Chapter VII · (continued 7)

Then her father comes, sees her with the veil, tears it from her head, and says that he does not mean her to be a nun, but to be the wife of Count Alan (“Pater meus cum me, quemadmodum dixi, velatam forte vidisset, furore succensus, injecta manu velum arripuit, et dissipans illud, odium Dei imprecatus est ei qui mihi illud imposuit, contestans se comiti Alano me potius in uxorem quam in contubernium sanctimonialium prædestinasse”).

Here we are not told how she came under her aunt’s care, nor what became of her after her father’s death. And there is something odd in the general reference to the “Normans,” unless it is meant as part of the outburst of special English feeling in the later months of the year 1100. Another version, instead of Normans in general, attributes the danger to a particular Norman whom we should hardly have looked for. This version is to be found in a most singular story, to which I have slightly referred in the text (see p. 32) and also in N. C. vol. v. p. 169, in the Narratio Restaurationis Abbatiæ S. Martini Tornacensis (D’Achery, ii. 893). The story is brought in at the same point at which it is brought in by Eadmer, at the time when Eadgyth—​if that is to be her name—​is sought in marriage by King Henry. The writer, Hermann, Abbot of Saint Martin’s, says that he had heard the story as a young man from Anselm himself. As Eadmer reports Eadgyth’s own statement, Hermann reports the statement of the abbess—“abbatissa in cujus monasterio puella illa fuerat nutrita.” If any trust can be put in the uncertified list of abbesses of Romsey in the Monasticon, ii. 507, the head of the sisterhood at that time would seem to have been an English Æthelflæd. The maiden herself also is without a name, and her brother is confounded with her father. She is “puella quædam, filia David regis Scotiæ.” The Abbess’s story is that the Scottish King entrusted his daughter to her care, not to become a nun, but simply for education (“Rex David pater ejus mihi eam commendavit, non ut sanctimonialis fieret, sed ut solummodo in ecclesia nostra propter cautelam cum ceteris puellis nostris coætancis suis nutriretur et literis erudiretur”). When the girl is about twelve years old (“cum jam adolevisset,” which is explained afterwards to mean “duodennis”), the Abbess hears that king William (defined as “rex Willelmus, domini mei regis Henrici germanus”) has come to see her (“propter eam videndam 601 venisse”). In the case of any decent king such a visit would surely have been neither scandalous nor wonderful. The King is at the abbey-gate with his knights, and asks to have it opened. The Abbess fears that he may conceive some bad purpose towards the maiden, but hopes that he will respect her if she wears the monastic veil. She therefore persuades Eadgyth to wear the veil for the time;

“Hæc audiens, nimiumque perterrita, ne forte ille, ut juvenis et rex indomitus, qui omne quod animo sibi occurrisset illico facere volebat, visa pulcritudine puellæ aliquam ei illicitam violentiam faceret, qui tam improvisus et insperatus propter eam videndam advenisset, in secretius cubiculum eam introduxi, rem ei sicut erat aperui, eaque volente velum unum capiti ejus imposui, quatenus eo viso rex ab illicito complexu revocaretur.”

The King goes into the cloister, as if to look at the flowers “quasi propter inspiciendas rosas et alias florentes herbas”). He sees Eadgyth with the veil, and goes away, showing, according to the Abbess, that his visit had been on her account only (“mox ut eam vidit cum ceteris puellis nostris velum capite gestantem, claustro exivit et ab ecclesia recessit, aperte ostendens se non nisi propter eam venisse”). Within a week King David came; seeing his daughter with the veil on her head, he was very angry; he tore it from her head, trampled it under-foot, and took his daughter away.

As the Abbot’s memory clearly failed him on one point, it may have failed him in others. This is, as far as I know, the only time in history or legend in which William Rufus is brought into connexion with the name of any woman. It may well be that Abbess Æthelflæd—​if that was her name—​did not know the secrets of the Red King’s court, and reckoned him among ordinary, instead of extraordinary, sinners.

The accounts of Orderic and Hermann assert, and that of Eadmer seems to imply, that Eadgyth at least, most likely Mary also, was sent to be brought up by their aunt when they were quite children. But there is something a little odd in the appearance of Malcolm both in Eadmer and in Hermann, where he is spoken of as if it were an every-day thing for a King of Scots to show himself at Romsey. We may here perhaps help ourselves to a date. The visit of 602 Malcolm must surely have been when he was in England in 1093. Eadgyth then, according to Hermann, was about twelve years old. Now, it seems from William of Malmesbury (iv. 389) that she had a godfather whom we should hardly have looked for in the person of Duke Robert. When could Robert have been godfather to a daughter of Malcolm and Margaret? Surely when he was in Scotland in the autumn of 1080 (see N. C. vol. iv. p. 671). That was therefore the time of Eadgyth’s birth; she would then be under thirteen when her father came into England. (Since this note was printed, I see that M. Gaston Le Hardy, p. 41, takes this date for granted.)

The fact that Malcolm and Margaret themselves sent their daughters into England seems to dispose of the account in Fordun (v. 21; see p. 30), according to which their uncle Eadgar somehow contrived to bring them to England after the death of their parents. The only way in which the two versions could be reconciled would be by supposing that, when Malcolm, according to Hermann, took Eadgyth away from Romsey, he took her back to Scotland.

In Eadgyth’s own statement in Eadmer, she says that her father meant her to marry Count Alan. So Orderic (702 A) says;

“Alanus Rufus Britannorum comes Mathildem, quæ prius dicta est Edith, in conjugem sibi a rege Rufo requisivit; sed morte præventus non obtinuit.”

Mr. Robertson (i. 152) makes merry over this passage, and takes the opportunity to sneer at Orderic. How, he asks, could Alan, who outlived Eadgyth-Matilda and died in 1119—she died in 1118—have been prevented by his own death from marrying her? He objects also that Alan married the second time (see N. C. vol. iv. p. 647) in 1093, “before Matilda could have sought refuge in England.” He adds, “Alan, however, was once a suitor for the hand of Matilda, but to her own father Malcolm (according to her own words), not to Rufus,” and goes on to tell about Orderic’s “gossip,” “infinity of error,” and what not. But though Orderic has made a slight slip, Mr. Robertson’s own error is much greater. There can be little doubt that the Alan meant is not the Alan of Britanny who married first Constance the daughter of the Conqueror and then Ermengarde of Anjou, but Alan the Black the second lord of Richmond (see N. C. vol. iv. p. 294, and Mrs. Green, Princesses, i. 25), a much more likely husband for the Scottish 603 King to think of for his daughter. Now this Alan died in 1093, just about the right time. Orderic has put Rufus instead of Niger, which is about the extent of his offence—​perhaps confounding Alan the Black with his brother Alan Fergeant, the first lord of Richmond. But Mr. Robertson quite forgot that Malcolm sent his daughters into England long before 1093. Thierry (ii. 152) saw clearly which Alan it was.

William of Malmesbury (v. 418) has a singular passage, where he tells us that “Matildis, filia regis Scotorum, a teneris annis inter sanctimoniales apud Wiltoniam et Rumesium educata, literis quoque fœmineum pectus exercuit. Unde, ut ignobiles nuptias respueret plusquam semel a patre oblatas, peplum sacratæ professionis index gestavit.”

But who could look on a marriage with Count Alan as “ignobilis”?

Tynemouth and Bamburgh.

The history of Tynemouth, and of Saint Oswine in relation to Tynemouth, comes largely from the Life of Saint Oswine in the Miscellanea Biographica published by the Surtees Society. This is the work of a monk of Saint Alban’s who went to Tynemouth in 1111. There are also several Saint Alban’s documents printed in the Monasticon, iii. 312. There is a large history of Tynemouth by Mr. W. S. Gibson, from which much may be learned, though the valuable facts and documents have largely to be dug out of a mass of irrelevant matter.

According to the Saint Alban’s writer, Eadwine built a wooden church at Tynemouth, and there his daughter Rosella took the veil. The name is strange enough, but we may perhaps see a confused tradition of a British name, when we read that “locus ubi nunc cœnobium Tinemuthense est, antiquitus a Saxonibus dicebatur Penbalcrag, i.e. caput valli in rupe. Nam circa hunc locum finis erat valli Severiani.” This building must be the same as that which is referred to in the Life, p. 11; “Delatus est ad ostium Tynæ fluminis, locum videlicet ab incolis regionis ob imminentis rupis securitatem ab hostibus celebrius frequentatum. Sed ob reverentiam gloriosæ 604 Virgini Mariæ inibi exhibitam tenerius amatum, ibique sepultus est in oratorio ejusdem Virginis, quod constructum erat ad aquilonem fluminis.” He goes on to tell how Oswald rebuilt the wooden church of stone, and how the monastery was more than once destroyed by the Danes. The Saint Alban’s writer (Mon. Angl. iii. 312) speaks more specially of the Danes. The biographer carries us at once to the time of Tostig;

“Memoria sancti martyris Oswini, obsoleta et penitus deleta, funditus ab hominum notitia evanuit. Jacuitque per multa annorum curricula gleba sancti corporis sub abjectiori cespite tumulata et usque ad tempora Thostii comitis et Ægelwini præsulis Dunelmi, incuriæ pariter et ignorantiæ neglectu, debita veneratione est fraudata.”

The writer has a curious remark to account for the neglect of the saint; “Genti prædictæ nunc fideles, nunc infideles principabantur, et juxta principum instituta, varia divinus cultus in subjectarum plebium studiis sensit dispendia.” This is doubtless true of Deira, hardly so of Bernicia, where no heathen prince reigned, though passing heathens did a good deal of damage.

He then gives a long account of the invention of the saint’s body, which came about through the vision of a monk named Eadmund. Judith, according to the character which she bears elsewhere (see N. C. vol. ii. p. 391), appears as “devota Deo famula,” “præpotens et devota femina,” “veneranda comitissa.” Of Tostig we are told that he succeeded Siward, “non testamenti beneficio, sed sancti regis Ædwardi dono regio.” He is described as beginning the new church which the monks of Saint Alban’s afterwards finished (p. 15); “Cujus tamen fundamenti initia, ut dicitur, comes Thostius jecerat, a fundamentis ædificaverunt.” But his deposition and death seem to be looked upon as a judgement for not being present in person at the invention (“Quia prædixtus comes Thostius interesse sanctæ inventioni in ditione sua factæ noluit, eodem anno culpis suis exigentibus ab Anglorum regno expulsus,” &c.), the exact date of which is given, March 15, 1065. It is added, “Thostio comite proscripto, hæreditas ejus devoluta est ad fiscum regium.”

