Barlow, , WR 152, 185–9, 207–11Google Scholar. He appears as witness to many royal charters, including two spuria for Durham, RAN i nos. *281, 349.

^{page 81 note 36} See I Cor. 16,20. According to Eadmer, HN 63 Rufus asked the bishops at Rockingham in 1095: nonne saltern … fraternae societatis amicitiam ei abnegare potestis?

^{page 82 note 37} Surely Ralph, not Roger, Paynel must be meant.

^{page 82 note 38} Presumably when they met in the summer; see note 20.

^{page 82 note 39} For the royal expedition against Bishop William, see lines 151–2. He was forced to quit Durham on n September: Unes 609–10, 649. The words ad tempus may suggest that he returned there later. But the libellus throws no clear light on the bishop's movements between 11 September and his appearance at Salisbury in November.

^{page 82 note 40} casati: tenants of the episcopal demesne seem intended.

^{page 82 note 41} tocius Anglie primatis: though s.a. 1066 ‘Florence’ describes Stigand thus, i, 228, this is not the title favoured by Canterbury under Lanfranc. Its archbishop was usually (and programmatically) called tocius Britannie primas. For examples, see Eadmer, , HN 12, 63Google ScholarThe Letters of Lanfranc Archbishop of Canterbury, ed. and trs. Clover, Helen and Gibson, Margaret, (Oxford, 1979), pp. 3–4, 46, 72–4, 78, 152Google Scholar. This was the form normally used by bishops in their professions to Canterbury, where the title tocius Anglie primas does not appear until 1177/1180 (there are later examples of tocius Britannie): Richter, M. (ed.), Canterbury Professions (Canterbury and York Soc. 67, 1975), nos. 113–4.Google Scholar

^{page 83 note 42} comes Rogerus: in this context Roger le Poitevin must surely be intended.

^{page 83 note 43} The bishop will plead before the king while still despoliatus only after an ecclesiastical judgement that he must.

^{page 83 note 44} Rogerus Bigotus: on him see GEC ix,575–9; Barlow, , WR 61–2Google Scholar. The greatest lay tenant in south and east Norfolk, he acted as sheriff in that county and also in Suffolk. Though he witnessed royal charters at the beginning of Rufus's reign, RAN i nos. 290, 291, 295, 296, he appears to have taken some part in the feudal disorders in 1088: ASC 1087 [1088]; Malmesbury, GR ii, 361Google Scholar; HR ii, 215Google Scholar. Possibly he was influenced by the example of Odo of Bayeux, whose tenant he was in England and Normandy: Bates, D. R., ‘The character and career of Odo, bishop of Bayeux (1049/50–1097) ‘, Speculum 50 (1975), 11CrossRefGoogle Scholar. But me extent of his involvement is difficult to estimate, Barlow, WR 81, n. 141, and soon after the capture of Rochester he must have made his peace with the king, for he witnesses RAN i no. 302. By 1091 he appears as royal steward, RAN i, p. xxiv, and in that year his name occurs as witness to a spurious Durham charter: RAN i no. 318; cf. DEC no. *6, pp. 48–53.Google Scholar

^{page 83 note 45} Hugo de Bellomonte: the problem of identifying a Hugh Beaumont at this date has baffled scholars from Freeman onwards, WR i, 98Google Scholar, n. 2. For a Beaumont of this name we have to await Hugo pauper, third son of Robert of Meulan, for whom King Stephen may have tried to create an earldom of Bedford in 1138, and this Hugh was born after 1104: GEC vii,526, n. 2. The libellus repeats the name in full, lines 313, 383–4; the given name only, lines 291, 391. Barlow, EC 284 and WR 75, 77 n. 111, attempts to get round the difficulty by supposing error in the textual transmission of the libellus; he feels entitled to convert on five occasions Hugo into Henricus: i.e. Henry, the brother of Robert of Meulan, who does indeed attest twice for Rufus during 1088: RAN i nos. 302, 325. But would not Henry de Beaumont have been styled earl of Warwick by November 1088? It seems generally accepted that his promotion to earl occurred before the end of 1088; cf. G. H. White in GEC xii.2 app. A, pp. 2–3; Crouch, David, The Beaumont Twins (Cambridge, 1986), 10CrossRefGoogle Scholar; though Barlow, , WR 93Google Scholar says ‘probably’ in 1089. Some opinion inclines to a date in or shortly after July 1088: cf. Vaughn, Sally N., Anselm of Bec and Robert of Meulan (Berkeley and Los Angeles, 1987), 99Google Scholar; Douglas, D. C., EHD ii, 616Google Scholar n. 1. Repeated scribal error, as postulated by Barlow, seems to me a less likely explanation of the difficulty than sheer confusion on the part of the author of the libellus. May he not have written (and continued to write) Hugo de Bellomonte by mistake for Hugo de Bellocampo? Hugh Beauchamp (of Bedford), Domesday tenant in Bedfordshire, Hertfordshire and Buckinghamshire (in which county he was sheriff), is a wholly plausible participant at the Salisbury curia; he witnesses Rufus's charters RAN i nos. 419, 446 (with Ralph Paynel), 477 (again with Ralph).

