The Danes had withdrawn from the region of the Humber, but they had not left the country. In the Isle of Ely, then more nearly an actual island than in modern times, was a bit of unsubdued England, and there they landed for a time. In this position, surrounded by fens and interlacing rivers, accessible at only a few points, occurred the last resistance which gave the Normans any trouble. The rich mythology which found its starting-point in this resistance, and especially in its leader, Hereward, we no longer mistake for history; but we should not forget that it embodies the popular attitude towards those who stubbornly resisted the Norman, as it was handed on by tradition, and that it reveals almost pathetically the dearth of heroic material in an age which should have produced it in abundance. Hereward was a tenant in a small way of the abbey of Peterborough. What led him into such a determined revolt we do not know, unless he was among those who were induced to join the Danes after their arrival, in the belief that their invasion would be successful. Nor do we know what collected in the Isle of Ely a band of men whom the Peterborough chronicler was probably not wrong, from any point of view, in calling outlaws. A force of desperate men could hope to maintain themselves for some time in the Isle of Ely; they could not hope for anything more than this. The coming of the Danes added little real strength, though the country about believed for the moment, as it had done north of the Humber, that the tide had turned. The first act of the allies was the plunder and destruction of the abbey and town of Peterborough shortly after the meeting of the council of Windsor. The English abbot Brand had died the previous autumn, and William had appointed in his place a Norman, Turold, distinguished as a good fighter and a hard ruler. These qualities had led the king to select him for this special post, and the plundering of the abbey, so far as it was not mere marauding, looks like an answering act of spite. The Danes seem to have been disposed at first to hold Peterborough, but Turold must have brought them proposals of peace from William, which induced them to withdraw at last from England with the secure possession of their plunder.

Hereward and his men accomplished nothing more that year, but others gradually gathered in to them, including some men of note. Edwin and Morcar had once more changed sides, or had fled from William's court to escape some danger there. Edwin had been killed in trying to make his way through to Scotland, but Morcar had joined the refugees in Ely. Bishop Ethelwin of Durham was also there, and a northern thane, Siward Barn. In 1074 William advanced in person against the "camp of refuge." A fleet was sent to blockade one side while the army attacked from the other. It was found necessary to build a long causeway for the approach of the army and around this work the fiercest fighting occurred; but its building could not be stopped, and just as it was finished the defenders of the Isle surrendered. The leaders were imprisoned, Morcar in Normandy for the rest of William's reign. The common men were mutilated and released. Hereward escaped to sea, but probably afterwards submitted to William and received his favour. Edric the Wild, who had long remained unsubdued on the Welsh borders, had also yielded before the surrender of the Isle of Ely, and the last resistance that can be called in any sense organized was at an end.

The comparatively easy pacification of the land, the early submission to their fate of so strong a nation, was in no small degree aided by the completeness with which the country was already occupied by Norman colonies, if we may call them so. Probably before the surrender of Ely every important town was under the immediate supervision of some Norman baron, with a force of his own. In all the strategically important places fortified posts had been built and regular garrisons stationed. Even the country districts had to a large extent been occupied in a similar way. It is hardly probable that as late as 1072 any considerable area in England had escaped extensive confiscations. Everywhere the Norman had appeared to take possession of his fief, to establish new tenants, or to bring the old ones into new relations with himself, to arrange for the administration of his manors, and to leave behind him the agents who were responsible to himself for the good conduct of affairs. If he made but little change in the economic organization of his property, and disturbed the labouring class but slightly or not at all, he would give to a wide district a vivid impression of the strength of the new order and of the hopelessness of any resistance.