Simeon in his History of the Church of Durham, iv. 4, puts the acts of Tostig and of Waltheof together under the head of Northumbrian 605 earls; “Ecclesiam sane sancti Oswini in Tinemuthe, jamdudum donantibus Northymbriæ comitibus, monachi cum adhuc essent in Gyrvum possederant, unde etiam ipsius sancti ossa ad se transferentes in ecclesia sancti Pauli secum non parvo tempore habuerunt, quæ postmodum ad priorem locum retulerunt.” He then goes on to record the confirmation by Earl Alberic, who “hoc donum renovavit, ipsamque ecclesiam cum suo presbitero ecclesiæ sancti Cuthberti perpetuo possidendam adjecit.”

It would seem that the fall of Tostig hindered the completion of his church, and that at the time of Waltheof’s grant it was still without a roof; for he goes on to say, “Quæ cum jam per quindecim annos velut deserta sine tecto durasset, eam monachi culmine imposito renovarunt, et per tres annos possederunt.” On receiving the confirmation of Alberic, a monk with a good Danish name was sent to put things in better order (Simeon, Gesta Regum, 1121); “Ex capituli totius sententia monachus noster Turchillus illuc mittitur, qui renovato ecclesiæ ipsius culmine, per multum tempus habitavit ibidem.”

I have referred to the charter of Waltheof and to the entry in Simeon (Gesta Regum, 1080) in N. C. vol. iv. p. 666. It is printed, along with a charter of Bishop William confirming it, dated April 27, 1085, in the time of Earl Alberic, whose confirmation is recorded, in the Surtees book called Historiæ Dunelmensis Scriptores Tres, pp. xviii, xix. The signatures to both are nearly all English, with the single exception of two to the charter of Waltheof. These are Gilbert, the nephew (see N. C. vol. iv. p. 665) of Bishop Walcher, and an unknown Walter. We meet with several other men that we know, as Morkere’s father Ligulf and his brother Uhtred, and Leofwine, written “Leobwinus,” the Dean of Durham. We notice also “Ernan Biscope sune,” and three Englishmen with the knightly title “Alwinus miles,” “Wlstanus miles,” and “Kinewlfus miles,” but I do not understand “signum Aldredi comitis.” Earl Ealdred, the common grandfather of Waltheof and young Morkere, had been murdered long ago, as the sons of Carl found to their cost. The story is told again in Simeon, Gesta Regum, 1121.

The next stage in the story is the taking away of Tynemouth from the church of Durham. It is amusing to contrast the ways in which this story is told at Durham and at Saint Alban’s. Simeon, in the chapter just quoted, tells us that Earl Robert made the gift 606 to Saint Alban’s “propter inimicitias quæ inter episcopum et ipsum agitabantur” (cf. Gesta Regum, 1121). The cause of their ill-will, a dispute about lands, comes out in the next chapter. Roger of Wendover (ii. 39), who is copied by Matthew Paris (Hist. Ang. i. 41, and Chron. Maj. ii. 31), tells us how Earl Robert—“vir quidem Deo devotus,” Matthew says—​gave Tynemouth to Saint Alban’s “divina inspiratione tactus.” The Gesta Abbatum (i. 57) add that it was done “regis et archiepiscopi Lanfranci benevolentia.” It would seem that under Durham rule Tynemouth had been simply an impropriate church, while in the hands of Saint Alban’s it became a cell. The judgement on Abbot Paul is recorded in the Durham History, iv. 4. The Gesta Abbatum, which record much about him, both good and evil, say nothing about this. The Life of Oswine, p. 15, gives a full account of the ceremony of the translation of Saint Oswine, with the date. Bishop Randolf of Durham was there, Abbot Richard of Saint Alban’s, and “Abbas Salesberiensis Hugo,” where we may see (see Mon. Angl. iii. 495) the old confusion (see N. C. vol. iv. p. 799) between Salisbury and Selby.

Tynemouth then, at the time when the revolt of Robert of Mowbray began (see p. 47), was already a monastery and a cell to Saint Alban’s, though the monks of Durham still held that they had been wrongfully deprived of it. But it appears from the narrative that, besides the monastery, there was also a castle. The account in the Chronicle is, “And þone castel æt Tinemuðan besæt oððet he hine gewann, and þæs eorles broðer þærinne and ealle þa þe him mid wæron.” Florence says, “Rex exercitu de tota Anglia congregato, castellum prædicti comitis Rotberti, ad ostium Tinæ fluminis situm, per duos menses obsedit; et interim, quadam munitiuncula expugnata, ferme omnes meliores comitis milites cepit, et in custodia posuit; dein obsessum castellum expugnavit, et fratrem comitis, et equites, quos intus inveniebat, custodiæ tradidit.” Florence seems to me to have confounded the sieges of Tynemouth and of the New Castle. By the “castellum ad ostium Tinæ” he would seem to mean the New Castle, and by his “munitiuncula” he would seem to mean the Earl’s fortress at Tynemouth. Now what was the relation between the castle and the monastery? As things now stand, castle and monastery are one. That is to say, the deserted church—​or more strictly the two deserted churches, monastic and parochial, once under one roof (see Archæological Journal, vol. xxxvii. 607 p. 250, No. 147, 1880)—standing on the northern promontory is now surrounded by military buildings and the great gate-house. I get my notion of the early arrangements of Tynemouth from several old plans collected by Mr. Gibson. There is one which seems to be of the sixteenth century, and, as the names are written in a curious mixture of English, Latin, and Italian, it struck me that it might be the work of an officer of those Italian mercenaries who were employed in the civil wars of Edward the Sixth. This is the only one which distinctly shows “the Castle,” on the southern promontory, though all mark that point as taken in within the lines of defence. It seems to me that the southern promontory must have been the site of the original castle, and that the name of Castle has shifted to the great gate-house, which fairly deserves it.

With regard to the order of the sieges, Orderic, who gives us so full an account of the siege of Bamburgh, tells us nothing about the others. I gather from the words of the Chronicle that the New Castle, which we find in the King’s hands directly after, was the point which was first taken; “Sona þes þe he þider [to Norðhymbran] com, he manege and forneah ealle þa betste of þes eorles hirede innan anan fæstene gewann, and on hæftene gedyde.” Florence, as I have said, seems to have misunderstood the words of the Chronicler, and to have confounded Tynemouth and the New Castle. This last would surely be, as the Chronicle implies, the first point of attack after the army entered Northumberland in the sense which that word now bears. Next in the narrative of the Chronicle follows the siege and capture of Tynemouth, and then the great siege of Bamburgh. Of this famous fortress I found something to say long ago in N. C. vol. i. p. 410, where Bamburgh appears as marking one stage in the art of fortification. Bæda (iii. 16) witnesses that the place took its name “ex Bebbæ quondam reginæ vocabulo;” so also the Northumbrian writer copied by Simeon of Durham, 774;

“Bebba civitas urbs est munitissima, non admodum magna, sed quasi duorum vel trium agrorum spatium, habens unum introitum cavatum, et gradibus miro modo exaltatum. Habet in summitate montis ecclesiam præpulcre factam, in qua est scrinium speciosum et pretiosum. In quo involuta pallio jacet dextera manus sancti Oswaldi regis incorrupta, sicut narrat Beda historiographus hujus gentis.”

608 The reference here is to Bæda, iii. 6, where he tells the story of Oswald’s bounty and the prophecy of Aidan, and adds how his hand and arm, cut off after his death in the battle by Penda, “in urbe regia quæ a regina quondam vocabulo Bebba cognominatur, loculo inclusæ argenteo in ecclesia sancti Petri servantur, ac digno a cunctis honore venerantur.” So again, iii. 12, where Bamburgh is simply “regia civitas.” He goes on to speak of the well; “Est in occidente et in summitate ipsius civitatis fons miro cavatus opere, dulcis ad potandum et purissimus ad videndum.” Florence also refers to the origin of the name; with him it is “Bebbanbyrig, id est, Urbs Bebbæ reginæ;” and Orderic (704 A) draws a little picture of the spot; “Munitissimum castrum, quod Babbenburg dicitur, obsederunt. Et quoniam illa munitio inexpugnabilis erat, quia inaccessibilis videbatur propter paludes et aquas, et alia quædam itinerantibus contraria, quibus ambiebatur, rex novam munitionem ad defensionem provinciæ et coartationem hostium construxit, et militibus, armis, ac victualibus implevit.” This last fact, the making of the Malvoisin, is recorded by the Chronicler and Florence, both of whom give the name. The Chronicler says; “Ac þa þa se cyng geseah þæt he hine gewinnan ne mihte, þa het he makian ænne castel toforan Bebbaburh and hine on his spæce Malueisin het, þæt is on Englisc yfel nehhebur, and hine swiðe mid his mannan gesætte, and syððan suðweard for.” So Florence; “Ante Bebbanbyrig in quam comes fugerat, castellum firmavit, id que Malveisin nominavit, et in illo militibus positis, in Suthymbriam rediit.” We may here note the way in which the Chronicler assumes French as the language of William Rufus, and also Florence’s somewhat archaic way of speaking of “Suthymbria,” where the Chronicler says simply “suðweard.” It is something like his mention of West-Saxonia in 1091 (see vol. i. p. 305).

The Malvoisin was clearly such a tower as we often hear of, temporary and of wood, but still not moveable, as is implied in Florence’s word “firmavit.” But the name seems afterwards to have been transferred to moveable towers; see Du Cange in Malveisin, where he refers to the passage about the siege of Dover in Roger of Wendover, iii. 380; “Misso prius ad patrem suum propter petrariam, quæ ‘Malveisine’ Gallice nuncupatur, qua cum machinis aliis Franci ante castrum locata muros acriter crebris ictibus verberabant.” In his account of the siege of Bamburgh 609 (ii. 46) Roger says, “Cum castellum inexpugnabile advertit, ante castellum illud castellum aliud ligneum construxit, quod Malveisin appellavit, in quo partem exercitus sui relinquens inde recessit.” Matthew Paris copies this in the Chronica Majora in the Historia Anglorum, i. 48; his words are, “Ante castellum illud aliud sed ligneum construxit, ad præcludendum illis exitum, quod patria lingua Maleveisine appellavit.” Viollet-le-Duc (Military Architecture of the Middle Ages, 24, Eng. trans.) seems to imply that moveable towers were known earlier than this time, but he seems (p. 30) to bring the petraria from the East.