^{page 84 note 46} If the libellus be accepted as a faithful witness, it implies (lines 282–9) this chronology: At the beginning of March or earlier Bishop William had advised Rufus to attack his enemies, Odo of Bayeux and Roger Montgomery, earl of Shrewsbury. When summoned to accompany the king, the bishop had agreed to do so with the seven knights he had with him and to send to Durham for more. But then St Calais fled from the royal court without leave, taking with him certain men of the royal household. All this must have happened before the proclamation of disseisin on 12 March; cf. line 30 and note 10.

^{page 84 note 47} For the parts played in the conspiracy against Rufus by Bishop Odo and Roger Montgomery, see ASC 1087 [1088]; ‘Florence’, ii,21–2; Orderic, iv,121–34; Malmesbury, GR ii, 360–1Google Scholar. Bates, ‘Character and career’ (as in note 44), 4, decries Orderic's account as ‘garbled and uninformed’.

^{page 84 note 48} septem militibus: were these the same as those mentioned in lines 189 and 347–8?

^{page 84 note 49} For Bishop Geoffrey, see Le Patourel, J., ‘Geoffrey de Montbray, Bishop of Coutances, 1049–1093’, EHR 59 (1944), 133–58Google Scholar. If Geoffrey's intervention amounted to support of Bishop William's plea for clerical privilege, as Le Patourel claims, p. 154, it was hardly forceful. Geoffrey himself had been involved in the revolt: ASC 1087 [1088]; ‘Florence’, ii, 24.

^{page 85 note 50} Cf. Pseudo-Isidore, Felix II, Decret, c.xii, ed. Hinschius, 485: Ut nemo episcopum penes seculares arbitros accuset, sed apud summos primates.

^{page 85 note 51} et prepositis et uenatoribus: Croc the huntsman, a Domesday tenant-in-chief in Hampshire, certainly attests charters for Rufus: RAN i nos. 319, 359, 361.

^{page 86 note 52} hiis septem hominibus: see lines 189, 286.

^{page 86 note 53} Lines 352–5 are redolent of Pseudo-Isidore. Cf. Julius, Decret, c.xii, ed. Hinschius, 469: Salva apostolicae aecclesiae auctoritate nullus episcopus extra suam provinciam ad iudicium devocetur, sed vocato eo canonicae [sic ed.] in loco omnibus congruo tempore synodali ab omnibus conprovintialibus episcopis audiatur, qui concordem super eum canonicamque proferre debent sententiam, quondam si hoc minoribus tam clericis quam laicis concessum est, quanto magis de episcopis serviri convenit? Nam si ipse metropolitanum aut iudices suspectas habuerit aut infensos senserit, apud primates dioceseos aut apud Romane sedis pontifices iudicetur.

^{page 86 note 54} absentibus omnibus comprouincialibus meis: Freeman, WR i, 105, thought the bishop's complaint ‘grotesque’; Hoffmann, Hartmut, ‘Zur Echtheit’, 439Google Scholar, justifies it as pleader's guile, ‘ein juristiche Trick’. At St Calais's own consecration suffragans of Canterbury had assisted Thomas of York; at that time only one bishop acknowledged York's authority, Ralph of Orkney, who had himself been consecrated by Thomas in 1073 with the aid of two Canterbury suffragans: C & S I.ii, 629–30; Letters of Lanfranc, ed. Clover and Gibson, nos. 12–13, pp. 78–85Google Scholar. It would be difficult to regard Fothadh, the last Celtic bishop of St Andrews, as even a nominal comprovincial of St Calais in 1088. There was indeed a later York claim that Fothadh professed obedience to Thomas of Bayeux: BL Harley ms. 433, f. 260r, edited for the Richard III Society by R. E. Horrox and P. W. Hammond, iii, (Gloucester, 1982), 84–5. But this seems very improbable.

^{page 86 note 55} et accusatores sunt simul et iudices: cf. Pseudo-Isidore, Damasus, Decret, c.xvi, ed. Hinschius, 504: Accusatores vero et iudices non idem sint, sed per se accusatores, per se iudices, per se testes, per se accusati, unusquisque in suo ordinabiliter ordine.

