The barons promptly stormed the ‘outer bailey’ of the castle (April 19),[117] and strove desperately to gain the keep, till, a week later, they fled suddenly at the news of the king’s advance on London.[118] But so vigorous were the siege operations by attack, battery, and mining, that they were on the point of succeeding when they had to raise the siege.[119]
Surely a ‘History of the Art of War’ should mention the above remarkable allusion to Simon’s mastery of siege operations, and to his teaching the English, who were then ignorant of the subject. But all that Mr. Oman tells us is that—
the massive strength of Gundulf’s Norman keep was too much for such siege appliances as the earl could employ. The garrison under John de Warenne, the Earl of Surrey, held their own without difficulty (p. 416).
We have seen that, on the contrary, the keep was on the point of being taken. But what are we to say to the words, “Gundulf’s Norman keep”? “It was long the custom,” as Mr. Clark wrote, “to attribute this keep to Gundulf, making it contemporaneous, or nearly so, with the Tower of London”; but, more than thirty years ago, it was shown by Mr. Hartshorne (in the ‘Archæological Journal’) that it was built in later days under William of Corbeuil (1126–1136).[120] No one, in the present state of our knowledge, could suppose that Gundulf was its builder; and it is obvious that a writer who does must have yet everything to learn on Norman military architecture.
I must lastly deal as briefly as possible with the subject of knight service. The view of modern historians has been that this was gradually evolved during the Norman period out of a pre-conquestual obligation to provide one armed man for every five hides held. As against this I have advanced the theory[121] that the whole arrangement was introduced de novo at the Conquest, when the Conqueror assessed the fiefs he granted in terms of the five-knight unit irrespective of hidation. Put in a less technical form my theory is that the Conqueror called on the holder of every considerable fief to furnish a contingent of five knights, or some multiple of five, to the feudal host.[122] And this he did arbitrarily, without reckoning the ‘hides’ that might be contained in the fief. Further, by the argumentum ad absurdum, I showed that if every five hides had to provide a knight, there would be nothing, or less than nothing, left for the tenant-in-chief.[123] It was of this new theory that Professors Pollock and Maitland observe, in their history of English Law (i. 238–9), that they regard it “as having been proved by Mr. Round’s convincing papers.”
Mr. Oman, however, leans to the now exploded theory, and holds that under Norman rule “the old notion that the five hides must provide a fully armed man was remembered;”[124] and that though “some lay tenants-in-chief” got off easily, “the majority were obliged to supply their proper contingent.”[125] He then proceeds:
It has been clearly shown of late, by an eminent inquirer into early English antiquities, that the hidage of the townships was very roughly assessed, and that the compilers of Domesday Book incline towards round numbers.
Now apart from the fact that this “eminent inquirer,” my friend Professor Maitland to wit, gives me full credit for having been first in the field[126]—a fact which Mr. Oman, with my book before him, of course carefully ignores—his words show that he cannot understand the simplest historical theory. Professor Maitland and I have dwelt on the antiquity of this assessment, with which “the compilers of Domesday Book” had no more to do than Mr. Oman himself, and which indeed the compilation of that book has almost utterly obscured.
From the fact of the five-hide unit Mr. Oman argues “that there was little difficulty in apportioning the military service due from the tenants-in-chief who owned them,”[127] though such apportionment, as I have shown, would result in an actual absurdity.[128] Indeed, Mr. Oman himself observes that the tenant-in-chief, to discharge his obligation, “might distribute the bulk of his estate in lots roughly averaging five hides to subtenants, who would discharge the service for him,”[129] although a moment’s consideration will show that this process would absorb not “the bulk,” but the whole of his estate.