[231] The medieval monarchy of Western Europe was an institution of surprising tenacity and elasticity. The truly remarkable strength which it exhibited for better than thirteen hundred years from Odoacer to Danton derived from many sources, but the most important of these was its ordinary recognition as the sole legitimate form of government. Only with the concept of the Christian empire, which it cannibalized, did the idea of kingship share any of the public recognition of its ultimate necessity as the guarantor of political stability and legal rights. The society of the medieval west understood the many kinds of power but it invariably turned from them eventually to the only source of proper authority it could conceive. Such reliance made the weakest of kings formidable and strong kings irresistible.

Within each kingdom the search for a justification and legalization, if not for justice, of what one had achieved through power led at every moment back to the crown. Every malcontent and every schemer sought eventual reconciliation on some terms with his king, and that permanent tendency operated inexorably as a balance wheel which restored the monarch to his rightful place at the center of political life. But that internal process was strengthened continually by the predilection of foreign monarchs and powers for dealing with the king alone in preference to the mightiest of his subjects. The crown's premier position in the political life of the realm could thus be enhanced by initiatives of its own in the realm of the international sphere or by responding to overtures from without.

Alfonso VI sought to improve his fortunes in the peninsula and in León-Castilla itself by exploiting this international recognition in 1092. One device he employed was an alliance with the Italian maritime powers of Genoa and Pisa for a joint descent upon the taifa of Valencia. The resources of these burgeoning naval powers of the western Mediterranean could be called into play by the crown and they were.

The late twelfth-century Muslim source which is our earliest authority [232] for the episode does not date the events beyond the year, but it must have been in spring or summer since fleets did not ordinarily sail during fall and winter in the Middle Ages. Moreover it is possible to place the court at Sahagún or León from early February until the beginning of April.⁽¹⁾ By May 1, 1092, Alfonso was at the monastery of Oña high in the upper valley of the Ebro. From that point he may have proceeded down the Ebro to Nájera and then through the territory of his tributary, al-Mustain of Zaragoza, who would scarcely have dared to oppose him openly.⁽²⁾

From the vicinity of Zaragoza the route to Valencia was a relatively easy one. Still, it was long enough to make it unlikely that the host could have arrived at Valencia before June 1 at the earliest. At the same time, Sancho Ramírez of Aragón and Berenguer Ramón of Barcelona were sitting down before the port of Tortosa farther up the coast. The aim of none of these allies could have been permanent conquest for which they lacked the resources at the time. Clearly the effort was aimed primarily at displacing the Cid from the position he had achieved in the Spanish Levant. The effort was a token of how formidable a power the Castilian adventurer had become. His ejection also would have allowed the allies to reclaim the parias, but we may be sure that they would have to have shared those with the Genoese and the Pisans. The price the Italian powers exacted would have included a share of those revenues as well as favorable trading privileges in both ports in return for their essential contribution.

Unfortunately for the allies, the coordination of all these elements was impossible to effect. Rodrigo Díaz himself refused to give battle but, slipping around the advancing armies and reaching Zaragoza, invaded the Rioja with the help of troops supplied by al-Mustain. Indeed, our Muslim source tells us that Alfonso's siege of Valencia endured but one day on that account.⁽³⁾ However that may be, it appears that the Italian [233] fleet was late in arriving and found that the Leonese monarch had already decamped. It then coasted north to join in the siege of Tortosa, but nothing was effected there either and the Aragonese and the Catalans also retired discomfited. Before summer's end, the Cid had regained his commanding position in the east from which nothing would dislodge him but death.

We cannot follow the movements of Alfonso VI during the remainder of the summer and early fall of 1092. It is likely that he returned from Valencia to León, but the court seems to have been in Oviedo at the end of July.⁽⁴⁾ When next it can be located it is at Sahagún on November 26, 1092.⁽⁵⁾ Menéndez Pidal put his faith in both a new reconciliation of Alfonso VI and the Cid and a major defeat of the former at Murâbit hands in Andalucía following the abortive Valencian campaign.⁽⁶⁾ As to the first, doubtless the king appreciated the fact that, if the parias of Valencia could not be his, at least they were supporting a captain with whom he had a certain commonality of interest. Yet there is no compelling evidence that this perception was embodied in any formal recognition. As to the latter, certainly the records of the year leave much to be desired, especially in view of the frequent confusion of the chroniclers. Nevertheless it is extremely doubtful that Alfonso VI could have kept an army in the field and on the march following what had to have been at least a two-month campaign against Valencia which had produced no booty worth mentioning.

The hard evidence rather suggests an unhindered mopping-up operation carried out by the Murâbits in the south. There one of the sons of Yusuf Muhammed ibn-Aisa, had been named governor of Almeria and Murcia. From that position he was able to force the surrender or the withdrawal of the Christian garrison in the castle of Aledo which had threatened the supply lines between the two since 1086.⁽⁷⁾ That success opened up the possibility of a serious attack against Valencia which city would become a major objective of the Murâbits over the next ten years.

[234] In the late summer and fall of 1092, Muhammed ibn-Aisa advanced north, taking Denia, Játiva, and Alcira in the southern reaches of the taifa of Valencia with little difficulty. This last was but thirty-five kilometers south of the city itself. With Murâbit assistance apparently so close a faction in the city was not slow to begin intrigues against the perpetually ineffectual al-Qadir. With the Cid still absent in Zaragoza and with the flight of some of the Christian forces who had been controlling the countryside, the conspirators seized control of the city and quickly, on October 28, 1092, decapitated al-Qadir, who had attempted to flee. Only the failure of the Murâbits to send a strong force to the aid of the rebels prevented the city from falling to them at this time. At it was, Valencia came to be governed for a time as an independent city under one of its former judges.⁽⁸⁾ Nevertheless the weakness of the taifa which had secured the eastern approaches to the realm of León-Castilla since 1085 remained a serious threat and one that was only to be repaired by the heroic efforts of Rodrigo Díaz de Vivar, not Alfonso.

If his best efforts in the field had come to naught in 1092, however, the king could draw on other possibilities inherent in the position of the crown to recoup his position. In that year he moved both to bolster his prestige and to remedy his lack of a male heir. This he did in the only way open to him at the moment, by taking a mistress from Andalucía who was to provide him with the son he had sought since 1074.

The life and even the death of the Muslim princess Zaida, widow of Fath al-Mamun of Córdoba since March of the previous year, are problematic. Our best authority, Bishop Pelayo, wrongly makes her the daughter rather than the daughter-in-law of al-Mutamid of Sevilla but does identify her as the mother of Alfonso's only son.⁽⁹⁾ Her sepulchral inscription, variously reported, informs that she died in childbirth on either Monday, the 13th of September, or Thursday, the 13th of September, without reporting the year of her death. We cannot even be sure that it was the birth of Sancho Alfónsez himself.⁽¹⁰⁾

Nonetheless the reasons for dating her relationship with Alfonso VI from late 1091 or 1092 are compelling. For the policy of that monarch, concerned with salvaging something out of the wreck of the taifas of Andalucía, only a date after the fall of Córdoba in March 1091 and before the fall of Badajoz in early 1094 would have made much sense. Zaida could only be of service there as a symbol of Alfonso's claim to be the protector of Spanish Islam against the African Murâbits. In addition, [235] if we consider that her son was to die in the battle of Uclés in May 1108, we are forced to predicate as early a date as possible. Even when one makes allowance for the early age accepted for maturity in the twelfth century, the risking of the only male heir of the realm on the field of battle would seem unlikely before he had attained the age of fourteen or fifteen at least. Then too, the death of Queen Constance in the fall of 1093 was perhaps preceded by serious illness, which would have facilitated the king's decision to take a mistress for reasons of state.

It is likely, then, that the negotiations for something like a "marriage" alliance began in the spring of 1091 at the instance of al-Mutamid of Sevilla and may or may not have been concluded before the fall of that king and his capital together in September of the same year. Jiménez de Rada in the thirteenth century connected the reception of the princess with a large cession of territory by the Sevillan monarch, which would have been likely under the circumstances, but, as has pointed out, the account is garbled badly.⁽¹¹⁾ Even if those negotiations had never been completed, the liaison would still have continued to be important for its value in reassuring the Muslim of Badajoz of the serious intents of the Leonese monarch in regard to Andalucía. It is likely, then, that it had become an established fact before al-Mutawakkil of Badajoz ceded Lisbon, Santarem, and Sintra to Alfonso in late April and early May of 1093.⁽¹²⁾ All of this diplomatic manuevering, which involved the likelihood and then the fact of considerable territorial gain in the south, would have strengthened the Leonese monarch materially within his own realm. It would have been even more effective if it were accompanied, as seems probable, by the pregnancy of Zaida in late 1092 or early 1093 and thus raised the possibility of a male heir to the throne.

Another avenue uniquely open to the assertion of the royal supremacy was the church, and Alfonso VI had been active in that arena more or less continuously but with special intensity since the Council of León of 1090. The particular focus for royal efforts was the frontier district of Portugal.

With all due respects to the arguments from anthropology and linguistics, Portugal appears first in the western European world as a frontier district of the kingdom of León created by the Reconquista. Naturally difficult to control from the interior of the peninsula, it had alternated between the Muslim and the Christian worlds from the ninth to the eleventh centuries. Then the definitive reconquest of Coimbra by [236] Fernando I in 1064 secured all of Portugal north of Santarem for León-Castilla. But isolated behind its mountains and easily approachable only from the north, that string of river valley facing the Atlantic presented the most onerous problems of government from this beginning.