^{page 86 note 56} To whom as St Peter's vicar could St Calais appeal effectively in 1088? Since Gregory VII's death in 1085 neither William I nor Rufus had recognised any of the contenders for the papacy: Brooke, Z. N., The English Church and the Papacy (Cambridge, 1931), 145Google Scholar; Southern, Anselm (1963), 145Google Scholar. St Calais here exposed himself to the sort of question from Rufus which Anselm was to hear in 1095 when he wished to go to the pope to receive his pallium: ‘A quo papa illud requirere cupis?’, Eadmer, HN 52Google Scholar. Though Brooke accepted that St Calais did appeal to Rome in 1088, he warned against regarding this action ‘too seriously. He was taking a very unusual step, but only to evade judgement, not as a matter of principle’ (p. 162). While the editors of C & S I.ii reject my views about the date of the libellus (p. 855, n. 1), they nevertheless remark (p. 635, n. 2): ‘It is certainly strange that a contemporary account should lay so much stress on the bishop's appeal to Rome without anyone on either side referring to the schism.’ Though St Calais, once across the Channel, did not proceed in person to Italy, he may have written to Pope Urban II about his treatment by Rufus, though it seems unlikely that the libellus was composed as an aide-mémoire for that purpose, as Hoffmann suggested, ‘Zur Echtheit’, 440. Urban's letter to Rufus, datable perhaps to April–June 1089, printed by S. Löwenfeld, Epp. Pont. Rom. ineditae (Leipzig, 1885), no. 129, p.65 (JL 5397), though it confirms that St Calais's complaints had come to Urban's ears, offers no guarantee that the libellus already existed, and there is no evidence that the letter ever reached Rufus; cf. C & S I.ii, 635. A. Becker, Papst Urban II (MGH Schriften 19/1, 1964) i, 173–6 surmised that Urban had heard about this affair from his legate in France, Cardinal Roger. Becker dismisses St Calais's appeal as a mere formal step, which the bishop never pursued seriously.

^{page 87 note 57} cuius dispositioni … reseruauit: literally from Pseudo-Isidore, either Sixtus II, Decret, c.ii, ed. Hinschius, 190, or Julius, Decret, c.xii, Hinschius, 467; cf. Sixtus I, Décret, v, Hinschius, 108. All stem from Innocentius papa I, Ep. ii. 3 (Victricio Rothomagensi), Migne, PL 20, 473; ib. 635.

^{page 87 note 58} On the face of it, this is the earliest account of how the Conqueror's action against Odo of Bayeux in 1082 was justified. The story as told by Orderic, iv,42, writing between 1130/1 and 1133 (see editor's note, p. xix), does not record an intervention by Lanfranc. Malmesbury however does impute responsibility for the reply to Lanfranc at GR ii,351, though not at GR ii, 334Google Scholar. Orderic's editor, Dr Marjorie Chibnall, remarks, Orderic, iv, xxvii–xxx, that Malmesbury, Orderic and the ‘Hyde’ chronicle appear to be independent of each other; she thinks that their accounts of the incident ‘tip the balance in favour of the distinction in capacities ‘having been made, if not in 1082, at least shortly afterwards’. But the ‘Hyde’ chronicle (post 1120), ed. Edward Edwards, RS, (1866), p. 296Google Scholar, is not very specific here, and Dr Chibnall's conclusion that if Lanfranc ‘did not make the distinction in 1082, almost certainly he made it retrospectively in 1088’ seems to postulate that the story in the libellus was contemporary with the events it purports to describe. If that cannot be sustained, then we may still envisage the libellus borrowing the distinction if not from the first version of Malmesbury's GR (completed in 1125: Thomson, R. D., William of Malmesbury (Woodbridge, 1987), pp. 3–4 )Google Scholar, then from Malmesbury's source.

^{page 88 note 59} See Gregory VII, Reg. viii. 21, cited in note 30.

^{page 88 note 60} Per uultum de Luca: reported as an oath favoured by Rufus in the 1090s by Eadmer, , HN 30, 39, 101, 110Google Scholar (cf. 116) and Malmesbury, GR 11, 364, 375Google Scholar; GP 80, 83nGoogle Scholar. The Volto Santo, a romanesque figured wooden crucifix of disputed date in the cathedral of San Martino at Lucca, is illustrated by Gustav Schnürer and Ritz, Joseph M., Sankt Kümmernis und Volto Santo (Forschungen zur Volkskunde, Heft 13–15, Düsseldorf, 1934)Google Scholar, pls. xi, xii, and by Barlow, WR pl. 4a, with discussion ib. 116–8. Webbe, Diana M., ‘The Holy Face of Lucca’, Battle 9 (1987), 227–37Google Scholar, admits the difficulty that while ‘the earliest independent Italian evidence for the existence and cult of the Vultus’ comes from the episcopate of Rangerius, bishop of Lucca c. 1097–1112 (the legend, ed. Schnürer and Ritz, 128–34, was written in the early twelfth century), Rufus's oaths in England ‘seem to furnish the earliest incontrovertible evidence’ for the existence of the Vultus. Accepting Eadmer's testimony (though not noticing the appearance of the oath in DIV), she inclines to the conclusion (p. 237) that Rufus ‘was bearing witness to the existence of a cult which had grown from obscure roots, although in favourable conditions, over the twenty or thirty years before 1090’. If the cult did indeed flourish as early as this, there would of course be little difficulty in suggesting how Rufus had come by his knowledge of it by 1088; a possible channel was his physician, Baldwin abbot of Bury (cf. Barlow, , WR 117Google Scholar). But it should be borne in mind that the English evidence for Rufus's use of this oath in the 1090s is not strictly contemporary: Eadmer did not put together the first four books of HN until 1109–1115, according to Southern, Anselm (1963), 299Google Scholar; Malmesbury, who may well have borrowed from Eadmer, is some ten years later; and the date of DIV, whose authority here Barlow, EC 67, n. 76 and WR 116, n. 73, esteems as no more than ‘probably’ independent, can be regarded as still at issue. It may be relevant to remark that Eadmer, HN 112–4, was vastly impressed by the conduct of Bishop Rangerius of Lucca at the Roman Easter council in 1099.