As for the details of the siege, the Chronicler and Florence tell us nothing till we come to the escape of Robert from Bamburgh. It is Orderic who gives us the picture of the state of mind of Robert and his companions, which, if it belongs to any period of the siege, must belong to the time before the King went southward. We see the loyal troops busily working at the making of the Malvoisin;

“Conscii autem perfidiæ et fautores eorum detegi verentes conticuerunt, et metu exsangues, quia conatus suos nihil valere perpenderunt, regiis cohortibus immixti, ejus servitium, cujus exitium optaverant, prompte aggressi sunt. Interea, dum rex in armis cum agminibus suis ad bellum promptus constaret, et chiliarchos ac centuriones, aliosque proceres Albionis, cum subditis sibi plebibus, operi novæ munitionis indesinenter insistere compelleret, Rodbertus de propugnaculis suis contrarium sibi opus mœstus conspiciebat, et complices suos alta voce nominatim compellebat, ac ut jusjurandum de proditionis societate conservarent, palam commonebat. Rex autem cum fidelibus suis hæc audiens ridebat, et conscia reatus publicati mens conscios et participes timore et verecundia torquebat.”

Then the King goes away; in Orderic’s phrase, “rege ad sua prospere remeante, et de moderamine regni cum suis amicis solerter tractante,” a rather odd description of the war in Wales. Now comes Robert’s escape from Bamburgh. Orderic, who seems to have no clear idea of any place except Bamburgh, merely says that Robert, “longæ obsidionis tædio nauseatus, noctu exilivit, et de castro in castrum migrare volens in manus inimicorum incidit.” The Chronicle is fuller; “Ða sona æfter þam þe se cyng wæs suð afaren feorde se eorl anre nihte ut of Bebbaburh towardes Tinemuðan, ac þa þe innan þam niwun castele wæron his gewær wurdon, and him æfter foran and onfuhton and hine gewundedon, and syððan 610 gelæhton, and þa þe mid him wæron sume ofslogan sume lifes gefengon.” But it is from Florence that we get the detailed account. His story runs thus;

“Post cujus discessum, comiti Rotberto vigiles Novi Castelli promisere in id se permissuros illum intrare, si veniret occulte. Ille autem lætus effectus, quadam nocte cum xxx. militibus ut id perageret exivit. Quo cognito, equites qui castellum custodiebant illum insequentes, ejus exitum custodibus Novi Castelli per nuntios intimaverunt. Quod ille nesciens, die dominica tentavit peragere cœpta, sed nequivit, deprehensus enim erat. Eapropter ad monasterium S. Oswini regis et martyris fugit, ubi sexto die obsessionis suæ graviter in crure est vulneratus dum suis adversariis repugnaret, quorum multi perempti, multi sunt vulnerati, de suis quoque nonnulli vulnerati, omnes sunt capti; ille vero in ecclesiam fugit, de qua extractus, in custodia est positus.”

Here now comes the obvious difficulty as to the way in which the Earl could have got into the monastery at Tynemouth after the castle had been taken. The Chronicler indeed does not necessarily imply that he got into Tynemouth at all. The fight which he describes might have happened somewhere else and not at Tynemouth. And if any one chooses to move the site of Robert’s resistance and capture from Tynemouth to some unknown spot, there is only the statement of Florence against him. That Robert was taken, and taken after a stout resistance, is plain.

With Robert’s capture, Orderic ends his story, as far as military operations are concerned. “Captus a satellitibus regis, Rodbertus finem belli fecit.” In a very general way this is not untrue; it was the capture of Robert which brought about the end of the war. But it is odd that he should have left out the striking story of the captive Earl being brought under frightful threats before the castle which his wife was defending. This stands out clearly in the Chronicle; “Ða þa se cyng ongean com, þa het he niman þone eorl Rotbeard of Norðymbram, and to Bæbbaburh lædan, and ægðer eage ut adon, buton þa þe þærinne wæron þone castel agyfan woldan. Hine heoldan his wif and Moreal, se wæs stiward and eac his mæg. Ðurh þis wearð se cartel þa agyfen.” Florence translates this.

Lastly comes the great difference of all as to Earl Robert’s last 611 days. The Chronicler and Florence merely record his imprisonment at Windsor, without saying how long it lasted. Florence says only, “Comes forti custodiæ mancipandus ad Windlesoram est ductus,” followed by the passage about Morel quoted in p. 55. He says nothing about the many accusations brought by Morel, or about the special summons of all the tenants-in-chief to the trial, of which the Chronicler speaks (see p. 56). The Chronicler, after recording them, says; “And þone eorl Rotbert hét se cyng to Windlesoran lædan, and þær innan þam castele healdan.” This is consistent with any later destiny, with release and monastic profession or with lifelong imprisonment. This last is asserted by several authorities. Thus Orderic (704 A) says; “Rodbertus…. fere triginta annis in vinculis vixit, ibique scelerum suorum pœnas luens consenuit.” He then sets forth the sad state of his wife; “Mathildis uxor ejus, quæ cum eo vix unquam læta fuerat, quia in articulo perturbationis desponsata fuerat, et inter bellicas clades tribus tantum mensibus cum tremore viri thoro incubuerat, maritali consolatione cito caruit, multisque mœroribus afflicta diu gemuit.” The Continuator of William of Jumièges (viii. 8), who has nothing to say about Matilda, equally bears witness to Robert’s lifelong imprisonment; “Captus a militibus Willelmi regis, ipsoque jubente in ipsis vinculis diutius perseverans; regnante jam Henrico rege, tandem in ipso ergastulo deficiens mortuus est.” So William of Malmesbury, iv. 319; “Captus et æternis vinculis irretitus est.”

On the other hand, there clearly was a story according to which Robert was released some time or other, and died a monk at Saint Alban’s. It is somewhat remarkable that there is no mention of this in any of the chief writings of Matthew Paris, neither in the Historia Major nor the Historia Anglorum, nor the Lives of the Abbots. But we find the story implied in the extract from his Additamenta in the Monasticon, iii. 312; “Ibidem [at Tynemouth] monachos congregavit de domo sancti Albani, tanquam ab electissima domo inter omnia cœnobia Angliæ, ubi etiam se vovit monasticum habitum suscepturum, et sepulturam in loco memorato. Quæ omnia, Deo sibi propitio, feliciter consummavit.” So in the Abbreviatio Chronicorum (Hist. Angl. iii. 175), a marginal note is added to the name of Earl Robert; “Sepultus est apud sanctum Albanum.” But, oddly enough, the most distinct statement that he became a monk comes, not from any Saint Alban’s 612 writer, but from one manuscript of the “De Regibus Saxonum Libellus” at the end of the Surtees Simeon, p. 214. King Henry keeps Robert of Mowbray some while in prison; then “rogatu baronum suorum eundem resolvens, concessit illi mutare vitam habitumque sæcularem. Qui ingressus monasterium Sancti Albani sub professione monastica ibidem vitam finivit.”

The story about Matilda’s second marriage and divorce comes from Orderic. His story runs thus; “Vir ejus, ut dictum est, in carcere vivebat, nec ipsa, eo vivente, secundum legem Dei alteri nubere legitime valebat. Tandem, permissu Paschalis papæ, cui res a curiosis enucleata patuit, post multos dies Nigellus de Albineio ipsam uxorem accepit, et pro favore nobilium parentum ejus, aliquamdiu honorifice tenuit. Verum, defuncto Gisleberto de Aquila fratre ejus, vafer occasionem divortii exquisivit, eamque, quia consanguinei sui conjux fuerat, repudiavit, et Gundream, sororem Hugonis de Gornaco, uxorem duxit.” If all this happened at all, it must have happened between 1099 and 1118, the years which mark the reign of Paschal.

Matilda of Laigle could not well have been the sister of William the Chaplain to whom Bishop Herbert Losinga writes his third letter (Ep. Herberti, p. 5). He there says; “De matrimonio sororis vestræ non aliud respondeo vobis, quam id quod præsens ex ore meo audivistis, suo videlicet ut vivente viro, secundum evangelium et secundum sanctorum canonum usum, alii viro nubere non potest.” But the person spoken of could hardly have been thinking of such a marriage, unless she had some special excuse, like this of Matilda.

The second wife of Nigel appears both as “Gundrea” and as “Gundreda.” There is a great deal about her husband Nigel and her son Robert, the founder of Byland Abbey, in the Monasticon, v. 346–351. The marriage of Nigel and Gundreda took place after Tinchebrai, and as King Henry gave Nigel the castle of Mowbray, and much else in Normandy and England which had belonged to Earl Robert, their son Roger called himself Roger of Mowbray. Such a description was likely to lead to confusion, and it may have led some to fancy that later bearers of the name of Mowbray had something to do with the famous Bishop and Earl of our story. The artificial Percy is indeed connected with the real one by grandmothers; but the artificial Mowbray was purely artificial. This Roger of Mowbray 613 appears also in the Continuator of William of Jumièges, viii. 8, who tells us that Nigel himself became a monk at Bec.

As Walknol has been casually mentioned in the text (p. 47) there may be some interest in a document in the Cartulary of Newminster published by the Surtees Society, p. 178. The date must be after 1137, the date of the foundation of Newminster. The number of English names, and specially the two bearers of scriptural names who are sons of English-named fathers, illustrate points of which I have often had to speak;

“De terra de Walknol in castro. Johannes filius Edwyni fabri, salutem. Sciatis me concessisse, dedisse, et hac præsenti carta mea confirmasse, Bartholomæo filio Edricii illam terram totam quæ jacet in australi parte cimiterii capellæ beati Michaelis, in longitudine a curtillo Eadmundi clerici usque ad terram quæ fuit Johannis Stanhard, et in latitudine a cimiterio capellæ beati Michaelis usque ad antiquam communem viam subtus versus austrum. Habendum et tenendum eidem Bartholomæo et hæredibus suis de me et hæredibus meis et assignatis in perpetuum, libere, quiete, et pacifice, pro duabus marcis arg. quas michi dedit idem Bartholomæus in manu in mea magna necessitate.”

The Conquest of Glamorgan.

I gave a note to the conquest of Glamorgan in the Appendix to vol. v. of the Norman Conquest, p. 820. I look, as I did then, upon the account in what I find it convenient to call the later Brut as thoroughly legendary in its details, though I am perhaps inclined to put rather more faith in the general story than I was then. And I am not so much inclined as I was then to draw the same wide distinction as Mr. Floyd draws between the expeditions led by the King himself and those which partook more or less of the character of private adventure. There was doubtless a difference, when it was King William who called the whole force of England to his standard, and when it was only either Earl Hugh or Robert Fitz-hamon who set out on an expedition on his own account. But 614 both processes were parts of the same general undertaking. Whatever individual lords conquered, they conquered with the King’s approval, to be held by them as his vassals and subjects. He himself stepped in only on great occasions, when the Welsh seemed to be getting too strong for the local lords. The same general work must have been going on all over the country. The only strange thing is that the conquest of Glamorgan, of whose general results there can be no doubt, and of which we have so very full a legendary account, is left out altogether in every really trustworthy history.

