During this period also, probably within a year after the death of his son William, Henry took measures to establish the position of one of his illegitimate sons, very likely with a view to the influence which he might have upon the succession when the question should arise. Robert of Caen, so called from the place of his birth, was created Earl of Gloucester, and was married to Mabel, heiress of the large possessions of Robert Fitz Hamon in Gloucester, Wales, and Normandy. Robert of Gloucester, as he came to be known, was the eldest of Henry's illegitimate sons, born before his father's accession to the throne, and he was now in the vigour of young manhood. He was also, of all Henry's children of whom we know anything, the most nearly like himself, of more than average abilities, patient and resourceful, hardly inheriting in full his father's diplomatic skill but not without gifts of the kind, and earning the reputation of a lover of books and a patron of writers. A hundred years earlier there would have been no serious question, in the circumstances which had arisen, of his right to succeed his father, at least in the duchy of Normandy. That the possibility of such a succession was present in men's minds is shown by a contemporary record that the suggestion was made to him on the death of Henry, and rejected at once through his loyalty to his sister's son. Whether this record is to be believed or not, it shows that the event was thought possible.[23]

Certainly there was no real movement, not even the slightest, in his favour, and this fact reveals the change which had taken place in men's ideas of the succession in a century. The necessity of legitimate birth was coming to be recognized as indisputable, though it had not been by the early Teutonic peoples. Of the causes of this change, the teachings of the Church were no doubt the most effective, becoming of more force with its increasing influence, and especially since, as a part of the Hildebrandine reformation, it had insisted with so much emphasis on the fact that the son of a married priest could have no right of succession to his father's benefice, being of illegitimate birth; but the teachings of the sacredness of the marriage tie, of the sinfulness of illicit relations, and of the nullity of marriage within the prohibited degrees, were of influence in the change of ideas. It is also true that men's notions of the right of succession to property in general were becoming more strict and definite, and very possibly the importance of the succession involved in this particular case had its effect. One may almost regret that this change of ideas, which was certainly an advance in morals, as well as in law, was not delayed for another generation; for if Robert of Gloucester could have succeeded on the death of Henry without dispute, England would have been saved weary years of strife and suffering.

The death of the young William was a signal to set Henry's enemies in motion again. But they did not begin at once. Henry's position was still unweakened. Very likely his speedy marriage was a notice to the world that he did not propose to modify in the least his earlier plans. Probably also the absence of Fulk of Anjou, who had gone on a pilgrimage to Jerusalem soon after his treaty of 1119 with Henry, was a cause of delay, for the natural first move would be for him to demand a return of his daughter and her dowry. Fulk's stay was not long in the land of which he was in a few years to be king, and on his return he at once sent for his daughter, probably in 1121. She returned home, but as late as December, 1122, there was still trouble between him and Henry in regard to her dowry, which Henry no doubt was reluctant to surrender.

About the same time, Henry's old enemy, Amaury of Montfort, disliking the strictness of Henry's rule and the frequency of his demands for money, began to work among the barons of Normandy and with his nephew, the Count of Anjou, in favour of William Clito. It was already clear that Henry's hope of another heir was likely to be disappointed, and Normandy would naturally be more easily attracted to the son of Robert than England The first step was one which did not violate any engagement with Henry, but which was, nevertheless, a decided recognition of the claims of his nephew, and an open attack on his plans. Fulk gave his second daughter, Sibyl, in marriage to William Clito, and with her the county of Maine, which had been a part of Matilda's dower on her marriage with Henry's son William. Under the circumstances, this was equivalent to an announcement that he expected William Clito to be the Duke of Normandy. Early in 1123, Henry sent over troops to Normandy, and in June of that year he crossed himself, to be on the spot if the revolt and war which were threatening should break out. In September the discontented barons agreed together to take arms. It is of interest that among these was Waleran of Meulan, the son of the king's faithful counsellor, Count Robert. Waleran had inherited his father's Norman possessions while his brother Robert had become Earl of Leicester in England.

In all this the hand of Louis, king of France, was not openly seen. Undoubtedly, however, the movement had his encouragement from the beginning, and very likely his promise of open support when the time should come. The death of the male heir to England and Normandy would naturally draw Henry's daughter Matilda, and her husband the emperor, nearer to him; and of this, while Henry was still in England, some evidence has come down to us though not of the most satisfactory kind. Any evidence at the time that this alliance was likely to become more close would excite the fear of the king of France and make him ready to support any movement against the English king. Flanders would feel the danger as keenly, and in these troubles Charles the Good abandoned his English alliance and supported the cause of France.