The instruments of royal authority there were three. In the first place were the merinos of the royal fisc. Those all-purpose estate officers are fairly well attested for the reign of Fernando.⁽¹³⁾ Then there were the chief magnates of the area who were tied to the royal dynasty by blood relationship but whom Fernando I, it is argued, replaced late in his reign with lesser nobles identifiable with the merinos of the fisc.⁽¹⁴⁾ However, in Alfonso VI's reign there is scattered evidence that single individuals far above that humble level held the royal jurisdiction in the region about Braga.⁽¹⁵⁾ In the territory to the south in and around Coimbra a separate jurisdiction under the local Mozárab Sisnando Davídez existed from the early reign of Alfonso VI if not perhaps from the time of Fernando I as has been asserted.⁽¹⁶⁾

The third prop to the royal authority were the restored bishoprics of the province. In the time of Fernando I the sole bishop in the Christian territories there was Sisnando of Oporto, who figured prominently in that monarch's diplomas and in those of García I as well. But when García restored the sees of Braga, Lamego, and Túy in 1070, the see of Oporto was subsumed into that of Braga.⁽¹⁷⁾ It would vanish until ecclesiastical politics brought about its restoration in 1113. A little later the [237] same fate was meted out to Lamego, whose bishop last appeared in a charter of July 29, 1071.⁽¹⁸⁾ Braga thus became the chosen instrument of the crown in Portugal between the Miño and Duero, and Bishop Pedro figured prominently in the charters making Alfonso VI's return to power in the fall of 1072.⁽¹⁹⁾

The rise of a second ecclesiastical center on the Mondego River in the south at Coimbra seems to have been a result of local forces operating on that far frontier. An interpolated document of Apríl 13, 1086, retrospectively attributes the appearance there of a Mozárab bishop named Paterno, formerly of Tortosa, to the initiative of Sisnando Davídez, but all we can say with assurance is that Paterno is reliably first cited as bishop on November 20, 1078.⁽²⁰⁾ He is last mentioned in that capacity on March 1, 1088, but never appears as confirmant of a royal document.⁽²¹⁾ Perhaps it is of some significance that in that final document the bishop was given permission to make a pilgrimage to the Holy Land.

Apparently a part of the royal agenda for the Council of Husillos of 1088 was the enhancing of royal control over Coimbra and its environs. The sole document of the council preserved to us was confirmed by a Martín, bishop-elect of that see, and later in the same year a private document of Coimbra cited him, although still as bishop-elect.⁽²²⁾ Pierre David identified him as prior of the cathedral chapter of Coimbra and as the protégé of Sisnando Davídez.⁽²³⁾ But if that magnate had proposed him and Alfonso had accepted him it is difficult to see why he was never consecrated. This obscure dispute continued for in the following year one "Julián" appears as bishop and in 1091 a "Juan."⁽²⁴⁾ They are probably the same person.

On August 25, 1091, Sisnando Davídez died after more than twenty years of semi-independent rule at Coimbra and was succeeded by his son-in-law, Martín Múñoz, scion of a magnate family of Portugal.⁽²⁵⁾ But Alfonso VI seems already to have lost patience with that state of [238] affairs and had secured the election of a new bishop at Easter time of 1091. The new bishop, Cresconio, was the former abbot of Saint Bartholomew of Túy and may be regarded as essentially a royal choice. He was consecrated by that pillar of royal authority, the archbishop of Toledo, with the assistance of the bishops of Túy and Orense.⁽²⁶⁾ The participation of these latter, as well as later events, suggests that Cresconio would have been acceptable to Count Raymond of Galicia also. But the consecration and installation of the royal candidate at Coimbra may have had to wait on the death of the magnate Sisnando.⁽²⁷⁾

Late in 1091 the king also acted against the audacious Pedro of Braga. Before the end of the year he had forced that prelate to retire permanently into a monastery for having secured the restoration of the Bracaran archiepiscopate from the antipope Clement III.⁽²⁸⁾ In June 1092 Archbishop Bernard of Toledo was representing the royal will when he consecrated Cresconio at Coimbra. He did so even more dramatically when he also consecrated the very cathedral of Braga itself on August 28, 1092, while its bishop languished in a monastery.⁽²⁹⁾ Thus at the very time when Alfonso VI was negotiating with al-Mutawakkil of Badajoz for the cession of Lisbon, Santarem, and Sintra he was also engaged in strengthening his hold over those Portuguese districts that would be crucial to the support and supply of the new acquisitions.

[239] These negotiations would have been in train while the court was un its customary Christmas location at Sahagún. The curia can be placed there from January until April 3, 1093.⁽³⁰⁾ Immediately after this latter date, however, the king must have left for Coimbra by way of Astorga, Orense, and Túy, for we find him there on April 22, 1093, confirming the fuero he had granted that city in 1085.⁽³¹⁾ The urgency of the matter is obvious when we realize that the king was on the road even on the feast of Easter, which fell that year on April 17. He probably had celebrated it at Túy.

Those who confirmed Alfonso's action at Coimbra suggest that the king had traveled with a very small court. In the document he is found in the presence of Count Raymond of Galicia, of Martín Múñoz of Coimbra, of Bishop Cresconio, and of Rodrigo, prior of the church of Braga. Doubtless the army he led was also composed chiefly of Galicians and Portuguese, for he must have led that force south from Coimbra to Santarem. The intervention of the Murâbits in the peninsula had led to a resurgence of Islamic militancy among the populations of the taifa states. Neither al-Mutawakkil nor his lieutenants in Portugal could afford to be seen openly cooperating with the Christian, so a pretense of coercion must be arranged to justify what had already been concerted. The matter went smoothly and Santarem surrendered on April 30, followd by Lisbon on May 5, and Sintra on May 8, 1093.⁽³²⁾ The whole of central Portugal north of the Río Tajo had passed into the hands of Alfonso VI.

At the same time the king rewarded his son-in-law, Count Raymond, so that the latter became the second most important man in the kingdom of León-Castilla. Santarem, Lisbon, and Sintra were placed under his command.⁽³³⁾ Then, or shortly thereafter, Martín Múñoz was removed from control at Coimbra. The latter seems to have held a lesser position for a time but then joined the Cid at Valencia and, after [240] Rodrigo's death, found hospitality at the court of Pedro I of Aragón.⁽³⁴⁾ Under the count, the north Portuguese magnate Soeiro Mendes held the frontier cities just acquired, and Bishop Cresconio at Coimbra and Prior Rodrigo at Braga were pillars of his rule in those towns respectively.⁽³⁵⁾ Raymond thus held the entire Atlantic coast of the peninsula from the bay at Lisbon to the Bay of Biscay in the north and inland to the mountains which formed a natural frontier at every point. It was a principality of truly kingly proportions and dwarfed the territories controlled by a García Ordoñez, a Pedro Ansúrez, or even a Cid. More than ever the Burgundian noble appeared to be the likely successor of Alfonso VI.

From central Portugal the king traveled up the Tajo River to Toledo, but from then until late July the documentary record fails.⁽³⁶⁾ On July 25, 1093, he granted a charter to the monastery of Valbanera in Rioja. He was perhaps at Nájera if not at the monastery itself, and the royal diploma was confirmed by many Castilian magnates. Even more significant is the confirmation of Count Raymond, who had now become a major figure in the royal court.⁽³⁷⁾

But after so much good fortune the prospects of the Burgundian noble were about to suffer two major reverses. Sometime in 1093, perhaps on September 13, a male heir was born to the king. This was the son of Zaida, Sancho Alfónsez.⁽³⁸⁾ Although much time would have to pass before it would become clear whether or not the infant would survive, for the mortality rate in the first year was on the order of one in four during the period, his birth at once distanced Raymond, Urraca, and their possible progeny from that succession to the throne which had recently seemed so secure.

At about the same time, the count also lost a most influential friend at court when Queen Constance, his aunt, died. The queen had confirmed her husband's charter of July 25, 1093, but that is her last known appearance in any document. When the king next becomes visible he is making two donations to the monastery of Sahagún on October 25, 1093. The lists of those who confirm indicate a major curia in process but Constance does not appear in either list. Both donations request [241] prayers for the souls of both the king and queen, but in neither of them is the queen associated with Alfonso in the act of giving even though in one of them what was being conceded was clearly Constance's property.⁽³⁹⁾ In another great gathering at Sahagún, which was the occasion of his next known diploma on November 22, 1093, the king again endowed that monastery with property of the queen, this time of a palace she had had constructed near that monastery.⁽⁴⁰⁾ These documents are the best indications that survive as to the date of Constance's death.⁽⁴¹⁾ It is at least probable that the charters of October 25 followed hard on her obsequies. She was buried at Sahagún alongside Queen Inés whom she had succeeded.⁽⁴²⁾ However much the king might have been personally attached to his late consort, politically her death would have been regarded as providential. Alfonso, still but a vigorous fifty-six, was now free to remarry in search of more sons and legitimate ones. In addition to that great advantage, he was now also better able to distance his ambitious son-in-law somewhat more from the seat of power.