^{page 89 note 61} presentibus centum meis multibus: if the bishop's household and stipendiary troops are included, this claim is perhaps not much exaggerated. By 1135 the bishops of Durham had enfeoffed 64 knights: Chew, H. M., English Ecclesiastical Tenants in Chief (Oxford, 1932), 19, 119Google Scholar. The total sum to be accounted for from the aid levied on the bishopric's knights in 1129 was £58 6s. 8d., PR 31 Henry I, ed. Joseph Hunter, (Record Commission, 1835), 132; on the nature of this levy, see Green, Judith A., The Government of England under Henry I (Cambridge, 1986), 77CrossRefGoogle Scholar. Assuming a rate of £1 per fee, this would amount to almost 60 fees.

^{page 89 note 62} comites: presumably, though perhaps mistakenly, including Roger le Poitevin.

^{page 89 note 63} Radulfus Piperellus: Ranulf Peverel, lord of Hatfield Peverel in Essex and also Domesday tenant in Berkshire, Oxfordshire, Norfolk and Suffolk, witnesses in early 1091 a royal charter and a royal confirmation in company with many others whom the libellus names as at Salisbury in November 1088: RAN i nos. 315, 319.

^{page 89 note 64} uetulus ligaminarius: this has puzzled the translators. Stevenson, Joseph, Church Historians of England III.ii (London, 1855), 743Google Scholar, offered ‘old turnkey’; Freeman, , WR i, 109Google Scholar, n. 3, ‘old gaoler’. Admitting that the meaning is uncertain, EHD ii, 619Google Scholar suggested ‘this trusty old liegeman’. Gibson, , Lanfranc of Bec, 161Google Scholar, n. 1, perhaps by aid of the Revised Medieval Latin Word List (1965)Google Scholar, arrived at ‘old bloodhound’. Possibly a shade of animus can be detected in Frank Barlow's repeated rendering ‘the old binder’, EC 286Google Scholar; The Norman Conquest and Beyond (London, 1983), 235Google Scholar. The variant vitulus for uetulus in two of the mss. hardly helps. To me the form ligaminarius suggests ‘lymerer’, the man who handled the leash or scenting hounds in a hunting pack.

^{page 90 note 65} R. et Odo comites: presumably Roger le Poitevin is intended.

^{page 90 note 66} christianam legem quam hie scriptam habeo: if St Calais was referring to a volume of canon law held in hand, it may well have been that copy of Lanfranc's collection now surviving in Cambridge, Peterhouse ms. 74. See Brooke, Z. N., English Church and the Papacy, 109, 162Google Scholar; Barlow, , EC 286Google Scholar and n. 52; Horst Fuhrmann, Einfluss und Verbreitung der pseudoisidorischen Fälschungen von ihrem Auftauchen bis in die neuere zeit (MGH Schriften 24/1, 1972) i,169, n. 61.

^{page 91 note 67} Constituta est ergo dies: apparently 14 November; see lines 549–50.

^{page 91 note 68} Lines 484–7 are reminiscent of how Fulbert of Chartres defined the obligations of fealty in his Ep. li (to Duke William V of Aquitaine, before 9 June 1021), ed. Behrends, F., Letters and Poems of Fulbert of Chartres (Oxford, 1976), 90–2Google Scholar. The definition appeared in various canonical collections before finding its way into Gratian, Decretum C.22, q.5, c.18; cf. Consuetud.Feud. ii, tit.6. Durham DCL ms. B.II.11, which contains Fulbert's letters, may have been copied for St Calais, who left it to the library: Mynors, DCM 38.Google Scholar

^{page 91 note 69} Dorobernium et Hastmgas, que iam pene perdiderat: if the claim is authentic, when did St Calais render these services? ASC and ‘Florence’, though admitting that the conspiracy against Rufus was prepared in Lent, place the outbreak of open hostilities after Easter (16 April), and St Calais must be assumed to have left the king's court before his first disseisin on 12 March; see note 10. Freeman, WR i,29, dated these efforts by St Calais ‘at the latest in the very first days of March’; so too Barlow, WR 75–6. The only possible alternative would seem to be during the putative meeting between the bishop and Rufus in the summer of 1088 (see line 112 and note 20), though this would necessitate the hypothesis of a temporary reconciliation between them at that time. It is generally accepted that Dorobernium here means Dover, as for example it quite plainly does in Henry of Huntingdon, Hist, anglorum, ed. Arnold, T., RS, (1879), 218Google Scholar, though forms such as Dubris, Dofra, Dofris, Doura, Doveria, Dovoria and Dovere seem better witnessed. But just possibly Dorobernium could be taken as a variant for Dorobernia, i.e. Canterbury. Both ‘Florence’, ii, 23 and Orderic, iv, 126 state that Canterbury and London were targets for the rebels from Rochester (in May 1088?).