Jestin ap Gwrgan must be accepted as a real man, on the strength of his real sons and grandsons (for his sons see N. C. vol. v. p. 821); but that is all that can be said of him. We can hardly carry our faith so far as Mr. John Williams ab Ithel, the Editor of the Brut in the Chronicles and Memorials, who asks us (xxiii) to “consider the great age of the prince of Glamorgan when he died. He is said to have married his first wife A.D. 994”—​it is perhaps prudent to mention the æra—“and to have died at the age of 111, according to others 129.” We Saxons do not venture to believe in the kindred tales of our own Harold and Gyrth. But we learn from Mr. Williams himself, at the very beginning of his Preface, that “the voice of Tradition would not lead us to suppose that the ancient Britons paid any very particular attention to the study of chronology previous to the era of Prydain, son of Aedd the Great, which is variously dated from the year 1780 to 480 before the nativity of Christ.” If centuries went for so little in the days of Prydain, it is not wonderful that decades did not go for much in the days of Jestin. Nor are we surprised to find that Mr. Williams knew the exact number of the descendants of Jestin, who were, like those of Attila, “pene populus.” All that we can say of Jestin’s story, in relation to Robert Fitz-hamon and his companions, is that there is no trustworthy evidence either for or against the story of his invitation to the Norman knights, but that the tale has a legendary sound, and that the date is in any case wrong. If we should be inclined, according to one or two indications (see p. 84), to place the conquest of Glamorgan several years earlier, perhaps even before the death of the Conqueror, we are only carried away yet further from the perfectly certain date of the death of Rhys ap Tewdwr. All that we can say is that the general story may be true, but that the list of settlers given in the 615 later Brut (72 to 75) is largely due to family vanity. The Stradling family, for instance, had nothing to do with the original conquest.

The best account of the whole matter is to be found in Mr. Clark’s first paper on “The Land of Morgan,” in the Archæological Journal, xxxiv. 11. I cannot however admit with him (p. 18) that “it seems probable that to the early Vikings, and not to the later settlements of Flemings or English, is due the Teutonic element which prevails in the topography of lower Pembroke and Gower.” I am quite ready (see p. 95) to admit a certain Scandinavian element; but the Flemish settlement in Pembrokeshire is undoubtedly historical (see N. C. vol. v. p. 855), while we have fair legendary evidence for making the settlement in Gower West-Saxon (see p. 103). The name of Worm’s Head given to the great promontory of Gower, in marked distinction to the Scandinavian Orm’s Head in North Wales, goes a long way to show that the Teutonic settlers in Gower were either Flemish or Saxon, and not Scandinavian.

Godwine of Winchester and his son Robert.

I gave a short note to the history of Robert son of Godwine in N. C. vol. v. p. 819. On going again more minutely through the story, I am even more struck than before by the singular way in which different notices of Robert and Godwine hang together. It is one of the best cases that I know of the argument from undesigned coincidences. Besides the interest of the story in itself, it teaches us, like many other stories, how, if we work with a proper caution, we may dig truth out of quarters where we should hardly have looked for it, and it may specially suggest matter for thought as to the value of those pieces of Scottish history which one hardly knows whether to call the writings of Turgot or Fordun, or of any one else. I suspect that, if we simply read the story of Godwine and Robert as it stands in Fordun, we should be inclined to cast it aside altogether. The story undoubtedly has a legendary air, and the details of the single combat are likely enough to have received some legendary colouring even at the time. Some might 616 even be a little startled at the appearance of Englishmen of knightly rank at the court of William Rufus. But we see from Domesday on the one hand, and from William of Malmesbury on the other, that Godwine and Robert were real men, and we see that the part which they play in Fordun’s story is exactly in accordance with their real position.

I have mentioned elsewhere (see N. C. vol. iv. p. 571; vol. v. p. 819) that there was a Godwine holding lands in Hertfordshire of the Ætheling Eadgar. We also have in two places in William of Malmesbury (iii. 251; iv. 384) notices of “Robertus Anglus,” “Robertus filius Godwini miles audacissimus,” who goes to the crusade with the Ætheling, and who does the exploits which I have spoken of in p. 122. Now if circumstantial evidence is ever good for anything, one can hardly doubt that the Godwine of Domesday is the same as the Godwine of William of Malmesbury and as the Godwine of Fordun, and that the Robert son of Godwine in Fordun is the same as the Robert son of Godwine in William of Malmesbury. The three accounts are wholly independent, but all bring Godwine and Robert into connexion with Eadgar. It is almost inconceivable that Fordun’s story should be mere invention, when it makes men of whom so little is known act exactly in character with the little that is known of them.

In the account in Fordun (ii. 22, Surtees Simeon 263), Ordgar, “Orgarus,” is described in the one text as “miles degener Anglicus,” in the other as “miles de genere Anglico,” which is clearly the better reading.

The name of Ordgar appears only twice in Domesday. In Oxfordshire, 161, Ordgar, a king’s thegn, holds two hides of the worth of forty shillings. He had two slaves on his domain, and half a carucate was held by two villains or churls. We then read, “Godwinus libere tenuit.” This is pretty sure to be our Ordgar, and it may very well be our Godwine, though we can say nothing for certain about so common a name. If they are the same, here is great likelihood, though no proof, that Godwine may have had other ground for willingness to fight Ordgar, besides his loyalty to the Ætheling. Ordgar, on the other hand, appears in Somerset, 93, as holding a hide which had passed to Robert of Courcelles, and which, with a good deal more, was held by Anschitil. Ordgar 617 was not the only Englishman who, among the endless forfeitures and grants—​to say nothing of ordinary sales, bequests, and exchanges, which went on T. R. W. as well as T. R. E.—​lost in one part of England and gained in another.

In Fordun’s story Eadgar is described as “clito Eadgarus, viz. genere gloriosus, nam sic ipsum nominabant.” “De genere gloriosus,” it will be marked, is a more literal translation of “Clito” than it is of “Ætheling.” William is inclined to hearken to Ordgar, “quia Eadgarus de regia stirpe fuerat progenitus, et regno, jure Anglico, proximus.” We then read, “nec incerta de Eadgaro jam poterat esse sententia, si crimen impositum probari potuisset.” Eadgar is in great trouble for fear of not finding a champion, when Godwine steps forward; “Miles de Wintonia, Anglicus natione, genere non ignobilis, nomine Godwinus, veteris parentelæ ipsius non immemor, opem se præstiturum in hac re tam difficili compromisit.”

The two knights now go forth, as I have described in the text, and we have a significant comment on the lack of English patriotism shown by Ordgar;

“Hinc etiam calumniatorem cum justa animadversione increpat, qui Anglicus genere existens naturæ videretur impugnator, quem enim ut dominum venerari debuerat, utpote de jure generis existens cui se et omnia sua debuisset.”

Then come the details of the combat. We hear no more of Godwine after his victory and reward, which last is thus told; “Superati hostis terras et possessiones hereditario jure rex ei concederet possidendas.” “Hereditario jure” most likely simply means, as usual, that the land was to go on to Godwine’s heirs. It need not refer to the probable fact that part at least of Ordgar’s lands had once belonged to Godwine.

Robert first appears in Fordun, v. 25, on the march to Scotland (see p. 119). He is introduced as “quidam miles, Anglicus genere, Robertus nomine, filius antedicti Godwini, paternæ probitatis imitator et hæres.” Then come his exploits and adventures in Britain, as I have told them in the text. Afterwards must come his crusading exploits as described by William of Malmesbury. In the earlier of his two accounts (see p. 122) one might almost have thought that King Baldwin had no companion except Robert. The second passage, which gives them four other companions, has 618 therefore the force of a correction; “Rex … quinque militibus comitatus, in montana rependo, insidiantes elusit. Militum fuit unus Robertus Anglus, ut superius dixi; cæteros notitiæ nostræ fama tam longinqua occubuit. Ille cum tribus comprehensus est; unus evasit cum rege.” Another point which is worth notice is that the period of the crusade at which Robert is brought in exactly agrees with the story of his doings in Scotland and Northumberland. A man who had difficulties with Flambard after he became bishop in 1099 could not have been with the first crusaders at Antioch and Jerusalem; he might have been quite in time to help Baldwin at Rama.

It would be worth the while of some Hertfordshire antiquary to see whether anything can be made out as to the descent of the lands held by Godwine, or as to any descendants of him and Robert. But I saw a little time back a newly published history of that county, which was eloquent about the grandmothers of various obscure persons of our own time, but which had not a word to say about the champion of Eadgar or the comrade of Baldwin.

The Expedition of Magnus.

The expedition of Magnus, which, by leading him to the shores of Anglesey, had a not unimportant bearing on English affairs, is not spoken of at any great length by our own writers. The Chronicler does not name the Norwegian king; but he does not fail to mention the death of Earl Hugh of Shrewsbury, and, what was practically its most important result, the succession of his brother Robert. His words are; “And Hugo eorl wearð ofslagen innan Anglesege fram ut wikingan and his broðer Rodbert wearð his yrfenuma, swa swa he hit æt þam cynge ofeode.” Florence is fuller;

“Eo tempore rex Norreganorum Magnus, filius regis Olavi, filii regis Haroldi Harvagri, Orcadas et Mevanias insulas cum suo adjecisset imperio, paucis navibus advectus illuc venit. At cum ad terram rates appellere vellet comes Hugo de Scrobbesbyria, multis 619 armatis militibus in ipsa maris ripa illi occurrit, et, ut fertur, mox ab ipso rege sagitta percussus … interiit.”

Florence, it will be seen, here makes the same confusion between the names Hardrada and Harfagra which he made in 1066, and which so many others made beside him. To the account in William of Malmesbury, iv. 329, I have referred in p. 134. He alone it is who mentions the presence of the younger Harold in the fleet of Magnus. His words, which I quoted in p. 124, seem to come from the same source as the account in Florence; but he gives the story a different turn by distinctly making Magnus design an attack on England;

“Jam Angliam per Anglesiam obstinatus petebat; sed occurrerunt ei comites, Hugo Cestrensis et Hugo Scrobesbiriensis; et antequam continentem ingrederetur, armis eum expulerunt. Cecidit ibi Hugo Scrobesbiriensis, eminus ferreo hastili perfossus.”