The remainder of the year seems to have been spent at Sahagún as well, surrounded by a court that now would have much to discuss indeed.⁽⁴³⁾ Another event of importance which may have taken place in November was the selection of a new bishop, Martín, for the see of Oviedo. Little is known about his early history except that he seems first to be mentioned in a document of late that month.⁽⁴⁴⁾

[242] While Alfonso spent 1093 negotiating an agreement that would try to bolster al-Mutawakkil of Badajoz against the Murâbits and simultaneously strengthening the western flank of his kingdom by vastly extending the authority of his son-in-law, Count Raymond, the initiatives of independent Christian powers were performing a parallel function on his eastern flank. The never-resting Sancho Ramírez of Aragón continued to preoccupy the taifa at Zaragoza with the raising of fortifications at Luna on the edge of the Riojan plain northeast of that City while his son, Pedro, similarly claimed the attentions of the taifa of Lérida-Tortosa by seizing the castle of Almenar on the northern approaches to Lérida.⁽⁴⁵⁾ Contemporary efforts of the Murâbit emir, Yusuf, to cultivate and encourage his coreligionists in the northeast limited themselves to diplomatic contacts with al-Mustain of Zaragoza.⁽⁴⁶⁾ Little more was done by his son, Muhammed ibn-Aisa, who watched without action the return of the Cid to the taifa of Valencia, the expulsion of the token force he had sent to aid the revolutionary regime there, and the resumption by the latter of the payment of parias to the Castilian. Even when this uneasy accord between the new Valencian regime and the Cid broke down in July 1093 and the latter undertook a siege of the City, the Murâbit response was slow and faint-hearted. An army from Andalucía marched into the southern reaches of the taifa aud then precipitately withdrew without offering combat.⁽⁴⁷⁾ Valencia was left to its fate.

But the activities of the Murâbits were quite other in relation to the "renegade" Muslim ruler of Badajoz, al-Mutawakkil. There the cousin of Emir Yusuf, Sir ibn-Abu Bakr, apparently took advantage of the late winter season, which prevented assistance from the Christian north, to act against the city on the Guadiana. Early in 1094 Badajoz fell prey to a combination of treachery and determined assault. Its former king, along with two of his sons, was murdered on the road from Badajoz to Sevilla by his captors.⁽⁴⁸⁾ The Murâbit empire was now coterminous [243] with Andalucía in Spain. An autonomous Spanish Islam survived only in the embattled taifas of the northeast.

In León-Castilla the year opened as usual with the court at Sahagún in January and early February.⁽⁴⁹⁾ On February 28, 1094, however, the king at least seems to have been in the monastery of San Millán de La Cogolla not far off the pilgrimage road from France to Santiago de Compostela. Presumably Alfonso had accompanied his friend, the archbishop of Toledo, that far on the latter's journey to the south of France.⁽⁵⁰⁾ By the middle of April the king and his court were back at León and remained in the León-Sahagún area until early May.⁽⁵¹⁾ After that, there is a lacuna in the documentary record which endures until late fall. The gap was filled with military activity for virtually the entire border between Christianity and Islam in the peninsula erupted into major hostilities.

The spring of the year saw the unquiet Sancho Ramírez of Aragón establishing a siege of the city of Huesca, the linchpin of the defense line of Zaragoza in the northeast. There, on July 4, 1094, that monarch was struck and killed by an enemy arrow. The siege had to be raised but the young Pedro I, his oldest son who had already been associated in the reign with his father from 1085, succeeded with no difficulty to both the realm and the policies of his father.⁽⁵²⁾

Less than three weeks before that misfortune, on June 15, 1094, the great city of Valencia had surrendered to the Cid.⁽⁵³⁾ The passage of that taifa into Christian hands was an event that much overshadowed the death of the king of Aragón in the eyes of Christian and Muslim alike. Rodrigo Díaz de Vivar had now truly become one of the princes of the peninsula and was shortly to give convincing proof that he was able to maintain himself as such. For, just as did the fall of Toledo, the fall of the taifa of Valencia was to provoke a major reaction in Murâbit North Africa. Yusuf ibn-Tashufin himself directed the embarkation of new forces from his port at Ceuta. Commanded by a nephew, Muhammed, this nucleus rallied around its reinforcements drawn widely from Andalucía, but it does not seem to have reached Valencia before September. [244] There in October it was defeated at Cuart de Poblet by the Cid, who staged an audacious sally and surprise attack from the city.⁽⁵⁴⁾

That victory solidified the Castilian's hold over Valencia and its territories in a fashion that even be could not have anticipated. Hence there is nothing inherently unlikely in Rodrigo's reported attempts to secure a reconciliation with Alfonso VI, though we need not accept the trapping of feudalism which form that overture's literary adornment.⁽⁵⁵⁾ These would logically have begun in June but would also have assumed pressing urgency as the Murâbit army began its advance toward the city. For his part, Alfonso had solid reasons for wishing his erstwhile vassal well. Thus mutual self-interest dictated that the Cid should ask the aid of the Leonese monarch and that the latter should respond. We may safely assume that Alfonso already had a force in the field in the summer of 1094 and that be responded to Rodrigo's appeal for help with the intention of operating in the rear of the besieging force. But when the Cid's surprising victory made that unnecessary he took advantage of the Muslim disarray to raid deep into the territory of Granada and to bear off large numbers of the Christian subject population there and to repopulate the lands around Toledo with them.⁽⁵⁶⁾

Despite these exploits the year was to end with a major Christian defeat in the far west. There the Murâbit governor of Sevilla, Sir, had followed up his conquest of Badajoz with a campaign in central Portugal, in the course of which be seems to have retaken Lisbon itself. Our source for what followed is the "Historia Compostelana" whose author, writing thirty-five years later, describes a Portuguese expedition in which Diego Gelmírez took part as a young man. An army of Galicians led by Count Raymond was surprised and surrounded in its camp near Lisbon by a Muslim force and suffered great casualties.⁽⁵⁷⁾ While the date of this calamity is not supplied it must have been in the latter part of November 1094. A charter of Count Raymond to the church of Coimbra put him there on November 13, 1094, together with a large concourse of Galicians including his alférez his majordomo, and Diego Gelmírez, who confirmed as the count's notary.⁽⁵⁸⁾[245] They were on their way south to the battle which turned out so disastrously.

The king and the major armies of the kingdom were not involved. Neither the chronicler nor the count's charter gives any indication of the presence of any save Galicians and Portuguese. Rather, the royal court was in winter quarters at Sahagún on November 29, 1094.⁽⁵⁹⁾

From that vantage point Alfonso could look back over a year of campaigning whose results were at least moderately satisfactory. If Lisbon had been lost and Count Raymond defeated in the west, the central frontier of Toledo had been reinforced in a major way by the infusion of Mozárab Christian settlers and the Muslim south weakened and impoverished by the same act. The eastern frontier was also much strengthened by the Cid's triumph at Cuart de Poblet and by the smooth succession of Pedro I to his father Sancho Ramírez in Aragón. But the king had been even more successful on the diplomatic front where he was able to secure results that bolstered his internal control in the realm.

The agent of this good fortune was Archbishop Bernard of Toledo whom we last saw in the north of Castilla at San Millán de La Cogolla with the king in February. After that time, Bernard was absent from all the documents until that of November 29, 1094, mentioned just above. The prelate had spent the spring and summer in the south of France, in good measure at the court of Urban II. Doubtless the archbishop had a variety of charges to execute, but the most important piece of ecclesiastical business he had to transact dealt with the see of Santiago de Compostela. That great shrine church was the focus of a pilgrimage whose popularity in the world of western Europe was growing at an enormously rapid rate in the eleventh century. Its patron, reputedly the only apostle buried in the west except for Peter and Paul at Rome, was the center of a popular cult that increasingly was transforming St. James the Great into the patron of the Reconquista, Santiago Matamoros. The literature of that mutation begins to appear in the early twelfth century but, as always, literature follows life.⁽⁶⁰⁾ All of these developments were turning that isolated, sleepy see into one of the greatest churches of the kingdom. Nevertheless, the bishopric there had been vacant since 1088.

Certainly that vacancy existed not by royal choice but out of the pertinacity of its bishop, Diego Peláez, deposed at Husillos in 1088 but unreconciled [246] to the loss of his see. We do not know where he was in 1094 but he will shortly appear in Aragón at the court of Pedro I. Urban II had rejected the action of the Council of Husillos and had summoned Bishop Diego to Rome for a canonical trial, but we do not know whether or not that doughty cleric actually had gone. Nevertheless, he had contrived somehow to keep his cause alive at Rome and his seat vacant at Compostela.⁽⁶¹⁾ Now the king had struck on a device to end the galling stalemate and it was Bernard's job to secure papal acquiescence.

The "Historia Compostelana" says that Alfonso, his son-in-law Raymond, and his daughter Urraca, with the counsel of the clergy and people of Compostela, chose a Cluniac monk to be the new bishop. This choice was approved by the abbot of Cluny and eventually by Rome.⁽⁶²⁾ The royal initiative in securing the pro forma "election" of the monk Dalmacio to the see was certainly concerted in advance with Abbot Hugh of Cluny. Archbishop Bernard would have had his former mentor's enthusiastic support, then, when he approached Urban II about the matter in the summer of 1094. Presumably he would also have had the support of Count Raymond of Toulouse, now married to Alfonso VI's daughter Elvira and also active at the papal court as it journeyed through his territories.⁽⁶³⁾ In addition, relations between Alfonso and Urban seem to have been good at the time. In the same year the king had agreed that the important Castilian monastery of Oña should come under papal protection and render an annual tribute to the pontiff.⁽⁶⁴⁾

Moved by all these considerations, and whatever merits of the case were presented to him by the Toledan archbishop, Urban II agreed to the election of Dalmacio. It is possible that he personally consecrated him bishop at some time in the late summer. Though we do not have any direct statement to that effect it would have been appropriate and fitting to have done so when Urban consecrated the great altar at Cluny on October 18, 1094.⁽⁶⁵⁾ Yet if Dalmacio's consecration took place so late he could hardly have reached Coimbra by November 13, 1094, where his confirmation as bishop first appears in a charter of Count Raymond.⁽⁶⁶⁾ At any rate the papal recognition of a new bishop for Santiago de Compostela enabled the king to demonstrate to his magnates the [247] strength of his influence and the number of his allies beyond the limits of the realm.