^{page 92 note 70} This seems the only record of unrest in London, though ‘Florence’, ii, 22 reports that Rufus went there at an early stage in the revolt belli tractaturus negotia. In 1066 ‘all the best men from London’ had submitted to the Conqueror, according to ASC D version and ‘Florence’, i, 228. But Liebermann's view, Die Gesetze der Angelsachsen II.i (Halle, 1906), 573Google Scholar, that the XII meliores ciues of DIV formed something like a permanent governing body for London in 1088 must be treated with caution, though it is not explicitly rejected by Weinbaum, M., Verfassungsgeschichte Londons 1066–1268 (Stuttgart, 1929), 15Google Scholar. It seems just as likely that the twelve were a chance ad hoc selection of prominent citizens; see Tait, James, The Medieval English Borough (Manchester, 1936), 266Google Scholar; Brooke, C. N. L. and Keir, Gillian, London 800–1216: the shaping of a city (London, 1975), 248–9.Google Scholar

^{page 92 note 71} Compurgatores and their oath as to the credibility of the accused are discussed X V.34.5, ed. Friedberg, col. 870.

^{page 92 note 72} Cf. the advice given by the bishops to Anselm in 1095, according to Eadmer, HN56: Verum si, remote omni aha conditione, simpliciter ad voluntatem domini nostri regis consilii tui summam transferre velles, prompta tibi voluntate, ut nobis ipsis, consuleremus.

^{page 92 note 73} citra mare: if Rufus was not looking at things from a Norman viewpoint, we must postulate a heedless author writing on the continent. At this stage it would not have made sense to demand further guarantees for the bishop's behaviour in England.

^{page 92 note 74} Duke Robert of Normandy.

^{page 93 note 75} Reginaldus Paganellus: No contemporary Reginald Paynel appears in the Paynel pedigree facing p. 1 of EYC vi (1939)Google Scholar, or seems known otherwise. Possibly a mistake for Radulfus; if so, it is uncertain whether the error is a scribe's or the author's. Reginaldus, though common enough by c. 1130–1140, seems an unusual form for 1088; at that time the name generally occurs as Rain- or Reinaldus.

^{page 93 note 76} Wilton, seat of a famous community of aristocratic nuns, struck Freeman as a rather odd choice of place for the bishop's house arrest. He queried, WR i, 112Google Scholar, n. 5: ‘should it be Wintonie?’

^{page 93 note 77} W. de Merlaio: As John Le Patourel remarked, ‘Geoffrey de Montbray’ (as in note 49), 154, we would like to know if this was William, Bishop Geoffrey's steward, who was his vassal in Bedfordshire, DB i, 210Google Scholar. Perhaps there was a connexion between this W. de Merlaio in DIV and the later Merlay lords of Morpeth in Northumberland. Of these a William de Merlay was charged on the PR 31 Henry I with a palfrey for his rights to his brother's lands, ed. Joseph Hunter, (Rec. Comm., 1833), 36. This William, who died before 5 September 1129, gave Morwick (par. Warkworth, Nb.) to St Cuthbert and his monks at Durham; DCD Cart. II, f. 251r = Dugdale, Monasticon 1,241. The suggestion that the Merlay family was established in Northumberland by Geoffrey's nephew, Robert Mowbray, while he was earl (c. 1086–95), seems plausible: see J. C. Hodgson in Northumberland County History v, 1899, 345; Hedley, W P., Northumberland Families i (Newcastle, 1968), 196–7Google Scholar. When and where this alleged depredation by St Calais's men from Durham can have taken place is far from clear. The libellus offers no clue to Bishop William's movements between his leaving Durham in early September and his appearance at Salisbury in November. Where could Geoffrey's cattle have come within reach of the Durham garrison during that time? Had he been attempting to restock his estates in the south from Scotland or Northumberland?

^{page 94 note 78} Walter of Aincourt (Ancourt, Seine-Maritime, arr. Dieppe, cant. Offranville: Loyd, Origins, 2), lord of Blankney in Lincolnshire, was also a Domesday tenant in Yorkshire, Northamptonshire, Derbyshire, Nottinghamshire and Cheshire. He witnessed a royai charter 1088–91: RANi no. 325. His son William is said to have died while being brought up at Rufus's court: Barlow, , WR 133–4.Google Scholar

^{page 94 note 79} Die uero crastina: 3 November 1088. The bishop chose Southampton as his port of departure.

^{page 94 note 80} Episcopus enim Baiocensis inde me castigauit: ‘the bishop of Bayeux taught me a sore lesson about that sort of thing’. Presumably a reference to the complications about the surrender of Rochester a few months earlier: see ASC 1087 [1088], Malmesbury, GR ii, 562Google Scholar; Orderic, iv, 126–38. As Dr Chibnall points out, iv, 126, n. 2, Ordene fails to mention ‘the apparent trick by which he [Odo] entered Rochester on the grounds of negotiating its surrender.’

^{page 94 note 81} 14 November 1088.

^{page 94 note 82} Gilberto uicecomiti: Gilbert of Bretteville was a Domesday tenant-in-chief in Hampshire, Berkshire and Wiltshire. He appears as sheriff of Berkshire 1090 × 1094: RAN i no. 359.

