[82] Norm. Conq., v. 251.
[83] "In a later stage, when the son of his rival was firm on the throne, the doctrine of female succession took root under a king who by the spindle-side sprang from both William and Cerdic, but who by the spear-side had nothing to do with either. Then it was that men began to find out that Stephen had been guilty not only of breaking his oath, but also of defrauding the heir to the crown of her lawful right" (ibid., p. 252).
[84] "Henrici regis filia, ... vehementer exhilarata utpote regnum sibi juratum ... jam adepta" (Cont. Flor. Wig., 130). But the above duplex character of her claim is best brought out in her formal request that the legate should receive her "tanquam regis Henrici filiam et cui omnis Anglia et Normannia jurata esset."
[85] "Conventio optimatum et baronum totius Angliæ apud Salesbyriam xiv. kalend. Aprilis facta est, qui in præsentiâ regis Henrici homagium filio suo Willelmo fecerunt, et fidelitatem ei juraverunt" (Flor. Wig., ii. 69).
[86] "Normanniæ principes, jubente rege, filio suo Willelmo jam tunc xviii. annorum, hominium faciunt, et fidelitatis securitatem sacramentis affirmant" (Sym. Dun., ii. 258).
[87] Oddly enough, the correct date must be sought from Symeon of Durham, though, at first sight, he is the most inaccurate, as he places the event under 1128 (a date accepted, in the margin, by his editor) instead of 1126, the year given by the other chroniclers. But from him we learn that the Christmas court (i.e. Christmas 1126) was adjourned from Windsor to London, for the new year, "ubi Circumcisione Domini" (January 1) the actual oath was taken. William of Malmesbury dates it, loosely, at Christmas (1126), but the Continuator of Florence, more accurately, "finitis diebus festivioribus" (p. 84), which confirms Symeon's statement.
[88] It is scarcely realized so clearly as it should be that the oath taken on this occasion was that to which reference was always made. Dr. Stubbs (Const. Hist., i. 341) recognizes "a similar oath in 1131" (on the authority of William of Malmesbury), and another in 1133 (on the authority of Roger of Hoveden). But the former is only incidentally mentioned, and is neither alluded to elsewhere, nor referred to subsequently by William himself; and the latter, which is similarly devoid of any contemporary confirmation, is represented as securing the succession, not to Matilda, but to her son. It is strange that so recent and important an oath as this, if it was really taken, should have been ignored in the controversy under Stephen, and the earlier oath, described above, alone appealed to.
[89] Henry of Huntingdon merely alludes to it, retrospectively, at Stephen's accession, as the "sacramentum fidelitatis Anglici regni filiæ regis Henrici" (p. 256).
[90] "Fecit principes et potentes adjurare eidem filiæ suæ et heredibus suis legitimis regnum Angliæ" (i. 93). This is, perhaps, somewhat confirmed by the words which the author of the Gesta places in the primate's mouth (p. 7).
[91] "In filiam suam, sororem scilicet Willelmi, ... regni jura transferebat" (p. 85). The oath to secure her this succession was taken "ad jussum regis" (p. 84). Compare with this expression that of Gervase above, and that (quantum valeat) of Roger Hoveden, viz. "constituit eum regem;" also the "jubente rege" of Symeon in 1120. It was accordingly urged, at Stephen's accession, that the oath had been compulsory, and was therefore invalid.