It is also entirely reasonable to believe that Archbishop Bernard had had yet another diplomatic mission, of a quite different nature but just as important. After all, he had gone abroad in March and Urban II did not arrive in France until August. The interim was spent by Bernard in negotiating a new marriage for his royal master. Bishop Pelayo tells us that Alfonso's third wife was "Bertram, Tuscia oriundam." The "Anonymous of Sahagún" calls her "otra muger de la nacion de Lombardia, llamada Berta."⁽⁶⁷⁾ The availability of this north Italian noblewoman may have come to the notice of Alfonso's ambassadors when they were negotiating the treaty that provided Genoese and Pisan naval assistance for his abortive siege of Valencia in 1092. Now she was to be married to the king during the Christmas season of 1094.⁽⁶⁸⁾ She may well have journeyed to León in the company of the archbishop in October and November.

This new marriage of the king continued his long preoccupation with guaranteeing the future of the dynasty by the provision of male heirs, but it also marked a surprising departure from his past practice. For the first time he had sought a bride other than one of the daughters of the south French feudal princes whose relatives might be of occasional help. It would be a mistake to see that choice as anything other than a deliberate act of policy. Alfonso had just buried a wife who was closely related to Count Raymond, Count Henry, and Abbot Hugh of Cluny. The king could just as easily have married some other product of that network of family alliances. That he chose not to do so argues his desire to diminish their influence in his court and realm. Yet such a change was a public act which must provoke a reaction and a crisis.

Count Raymond of Galicia and Portugal had seen his prospects wax steadily since his arrival in the kingdom. First as the betrothed and then as the husband of Urraca, the eldest daughter of a king who had no sons, his position as potential successor was patent to all observers from the beginning. The death of the king's only living brother, García, in March 1090 made the count's claim even stronger. García had a [248] son, Fernando García, who became quite an influential figure during the reign of his aunt, Queen Urraca.⁽⁶⁹⁾ The complete silence of the chroniclers in regard to him, however, suggests strongly that he was illegitimate and born of a liaison during the period of García's imprisonment. The Historia Silense reported that Alfonso considered naming García, not his son, as his successor.⁽⁷⁰⁾ Yet it was to Raymond that the "Chronicon Compostellanum" says Alfonso promised the kingdom.⁽⁷¹⁾ Even if the promise remained implicit, rather than formal and explicit as the chronicler states, after the death of García in 1090 its content seemed destined to fulfillment.

Yet events had taken a disturbing turn for the Burgundian since the middle of 1092 when Alfonso had taken Zaida as a mistress. The birth of a son to that Muslim princess in September 1093 cast a long shadow over the hopes of the count and his wife. Still, little Sancho Adefónsez was illegitimate, born of a Muslim mother, and quite possible would not live out his first year as so many infants did not. In that very spring the king had extended the sway of Raymond from Galicia and the county of Portugal to the south at Coimbra, Lisbon, Sintra, and Santarem. The count could afford to wait. But then Queen Constance, his aunt, had died in the fall of 1093. Discussions must have begun in the Christmas court of that year, if not earlier, as to the choice of a new queen and evidently did not take a direction congenial to the count despite his presence.⁽⁷²⁾ Discussions and even negotiations still are not a marriage so that he could afford to continue to wait and to rely on chance or the gentle suasions of his great-uncle Abbot Hugh of Cluny or the French Cluniac Urban II to prevent the ultimate realization of such plans.

In early 1094 the king and Raymond were still able to cooperate as evidence by their agreement on the Cluniac Dalmacio for the see of Compostela. Sometime in that same year the count's jurisdiction may even have further extended to Zamora on the western edge of the meseta itself. A fuero, or charter of privileges, that Raymond and Urraca granted to the inhabitants of a village just outside that town is dated to that year. There is little reason not to accept it except that no other notices exist of the count's authority in that area until 1096.⁽⁷³⁾ Before the [249] end of the year he was obviously in control of some of the fisc lands in Portugal as evidenced by his grant to the church of Coimbra on November 13, 1094.⁽⁷⁴⁾ Certainly he would have had control of the fisc lands assigned to the infantaticum, that is, set aside for the support of an infanta, since his marriage to Urraca. A further authority over fisc lands assigned to the comitatum, that is, the maintenance of the comital power, had been his since his appointment to the countship in Galicia and Portugal. Whether or not he had come to control all land of the royal fisc in these territories is impossible to say on the basis of the extant evidence.⁽⁷⁵⁾ The Coimbra charter also gives the first evidence that Raymond now was surrounded by a court of his own, boasting a majordomo, an alférez, and a notary.

But none of these trappings or authority sufficed either to prevent the marriage of his king to the Italian Berta in December 1094 or to content the count in acquiescence to that development. He was not at court for the royal marriage. Neither the one document we possess for late 1094 nor the numerous documents for early 1095 through late April disclose the Burgundian's presence in the royal entourage at Sahagún.⁽⁷⁶⁾ In February we glimpse the count in Galicia and Portugal. On February 11, 1095, he granted a charter to the bishopric of Túy while visiting that city.⁽⁷⁷⁾ By February 25, 1095, he had moved south to Coimbra where he issued another charter to the monastery of Montemor.⁽⁷⁸⁾ Such a deliberate and unforgivable slight, for Infanta Urraca was with him, would have alerted everyone at court to the trouble that was brewing.

Indeed there seems already to have been some choosing of sides for [250] the bishops of Túy and Coimbra were implicated by the charters just mentioned. The new bishop of Santiago de Compostela, Dalmacio, was conspicuous by his absence from the comital entourage while the future bishop of that see, Diego Gelmírez, continued to serve as Raymond's notary. The commander at Santarem, Soeiro Mendes, and the commander around Coimbra, Soeiro Fromariques, both appear there. More ominously perhaps, so does Count Froila Díaz of Astorga and the Bierzo, who was in a position to control the passages from León to Galicia-Portugal.

In the meanwhile Alfonso VI and his new queen had undertaken a progress through the realm. A royal charter of May 7, 1095, seems to place them already in Toledo.⁽⁷⁹⁾ On May 21, 1095, the court was at Valladolid, on its way north again, where it participated in the consecration of the splendid new collegiate church constructed there under Count Pedro Ansúrez's patronage.⁽⁸⁰⁾ This celebration and council of the realm attended by seven bishops and eight counts had obviously been preceded by serious negotiation for Count Raymond was there as well. What was discussed is unknown, but subsequent developments indicate that no resolution was effected. On June 2, 1095, Alfonso was back at Sahagún but Raymond appears not to have been.⁽⁸¹⁾ The progress of the king and his bride continued with a trip north to Oviedo where they are found on July 16, 1095.⁽⁸²⁾ From July 23, to July 31, 1095, the court was back again in Sahagún.⁽⁸³⁾

[251] While the king was playing his game by canvassing the realm for support and displaying his prestige as embodied in yet another foreign bride, Count Raymond had yet some counters of his own to play. That strange document, the pact of succession, was probably a product of the period between December 1094 and July 1095. The text survives only in a late copy of unknown but probably Cluniac provenance and is itself undated. It takes the form of a brief note to Abbot Hugh of Cluny in which counts Raymond and Henry include a copy of the oaths they have mutually sworn as Hugh had bid them and which note is entrusted to the abbot's convoy, Dalmacio Geret. In it Henry swears to support Raymond faithfully to help him obtain after Alfonso's death "hanc totam terram regis Aldephonsi," and to surrender to Raymond two-thirds of the treasury of Toledo if Henry seizes it first. In return, Raymond swears to support Henry faithfully, to grant to him "Toletum terramque totam subjacentem ei" under the customary conditions after the death of Alfonso, and Henry will thereupon surrender to Raymond "omnes terras de Leon et de Castella." Raymond further swears to give Henry a third of the treasury of Toledo and to grant "Galaeciam" to him, upon Henry's helping him to secure León and Castilla and then surrendering them to him, if Raymond then finds himself unable to grant Henry Toledo.

Ever since the pact first came to light, its proper date has been the subject of argument among historians. Its most recent editor has surveyed the literature and opts merely to assign it to the period 1095-1107.⁽⁸⁴⁾ Since Azevedo did his masterly edition, however, the argument has been taken up once more by Charles Julian Bishko in a careful and extended study and the pact dated to 1105 on the basis of the conjunction of a number of complex assumptions.⁽⁸⁵⁾ At the proper points we shall have to encounter all those arguments, but for now it may suffice to point out that Bishko's dating itself depends on redating one document from 1108 to 1105 and accepting the date of 1106 for another, which must have been at least seven years earlier.⁽⁸⁶⁾

But by the end of 1094, as already shown, the motivations existed that would produce the pact. Raymond and Henry were faced by the fact that Alfonso now had a male heir, that Queen Constance was dead, and that their chief support at court had been replaced by an Italian consort who was a living symbol of the king's desire to avoid excessive dependence on the Burgundians. Abbot Hugh of Cluny, whose initiative [252] the text makes so apparent, was faced by the loss of the annual cens as the Murâbits overran the taifas one by one and León-Castilla lost the parias that had facilitated its payment. Moreover, the pact cannot be dated later than 1095 for it is apparent that it was made before Henry had become count of Portugal.