Henry of Huntingdon would seem to translate the Chronicle; but he makes a confusion as to the persons by whom Earl Hugh was slain; “Hugo consul Salopscyre occisus est ab Hibernensibus. Cui successit Robertus de Belem frater ejus.”

If we could suppose that the Archdeacon of Huntingdon had paid so much attention to British affairs, we might fancy that he confounded the fleet of Magnus with the wikings from Ireland whom Cadwgan and Gruffydd hired a little time before. See p. 128.

The Welsh writers naturally tell the tale as part of their own history. The Earls have come into Anglesey; then comes Magnus. There are two different accounts in two manuscripts of the Annales Cambriæ; that which the editor follows in the text runs thus;

“Francis in insula morantibus, Magnus rex Germaniæ cum exercitu venit in insulam volens. Sed ei nolenti Franci ei occurrentes se invicem sagittis salutaverunt, hi de terra, illi de mari, alter comes sagitta in facie percussus occubuit. Quo facto, Magnus abivit.”

The other manuscript reads;

“Francis in insula morantibus, Magnus rex Germaniæ ad insulam Mon venit et prœlium cum consulibus commisit; sed alter consulum vulneratus in facie cecidit; alter vero cum majoribus insulam dereliquit. Postea vero Magnus rex insulam Mon repente reliquit.”

620 The Brut says;

“The French entered the island, and killed some of the men of the island. And whilst they tarried there, Magnus, King of Germany, came, accompanied by some of his ships, as far as Mona, hoping to be enabled to take possession of the countries of the Britons. And when King Magnus had heard of the frequent designs of the French to devastate the whole country, and to reduce it to nothing, he hastened to attack them. And as they were mutually shooting, the one party from the sea, and the other party from the land, Earl Hugh was wounded in the face, by the hand of the King himself. And then King Magnus, with sudden determination, left the borders of the country.”

It will be seen that both versions of the Annals call Magnus “rex Germaniæ.” In the text of the Brut he is “Magnus brenhin Germania.” Another manuscript, worse informed as to his name, better informed as to his kingdom, calls him “Maurus brenhin Norwei.” This odd description of a Norwegian king as king of Germany has been met with before in the Brut, 1056; but it is not found in the Annals for that year. But it must have been by a kindred flight that the annalist in 1066 called Harold Hardrada “rex Gothorum.”

Our fuller accounts of the course of Magnus come from Orderic, from the Manx Chronicle, and from the Saga of Magnus Barefoot (Johnstone, 231; Laing, iii. 129). Orderic, as we have seen, looks upon the expedition as being directly designed against Ireland. The Norwegian writer mentions Ireland only quite incidentally. Magnus plunders in Ireland, as everywhere else, on his way to Man, but the object of the expedition is clearly marked as being Man and the other islands which were so closely connected with it, a connexion which is also most strongly set forth in the pompous words of Orderic (767 D). We can have little doubt in accepting the Manx writer’s version of the history of his own island, rather than that of the Norwegian writer, to whom the internal affairs of the island were of no great interest, or the wild statement of Orderic (see p. 141) that Man was at this moment a desert island. On the other hand, the Saga is the best authority for the actual voyage of Magnus, though it is the Manx writer who preserves the fact or legend of the irreverent dealings of 621 Magnus towards his sainted kinsman. As to what happened in Anglesey, I have already quoted the accounts of the English and Welsh writers, and the Manx chronicler does not go into any greater detail;

“Ad Moiniam insulam Walliæ navigavit, et duos Hugones comites invenit in ea; unum occidit, alterum fugavit, et insulam sibi subjugavit. Wallenses vero multa munera ei præbuerunt, et valedicens eis ad Manniam remeavit.”

The detailed accounts of the death of Earl Hugh come from the Saga and from Orderic. Orderic, it must be remembered, is writing on a subject of special interest to him, on account of his close connexion from childhood with the house of Montgomery. On the other hand, as we have seen (see p. 143), he does not well understand the geography, and seems to fancy that Dwyganwy was in Anglesey. But it will be at once seen that he conceives the death of Earl Hugh in a quite different way from the author of the Saga. In Orderic’s story, though there is a great deal of preparation for fighting, there is no actual fighting at all, except the one shot sent from the bow of the Norwegian King. His version stands thus;

“Quadam vero die, dum supra littus indigenæ turbati discurrerent, seque contra Nordicos, quos in navibus suis sævire contra Anglos videbant, præpararent, Hugo comes, equum calcaribus urgens, cœtus suos congregabat, et contra hostes, ne sparsim divisi invaderentur, principali rigore coercebat. Interea barbarus Nordwigena, ut comitem agiliter equitantem prospexit, instigante diabolo stridulum missile subito direxit, egregiumque comitem, proh dolor! percussit. Qui protinus corruit, et in fluctibus maris jam æstuantis exspiravit. Unde dolor ingens exortus est.”

This really seems hardly possible, and the Welsh account, as well as the Norwegian, distinctly records fighting and shooting of arrows on both sides. The Saga gives us the details, both in prose and verse. The shooting of the King and the other archer is described in prose as I have told it in p. 144, and both the death of Earl Hugh and the general picture of the battle are given in vigorous verse from the minstrelsy of Biorn Cripplehand (Biörn inn Krepphendi). Besides the verses which Laing translates, the Saga gives others from another poet, Gisl, who vigorously describes the fight between the King and those whom he calls the 622 Welsh Earls (Valsea Jarla), meaning doubtless rather Gal-Welsh than Bret-Welsh;

The relations between Magnus and the Irish King Murtagh are very puzzling. Orderic must have made some mistake when he attributes the expedition of Magnus to a dispute with an Irish king whose daughter he marries and sends back again (767 C, D). This must surely be a confusion between Magnus himself and his son Sigurd, who, according to the Saga, did marry the Irish king’s daughter. But it is possible that Orderic’s story about the Irish princess being sent back again, because her father did not fulfil the marriage contract, may be true of Sigurd, though not of his father. We should thus better understand the transactions which go on a little later about the marriage of a daughter of Murtagh, seemingly the same, to Arnulf son of Earl Roger (see p. 442). The Manx writer has nothing to say about these marriages, but he fills up the space between this expedition of Magnus and that in which he fell with some very strange dealings between Magnus and Murtagh. Magnus sends his shoes to the Irish king, bidding him bear them on his shoulders in public as a sign of subjection to their owner (“Murecardo regi Yberniæ misit calceamenta sua, præcipiens ei ut ea super humeros suos in die natalis Domini per medium domus suæ portaret in conspectu nunciorum ejus, quatinus intelligeret se subjectum esse Magno regi”). The Irish are naturally angry; but their king takes matters more quietly. He would willingly not only carry the shoes but eat them, sooner than a single province of Ireland should be laid waste. So he did as he was bid (“rex, saniori consilio usus, non solum, inquit, calceamenta ejus portare, verum etiam manducare mallem, quam Magnus rex unam provinciam in Ybernia destrueret. Itaque complevit præceptum et nuncios honoravit”). The Irish writers of course know nothing about the shoes; but 623 the Chronicon Scotorum records a year’s peace made in 1098 between Murtagh and Magnus (“Magnus ri Lochlainne”). The Manx chronicler also goes on to say that a treaty followed the ceremony of the shoes, but that the ambassadors of Magnus gave such a report of the charms of Ireland, that he determined to invade it again in breach of the treaty.

This brings us to the date of the last expedition of Magnus. The Chronicon Scotorum records the death of Magnus (“Magnus ri Lochlainne ocus na Ninnsit”) in 1099 in an attack on Ulster. But this date must be too early. The Norwegian account places the second expedition of Magnus nine years after his accession in Norway (Laing, iii. 143, Johnstone, 239). This would fix its date to 1102. This is the date commonly given, with 1103, as the year of his death. The Manx writer places the death of Magnus six years after his first expedition (“regnavit in regno insularum sex annis,” p. 7), which would put his death in 1104. But he gives 1102 as the date of his successor in the island kingdom, Olaf the son of Godred Crouan (see p. 137). He was, it seems, at the English court; “Quo [Magno] mortuo, miserunt principes insularum propter Olavum filium Godredi Crouan, de quo superius mentionem fecimus, qui tunc temporis degebat in curia Henrici regis Angliæ filii Willelmi, et adduxerunt eum.”

The date of 1102 exactly falls in with the account of the attempt of Robert of Bellême to obtain help from Magnus in that year (see p. 442). For this I have followed the account in the Brut (1100; that is 1102). But it would seem that the Welsh writer was mistaken in saying that Magnus “sent over to Ireland, and demanded the daughter of Murchath for his son; for that person was the chiefest of the Gwyddelians; which he joyfully obtained; and he set up that son to be king in the Isle of Man.” His death is recorded in the next year, 1101 (1103), when “Magnus King of Germany” (“Vagnus vrenhin Germania”) is made to invade Britain and be killed by the Britons, who are said to have come “from the mouths of the caves in multitudes like ants in pursuit of their spoils.” Another manuscript for “Prydein” reads “Llẏchljẏn,” that is Denmark, which does not make matters much better. The followers of Magnus are called in the one manuscript “Albanians” (“yr Albanóyr”), meaning doubtless Scots; in the other manuscript they are men of Denmark (“gwyr Denmarc”). The Annales Cambriæ 624 do not mention the dealings between Robert of Bellême and Magnus; but there is an entry under 1103; “Magnus rex apud Dulin [Dublin?] occiditur.”

The death of Magnus in his second Irish expedition is told with great detail in the Saga (Johnstone, 239–244; Laing, iii. 143–147). Orderic also tells the story in p. 812. The Irish, according to this account, call in Arnulf of Montgomery to their help; but, when Magnus is killed, the Irish try to kill Arnulf and his Norman companions. Murtagh now takes away his daughter from Arnulf, and marries her, according to the irregular fashion of the country, to a kinsman (“ipsam petulantem cuidam consobrino suo illicite conjunxit”). But twenty years later, Arnulf, by that time an old man, is reconciled to Murtagh, marries his daughter, and dies the next day. This carries us beyond the range of my story, and I must leave Irish, Norwegian, and Norman enquirers to see to it. It concerns me more that it is now that Orderic mentions the great treasure which Magnus had left with a rich citizen of Lincoln. (See p. 134.) The Lincoln man seems to have thought that the death of the Norwegian king gave his banker a right to his money; but King Henry thought otherwise, and took the twenty thousand pounds to his own hoard.

The Relations between Hildebert and Helias.