^{page 94 note 83} 21 November 1088.

^{page 94 note 84} Though Robert count of Mortain could be called Robert de Conteville, he cannot be intended here. But it is naturally tempting to try to link this Robert in the libellus in some way with Herluin vicomte of Conteville (Eure, arr. Bernay, cant. Beuzeville), father by Arletta of Odo of Bayeux and Robert of Mortain. By his second wife, Fredesendis, Herluin had two more sons, Ralph and John, the former of whom may be the Radulfus de Contivilla who held land in Somerset and Devon according to the Exon Domesday, or the Radulfus filius Herluini, tenant of Roger Bigot in Norfolk: see Orderic, iv, 98, n. 1; David R. Bates, ‘Notes sur l'aristocratie normande’, Annales de Normandie 23 (1973), 26, 29–33. Possibly the Robertus of DIV is author's or scribal error for Radulfus. But it must be noted (a) that Robert of Mortain's second son was named Robert, and might conceivably have been called de Conteville; and (b) that there are other Contevilles in Normandy, not connected with Herluin at all.

^{page 95 note 85} Ivo Taillebois, first husband of Lucy of Bolingbroke, benefactor of Peterborough and St Nicholas, Angers, and frequent witness to royal charters under William I and II, appears as royal dapifer in 1091: see RAN, i p. xxiv and nos. 129, 143, 177, 229, 233–6, 2883, 302, 315, 319, 326, 328, 370, 386, 403–4, 406–10. Domesday tenant-in-chief in Lincolnshire and Norfolk, and perhaps sometime sheriff of Lincolnshire, before his death c. 1094 he had become established as a great landowner in northern England, on the upper Eden and in a block of territory stretching from upper Ribbesdale to Lonsdale and Kendal: see Sanders, , English Baronies, 56Google Scholar; Mowbray Charters, p. xxii.Google Scholar

^{page 95 note 86} Erneis de Burun, sheriff of Yorkshire towards the end of the Conqueror's reign, was perhaps succeeded in that office by Ralph Paynel under Rufus; see Farrer, W., EHR 30 (1915), 282Google Scholar and note 13 above. Domesday tenant-in-chief in Lincolnshire and Yorkshire, where he held in effect the barony of Hunsingore, that is Gospatric's former lands in Nidderdale in chief and those in and about Masham from Count Alan, lord of Richmond; see EYC x, 1–2, 23–30Google Scholar; Sanders, , English Baronies, 56Google Scholar; Mowbray Charters, p. xxii.Google Scholar

^{page 95 note 87} 14 November 1088.

^{page 95 note 88} Identified by Freeman, , WR i, 114Google Scholar, n. 4 as Heppo the Balistarius, tenant-in-chief in Lincolnshire, DB i, 369.Google Scholar

^{page 95 note 89} Calendared RAN i no. 298.

^{page 95 note 90} 21 November 1088; for Gilbert the sheriff and Robert, see notes 82 and 84.

^{page 96 note 91} If the name has been transmitted correctly, this seems the only indication that Roger Montbray, brother of Geoffrey of Coutances and father of Robert earl of Northumberland, had been present in England in 1088, or had indeed survived so long. Roger, said to be a kinsman of Nigel of St-Sauveur, vicomte of the Cotentin, was among the Norman notables of 1066 eulogized by Orderic, ii, 140, and in that year was a benefactor of Holy Trinity, Caen: Fauroux, M., Recueil des actes des ducs de Normandie de 911 à 1066 (Caen, 1961), no. 231, p. 445Google Scholar. His Norman estates were forfeited by his son Robert in 1095: Hollister, C. W., ‘The greater Domesday tenants-in-chief’, in Domesday Studies, ed. Holt, J. C., (Woodbridge, 1987), 229Google Scholar. Only very much later does the next Roger Mowbray appear in English records. He was the son of Nigel d'Aubigny and Gundreda de Gournay, and not born till c. 1119/1120: Mowbray Charters, p. xxvi and n. I. Barlow, WR 89 and n. 171, 92, 168, attempts to get round the difficulty by proposing to read Robertus for Rogerus.

^{page 96 note 92} 26 November 1088.

^{page 96 note 93} 27 November 1088.

^{page 96 note 94} Osmund, bishop of Salisbury, 1078–99.

^{page 96 note 95} Robert de Lisle witnesses Rufus's confirmation of a Salisbury charter in 1091: RAN i no. 319. His relationship, if any, to Humphrey de Lisle, who witnesses the same document, is uncertain. Humphrey was a Domesday tenant-in-chief in Wiltshire: DB i, 64b, 70b.Google Scholar

^{page 96 note 96} Ricardus de Cultura: not identified. It may just be worth recalling that Richard de Courceie (Courcy) witnesses Rufus's charters at this time, RAN i nos. 310, 334, 349, 433; his name also appears on two spurious charters of William I for Durham, ib. nos. *205, *286.

^{page 96 note 97} 1 December 1088.