Although Raymond does not explicitly refer to himself as count of Galicia, the text makes clear he was. He was in a position to grant Galicia to Henry if he must, and the whole point of the agreement was to provide a means to take possession of León, Castilla, and Toledo on Alfonso's death. Galicia he already held! Of Portugal there is no mention at all and the argument, dearly beloved by Portuguese historians who prefer a constitutional basis for the eventual independence which that territory would win, that this silence was simply due to Alfonso VI having granted it in perpetuity to Henry is simply incredible.⁽⁸⁷⁾ To imagine that an agreement to divide an entire kingdom between two cousins would have totally ignored an extant major holding of one out of respect for legal niceties ignores the dynamics of such plots. Portugal was not mentioned separately because it was then nothing more than a part of Galicia and therefore of Raymond's holdings! It is also noticeable in the language of the pact that it is anticipated that Henry will be important precisely in securing León and Castilla for his cousin. The expectation is not that, as was the fact eventually, he will be in Portugal and hence isolated and handicapped in seizing control of the meseta. The proper inference is that Raymond will likely be far away in Galicia and that Henry will be in León and in a position to seize the initiative.

The pact of succession of 1095 was an act of treason. Ir was not a legal brief that argued Raymond's right to the throne as against others. The pact assumed his right and made plans for action to realize it upon the king's death. Although Alfonso may not have known the terms, the importance of the three principals made it inevitable that be would at least quickly come to know of their negotiations and agreement. He could guess its general import well enough, but the conspirators could not be simply attacked frontally without major damage to the realm and to himself.

[253] The Leonese monarch nevertheless had to maintain the initiative against such a challenge, and the deference and respect due to the "anointed of God" could be utilized to ensure the most audacious actions to that end. The progress he had been making through the realm with his new queen furnished the formal vehicle for Alfonso's riposte. In midsummer he and queen entered the Galician territories of Raymond and on August 10, 1095, met with his son-in-law, with the latter's supporter, Count Froila Díaz, and with the Galician bishops of Santiago de Compostela, of Lugo, and of Mondoñedo, at the monastery of Samos.⁽⁸⁸⁾ Negotiations there must have been at least partially successful for the party now continued into the very heart of Count Raymond's domains. On September 24, 1095, the count issued a charter of safe conduct to the merchants of Santiago de Compostela and both Alfonso and Queen Berta confirmed it.⁽⁸⁹⁾ The crisis had now passed. Royal panache had overridden the challenge to its authority, and royal politics would now nullify the pact of succession.

Presumably the royal court returned to León or Sahagún in late September or early October while good weather in the northwest still made that journey somewhat comfortable. In fact, we cannot locate Alfonso for the remainder of the year for the lack of documents. The well-known fuero of Santarem, dated to November 13, 1095, does not place the court in that frontier town nor could it have been issued even at León in that year. That charter must be redated either to 1093 or 1094 for only then, among the possible datings, would the archbishop of Toledo have been able to confirm it as he did.⁽⁹⁰⁾ In November 1095 Bernard of Toledo was in France attending the Council of Clermont.

Sometime during the ensuing year, Alfonso was able to write a definitive end to the prospect of the Burgundian courts and cousins, Raymond and Henry, combining against him. The price was high but he [254] was able to transform the two from allies into rivals by the expedient of marrying his natural daughter Teresa to Henry and then bestowing upon the latter all of the Portuguese lands from the Río Miño in the north to Santarem its the south. If Teresa was the second daughter of the royal concubine Jimena Múñoz, as suggested by the order of her mention by Bishop Pelayo, then she could not have been much more than eleven years of age at the most if the liaison of the former with the king had begun in 1081.⁽⁹¹⁾ Nonetheless, her royal blood effectively placed Count Henry almost alongside Raymond as a contender for the throne of Alfonso VI while her dowry reduced by half the territories under Raymond's control.

Whether Urraca's husband was present for these royal maneuvers and had even had to participate in their ceremonial aspects with every appearance of good grace, we cannot know for lack of documentation. Certainly a charter of Count Raymond to the Galician monastery of San Lorenzo of Carboeiro dated to January 11, 1096, must be redated to after 1099 on the basis of its confirmation by the bishops Diego Gelmírez of Compostela and Diego of Orense.⁽⁹²⁾ The royal court can be confidently placed in the Sahagún area from early February through early April of 1096, but neither Count Raymond nor Count Henry appears to be in residence.⁽⁹³⁾

If a very problematic charter of Count Raymond to the monastery of Sahagún could be dated to May 1, 1096, then it would seem that the count was at court by that date.⁽⁹⁴⁾ Unfortunately the document has been reworked in some degree and, as it stands, has been confirmed by Archbishop Bernard of Toledo who only three weeks later was in Toulouse. If the general facts of the document were to be accepted, however, it would evidence that by this date Henry had not yet become [255] count in Portugal and that negotiations between Alfonso and the two Burgundians were still likely in train.

As late as October 1, 1096, a private document will cite Henry as count in Tordesillas and Raymond as holding Coria and Zamora.⁽⁹⁵⁾ It is dubious that the scribe would have ignored the grant of the county of Portugal to Henry, and so that might not have yet occurred. On the other hand it is obvious that Count Henry began very energetically to see to the proper ordering of his new domains which process was as crucial to the defense of the realm as it was to its internal tranquility. The fueros issued to Guimarães and Constantim de Panoias in 1096 illustrate the sort of administrative business and detail that began to claim his attention immediately upon his accession to practical power on this frontier of the realm.⁽⁹⁶⁾

For the remainder of May there are notices of the court at Sahagún.⁽⁹⁷⁾ Then by July 9, 1096, it had moved east to Burgos where, it should be remarked, disputants from the Galician monastery of Samos and the Galician Count Pedro Fróilaz sought him out for judgment.⁽⁹⁸⁾ However strong Count Raymond might be, this incident demonstrates that his province of Galicia was never simply closed to royal authority. Julio González thought that the king then moved south to Buitrago, twenty kilometers south of the pass of Somosierra in the Gudarrama chain, where he authorized repopulation of the district on July 18, 1096.⁽⁹⁹⁾ But by July 23, 1096, Alfonso was in Oviedo where he granted a charter to Bishop Martín and where members of his court also confirmed a private document on July 31, 1096.⁽¹⁰⁰⁾ It is strictly impossible, then, that [256] could have been in Buitrago only five days before the first of these documents.

From Oviedo, the king moved on to Lugo by August 15, and then to Compostela by August 21, 1096, where be confirmed a charter of Count Raymond to Bishop Gonzalo of Mondoñedo.⁽¹⁰¹⁾ Either the seriousness of his business there or the refreshing coolness of Galicia after the beat of the meseta kept Alfonso there as late as September 10, 1096, when be is at the monastery of Samos, again with Count Raymond.⁽¹⁰²⁾ The count continued at the royal court which by September 17, 1096, had returned to Sahagún.⁽¹⁰³⁾ There the monarch probably remained through the Christmas season as was customary with him.⁽¹⁰⁴⁾

From the end of 1096 Alfonso VI was to face the most complex military and political problems that the realm of León-Castilla had experienced since its reconstitution at his bands in 1072. During those twenty-four years the kingdom had grown spectacularly, first by the annexation of the Rioja which had pushed its frontier east to the Ebro from Haro to Calahorra, then by the conquest of Toledo which had trust its southern limes well beyond the Tajo, and finally by the cession of Santarem which moved the border in the southwest to the hills overlooking the Tajo there. Still, this remarkable series of advances had been the result of the most determined exploitation of chance opportunity and the close calculation of the necessary risks. Only the Rioja, however, had offered an easily accessible, contiguous territory, relatively easy of defense and with a fairly assimilable population.

The former taifa of Toledo lay beyond the reaches of the trans-Duero whose roiling plains would scarcely begin to be repopulated fully before another forty years. Worse, that isolated advance post faced a newly united Muslim Andalucía whose access to it was quite as easy as was Alfonso's. If the population of the former taifa in 1096 was better than half Christian, that Christian majority was also still overwhelmingly Mozarab and so presented peculiar problems of government to the Leonese who bad become its masters. This problem was replicated [257] on a smaller scale in the west at Santarem. As far as the full occupation and repopulation of the land there was concerned, southward migration was a painstakingly slow process, and it was initially confined to the narrow litoral along the Atlantic. The hills and mountains of Portugal just to the west would remain for many decades yet the home of pastoralists, ungoverned and dangerously hostile to the settled, agricultural world of the coast whether the latter were Christian or Muslim. Under these circumstances, Santarem with its probably quite slender Mozarab majority was also an isolated frontier post.

This whole fragile structure of new acquisitions had now to be defended continuously against a united, Murâbit Andalucía which might be reinforced at any time from North Africa. Worse yet, it must now be defended by essentially volunteer armies which could not easily be summoned or kept in the field for more than six to eight weeks. The Murâbit hegemony in the south had ended those payments, or parias, which had allowed Spanish Islam to be terrorized with its own resources by supporting a paid Leonese army almost continuously in the field during the campaigning season. Under the new circumstances, increasingly, an effective defense would be a local defense and the power of local authorities would grow correspondingly.

In the east the situation was better but exceedingly delicate nonetheless. There Alfonso VI had no enemies. The task of his policy there was to see that no one of his friends became too powerful. The small kingdom of Aragón had been essentially friendly since it had shared with León in the partition of Navarra in 1076. The taifa of Zaragoza was a vassal state, in good measure out of its fear of Aragonese aggression. At Valencia the Cid ruled by virtue of his unique military prowess, supported by a paria-paid, mercenary Christian army and surrounded by a hostile, massively Muslim majority. The county of Barcelona was far away indeed. The taifa of Lérida-Tortosa lived in everlasting fear of these other four who surrounded it.

The internal political circumstances of the realm were not less complex. There too the cessation of the parias had deprived the crown of a source of patronage which had been much of the stuff of royal power in the past half century. Worse, Alfonso VI was a king of fifty-nine years with a male heir of perhaps three years. Any prudent magnate must expect a regency of fair duration in the not too distant future, and chief among the regents would be counts Raymond and Henry, each married to a daughter of the king and perhaps with sons of their own by that time. The king had taken a new Italian wife, of course, but two years had passed without the issue of an heir from that quarter so that the significance of the marriage diminished monthly.

