when a confederation of the western towns, with the great city of the district at their head, suddenly started into life to check the progress of the Conqueror, it shows that a spirit had been kindled, etc., etc.... It is worth while to stop and think how near England once was to running the same course as other lands, etc., etc.[16]
Returning now to sober fact, let us ask how the city of Exeter came into William's hands. This is the pivotal point on which the whole story revolves. On this point Mr Freeman spoke with no uncertain sound: the city was 'taken by means of a mine'.[17] It was, he wrote, 'by undermining the walls that William at last gained possession of the city', the citizens being thus forced 'to submit unreservedly'.[18] He added, contrasting the success of William with the failure, in 1003, of Swend:
William might have been beaten back from Exeter as Swend had been, if the military art of Normandy in William's days had not been many steps in advance of the military art of Denmark in the days of Swend.
This allusion to 'Swend' involves a perfect tangle of confusion. Turning back a couple of pages, we are reminded that on Penhow, 'sixty-seven years before (1001), Swend, of Denmark, driven back from the city, had found his revenge' (p. 154). Guided by a footnote, we turn for information to the earlier volume to which the author refers us, only to learn that it was not Swegen, but the adventurer Pallig who was driven back from Exeter in 1001 (i. 307), while 'of Swegen himself we hear nothing in English history for nine years (994-1003)'.[19] Moreover, when Swegen did come—in 1003—invading England to avenge the massacre of Saint Brice, he was not 'driven back from the city', but, on the contrary, 'stormed and plundered it' (p. 315), for 'the citizens who had beaten back Pallig had no chance of beating back Swegen' (Exeter, p. 27). Moreover, the suggestion that the Danes would not have been able to attack and breach the city wall is in direct conflict with the evidence quoted by Mr Freeman himself. Not only did Pallig, in 1001, direct his attack against the wall,[20] but 'Swegen', we read, in 1003, 'Civitatem Exanceastram infregit'.[21] Now, speaking of 1063, Mr Freeman wrote that 'the expression of Florence "infregit" seems to fall in with' his view that William breached the wall. That is to say that, according to Mr Freeman, 'Swend' was 'beaten back' (which he was not), because he could not breach the walls, which is precisely what, on his showing, Swegen succeeded in doing. Could confusion further go?
For his statement that 'William's mine advanced so far that part of the wall crumbled to the ground, making a practicable breach' (p. 156), Mr Freeman relied on an ingenious combination of Orderic's statement that the Conqueror 'obnixe satagit cives desuper impugnare et subtus murum suffodere' with William of Malmesbury's assertion that he triumphed 'divino scilicet adjutus auxilio, quod pars muralis ultro decidens ingressum illi patefecerit'. He argued that, on the supposition that 'Exonia' is the right reading in William of Malmesbury, his 'story, allowing for a little legendary improvement, fits so well into Orderic's as to support the theory of a breach'. The argument is ingenuous and plausible, nor can it be lightly dismissed. But whether the words of Orderic imply, of necessity, a mine or not,[22] the real point is that he does not mention a breach. He speaks of William's efforts, but he does not say they were successful. It is difficult to suppose that William of Poitiers, of whom Orderic is here the mouthpiece, would not have mentioned his hero's success, had success rewarded his efforts. We are reduced then, as the sole and unconfirmed authority for Mr Freeman's absolute statement—or rather as the legend from which he 'infers' the facts he states—to the words of William of Malmesbury. Now William was classed, by Mr Freeman himself, among those writers whose 'accounts are often mixed up with romantic details', so that 'it is dangerous to trust them' (i. 258); and he pointed out of the murder of Edward that:
In the hands of William of Malmesbury the story becomes a romance.... The obiter dictum of William of Malmesbury that Ælfhere had a hand in Edward's death is contrary to the whole tenor of the history ... (i. 265).
If there is thus, on Mr Freeman's showing, need for accepting with some caution a statement made by William alone, there is further, in this special case, the consideration that even if his story does refer to Exeter, the phrase, 'leviter subegit' is justly queried by Mr Freeman;[23] and that William here deals in hyperbole and miracle. Indeed, when we find Mr Freeman writing: 'I infer this from William of Malmesbury', we are reminded of his words on his predecessor's treatment of the legend of Siward: 'Such stuff would not be worth mentioning, had not Sir Francis Palgrave inferred from it the existence of an historical Tostig, Earl of Huntingdon' (iv. 768-9). I will not express an opinion of my own, but will quote from Mr Freeman's able essay on 'The Mythical and Romantic Elements in Early English History'.[24] In it he expressly disclaimed
sympathy with the old pragmatizing or euhemeristic school of mythological interpretation.... The pragmatizers take a mythical story; they strip it by an arbitrary process of whatever seems impossible; they explain or allegorize miraculous details; and having thus obtained something which possibly may have happened, they give it out as something which actually did happen.... It will never do to take the tale of Troy, to leave out all intervention of the gods, and to give out the remnant as a piece of real Grecian history (p. 3).
This criticism would seem to apply to the 'legendary' tale that the walls of Exeter fell down, like those of Jericho, by supernatural intervention. At least, we may say of the breaching of the walls, when given out 'as something which actually did happen', what was said of the possible siege of Oxford, this same year, by Mr Freeman:
The direct evidence for a siege of Oxford is so weak that the tale cannot be relied on with any certainty (iv. 188).