The contest which followed between the king and his revolted barons is hardly to be dignified with the name of war. The forced surrender of a few strongholds, the long siege of seven weeks, long for those days, of Waleran of Meulan's castle, of Pont Audemer and its capture, and the occupation of Amaury of Montfort's city of Evreux, filled the remainder of the year 1123, and in March of 1124 the battle of Bourgtheroulde, in which Ralph, Earl of Chester, defeated Amaury and Waleran and captured a large number of prisoners, virtually ended the conflict. Upon the leaders whom he had captured Henry inflicted his customary punishment of long imprisonment, or the worse fate of blinding. The Norman barons had taken arms, and had failed without the help from abroad which they undoubtedly expected. We do not know in full detail the steps which had been taken to bring about this result, but it was attributed to the diplomacy of Henry, that neither Fulk of Anjou nor Louis of France was able to attack him.

Henry probably had little difficulty in moving his son-in-law, the emperor Henry V, to attack Louis of France. Besides the general reason which would influence him, of willingness to support Matilda's father at this time, and of standing unfriendliness with France, he was especially ready to punish the state in which successive popes had found refuge and support when driven from Italy by his successes. The policy of an attack on Louis was not popular with the German princes, and the army with which the Emperor crossed the border was not a large one. To oppose him, Louis advanced with a great and enthusiastic host. Taking in solemn ceremony from the altar of St Denis the oriflamme, the banner of the holy defender of the land, he aroused the patriotism of northern France as against a hereditary enemy. Even Henry's nephew, Theobald of Blois, led out his forces to aid the king. The news of the army advancing against them did not increase the ardour of the German forces; and hearing of an insurrection in Worms, the Emperor turned back, having accomplished nothing more than to secure a free hand for Henry of England against the Norman rebels.

Against Fulk of Anjou Henry seems to have found his ally in the pope. The marriage of William Clito with Sibyl, with all that it might carry with it, was too threatening a danger to be allowed to stand, if in any way it could be avoided. The convenient plea of relationship, convenient to be remembered or forgotten according to the circumstances, was urged upon the pope. The Clito and his bride were related in no nearer degree than the tenth, according to the reckoning of the canon law, which prohibited marriage between parties related in the seventh degree, and Henry's own children, William in his earlier, and Matilda in her later marriage, with the sister and brother of Sibyl, were equally subject to censure. But this was a different case. Henry's arguments at Rome—Orderic tells us that threats, prayers, and money were combined—were effective, and the marriage was ordered dissolved. Excommunication and interdict were necessary to enforce this decision; but at last, in the spring of 1125, Fulk was obliged to yield, and William Clito began his wanderings once more, followed everywhere by the "long arm" of his uncle.

At Easter time in 1125, probably a few days before the date of the papal bull of interdict which compelled the dissolution of the marriage of William and Sibyl, a papal legate, John of Crema, landed in England. Possibly this departure from Henry's practice down to this time was a part of the price which the papal decision cost. The legate made a complete visitation of England, had a meeting with the king of Scots, and presided at a council of the English Church held in September, where the canons of Anselm were renewed in somewhat milder form. On his return to Rome in October, he was accompanied by the Archbishops of Canterbury and York, who went there about the still unsettled question of the obedience of the latter. Not even now was this question settled on its merits, but William of Corbeil made application, supported by the king, to be appointed the standing papal legate in Britain. This request was granted, and formed a precedent which was followed by successive popes and archbishops. This appointment is usually considered a lowering of the pretensions of the Archbishop of Canterbury, and an infringement of the independence of the English Church, and to a considerable extent this is true. Under a king as strong as Henry I, with an archbishop no stronger than William of Corbeil, or, indeed, with one not exceptionally strong, the papal authority gained very little from the arrangement. But it was a perpetual opportunity; it was a recognition of papal right. Under it the number of appeals to Rome increased; it marks in a legal way the advance of papal authority and of a consciousness of unity in the Church since the accession of the king, and it must have been so regarded at Rome. The appointment gave to Canterbury at once undoubted supremacy over York, but not on the old grounds, and that question was passed on to the future still unsettled.

In the spring of 1125 also occurred an event which again changed the direction of Henry's plans. On May 23, the emperor Henry V died, without children by his marriage to Matilda. The widowed Empress, as she was henceforth called by the English though she had never received the imperial crown, obeyed her father's summons to return to him in Normandy with great reluctance. She had been in Germany since her early childhood, and she was now twenty-three years of age. She could have few recollections of any other home. She loved the German people, and was beloved by them. We are told even that some of them desired her to reign in her husband's stead, and came to ask her return of Henry. But the death of her husband had rendered her succession to the English throne a matter of less difficulty, and Henry had no mind to sacrifice his own plans for the benefit of a foreign people. In September, 1126, he returned with Matilda to England, and in January following, at a great council in London, he demanded and obtained of the baronage, lay and spiritual, an oath to accept Matilda as sovereign if he should die without a male heir. The inference is natural from the account William of Malmesbury gives of this event, that in the argument before the council much was made of the fact that Matilda was a descendant of the old Saxon, as well as of the Norman, line. It is evident, also, that there was hesitation on the part of the barons, and that they yielded reluctantly to the king's demand.