[258] Thus far Alfonso had been able to prevent or fail the rise of any dangerous faction or party. Considering its past history, Castilla had been remarkably quiescent. Asturias and León remained the central pillars of the royal power as always. Only in the border district of the Bierzo had the king suffered a defection where Count Froila Díaz had openly espoused the party of Raymond. In fact, although Alfonso had been unable to prevent some devolution of power, he had been able to divide and isolate the most ambitious of his nobles, confining them to the far reaches of the kingdom rather than to its heart.

If there had briefly been the prospect of a French party, held together by its adherents' defensive reaction to a natural native xenophobia, that spectre had been laid by Alfonso's strong initiatives. Count Henry was now more a rival than an ally to his cousin Raymond. For the ambition that came to each as consorts of royal daughters made alliance with the king the easiest road to preferment. Just so the establishment of one and the other in the providences of Galicia and of Portugal heightened their dependence on the king as well. Since, on the evidence of the documents, neither had brought to the peninsula with them any extensive group of retainers, only the royal support could sustain their authority among the nobles of those highly particularist districts. At the same time that regal aura could be purchased only by a considerable adherence to the purposes of the crown, and in this manner Alfonso had realized a marked gain in the efficacy of royal authority in those distant province of his increasingly unwieldly kingdom.

Such a policy had its dangers, but then every policy has its dangers. To the extent that Raymond and Henry attempted to establish themselves as independent powers by accommodating themselves to local ambitions and magnates, they also alienated themselves from the court and the aims of the magnates of the meseta. In addition that same process carried the two counts further apart for the hopes of supporters so gained by no means ran parallel, especially when it came to the hopes of their respective churches.

Braga looked to revive an ecclesiastical empire over the churches of Galicia. Santiago de Compostela enjoyed its new-found independence and dreamed of a dominion suitable to the only apostolic see of the west except Rome itself. Both wished to prevent the newly revived primatial status of Toledo from acquiring real content. Yet the fulfillment of these hopes could only be achieved by Roman dispensation, and the road to Rome ran through León.

So it was that at the center of dozens of incipient intrigues sat the king, old and even venerable by the standard of the day, but nonetheless the undeniable and absolutely essential factor in every hope. Vested [259] with the sanctity that always cloaked "the anointed of the Lord," with the respect that flowed from a lifetime of spectacular successes if some defeats, and with the nostalgic affection that was the natural concomitant of a thousand shared perils and privations, Alfonso could be coerced but never ignored. But to bend the old king to their common will the ambitious would first have to achieve one. As against that difficult and highly dangerous enterprise, each partisan had merely to find the favor of the king in pursuit of one's particular ambition. To rule so was a tedious, tiring business and a business slow to find a bard to celebrate it. Nevertheless Alfonso would rule unchallenged down to his death. Only after those splendid obsequies would even the mightiest of his subjects venture to dispute his essential dispositions for León-Castilla.

1. Feb. 11, 1092. AHN, Códices, 989B, fol. 89v. Mar. 15, 1092. AC León, Códice 11, ff.87r-88r; pub., ES 36:76-78 append. and Rodríguez Fernández, La judería de la ciudad de León, pp. 186-88. Apr. 1, 1092. AHN, Códices, 989B, fol. 76r.

2. May 1, 1092. Alamo, ed., Colección diplomática de San Salvadorde Oña 1:127-29. 1092. Serrano, El obispado de Burgos 3:83-85; and late copies unknown to him in BN, Manuscritos, 720, fol. 8ir, and 222r-v. Also pub. Antonio Suárez de Alarcón, Relaciones genealógicas de la casa de los Marqueses de Trocifal (Madrid, 1656), pp. 1-2 append. This private donation of Rodrigo Ordóñez is dated only by year but was likely executed at Nájera about this time. May 1, 1092. Quintana Prieto, El obispado de Astorga, pp. 604-605, published an agnitio of this date which would seem to place the king in the Bierzo. It should probably be redated.

3. Al-Kardabus, "Kitab al-Iqtifa" 2:xxxviii append. C, is the only early source. In addition, my account follows that of Menéndez Pidal, España del Cid 1:416-21, and 2:773-77, who utilized the late Crónica de 1344. The earlier Christian sources are silent on this campaign and the later ones are much influenced by the literature of the Cid cycle. Diego Catalán, "De Alfonso X al conde de Barcelos" (Madrid, 1962), has well begun the exploration of the relationships of these romance chronicles.

4. July 31, 1092. García Larragueta, Documentos de Oviedo, pp. 285-86. It is a private donation to the church of Oviedo.

5. AHN, Códices, 989B, fol. 74r; pub. Prieto Prieto, "Documentos de Sahagún," p. 529. The document dated to 1092 by Santiago Estefanía, "Memorias," pp. 209-11, is a composite of two other documents and cannot be safely employed.

6. España del Cid 1:420-21.

7. Ibid. 2:765-76, and Huici Miranda, Historia musulmana de Valencia 2:44-45.

8. Ibid., pp. 47-58.

9. Sánchez Alonso, ed., Crónica del Obispo Don Pelayo, p. 87.

10. See the discussion in Menéndez Pidal, España del Cid 2:760-64.

11. Jiménez de Rada, "De rebus Hispaniae," p. 143. See Julio González, Repoblación de Castilla la Nueva 1:190-91.

12. "Chronica Gothorum," in David, Études historiques, pp. 300-301.

13. Sept. 21, 1045. PMH, Diplomata, pp. 209-11. June 20, 1049. AHN, Lisbon, Corporaçóes Religiosas, Colegiada de Guimaraes, Caixa 12, Maço 1, no. 1; and Coleccáo Basto, no. 40, Livro de Mummadona, ff. 39v-40r; pub. PMH, Diplomata, pp. 226-27. Jan. 12, 1052. AHN, Lisbon, Coleccões Religiosas, Suplemento, Sala 16, S. Pedro de Pedroso, Caixa 1, Maço 1, no. 12; pub. PMH, Diplomata, p. 234, but dated to 1053. June 10, 1056. AHN, Códices, 986B, fol. 7v; also BN, Manuscritos, 712, fol. 403r-v; pub. Serrano y Sanz, ed., "Documentos del monasterio de Celanova," pp. 14-15. Mar. 10, 1065, AHN, Lisbon, Corporaçóes Religiosas, Cabido de Sé da Coimbra, Mago 1, no. 12; pub. PMH, Diplomata, pp. 273-74, dated to 1063, and López Ferreiro, Historia de Santiago de Compostela 2:242-44.

14. Torquato de Sousa Soares, "Refexões a volta da segunda reconquista de Coimbra aos Mouros," Homenaje a Fray Justo Pérez de trbel, vol. 1 (Silos, 1976), p. 188, and Serrao, História de Portugal, 1:64.

15. Aug. 25, 1072. Braga. Arquivo Disirital, Liber Fidei, ff. 102r-103v, and ff. 112v-113v. "Princeps ipsius terre ferrandus menendiz." Apr. 19, 1107. DMP 2:216. "Imperante inter durio et tamice sarrazino osoriz."

16. His career has been traced most exhaustively by García Gomez and Menéndez Pidal, "El conde mozárabe Sisnando." More recently the documents they used have been called into question by Gérard Pradalié, "Les faux de la cathédrale." Independent of the Coimbra documents, his earliest appearanee is May 1, 1070. PMH, Diplomata, p. 303.

17. See chapter 2, notes 45 and 46. Also Costa, O Bispo D. Pedro 1:27 and 44.

18. See chapter 2, note 65, and Costa, O Bispo D. Pedro 1:274.

19. See chapter 5, note 4.

20. See chapter 8, notes 30 and 29.

21. PMH, Diplomata, pp. 419-20.

22. See chapter 10, note 39, and Sept. 1088, PMH, Diplomata, p. 427.

23. "Regula Sancti Augustini, à propos d'une fausse chartre de fondation du chapitre de Cöimbre," RPH 3 (1947): 29. David also accepted the Peter who confirmed the royal charter of May 8, 1080, as another contender for the see, but that is a simple scribal error and it is Peter of Braga who confirmed. See chapter 6, note 66.

24. Oct. 1089. PMH, Diplomata, p. 434, and Oct. 12, 1091. Ibid., p. 454.

25. David, Études historiques, p. 300, and Mattoso, A nobreza medieval protuguesa (Lisbon, 1981), pp. 262-63.

26. The date of Cresconio's accession to Coimbra has been confused by a note in the Livro Preto of Coimbra, PMH, Diplomata, p. 461, which says that he was elected at the Couneil of Husillos on April 13, 1092, and ordained in the octave of Pentecost at Coimbra. All of that is patently erroneous but it has misled many who did not know all of the documents. Even Peter Feige, "Die Anfänge des portugiesisehen Königtums and seiner Landeskirche," GAKS 29 (1978): 26-27, clings to the notion of a second council at Husillos in 1092. Feb. 26, 1095. PMH, Diploinata, pp. 485-86, gives Apr. 15, 1091, as the beginning of his pontificate. That is two days after Easter. "Anno episcopatus supradicti presulis III, Mense X, die mensis xii." Mar. 3, 1095. Ibid. pp. 486-87, corroborates that year. Fortunato de Almeida, História da Igreja em Portugal, vol. 1, new ed. (Porto, 1967), p. 268, n. 7, cites a document mentioning him as bishop in 1091.

