From this flight the Normans were rallied by the desperate efforts of the Duke himself, who, as is usual at such moments, was believed to have fallen. I deem this episode a fixed point, and it conveniently divides the battle. All our four leading authorities—the Tapestry, William of Poitiers, Bishop Guy, and Baudri—are here in complete agreement. William describes the Duke as 'nudato insuper capite'; Guy tells us that 'iratus galea nudat et ipse caput'; Baudri writes 'subito galeam submovet a capite'; in the Tapestry, 'William (writes Dr Bruce), when he wishes to show himself in order to contradict the rumour that he has been killed, is obliged to lift his helmet almost off his head' (p. 98). It is singular that so striking and well-established an episode is wholly ignored by Wace.

THE FOSSE DISASTER

The serious character of the assailants' flight is duly recognized by Mr Freeman.[120] We could have no more eloquent witness to the fact than the admission even by William of Poitiers that the Duke's Normans themselves gave way, or the description of them by Bishop Guy as 'gens sua victa'. The only point in question here is whether what I call 'the fosse disaster' was an incident of this headlong flight or happened at a later stage of the battle. Mr Freeman, discussing 'the order of events',[121] faced the difficulty frankly, observing that Guy had placed the feigned flight before what I have termed above the dividing incident of the day, and that this view 'may be thought to be confirmed by the Tapestry', etc., etc. We have here perhaps the most difficult problem raised in the course of the battle, and one which it would be easier and safer to pass over in silence. As to Guy, I suggest, as a possible solution—it does not profess to be more—that what he was describing was not the great feigned flight but the lesser manœuvres of the same character described by Baudri above. He may, of course, have transferred to these the importance of the later episode. On the real flight, at least, he is sound. Of the Tapestry I would speak with more confidence. 'In the nature of things,' Mr Freeman wrote, 'exact chronological order is not its strongest point' (p. 768). But in this case there was nothing to make it depart from that order, no reason why it should not place the incident of 'the fosse disaster' after the central incident of the day, instead of before, if that were its right position. Moreover, it is here, we find, in the closest agreement with Wace; and though I claim, as did Mr Freeman, the right of rejecting his testimony when wholly unsupported (as still more, when opposed to probability), yet such marked agreement as this is not to be lightly cast aside.

In any case, nothing can be more unfortunate than Mr Freeman's treatment of what he describes as the 'great slaughter of the French in the western ravine' (p. 489). This is a scene invented by Mr Freeman alone, and illustrates the peculiar use he made, at times, of his authorities. There is no question that the Norman knights suffered, in the course of the day, at least one such disaster as the nobles of France at Courtrai (1302) or her cuirassiers at Waterloo. But five authorities, so far as one can see, place the incident in the thick of the battle, while three others assign it to the pursuit of the defeated English. It is not strange, therefore, that some writers should have held that there was but one such incident: Mr Freeman, however, holds that there were two; and I expressly disclaim questioning his view, the matter being one of opinion. Assuming then, as he does, that the episode occurred in the course of the battle, I turn to the spirited version of Wace, as Mr Archer defies me to 'impeach Wace's authority' (p. 346). The 'old Norman poet' is here very precise. He first tells us (ll. 7869-70, 8103-6) that the English had made a 'fosse', which the Normans had passed unnoticed in their advance.[122] These passages Mr Freeman accepts without question (p. 476). But then Wace proceeds to state (ll. 8107-20) that the Normans, driven back, as we have seen, by the English, tumbled, men and horses, into this treacherous 'fosse' and perished in great numbers. Now Wace, far from standing alone, is here in curiously close agreement with the Tapestry of Bayeux. Two successive scenes in that 'most authetic record' are styled 'Hic ceciderunt simul Angli et Franci in prœlio; hic Odo episcopus baculum tenens confortat pueros.' Wace describes these scenes in thirty-six lines (ll. 8103-38), devoting eighteen lines to the first and the same number to the second. Actual comparison alone can show how close the agreement is. Henry of Huntingdon, we may add, independently confirms the statement that English as well as French perished in the fatal fosse.[123]

Now all this is quite opposed to Mr Freeman's 'conception of the battle'. He had, therefore, to adapt, with no gentle hands, his authorities to his requirements. Cinderella's stepmother, when her daughter's foot could not be got into the golden shoe, armed herself, we read, with axe and scissors, and trimmed it to the requisite shape. With no less decision the late Professor set about his own task. Wace's evidence he simply suppressed; Henry of Huntingdon's he ignored; but that of the Bayeux Tapestry could not be so easily disposed of. I invite particular attention to his treatment of this, his 'highest authority'. Retaining in its natural place (pp. 481-2) the second of the two scenes we have described, he threw forward the one preceding it to a later stage of the battle (p. 490). Nor did his vigorous adaptation stop even here. The scene thus wrenched from its place depicts a single incident: mounted Normans are tumbling headlong into a ditch at the foot of a mound, on which 'light-armed' English stand assailing them with their weapons. The fight is hand to hand; the bodies touch. And yet the Professor treats this scene as a description of two quite separate events happening at a distance from each other. These he terms (p. 489) the 'stand of the English at the detached hill'; and the 'great slaughter of the French in the western ravine'. But on referring to his own ground-plan, we find that this 'ravine' and the 'detached hill' were a quarter of a mile apart, with the slopes of the main hill between them.

My criticism here is twofold. In the first place, Mr Freeman endeavoured to conceal the liberties he had taken with his leading authority. No one would gather from his narrative of the battle that any such violence had been used; nor would anyone who read of the 'hill' episode that 'the scene is vividly shown in the Tapestry' (p. 489), and, subsequently, of the 'ravine' disaster, that 'this scene is most vividly shown in the Tapestry' (p. 490), imagine that 'the incidents of the ravine and the little hill' (p. 768) are in the Tapestry one and the same. In the second place, the large part which the writer's own imagination plays in his narrative of the fight is here clearly seen. There is nothing, for instance, in any authority to connect 'the western ravine' with 'the great slaughter of the French'. It is placed by those who mention it in a 'fosse', 'fossatum', or 'fovea'. 'If Wace is any authority,' to quote Mr Archer's words, 'the question is settled once and for all';[124] the slaughter took place not in the 'ravine', but in a ditch which according to him, the English had dug to the south of the hill, and which, according to Henry of Huntingdon, they had cunningly concealed. Mr Freeman produces no authority in support of his own fancy; his only argument is that the slaughter

must have happened somewhere to the south or south-west of the hill. The small ravine to the south-west seems exactly what is wanted (p. 771).

The 'western ravine' however, does not fulfil these requirements (see ground-plan, where it lies to the north-west of the hill); while Wace's 'fosse', which—though here ignoring it—he had already accepted, lay, as required, to the south of the hill. Wace mentions another instance (ll. 1737-50) in which this stratagem was adopted,[125] but whether our ditch was dug, as he states, expressly or not, the fact of its existence does not depend on his evidence alone.