^{page 96 note 98} Gaufrido monacho suo: No Geoffrey is named among the early monks at Durham in the list given in Durham UL, Cosin's ms. V.ii.6, ff. 7r-8v. But in the similar list in LV, f. 42r col. 2, after the entries for Bishops Walcher, William of St Calais and Rannulf and for Priors Aldwin and Turgot, Gosfridus has been interlineated before the next name in the original hand, that of the monk Aelfwius. Bishop William's high regard for Geoffrey is indicated, not perhaps wholly favourably, by Symeon, HDE 132. Here Symeon inserts into his history a kind of purgatorial vision by an episcopal knight called Boso, who claimed when in a rapture to have seen the bishop thrusting his head out of the doors of an iron building sited in the midst of a desolate waste, and to have heard him asking: ubinam Gosfridus monachus esset.… ‘Hic enim’, inquit, ‘hic ad placitum mecum adesse deberet.’ Boso's narrative continues: Hunc namque episcopus procuratorem sui episcopatus constituerat. This clearly denotes Geoffrey's function as general viceregent for managing business in the bishopric. The reference to placitum seems to betray the presence of a calculated barb, for here the word most probably means trial, rather than court (as Barlow, WR 356 translates). Perhaps brought to Durham by St Calais from a Norman monastery, Geoffrey may well be the G. Dunelmensis who seems to have acted as custodian of the temporalities of the see after the bishop's death: DCD 1.1. Reg. 8, illustrated as pl. x in Bishop and Chaplais, ERW, and printed EYC ii no. 930. On four occasions between March and June 1088 the libellus shows St Calais using an unnamed monk as his envoy to Rufus: lines 56, 84, 128, 151; whether this was Geoffrey is uncertain. The libellus implies that Geoffrey was with the bishop in early December at Southampton, whence he was to be sent back to Durham (line 619), but he does not appear to have been at the Salisbury court a month earlier. Presumably he had been in charge of Durham castle after the bishop had come to terms with Rufus's army in September (see note 107) and had quitted Durham in order to rejoin St Calais after the royal seizure of the castle on 14 November (lines 549–50). Though it cannot be proved, it is very tempting to suppose that this Geoffrey is the monk of that name from St Calais (cuidam monacho sancti Carileffi Gausfrido nomine) to whom Henry I gave custody of Battle abbey after the death of Abbot Henry in June 1102. The Battle chronicler says that though Geoffrey lacked learning, he was highly shrewd, prudent and skilled in worldly affairs, as he showed in defending the monastery's temporal interests until his death on 16 May 1105: The Chronicle of Battle Abbey, ed. & trans. Searle, Eleanor (Oxford, 1980), 108–117.Google Scholar

^{page 97 note 99} municionem castelli: the castle's stores and provisions.

^{page 97 note 100} Ipse: Rufus.

^{page 97 note 101} The libellus thus makes St Calais argue that he was not disseized of his church, castle and all his lands until 14 November (cf. lines 565–7), despite the disseisin from his Yorkshire possessions on 12 March (lines 29–30). Yet the tractate in which the libellus is framed reckons the date at which William was driven from his see as 11 September: lines 649–50 below.

^{page 98 note 102} Walkelin, bishop of Winchester, 1070–1098.

^{page 98 note 103} Hugh of Port-en-Bessin (Calvados, arr. Bayeux, cant. Ryes: Loyd, Origins, 77Google Scholar) was a tenant-in-chief in Hampshire, Berkshire, Dorset and Cambridgeshire, and held lands in England and Normandy from Odo of Bayeux; he appears as sheriff of Hampshire: GEC xi, 316–7; RAN i nos. 143, 267, 270, 284, 379.

^{page 98 note 104} Geoffrey de Traileio, perhaps from Trelly (Trailliacum) in Normandy, (dép.) La Manche, cant. Montmartin-sur-Mer: Loyd, Origins, 106Google Scholar. Goisfridus de Traillgi or Tralgi appears as a Domesday tenant of Geoffrey of Coutances in Bedfordshire1. DB i, 210Google Scholar. Presumably it was a descendant, also named Geoffrey, who married Albreda, sister and joint heir of Walter Espec, and so came into some of the lands in Bedfordshire held in 1086 by William Espec: Sanders, , English Baronies, 133–4Google Scholar. Nicholas of Trailli, nephew of Walter Espec of Helmsley, occurs from c. 1143 onwards as canon of York: Clay, C. T., York Minster Fasti ii (Yorks. Arch. Soc. Record Series 124, 1959), 70–1.Google Scholar

^{page 98 note 105} Here Roger le Poitevin seems reckoned as a count.

^{page 98 note 106} transitum: on the evidence of the libellus this must have been well on in December. After receiving the royal sumons dated 1 December (line 595), St Calais had sent his knight to the king (lines 614–15); Rufus had dispatched Bishop Walkelin and his companions to St Calais (lines 617–20), causing him to appeal to Earl Alan and his fellow guarantors, who will have needed time to work on Rufus. Possibly the bishop did not reach Normandy before the new year opened.