There is a remarkable difference of tone between Orderic and the Biographer of the Bishops of Le Mans in their way of speaking of Helias. That the Count should be blamed for making Bishop Howel a prisoner (see p. 198) is in no way wonderful; the thing to be noticed is the way in which he several times speaks of Helias during the episcopate of Hildebert; still more remarkable is the way in which Hildebert speaks himself. Orderic always puts the acts of Helias in the best light; the Biographer, during certain parts of his story at least, seems well-pleased to throw in any little insinuation against him. Perhaps the strangest case of all is the way in which he leaves out all mention of the double appointment to the see of Le Mans on the death of Howel 625 (see p. 211), and of the action of Helias in that matter. One would have thought that, even from an ecclesiastical point of view, the story told more for Helias than against him. He put forth a claim which any other prince of his time would have equally put forward; he withdrew it in a way in which very few princes of his time would have withdrawn it. But the Biographer (see p. 297) lets us into the fact that there had been an opposition to Hildebert’s election in the Chapter itself. Could his enemies have been special partisans of Helias, and supporters of his candidate? If so, it is rather strange, though quite possible, that they should have been the accusers of Hildebert to Rufus, when the charge brought against him was that of being a confederate with Helias.

The Biographer is quite loyal to Helias during the campaign of 1098. He brings out prominently (see p. 213, note) the cause of the war, namely the election of Hildebert by the Chapter and his acceptance by the Count, without any regard to the alleged claims of the Norman Dukes. Helias was in fact fighting on behalf of Hildebert. When Helias is taken prisoner, he raises a wail—“proh dolor” (see above, p. 223)—which almost reminds us of Florence’s wail over the death-wound of Harold. He brings out strongly the Red King’s wrath against Hildebert, as shown in his ravages at Coulaines (see p. 234). He brings out also, what Orderic does not mention, the friendly relations between Hildebert and Helias which are shown in the negotiations which led to the Count’s release (see p. 238). We may perhaps infer that, during this stage, the friendship between the Count and the Bishop remained unbroken, and that the Biographer remains the Count’s friend so long as the Bishop does.

During the campaign of 1099 the Biographer’s tone becomes quite different. He has not a word to say about the zeal of the citizens of Le Mans on behalf of Helias, which comes out so strongly in Orderic, and after him in Wace (see p. 279). He rather implies that they fought against him. The enemies who meet him at Pontlieue are “milites regis cum populo” (see p. 278, note 2). It is quite possible that, as the Normans had military possession of the city, its levies may have been made, even against their will, to take their place in the Norman ranks, and the presence of such unwilling allies may have very likely helped to bring about the 626 Norman defeat. Still the insertion of the words without any comment or qualification gives the Biographer’s story a different turn from that of Orderic. Yet the Biographer himself after all allows that Helias entered Le Mans with the good-will of the citizens, when he allows (see p. 297) the accusers of Hildebert to say “quando Helias comes consentientibus civibus civitatem occupavit.” He next leaves out the fact recorded by Orderic (see p. 297) that, before William Rufus had crossed the frontier, Hildebert met him and was received to his peace, on affirming that he had no share in the enterprise of Helias. There is nothing wonderful in this. It is a case which often happens. The original cause of a war is forgotten, and the fault of the original enemy is forgiven, when a new enemy has given fresh offence. William was so wroth at Helias for seizing Le Mans, that he forgot any quarrels of earlier date. If Hildebert was clear on that score, William could pass by all that had gone before. He was therefore at this moment ready to forgive Hildebert in his wrath against Helias. But the old enemies of Hildebert in the Chapter were ready, for the sake of the old grudge, to turn against Helias. The chances are that Hildebert had nothing to do with the return of Helias, but that the towers of the cathedral were turned by Helias to military uses. Hildebert most likely deemed—​and, as events proved, more wisely than either the Count or the citizens—​that the enterprise of Helias was rash, and therefore unjustifiable. This would turn him, at least for the time, into an enemy of Helias, if not into a partisan of Rufus. The Biographer takes up this tone. It may be with a little feeling of spite that he records (see p. 281) the way in which the loyalty of the citizens towards the Count not unnaturally cooled after the fire. There is certainly such a feeling in the passage (see p. 287) where he speaks of Helias as flying, “saluti suæ consulens,” while Orderic rather describes him as swept away in a general flight. But this tone lasts only through the year 1099. When Helias comes back in 1100, all seems to be made up again; we now hear (Vet. An. 309, 311) of the “liberalitas” of the “liberalis comes;” the Normans are “hostes” and Helias brings back peace. That is to say, as the story shows, the Count and the Bishop were again reconciled, and the Biographer follows the lead of the Bishop.

But we need not wonder at the tone of the Biographer, if we 627 know the tone of the Bishop himself. In a letter printed in Duchèsne’s French collection, iv. 247, Hildebert speaks of a space of three years, “peractum triennium,” within which time Le Mans has had six counts, all of them enemies to peace (“tam modico tempore sex in urbe sustinuimus consules, quorum nullus pacificum prætendens ingressum, gladiis et igne curtam sibi vendicavit potestatem.” It is certainly very hard to reckon up six counts in three years, seemingly the years 1096–1099. In twelve years (1087–1099) not more than five counts—​William the Great, Robert, Hugh, Helias, William Rufus—​can be made out, unless Helias, with his two reigns, is reckoned twice over. Hildebert then goes on;

“Plebe coacta in favorem, tyrannum suscepit ex necessitate, non ducem ex lege: in susceptum studia simulavit, non exhibuit. Fidem reperit in ea, quia superior. Consul vero tanto gravius dominatus est quanto brevius. Miles ejus simulatis usus injuriis, eos scelerum judicavit expertes quos rerum. Et quia non parcit populis regnum breve, finem rapinis inopia posuit, non voluntas.”

This certainly reads most like a description of the reign of Hugh; but in what follows we surely see the events of 1098 and 1099;

“Ea clades usque ad sanctuarium Domini pervagata est, et primo quidquid extra muros nostræ fuerat potestatis, vel evanuit in favillas vel dissipatum est in rapinam. Deinde similibus cecidere præjudiciis episcopales domus et ecclesiæ non paucæ. In reliquis quibus ignis pepercit æque periclitata est et facultas pauperum et reverentia sacerdotum. Omnia confracta sunt, omnia direpta, omnia contaminata. Nihil eorum manus evasit qui gratis ad flagitium discurrunt, ad honestum nec pretio.”

To what does all this refer? It reads most like a description of the Red King’s harryings at Coulaines in 1098 (see p. 234); but no one is mentioned, whereas the “Rex Anglicus” and his “tyrannis” are openly spoken of further on in the letter. And it is strange, if in all this there is no reference to the fire of 1099. Did Hildebert attribute the fire to Helias, and does that account for any enmity towards him? Yet the version of the Biographer as clearly makes the fire the work of the Normans as the version of Orderic. Helias is not mentioned by name, nor is any recorded act of his distinctly mentioned. The passage is obscure, most 628 likely purposely obscure. It might be so construed as to attribute all mischief to Helias; it might be so construed as not to lay any particular act to his charge. But in any case Helias would at least come under the general condemnation which is pronounced upon all the counts of Maine, be they six or fewer. No friend of Helias could have so spoken; and it is plain that, when Bishop Hildebert wrote the letter, he was—​very naturally—​not a little angry, if not with Helias in particular, yet at least with a class of men among whom Helias must be reckoned.

Of the rest of the letter I shall have to speak in another Note.

The Surrender of Le Mans to William Rufus.

It is not very easy at first sight to reconcile our accounts of the negotiations which led to the surrender of Le Mans in August 1098. Yet there seems to be no direct contradiction of any moment. It seems not impossible that the difference is merely one of those cases where one writer gives prominence to some feature in the story which another writer leaves out.

According to all accounts, Le Mans was at this time in the possession of Fulk of Anjou. Orderic (see p. 237) makes him personally present in the city; the Biographer of the Bishops does not say whether he was there or not. But in any case the city had admitted his authority in May and had not yet thrown it off. Fulk was therefore fully in a position to negotiate with William, while Helias, who was a prisoner in William’s hands, was not strictly in a position to negotiate with anybody. Yet the Biographer makes no mention of Fulk as an actor or a party to the treaty, but only as one of whose devices Helias was afraid. In his version Bishop Hildebert and some of the chief men of Le Mans first, by the King’s leave, visit the captive Count, and agree on terms with him; then they draw up a treaty with the King according to those terms. The tale runs thus (Vet. An. 306);

“Helias timens ne Fulco comes proscriptioni ejus intenderet, manduvit ad se episcopum et quosdam ex primoribus civitatis ex 629 consensu regis, et cœpit agere cum eis, eosque suppliciter deprecari, quatenus casibus illius condolentes, modis omnibus niterentur, qualiter civitatem regi traderent, ipsumque a vinculis liberarent. Timebat enim quod Fulco comes, regis deceptus muneribus, cum eo pacem faceret, atque civitate tradita perpetuo damnaretur exsilio. Episcopus autem et qui cum eo venerant, ejus angustias miserantes, cum rege de ejus liberatione locuti, cum eo tale pactum fecerunt, ut si eorum consilio atque ingenio sibi civitas traderetur, ipse Heliam comitem quietum et liberum abire permitteret.”

He adds, hurrying matters a little; “Quod negotium industria præsulis celerius quam sperabatur effectum, eodemque tempore et regi civitas et consuli abeundi libertas reddita est.”

Orderic, on the other hand (772 D), has a version in which there is no mention of any dealings with Helias, but which makes William and Fulk—​the latter, it would seem, under some pressure—​agree on terms substantially the same as those stated in the other account. His version runs thus;

“Andegavenses autem cum Cenomannis consiliati sunt, et sese Normannis in omnibus inferiores compererunt, unde colloquium inter regem et consulem procuraverunt. Ibi tunc, auxiliante Deo, necessaria pax inter eos facta est, et inde multis pro pluribus causis utriusque populi gaudium ingens exortum est. Requisitum est et concessum ut Helias comes et omnes qui capti fuerant ex utraque parte redderentur, et Cenomannis et omnia castra quæ Guillelmus rex habuerat Rufo filio ejus subjugarentur.”

The joy of which Orderic speaks clearly did not extend to Angers. The Chronicle of Saint Albinus (1098) puts things in quite another light; “Quam [Cenomanniam urbem] tribus mensibus retentam, Cenomanensibus, more suo, sibi fraudantibus et a se deficientibus, reddidit eam in amicitia præfato regi Anglorum, qui ipsam urbem magis pecunia quam viribus impugnabat jamque pene possidebat.”