Already Norman families, who were to make so much of the history of the coming centuries, were rooted in the land. Montfort and Mortimer; Percy, Beauchamp, and Mowbray; Ferrets and Lacy; Beaumont, Mandeville, and Grantmesnil; Clare, Bigod, and Bohun; and many others of equal or nearly equal name. All these were as yet of no higher than baronial rank, but if we could trust the chroniclers, we should be able to make out in addition a considerable list of earldoms which William had established by this date or soon afterwards, in many parts of England, and in these were other great names. According to this evidence, his two half brothers, the children of his mother by her marriage with Herlwin de Conteville, had been most richly provided for: Odo, Bishop of Bayeux, as Earl of Kent, and Robert, Count of Mortain, with a princely domain in the south-west as Earl of Cornwall. One of the earliest to be made an earl was his old friend and the son of his guardian, William Fitz Osbern, who had been created Earl of Hereford; he was now dead and was succeeded by his son Roger, soon very justly to lose title and land. Shrewsbury was held by Roger of Montgomery; Chester by Hugh of Avranches, the second earl; Surrey by William of Warenne; Berkshire by Walter Giffard. Alan Rufus of Britanny was Earl of Richmondshire; Odo of Champagne, Earl of Holderness; and Ralph of Guader, who was to share in the downfall of Roger Fitz Osbern, Earl of Norfolk. One Englishman, who with much less justice was to be involved in the fate which rightly befell these two Norman earls, was also earl at this time, Watheof, who had lately succeeded Gospatric in the troubled earldom of Northumberland, and who also held the earldoms of Northampton and Huntingdon. These men certainly held important lordships in the districts named, but whether so many earldoms, in form and law, had really been established by the Conqueror at this date, or were established by him at any later time, is exceedingly doubtful. The evidence of the chroniclers is easily shown to be untrustworthy in the matter of titles, and the more satisfactory evidence which we obtain from charters and the Domesday Book does not justify this extensive list. But the historian does not find it possible to decide with confidence in every individual case. Of the earldoms of this list it is nearly certain that we must drop out those of Cornwall, Holderness, Surrey, Berkshire, and Richmond, and almost or quite certain that we may allow to stand those of Waltheof and William Fitz Osbern, of Kent, Chester, and Shrewsbury.

Independently of the question of evidence, it is difficult to see what there was in the general situation in England which could have led the Conqueror to so wide a departure from the established practice of the Norman dukes as the creation of so many earls would be. In Normandy the title of count was practically unknown outside the ducal family. The feudal count as found in other French provinces, the sovereign of a little principality as independent of the feudal holder of the province as he himself was of the king, did not exist there. The four lordships which bore the title of count, Talou or Arques, Eu, Evreux, and Mortain, were reserved for younger branches of the ducal house, and carried with them no sovereign rights. The tradition of the Saxon earldom undoubtedly exercised by degrees a great influence on the royal practice in England, and by the middle of the twelfth century earls existed in considerable numbers; but the lack of conclusive evidence for the existence of many under William probably reflects the fact of his few creations. But in the cases which we can certainly trace to William, it was not the old Saxon earldom which was revived. The new earldom, with the possible exception of one or two earls who, like the old Prankish margrave, or the later palatine count, were given unusual powers to support unusual military responsibilities, was a title, not an office. It was not a government of provinces, but a mark of rank; and the danger involved in the older office, of the growth of independent powers within the state under local dynasties which would be, though existing under other forms, as difficult to control as the local dynasties of feudal France, was removed once for all by the introduction of the Norman centralization. That no serious trouble ever came from the so-called palatine earldoms is itself evidence of the powerful monarchy ruling in England.

This centralization was one of the great facts of the Conquest. In it resided the strength of the Norman monarchy, and it was of the utmost importance as well in its bearing on the future history of England. Delolme, one of the earliest of foreign writers on the English constitution, remarks that the explanation of English liberty is to be found in the absolute power of her early kings, and the most careful modern student can do no more than amplify this statement. That this centralization was the result of any deliberate policy on the part of William can hardly be maintained. A conscious modification of the feudal system as he introduced it into England, with a view to the preservation of his own power, has often been attributed to the Conqueror. But the political insight which would have enabled him to recognize the evil tendencies inherent in the only institutional system he had ever known, and to plan and apply remedies proper to counteract these tendencies but not inconsistent with the system itself, would indicate a higher quality of statesmanship than anything else in his career shows him to possess. More to the purpose is the fact that there is no evidence of any such modification, while the drift of evidence is against it. William was determined to be strong, not because of any theory which he had formed of the value of strength, or of the way to secure it, but because he was strong and had always been so since he recovered the full powers of a sovereign in the struggles which followed his minority. The concentration of all the functions of sovereignty in his own hands, and the reservation of the allegiance of all landholders to himself, which strengthened his position in England, had strengthened it first in Normandy.