^{page 99 note 107} After the end of the libellus, this concluding section of DIV is mostly based on Symeon, HDE: lines 648–56 HDE 128–9Google Scholar, and unes 658–68 on HDE 133–5. But HDE does not give the information at lines 646–8 about the restoration of the bishop to Durham by Rufus on 11 September 1091, nor that at lines 656–7 about the collaboration of Malcolm king of Scots in laying the first stones of the new cathedral on 11 August 1093. These entries do appear, however, in the Durham Historia regum, 218, 220, as additions to HR's habitual source for this period, ‘Florence’ of Worcester. David, ‘A tract’, 384 argued that here HR borrowed from DIV. Undoubtedly the compiler of HR knew the libellas. But the framework of DIV (lines 3–28, 644–68) in which the libellus proper is presented was probably composed separately; it is not impossible that it borrowed from HR.

Anno sui episcopatus octauo: cf. HR 217Google Scholar. Calculating William's episcopal years from his ‘election’ on 9 November 1080 (as the total length of his episcopate is calculated, lines 664–5), this might imply that the bishop was driven out of England before 9 November 1088, which is plainly at variance with the libellus. Possibly expulsus refers to the sentence of the Salisbury court early in November.

tocius Normannie curam suscepit: HR 216–7Google Scholar has: a Rodberto comite totius provinciae curam suscepit. But HR appears to have muddled the order of the last two sentences in its entry for 1088, thus leaving a half-suggestion that Odo of Bayeux was being referred to. The language of ASC 1087 [1088] on the eve of the conspiracy against Rufus is similarly ambiguous: ‘So generously did the king behave to that bishop that all England was governed on his advice and direction.’ Though the Waverley annals, ed. H. R. Luard, RS (1865), 198Google Scholar explicitly refer this to Bishop Odo, Dr Cecily Clark thinks that ASC probably means St Calais, not Odo (ASC, 77); cf. ‘Florence’, ii, 22 on St Calais's influence early in Rufus's reign: eiusque consiliis totius Anglie tractabatur respublica; Malmesbury, GR, ii, 360 tells the same story. Barlow, , WR 61Google Scholar, n. 35 thinks it likely that ‘Florence’ and Malmesbury give the correct meaning of what ASC had garbled.

^{page 99 note 108} repacificatus regi: David, , RC 59Google Scholar, n. 79 regards it ‘as a plausible hypodiesis’ that St Calais had something to do with bringing about peace between Rufus and Robert Courthose at Rouen after Rufus's crossing to Normandy in February 1091; cf. Symeon, , HDE 128Google Scholar. According to the agreement between Rufus and Robert, those who had lost their lands in England for supporting Robert were to recover them: ASC 1091; ‘Florence’, ii, 27Google Scholar. Perhaps St Calais was regarded as included in these terms, but note Barlow's skepticism about this, WR 282Google Scholar. It is not known precisely when the bishop returned to England; Rufus came back in August 1091 according to ‘Florence’, ii, 28Google Scholar; Orderic, , iv, 236.Google Scholar

^{page 99 note 109} For Rufus's expedition in company with his brother Robert against Malcolm III (Canmore) of Scotland, see ASC 1091; ‘Florence’, ii, 28Google Scholar; Orderic, , iv, 254, 268–70Google Scholar; Barlow, , WR 291–5.Google Scholar

^{page 99 note 110} As HR 218Google Scholar, which gives the same date, 11 September 1091. Perhaps DIV's claim that St Calais had been driven from Durham on 11 September 1088 refers to die date when he capitulated to Rufus's expedition against him; see note 101. Bishop William's charter confirming the Allertonshire churches to the monks of Durham, often connected with this episode because of its famous list of witnesses, which includes Rufus, Robert Courthose and their brother Henry, is not authentic: RAN i no. 318; see DEC no. *6, pp. 48–53, where the dating limits suggested should be corrected in the light of Barlow's criticisms, WR 287Google Scholar, n. 106: 294, n. 136.

^{page 100 note 111} Thursday, 11 August 1093.

^{page 100 note 112} Neither Symeon, HDE nor the tendentious rehandling of it in LV, f. 46v mentions Malcolm's presence. HR 220Google Scholar does, and is followed by Chron. Melrose, f. 16r, facs. edn., 1936, p. 29. If indeed Malcolm was at Durham on n August, presumably he was on his way to attempt to see Rufus at Gloucester on 24 August: ‘Florence’, ii,31; Barlow, , WR 309–10.Google Scholar

^{page 100 note 113} 25 December 1095.

^{page 100 note 114} a sancte memorie Anselmo: Anselm died on 21 April 1109.

^{page 100 note 115} Thomas of Bayeux, archbishop of York, 1070–1100; Walkelin, bishop of Winchester, 1070–1098; John of Tours, bishop of Bath, 1088–1122.

^{page 100 note 116} Wednesday, 2 January 1096. ASC gives 1 January; ‘Florence’, ii, 39Google Scholar, Wednesday, 1 January.

^{page 100 note 117} DIV's reckoning gives an episcopate of 16 years and 54 days (2 × 28 – 2) from William's ‘election’ on 9 November 1080 to 2 January 1096. This is better than Symeon, HDE 134–5Google Scholar, which says 15 years and 53 days.

^{page 100 note 118} Wednesday, 16 January 1096.