Here we have no mention of Helias or of any dealings with him, nothing of any agreement between Fulk and William. The citizens of Le Mans fall away from the Angevin Count and betray their city to the King. And they fall away through the temptation which the Red King knew well how to bring to bear upon his other enemies, but of which there is no recorded instance in the whole history of the war of Maine. See p. 290. 630

The tone and effect of these stories is very different, and yet they seem quite capable of being put together. It is simply that each writer enlarges on the persons and things which he cares most about. The Biographer of the Bishops of course enlarges on the part taken by Hildebert; next to Hildebert, he has to tell of Helias. A mission of Hildebert to Helias was a thing which he could not leave out; the fact that the terms were settled between his own Bishop and his own Count was more interesting to him than the fact that those terms were put in the form of a formal treaty between two foreign princes. He cannot leave out the Norman king, but he can and does leave out the Angevin count. He speaks of a treaty between William and Fulk as a thing which was likely to happen; he leaves out the fact that it actually did happen. The Angevin Chronicler is angry at the loss of Le Mans, and is glad to speak of its loss as due altogether to Cenomannian treason or fickleness. Orderic alone, who is, more strictly than either of the others, telling the history of the campaign, and who is less influenced by local passion one way or another, brings out the diplomatic fact that the treaty was formally agreed to in a meeting between King William and Count Fulk. It must have taken the shape of an agreement of some kind between them, unless Fulk and his troops had been driven out of Le Mans by force. But this in no way shuts out the possibility of the dealings between Hildebert and Helias which are described by the Biographer. The state of things would seem to be this. The people of Le Mans, tired of Fulk, unable to have Helias, think that the best thing is to submit to William, but on terms which will secure at least the personal freedom of their native prince. Hildebert and his companions are allowed by William to confer with Helias. The results of the conference are put into the shape of a treaty between William and Fulk. Fulk is in no condition to resist William and the Cenomannian people together; he therefore accepts the treaty, doubtless against his will. Thus the accounts of Orderic and the Biographer seem simply to fill up gaps in one another. The Angevin chronicler simply gives a short and snarling summary of the actual result.

631

The Fortresses of Le Mans.

A great deal about the walls and the castle of Le Mans, as well as about several other points in the county of Maine, will be found in M. Hucher’s book, Études sur l’Histoire et les Monuments du Département de la Sarthe (Le Mans and Paris, 1856). M. Hucher however hardly carries his researches beyond the city itself; so that, while his remarks and the documents which he quotes tell us much about the “regia turris,” the castle close to the cathedral, he has but little to tell us about the fortress of Mont-Barbé, which is for our purpose of at least equal interest.

I have quoted elsewhere (N. C. iii. 207) some of the passages which record the building of at least two castles by the Conqueror, the royal tower and that of Mont-Barbé. In the extract from William of Jumièges for “ponte Barbato” we must read “monte.” Benoît, oddly enough, knew the name of Mont-Barbé, but did not know that of the royal tower (35735);

Wace, on the other hand (15014), in his wild chronology of all Cenomannian matters, makes William Rufus build this castle in the expedition of 1099;

But this story, though utterly out of its place, may possibly preserve a fact. The royal tower was undoubtedly built by the Conqueror after he had taken Le Mans in 1063 in order to secure the possession of the city. But Mont Barbé looks rather like one of the besieging castles made in order to get possession. Nothing is now left but the mound. William may conceivably have found this mound ready made. If not, his building of 1063 must have been of wood, though it may very likely have had a stone successor. The mound, not far from Saint Vincent’s abbey, stands in a private garden, and the visitor to Le Mans, unless he has local guidance, may very likely fail to 632 find it. I missed it at my first visit in 1868, which must be my excuse for the rather vague language in the third volume of the Norman Conquest. I saw it for the first time in 1876, through the kindness of M. Henri Chardon, and again in 1879 with Mr. Parker and Mr. Fowler.

The question remains, Was there a Mons Barbatulus as well as a Mons Barbatus? The passages quoted from Orderic and William of Jumièges (N. C. vol. iii. p. 207) seem to imply it; only the odd thing is that the words of William of Jumièges seem to leave out the royal tower, and to speak of Barbatus and Barbatulus only. And one might take the words of Wace, “La mote devant Barbé,” to mean Barbatulus rather than Barbatus; only it would be hard to find another mota. Barbatulus is conjecturally, but with every likelihood, placed on the site of the present Lyceum, between Barbatus and the city.

The royal tower was built just outside the Roman wall, two of whose bastions, known as La Tour Margot—after Margaret, the promised bride of Robert?—and La Tour du Cavalier, were taken into its precinct. All these must be distinguished from the palace of the Counts (see N. C. vol. iii. p. 205) which stands on the Roman wall, almost in a line with the east end of the cathedral. It contains a window of the twelfth century, of great width, a feature characteristic of Le Mans. In this palace was the sainte chapelle of the Counts.

The Dates of the Building of Le Mans Cathedral.

I have more than once, in the History of the Norman Conquest, had to speak of the dates of the various parts of the church of Saint Julian at Le Mans. The subject is so closely connected with so many names which appear in our story that an inquiry of this kind can hardly be thought out of place. My later visits to Le Mans have enabled me to examine and consider several points again; and I am now inclined to think that there is very little, if anything, standing in the present church of an earlier date than William the Conqueror’s first taking of Le Mans in 1063. I have got some 633 help from a local book, called “Recherches sur la Cathédrale du Mans. Par L’Abbé….” (Le Mans, 1872); but its architectural criticism is not of a high order. Another local book, “L’Ancien Chapitre Cathédral du Mans, par Armand Bellée, Archiviste de la Sarthe” (Le Mans, 1875), is a very thorough piece of capitular history, but it throws little light on the architecture.

The earliest church of which we have any certain account was a basilica of the ninth century. Saint Aldric, bishop from 832 to 856, rebuilt the cathedral church, of which he consecrated the eastern part in 834 and the rest in 835. I have for these dates to trust the author of the “Recherches sur la Cathédrale du Mans,” who quotes from a manuscript life of Aldric in the library at Le Mans. (I have seen the volume, and I could wish that it was in print.) The time allowed for the building is wonderfully short; but Aldric, if he did all that is attributed to him by the Biographer of the Bishops (Vet. An. 276), must have been a man of wonderful energy. There is nothing said directly of his works at Saint Julian’s; but they might almost be taken for granted when we hear of the many churches which he built and restored (“Ædificia quæ prædictus pontifex multipliciter a novo operatus est, et ecclesias sive nonnulla monasteria quæ a novo fundavit atque perficere et ornare studuit, necnon et restaurationes aliorum monasteriorum et ceterarum ecclesiarum,” &c. &c. &c). In the days of the next Bishop Robert (856–885) Le Mans was sacked by the Northmen and the church burned. We are of course met by the usual difficulty as to the amount of destruction which is implied in words of this kind; but it naturally led to a restoration, and to a new dedication, on which last point however it seems to have been thought needful to consult the Pope (“Matrem ecclesiam, a paganis incensam, diligenti studio renovavit, et ex consilio Romani antistitis jam denuo celeberrime consecravit;” 287*). We hear again (296*) of a dedication under Bishop Mainard (940–960); but not of any rebuilding, just as in some of the intermediate episcopates (Vet. An. 288* et seqq.) we hear a good deal about havoc and desecration, but nothing about actual destruction. The church of Aldric, allowing for the restorations of Robert and any later repairs, seems plainly to have stood till the days of Vulgrin (1055–1067), the earliest Bishop of Le Mans who has even an indirect share in the building of the present church. No work of his, unless 634 possibly the merest fragments, seems to be now standing; but he was the beginner of a great work of rebuilding which gave us what we now see.

In the Life of Vulgrin (Vet. An. 312*) we are simply told that in 1060 he began the foundations of a new church on a greater scale (“Quinto ordinationis suæ anno fundamenta matris ecclesiæ ampliora quam fuerant, inchoavit, sed morte inopina superveniente perficere non potuit”). His foundations were badly laid and his work was unskilful; so that, while attempts were making under his successor Arnold (1067–1082) to prop it up, it fell down. Arnold accordingly destroyed the whole work of Vulgrin, and began again from a new foundation. The extent of his work is clearly marked. He finished the eastern limb, as far as its walls and outer roof were concerned; its internal adornments he left for his successor. Of the transepts with their towers he merely laid the foundations;

“Fabrica novæ ecclesiæ quam præsul Vulgrinus inchoaverat, fundamentorum mobilitate atque lapidum debilitate corrupta, innumera crepidine ruinam suam cœpit terribiliter minitari; quam dum artifices fulcire conantur, repentino fragore nocturno tempore collapsa est…. Inde … episcopus totam cœpti operis fabricam usque ad ima fundamenta destruens, denuo ipsam ecclesiam fundamento firmiori et solidiori lapide construere cœpit, et parti superiori quæ vulgo cancellum nominatur etiam tectum imposuit, membrorum quoque quæ cruces vocantur atque turrium solidissima fundamenta antequam moreretur instituens” (313*).

That he added only the outer roof is plain from what we read of his successor Howel (Vet. An. 289). As Howel adorned the “cancellum” with a pavement and stained glass windows, he also added a painted ceiling;

“Cancellum quod ejus antecessor construxerat pavimento decoravit et cœlo, vitreas quoque per ipsum cancellum, per quod cruces circum quoque laudabili sed sumptuosa nimium artis varietate disponens.”

So again, p. 299;

“Cœpit … superiores partes ejusdem basilicæ diligenti sollicitudine laborare, oratorium scilicet quod chorum vocitant sedemque pontificalem, altaria congrua dimensione disponere, pavimenta substernere, columnas ac laquearia gratissima varietate depingere, parietes per circuitum dealbare.”

635 Howel also finished the transepts and towers of which Arnold had merely laid the foundations (Vet. An. 289);

“Fabricam novæ ecclesiæ … tanto studio aggressus est consummare ut cruces atque turres, quarum antecessor ipsius … jecerat fundamenta brevi tempore ad effectum perduxit.”

We see then what the work of Vulgrin and Arnold was. It touched the eastern part only; Aldric’s nave was left alone. The original church was a basilica, most likely with three apses, but without transepts. The new design was to rebuild the eastern part on a greater scale with transepts, transept towers (like Geneva and Exeter), and a choir ending in an apse with a surrounding aisle and chapels—​as is shown by the mention of many altars. The arrangement was that of the two other great churches of Le Mans, La Couture and Saint Julian in the Meadow, with the single exception of the towers, which do not appear in either of those churches. Arnold built the choir, and began the transepts and towers; Howel adorned the choir and finished the transepts and towers. There is nothing to imply that either of them touched the nave. The arcades of Aldric’s basilica were therefore still standing when William the Great came in 1063 and again in 1073. The work of Vulgrin in the eastern part was doubtless going on at the earlier of those two dates, and that of Arnold at the later.