27. If Cresconio dated his pontificate from the time of his election, his consecration could have taken around Pentecost of either 1091 or 1092. In both years that would have fallen in the second week of June. Since the note in the Livro Preto cites the approval of Martín Múñoz rather than of Sisnando perhaps 1092 is to be preferred. June 11, 1092. PMH, Diplomata, p. 463, is the first reliable document that cites Cresconio as bishop.

28. Costa, O Bispo D. Pedro, pp. 248-53, based on the account in the "Vita Geraldi" in the PMH, Scriptores, p. 54.

29. Aug. 28, 1089. Braga, Arquivo Distrital, Gaveta 2 das propriedades do Cabido, no. 138; pub. Costa, O Bispo D. Pedro 2:411, who calls it an original but in fact it is a copy from the early twelfth century to judge by the script. The copyist has the date wrong for Bernard of Toledo was at court in León on Aug. 25, 1089, according to an original royal charter. See chapter 11, notes 8 and 9. There would be no such impediment in 1092, and the bishops of Túy and Orense are said to have assisted at both the consecration of Cresconio and of the cathedral.

30. Jan. 4,1093. AHN, Códices, 989B, ff. 116v-117r; pub. Prieto Prieto, "Documentos de Sahagún," p. 530. Feb. 5, 1093. AC León, Códice 11, ff. 86v-87r; pub. ES 36:79-80 append. Feb. 8, 1093. AHN, Clero, Carpeta 887, no. 7. Feb. 8, 1093. AHN, Códices, 989B, ff. 94v-95r. Feb. 18, 1093. Ibid., fol. 104v. Feb. 22, 1093. Pub. Seraano, "Los Armudez de Toledo y el monasterio de Tórtoles," BRAH 103 (1933): 108-10, who redated it to 1097 on the basis of its confirmation by Bishop García rather than Gomez of Burgos. A scribal error in the name is likely for the others who confirm support 1093. Mar. 23, 1093. Pub. Mañueco Villalobos and Zurita Nieto, eds., Documentos de Valladolid 2:21-23. This document probably was issued at Sahagún as would have also the immediately preceding one. Neither the time nor the circumstances would have permitted a trip south and back. Apr. 3, 1093. AHN, Códices, 989B, ff. 9v-l0r.

31. See chapter 9, note 42.

32. David, Études historiques, pp. 300-301.

33. Ibid.

34. Aug. 10, 1094. PMH, Diplomata, pp. 481-82. "Tenente arauca martino monniz." Also Pradalié, "Faux de Cöimbre," p. 80.

35. See note 32, and Mattoso, A nobreza medieval portuguesa, pp. 203-204, and 212-13.

36. See note 32.

37. AHN, Clero, Carpeta 1.064, no. 1, a fourteenth-century copy dated to July 25, 1081. The confirmants necessitate a date between 1088 and 1093 and only in the latter year does July 25 fall on a Monday as called for by the dating formula.

38. See note 10.

39. AHN, Códices, 989B, fol. 4v; and 988B. ff. 17v-18r; pub. Escalona, Historia de Sahagún, pp. 490-91. AHN, Clero, Carpeta 887, nos. 10, 11, 12, 13, 14, and 15. The earliest of these copies is no. 12 from the twelfth century. Also AHN, Códices, 989B, fol. 7r-v; and a late copy in Acad. Hist., Colección Salazar, 0-22, ff. 30-31. In this second charter, "Quod monasterium cum omnibus villis et hereditatibus sive possessionibus suis sicut eum obtinuit uxor mea."

40. AHN, Códices 989B, fol. 4v; Acad. Hist., Colección Salaar, 0-22, fol. 30r; pub. Escalona, Historia de Sahagún, pp.493-94.

41. The only contemporary witness reports "En el anno de veinte y siete de su rreino, la rreina doña Costança, su muger, cerro el su postrimero dia." Puyol y Alonso, eds .," Las crónicas anónimas de Sahagún," BRAH 76 (1920): 116.

42. Ibid.

43. Dec. 2,1093. AC León, reales, nos. 1.371 and 1.372; Códice 11, ff. 28r-29r; pub. ES 36:81-84 append., and Hinojosa, ed., Documentos de León y de Castilla, pp. 40-43. Dec. 27, 1093. AHN, Códices, 989B, fol. 116r-v; pub. Prieto Prieto, "Documentos de Sahagún," pp. 531-32.A document of Alfonso VI dated only to 1093, pub.Quintana Prieto, El obispado de Astorga, p. 605, might have been drawn up at almost any time.

44. Nov. 30, 1093. AHN, Clero, Carpeta 1.240, no. 1. An original, private document of Samos. His predecessor, Ariás, last appears in a document of July 31, 1092. See note 4. A mention of Martín in a private document of June 1092, pub. Pedro Floriano Llorente, ed., Colección diplomática del monasterio de San Vicente de Oviedo (Oviedo, 1968), pp. 185-87, should be regarded with suspicion for this copy is very rough. If a document of Nov. 26, 1087, pub. Fita, "San Miguel de Escalada," pp. 479-81, can be redated to 1092, on grounds of its confirmation by Gomez González as alférez, and that Friday fell on that day in the latter year, then Bishop Arías was still in office.

45. Aubieto Arteta, Historia de Argón, pp. 108-12. It is possible, as the author believes, that the Aragonese monarch made a strong feint against Tortosa while Pedro concentrated on Almenar, but it is more likely that the evidence he cites relates to the joint expedition with Alfonso VI of the prior year.

46. Turk, Reino de Zaragoza, pp. 167-68.

47. As usual, Huici Miranda, Historia musulmana de Valencia 2:57-81, offers a careful account of developments, a critique of the literature of the Cid cycle, and a refutation of Menéndez Pidal's use of the latter.

48. Abd Allah, El siglo XI en 1^a persona, pp. 293-96.

49. Jan. 5,1094. AHN, Clero, Carpeta 887, no. 16. Feb. 2, 1094. AHN, Códices, 989B, ff. 150v-151r. Feb. 17, 1094. AHN, Clero, Carpeta 887, no. 18; copy in ibid., Códices, 989B, fol. 127r; pub. Prieto Prieto, "Documentos de Sahagún," p. 532.

50. Serrano, ed., Cartulario de San Millán, pp. 283-85.

51. Apr. 13, 1094. AC León, no. 995, an original royal charter; a copy in Códice 11, fol. 73r-v. May 2, 1094. AHN, Códices, 989B, fol. 125r. May 4,1094. Ibid., fol. 119v.

52. Ubieto Arteta, Historia de Aragón, pp. 116-18. Turk, Reino de Zaragoza, pp. 168-70. The new king is the only peninsular monarch of the period whose charters have been critically edited. Ubieto Arteta, ed., Colección diplomática de Pedro I.

53. Huici Miranda, Historia musulmana de Valencia 2:82-114.

54. Ibid., pp. 114-37.

55. Menéndez Pidal, España del Cid 1:495-506.

56. Ibn-Idari, Al-Bayân-al-mugrib, pp. 84-85. Huici Miranda, Historia musulmana de Valencia, p. 118, rejects the story of this Alfonsine expedition, alleging that Alfonso would not have had time to raise an army. He has no firmer basis for doing so than that only Ibn-Idari relates it.

57. ES 20:360. For the date of composition, see Reilly, "The 'Historia Compostelana,' p. 83.

58. AHN, Lisbon, Sé da Coimbra, Maço 1, nos. 1 and 2, both copies of the fourteenth century; pub. PMH, Diplomata, pp. 484-85.

59. AHN, Clero, Carpeta 887, no. 22; and Códices, 989B, ff. 27r-28r.

60. Pérez de Urbel and González Ruiz-Zorilla, eds., Historia Silense, pp. 191-93, is the earliest manifestation of which I am currently aware and it dates between 1110 and 1120.

61. Ubieto Arteta, "El destierro del obispo compostelano Diego Peláez," pp. 43-51, is a useful study of the question.

62. ES 20:20

63. Hill and Hill, Raymond VI de Saint-Gilles (Toulouse, 1959), pp. 19-20.

64. Alamo, ed., Colección diplomática de San Salvador de Oña 1:131-32.

65. Rivera Recio, La Iglesia de Toledo, p. 144.

66. See note 58.

67. Sánchez Alonso, ed., Crónica del Obispo Don Pelayo, p. 86. Puyol y Alonso, ed., "Las crónicas anónimas de Sahagún," BRAH 76 (1920): 116.

68. Ibid. The "Anonymous" reports that in 1100 "la rreina doña Berta, apenas conplidos seis annos del matrimonia, pago la dueda." Jan. 25, 1100. AHN, Códices, 989B, ff. 10v-11r; Clero, Carpeta 890, no. 4, with a date of Jan. 15, 1100; pub. Escalona, Historia de Sahagún, pp. 499-501, with the erroneous date of Feb. 1, 1100; and Bruel, Recueil des chartes 5:83-86, from a French copy with the date of Jan. 25, 1100. The first document that mentions her as queen is dated Apr. 28, 1095. AHN, Códices, 989B, ff. 28v-29r.

69. See Reilly, León-Castilla under Queen Urraca, pp. 221-22.

70. Ubieto Arteta and Ruiz-Zorrilla, eds., p. 124. "Considerabat namque Adefonsus hunc interim salva pace post se regnaturum.