Intentional weakening of the feudal barons has been seen in the fact that the manors which they held were scattered about in different parts of England, so that the formation of an independent principality, or a quick concentration of strength, would not be possible. That this was a fact characteristic of England is probably true. But it is sufficiently accounted for in part by the gradual spread of the Norman occupation, and of the consequent confiscations and re-grants, and in part by the fact that it had always been characteristic of England, so that when the holding of a given Saxon thane was transferred bodily to the Norman baron, he found his manors lying in no continuous whole. In any case, however, the divided character of the Norman baronies in England must not be pressed too far. The grants to his two half brothers, and the earldoms of Chester and Shrewsbury on the borders of Wales, are enough to show that William was not afraid of principalities within the state, and other instances on a somewhat smaller scale could be cited. Nor ought comparison to be made between English baronies, or earldoms even, and those feudal dominions on the continent which had been based on the counties of the earlier period. In these, sovereign rights over a large contiguous territory, originally delegated to an administrative officer, had been transformed into a practically independent power. The proper comparison is rather between the English baronies of whatever rank and those continental feudal dominions which were formed by natural process half economic and half political, without definite delegation of sovereign powers, within or alongside the provincial countships, and this comparison would show less difference.

If the Saxon earl did not survive the Conquest in the same position as before, the Saxon sheriff did. The office as the Normans found it in England was in so many ways similar to that of the viscount, vicecomes, which still survived in Normandy as an administrative office, that it was very easy to identify the two and to bring the Norman name into common use as an equivalent of the Saxon. The result of the new conditions was largely to increase the sheriff's importance and power. As the special representative of the king in the county, he shared in the increased power of his master, practically the whole administrative system of the state, as it affected its local divisions, was worked through him. Administrator of the royal domains, responsible for the most important revenues, vehicle of royal commands of all kinds, and retaining the judicial functions which had been associated with the office in Saxon times, he held a position, not merely of power but of opportunity. Evidence is abundant of great abuse of power by the sheriff at the expense of the conquered. Nor did the king always escape these abuses, for the office, like that of the Carolingian count, to which it was in many ways similar, contained a possibility of use for private and personal advantage which could be corrected, even by so strong a sovereign as the Anglo-Norman, only by violent intervention at intervals.

Some time after the Conquest, but at a date unknown, William set aside a considerable portion of Hampshire to form a hunting ground, the New Forest, near his residence at Winchester. The chroniclers of the next generation describe the formation of the Forest as the devastation of a large tract of country in which churches were destroyed, the inhabitants driven out, and the cultivated land thrown back into wilderness, and they record a contemporary belief that the violent deaths of so many members of William's house within the bounds of the Forest, including two of his sons, were acts of divine vengeance and proofs of the wickedness of the deed. While this tradition of the method of making the Forest is still generally accepted, it has been called in question for reasons that make it necessary, in my opinion, to pronounce it doubtful. It is hardly consistent with the general character of William. Such statements of chroniclers are too easily explained to warrant us in accepting them without qualification. The evidence of geology and of the history of agriculture indicates that probably the larger part of this tract was only thinly populated, and Domesday Book shows some portions of the Forest still occupied by cultivators.[9] The forest laws of the Norman kings were severe in the extreme, and weighed cruelly on beasts and men alike, and on men of rank as well as simple freemen. They excited a general and bitter hostility which lasted for generations, and prepared a natural soil for the rapid growth of a partially mythical explanation to account in a satisfactory way for the dramatic accidents which followed the family of the Conqueror in the Forest, by the direct and tangible wickedness which had attended the making of the hunting ground. It is probable also that individual acts of violence did accompany the making, and that some villages and churches were destroyed. But the likelihood is so strong against a general devastation that history should probably acquit William of the greater crime laid to his charge, and refuse to place any longer the devastation of Hampshire in the same class with that of Northumberland.

After the surrender of Ely, William's attention was next given to Scotland. In 1070 King Malcolm had invaded northern England, but without results beyond laying waste other portions of that afflicted country. It was easier to show the Scots than the Danes that William was capable of striking back, and in 1072, after a brief visit to Normandy, an army under the king's command advanced along the east coast with an accompanying fleet. No attempt was made to check this invasion in the field, and only when William had reached Abernethy did Malcolm come to meet him. What arrangement was made between them it is impossible to say, but it was one that was satisfactory to William at the time. Probably Malcolm became his vassal and gave him hostages for his good conduct, but if so, his allegiance did not bind him very securely. Norman feudalism was no more successful than the ordinary type, in dealing with a reigning sovereign who was in vassal relations.