It must be plain to every one who has seen the building that the work of these bishops in the eastern part of the church has given way to the later choir and transepts. The choir was built between 1218 and 1254, and its great extension to the east involved, as at Lincoln, the destruction of part of the Roman wall. The transepts were built at several times from 1303 to 1424. They are among the very noblest works of the architecture of those centuries; but we may be allowed to rejoice that, as the works of Vulgrin and Arnold left Aldric’s nave standing, so the great works of the thirteenth century and later have left the nave which succeeded that of Aldric. With all its artistic loveliness, the work of the later day cannot share the historic interest of the works of the times of William and Howel, of Helias and Hildebert.

In the present nave it is plain at the first glance that there are two dates of Romanesque; a further examination may perhaps lead to the belief that there are more than two. It is easy to see outside that the aisles and the clerestory are of different dates. The 636 masonry of the aisles is of that Roman type which, in places like Le Mans, where Roman models were abundant, remained in use far into the middle ages, and which in some places can hardly be said to have ever gone out of use at all. The masonry of the clerestory is ashlar. The difference is equally clear between the plain single windows of the aisles and the highly finished coupled windows of the clerestory. Inside, the eye soon sees that the design has undergone a singular change. Without the pulling down of any part, the church put on a new character. Columns supporting round arches after the manner of a basilica were changed into a series of alternate columns and square piers supporting obtusely pointed arches. Each pair of arches therefore forms a couplet, and answers to a single bay of the pointed vaulting and a single pair of windows in the clerestory. The object clearly was to give the building as nearly the air of an Angevin nave, like that of La Couture (see N. C. vol. v. p. 619), as could be given where there were real piers and arches. Now this reconstruction, one which brings in the pointed arch, cannot possibly be earlier than the episcopate of William of Passavant, Bishop from 1143 to 1187. He was a great builder; he translated the body of Saint Julian (Vet. An. 366); he celebrated a dedication of the church (Ib. 370), which my local book fixes, seemingly from manuscripts, to 1158, a date a little early perhaps for such advanced work, but not impossible. To William of Passavant then we must attribute the recasting of the nave, and whatever else seems to be of the same date. To this last head belongs the great south porch, and, I should be inclined to add, the lower part of the southern, the only remaining, tower, though some assign it to Hildebert. The question now comes, What was the nave which William of Passavant recast in this fashion, and whose work was it?

We have seen that we cannot attribute any work in the nave to any prelate earlier than Howel. He must have found the nave of the ninth century still standing. Did he do anything in that part of the church? He performed a ceremony of dedication in 1093 (Vet. An. 300); but that would be fully accounted for by his works in the eastern part. On the other hand, Hildebert celebrated in 1120 (Vet. An. 320) a specially solemn dedication, and the words used seem to imply that the church was now complete in all its parts. The words of Orderic (531 D) seem 637 express. Howel began to build the church (“episcopalem basilicam … condere cœpit”); Hildebert finished it (“basilicam episcopii quam prædecessor ejus inchoaverat, consummavit, et cum ingenti populorum tripudio veneranter dedicavit”). It is doubtless not strictly true that Howel began the church, words which shut out the work of Vulgrin and Arnold; but the time when Orderic wrote makes him a better authority for Hildebert’s finishing than for Howel’s beginning, and the expression might easily be used if Howel began that particular work, namely the nave, which Hildebert finished. I do not think that we need infer from certain expressions of the Biographer that Hildebert left the nave, or any essential part of the building, unfinished. He says indeed (Vet. An. 320);

“Hildebertus opus ecclesiæ, quod per longa tempora protractum fuerat, suo tempore insistens consummare, dedicationem ultra quam res exposcebat accelerans, multa inibi necessaria inexpleta præteriit.”

Comparing this with the words of Orderic, this surely need not mean more than that, though the fabric was perfect, yet much of the ornamental work was left unfinished. Hildebert, in short, left the nave much as Arnold left the choir. At least the nave was in this case when he dedicated the church. For he had time after the dedication to make good anything that was imperfect.

We should then infer from Orderic that the nave which William of Passavant recast was begun by Howel and finished by Hildebert. This may give us the key to a passage in the Biographer on which we might otherwise be inclined to put another meaning. After describing Howel’s building of the transepts in the words quoted above in p. 635, he goes on (289);

“Eisque [crucibus] celeriter culmen imponens, exteriores etiam parietes, quos alas vocant, per circuitum consummavit.”

One might have been tempted to take this of transept aisles; but, weighing one thing with another, it seems to be best understood as meaning that Howel rebuilt the whole of the outer walls of the nave and its aisles. This would give to him the whole extent of the quasi-Roman work of the aisles, together with the great western doorway. The interior work of the aisles seems also to agree with his date. We must therefore suppose that Howel rebuilt the nave aisles only, still leaving the arches of Aldric’s basilica. Then Hildebert rebuilt or thoroughly restored the nave 638 itself, with the columns and arches and whatever they carried in the way of triforium and clerestory. We may therefore suppose that the existing columns, as distinguished from the square piers, are his work, though the splendid capitals of many of them must have been added or carved out of the block in the recasting by William of Passavant.

There is however one fragment of the nave arcades which is older than Hildebert, very likely older than Howel. This is to be seen in the first pier from the east. I need not say that the eastern bay of a nave often belongs to an older work than the rest, being in truth part of the eastern limb continued so far—​perhaps for constructive reasons, to act as a buttress—​perhaps for ritual reasons, to mark the ritual choir—​very often for both reasons combined. One of the best examples is that small part of the nave of Durham abbey which belongs to the work of William of Saint-Calais (see N. C. vol. v. p. 631). At this point then in the nave of Le Mans, we find half columns with capitals and bases of a strangely rude kind, more like Primitive Romanesque (see N. C. vol. v. pp. 613, 618, 628) than anything either Norman or Angevin. These are assuredly not the work of Hildebert. There is one argument for assigning them to Howel, namely that something of the same kind is to be found in the remains of the northern tower of which I shall speak in another Note (see below, Note RR). But if any one holds them to be the work of Arnold or of Vulgrin, or even looks on them as a surviving fragment of the basilica of the days of Lewis the Pious, I shall not dispute against him.

I must add however that, between Hildebert and William of Passavant, we have, according to the use of Le Mans, to account for two fires—“solita civitatis incendia,” as the Biographer (Vet. An. 349) calls them—​and their consequences. In 1134 there was a fire which, according to the Biographer (350), was more fearful than any which had ever happened at Le Mans since the city was built, not even excepting the great one of 1098. Everything perished. “Tota Cenomannensis civitas cum omnibus ecclesiis quæ intra muros continebantur, evanuit in favillas.” We read of the “matris ecclesiæ destructio” and “combustio,” all the more lamentable because of its beauty—“ipsa enim tam venustate sui quam claritate tunc temporis vicinis et remotis excellebat ecclesiis.” So Orderic (899 B); “Tunc Cenomannis episcopalis basilica, quæ 639 pulcherrima erat, concremata est.” The then Bishop, Guy of Étampes (1126–1136), spent two hundred pounds in trying to repair the damage; “Ad cujus restaurationem cc. libras Cenomannenses dedit, sine mora contulit, et omnibus modis desudavit quomodo ipsa ad perpetuitatem decenter potuisset restaurari.” Under the next Bishop, Hugh of Saint-Calais (1146–1153), there was another fire, the account of which is very curious (Vet. An. 349);

“Ignis circa meridiem a vico sancti Vincentii prosiliens, sibi opposita usque ad muros civitatis et domos episcopales, tegmenque sacelli beati Juliani adhuc stramineum, cum fenestris vitreis concremavit et macerias, et in summis imagines sculptas lapidibus deturbavit.”

The people break open the shrine of Saint Julian in order to save his body, which they carry to the place where the Bishop was. The Bishop seems to have repaired the episcopal buildings before he touched the church, and the details have some interest in the history of domestic architecture (“domum petrinam ex parte sancti Audoëni positam, decenti solariorum interpositu numerosas fenestras habentium cum sua camera continuavit”). Presently we read;

“Beatissimum patrem nostrum Julianum ipso die a lignea basilica in occidentali membro ecclesiæ intra macerias facta, post incendium in qua fere triennio requieverat, in redivivam sollenniter, clero cantibus insultante, populo congaudente, transtulerunt ecclesiam.”

We do not hear of any more building, but there is a long list (Vet. An. 354) of the ornaments which Bishop Hugh gave to the Church.

Some of the expressions used in these passages are very odd. “Sacellum beati Juliani” is a strange phrase for the cathedral church, and yet the thatched roof and the glass windows must be spoken of a building and not of a mere shrine. It is Saint Julian’s church itself whose roof and windows are spoken of. But the phrase “lignea basilica,” which makes one think of Glastonbury, must not lead us to think that any wooden church of early days was then standing at Le Mans. The whole story seems quite intelligible, without supposing any really architectural work between Hildebert and William of Passavant. The language of the Biographer in describing the fire of 1134 is, as so often happens, very 640 much exaggerated. His own account shows that the walls of the church were left standing. It looks on the whole as if the roof was destroyed in 1134. It was hastily repaired with thatch. It was burned again, and the clerestory (“fenestræ vitreæ”) with it, at the next fire in 1146–1153. The whole church perhaps remained for a while unfit for divine service. Then some wooden structure (“lignea basilica”) was raised within the walls of the nave (“in occidentali membro ecclesiæ intra macerias facta”). Meanwhile Bishop Hugh repaired the choir (“rediviva ecclesia”), seemingly doing nothing to the nave. Bishop William, finding things in this state, rebuilt the clerestory and vaulted it Angevin fashion. So to do required that every alternate column of the nave should be built up into a square pier. This again required a change in the line of the arches, and, according to the fashion just coming in, they were made obtusely pointed. If any one thinks that the superb foliage of the nave capitals must be later than 1158, he may hold that they were cut out afterwards, or he may even hold that Bishop William’s dedication in that year belongs only to the eastern parts—​where something was clearly done in his time or thereabouts—​and that the whole recasting of the nave came later in his long episcopate.

I am not writing an architectural history of the church of Saint Julian, and I have perhaps, as it is, gone more into detail than my subject called for. I think that any one who has been at Le Mans will forgive me. But there are many architectural points in this wonderful church on which I have not entered. There is much also in the other two minsters of Le Mans which throws much light on the work at Saint Julian’s. I have merely tried in a general way to assign to their most probable dates and founders the different parts of a church which so often meets us in our present history.