71. ES 20:611. "cui totum suuni Regnum jurejurando pollicitus fuerat."

72. The documents indicate his presence at court from November 1093 to February 2, 1094. See notes 40 and 49.

73. AC Zamora, Tumbo Negro, ff. 121v-122r; Tumbo Blanco, fol. 61v; BN, Manuscritos, 714, fol. 164v; Acad. Hist., Colección de documentos y privilegios, 1, fol. 2; pub. Muñoz y Romero, ed., Collección de fueros municipales y cartas pueblas, pp. 332-33. The twenty-four charters of Raymond known to me at present are completely unedited and are so few in any event that paleographic or diplomatic norms have not been established which would permit a sure critique. It is obvious that they do not follow the usages of the Alfonsine chancery except in its very broadest outlines.

74. See note 58.

75. For the utilization of the fisc lands see Reilly, León-Castilla under Queen Urraca, pp. 259-69.

76. Nov.29, 1094. See note 59. Jan. 11,1095.AHN, Clero, Carpeta 888, no. 2. Jan. 13, 1095. Ibid., no. 3. Feb. 15, 1095. Pub. Jésus Muñoz y Rivero, Manual de paleografia diplomática española de los siglos XII al XVII, 2d ed. (Madrid, 1917), pp. 152-53. Feb. 27, 1095. AHN, Códices, 989B, ff. 66v-66r. Mar. 9,1095. AHN, Clero, Carpeta 959, no. 16, and a copy in Carpeta 960, no. 11, dated to Mar. 13. Apr. 11, 1095. AHN, Códices, 989B, fol. 81r; pub. Escalona, Historia de Sahagún, pp. 496-97. Apr. 28, 1095. AHN, Códices, 989B, ff. 28v-29r.

77. AC Túy, 4/1; also copies in 3/1, 4/4, and Becerro 1; pub. Pascual Galindo Romeo, Túy en la Baja Edad Media (Madrid, 1950), pp. iii-vi. Another later copy exists in AG Simancas, Sección de Gracia y Justicia, legajo 1.672, varios, Túy, no. 14, ff. 18v-20v.

78. AHN, Lisbon, Coleccão Basto, no. 2, fol. 57r-v.

79. AHN, Microfilmas, Palencia, rollo 1.658, no. 10; and a copy in Acad. Hist., Catedrales de España, Palencia, 9-25-1-C-6, ff. 32r-34r; pub. Fernández de Pulgar, Historia secular y eclesiástica de Palencia 1:139-40. The charter is one of the few of Alfonso VI to mention the place of issue and the patronymic of the alférez is misconstrued by the copyist, but the diplomatic seems sound. But the royal court would have to have left Sahagún before April 28 (see note 76) to have reached Toledo by May 7.

80. A donation of Pedro Ansórez, pub. Pulgar, Historia de Palencia 1:135-37; and Mañueco Villalobos and Zurita Nieto, eds., Documentos de Valladolid 1 :24-54; and another copy in Acad. Hist., Catedrales de España, Valladolid, 9-25-1-C-4, ff. 3r-6v, dated to May 21, 1094. A donation of Bishop Raymond of Palencia, pub. Mañueco Villalobos and Zurita Nieto, eds., Documentos de Valladolid, 1:55-58.

81. AHN, Códices, 989B, fol. 128v.

82. Pub. García Larragueta, Colección de Oviedo, pp. 290-92. Fernández Conde, El Libro de Testamentos, pp. 310-13, believes this private charter to be falsified, but I suspect that the confirmants were taken from a genuine document of this date.

83. AHN, Códices, 989B, fol. 95r-v, and ff. 129v-130r. A visit to Castilla is also argued sometime during this year, probably in June, by the fuero of Logroño issued by Alfonso sometime in 1094 and also by the king's confirmation of a private donation to San Millán de La Cogolla in the same year. See BN, Manuscritos, 9.194, ff. 111-12v; pub. Muñoz y Romero, ed., Colección de Fueros, pp. 334-43; and Serrano, Cartulario de San Millán, pp. 287-88, respectively.

84. DMP, 1, pt. 1:1-2; and 1, pt. 2:547-53.

85. "Count Henrique of Portugal, Cluny, and the antecedents of the Pacto Sucessório," RPH 13 (1970): 155-88.

86. Ibid., pp. 185-86.

87. Pierre David, "Le pacte successoral entre Raymond de Galice et Henri de Portugal," BH 50 (1948): 275-90, accepts this hypothesis and places the pact about 1105-1107. He bolstered it by demonstrating that Dalmacio Geret was not known to have been in the peninsula before 1100. The argument from silence is very weak indeed in a period of such few documents and the laconic character of those which do survive. His viewpoint has been widely accepted, however. See Defourneaux, Les français en Espagne, pp. 198-99; Maur Cocheril, Etudes sur le monachisme en Espagne et au Portugal (Lisbon, 1966), pp. 122-23; and A. H. de Oliveira Marques, History of Portugal, vol. 1 (New York, 1972), p. 38.

88. BN, Manuscritos, 18.387, fol. 300r, is a notice of a private charter which bears their confirmations. Other private document of Aug. 9, 1095, and Aug. 13, 1095, reveal the presence as well of the bishop of León and of the royal majordomo. Ibid., fol. 300v, and ES 40:189.

89. AC Compostela, Tumbo A, fol. 28v; and Chartularum, ff. 66v-67v; pub. López Ferreiro, Historia de Compostela 3:36-39 append.

90. AHN, Lisbon, Cabido da Sé de Coimbra, Maço 1, nos. 43 and 44. Both are copies from the late twelfth century. Pub. PMH, Diplomata, pp. 348-50. The document must postdate the acquisition of Santarem in April 1093 and predate the death of Bishop Gomez of Burgos, who also confirmed, in February 1097. In November 1096 Archbishop Bernard of Toledo was again abroad. Of 1093 and 1094 the former is the most likely on a number of counts. For one, the language of the document suggests a recent acquisition of the town. "tradidit civitatem sancte herene in manibus meis quod incredibile ab omnibus aliquando erat."

91. Sánchez Alonso, ed., Crónica del Obispo Don Pelayo, pp. 86-87.

92. AHN, Clero, Carpeta 1.784, no. 2; pub. Lucas Alvarez, "La colección diplomática de San Lorenzo de Carboeiro," pp. 276-77. Sánchez Belda, ed., Documentos reales de la Edad Media, no. 170, simply regards it as a forgery.

93. Feb. 3, 1096. AHN, Clero, Carpeta 1.740, no. 9. Feb. 5, 1096. AHN, Códices, 989B, fol. 83v. Mar. 5, 1096. Ibid., ff. 138v-139r. Mar. 6, 1096. AC León, Códice 11, ff. 81v-8zv; pub. ES 36:84-87 append. Apr. 1,1096. Pub. Escalona, Historia de Sahagún, p. 497.

94. AHN, Códices, 989B, ff. 172v-173r. Escalona, Historia de Sahagún, p. 506, published it under date of May 1, 1106, which date was accepted by Bishko, "Pacto Successório," p. 185, and used to bolster the latter's dating of that pact. That date is far too late, however, for by that time Diego Gelmírez had been bishop of Compostela for six years but he confirms this document as "clericus sci. iacobi." The date as it appears in the extant "Becerro Gotico" in the AHN is partially obliterated.

95. AHN, Clero, Carpeta 889, no. 3.

96. If a document of December 18, 1095, could be accepted, the date of Henry's accession at Coimbra could be pushed back yet further. But Rui Pinto de Azevado, "Observações de diplomática," RPH 12 (1969): 147-67, suggests that it is more reasonably dated to 1098. Also the same historian and editor, in DMP 1, pt. 1:1-6, and 1, pt. 2:541-46, dates the charter of Henry to Guimnarães to 1095 but that document is undated as we have it. Azevado regards it as prior to the charter to Constantim de Panóias, which is dated simply to 1096 in the text. Clearly both could still easily be of the same year. Just so, the confirmation of Henry and Teresa of Count Raymond's charter to Túy of Feb. 11, 1095, is of a later date. See note 77. Feige, "Anfange des portugiesisehen Königtums," pp. 110-13, accepts 1096 as the date of Henry's accession. A private document of Aug. 1, 1096, pub. PMH, Diplomata, pp. 497-98, cites Raymond as ruling in Coimbra still.

97. May 20, 1096. AHN, Códices, 989B, fol. 100r-v. May 23, 1096, Ibid., fol. 139r-v. May 31, 1096. Ibid., fol. 140v.

98. Acad. Hist., Manuscritos, 9-27-2-E-50. Estefanía, "Memorias," pp. 215-17.

99. Repoblación de Castilla la Nueva 1:126-27. BN, Manuscritos, 2.190, pp. 33-38, has a notice in Spanish of a charter to this effect, but it is garbled or has been tampered with. For one thing, it is confirmed by Archbishop Bernard of Toledo, then in France.

100. Pub. ES 38:338-40, and García Larragueta, Colección de Oviedo, pp. 296-97, respectively. July 31, 1096. AC León, Codice 11, ff. 88v-89r, clearly must be reassigned to that date in 1100 on the basis of the confirmants.

101. See note 99 and BN, Manuscritos, 5.928, fol. 27r-v; and Acad. Hist., Catedrales de España, Mondoñedo, 9-25-1-C-3, ff. 13r-14r; pub. ES 18:340-42, respectively. A charter of Alfonso VI to the bishop of Burgos dated August 20, 1096, should be redated to the same day in 1100. See Serrano, Obispado de Burgos 3:110-12.

102. See note 98.

103. AHN, Clero, Carpeta 889, no. 1; and a copy in AHN, Codices, 989B, fol. 52v.

104. Nov. 1096, AHN, Microfilmas. León, Gradefes, rollo 6.311, no. 10. Dec. 15, 1096. AHN, Códices, 989B, ff. 151v-152r. Dec. 20, 1096, Ibid., fol. 19v; pub. Prieto Prieto, "Documentos de Sahagún," pp. 535-36.