[303] Before the end of 1101 Alfonso VI of León-Castilla would have seen fifteen years of battling the combined might of Murâbit North Africa and Muslim Spain. Since Zalaca he had taken the offensive whenever the opportunity offered, but most often he had been forced back on a dogged defense which delayed the enemy's advance as long as humanly possible. At first he had worked to defend the independence of the Iberian taifas. When the last of those independent states of Andalucía had bowed to Yusuf ibn-Tashufin, Alfonso had then maintained his own domains as far to the south of the central meseta as possible, extracting full cost and buying time with each kilometer surrendered. Still the enemy came on, and the sixty-four-year-old monarch would spend the remainder of his life as he had the previous fifteen. Only just after his death with the occupation of Zaragoza in 1110 and the capture of Santarem in 1111 would the Murâbit breaker crest.

At no time in the last eight years of his life is there any sign that the determination of the king of León-Castilla flagged. Seemingly always at bay, he found the resources to keep one more withdrawal from becoming a rout or a collapse. Through some royal legerdemain Alfonso continually managed to replace an army defeated one year with an army ready to fight in the next year. Never was the Murâbit enemy allowed simply to capitalize on a victory; but rather they found its full fruits denied them by a rapid adjustment or a sudden counterstroke of their stubborn enemy. Alfonso might often be beaten but he was never defeated.

Nonetheless the Leonese king could not prevent the lineaments of later medieval Spain and, in a real sense, modern Spain from taking form while he was preoccupied by this assault. That is, in these same years the medieval kingdom of Aragón was transforming itself from a tiny Pyrennean principality into the second power of Christian Spain by its encroachment on Zaragoza. Although he is usually overshadowed by his brother and successor, Alfonso I el batallador (1104-1134), the contribution of the brief reign of Pedro I (1096-1104) is central to [304] this transformation. His grasp of the possibilities inherent in the age seems to have been faultless.⁽¹⁾

Even before the capture of Barbastro Pedro had requested the papacy to transfer the mountain see of Roda there. Consequently already on January 25, 1101, Bishop Pons of Barbastro can be found endowing the papal legate, Richard of Saint-Victor, and his abbey at a meeting in his new see.⁽²⁾ The same determined policy had previously seen the Aragonese see of Jaca slide south to Huesca after the conquest of the latter. Pedro I envisioned no retreat.

In this light the assertion that the Aragonese monarch took the cross and intended to proceed to Jerusalem in early 1101 seems fabulous, and that he was forbidden to do so by Paschal II and directed instead to crusade against Zaragoza is simply impossible in light of the chronology involved.⁽³⁾ What Pedro did do, which is totally in character, was to launch a furious assault against that taifa which lasted the entire year.

This campaign, or some phases of it, very likely had the support and aid of French knights from beyond the Pyrennees. On February 12, 1101, the scribe of a Leonese private document mentioned in the dating formula "Petrus quoque rex aragonensis eum infinita armorum multitudine Cesaraugustam civitatem cum Christi vexillo preliante." It is doubtful that merely an army of Aragonese would so have impressed so distant a witness.⁽⁴⁾ The presence of some elements of the nobility of Catalonia might also be indicated by the presence of Bishop Berenguer of Barcelona in Barbastro on May 5, 1101, when the mosque of that city was formally reconsecrated as the Christian cathedral.⁽⁵⁾

Whatever its composition, Pedro's army was operating in the environs of Zaragoza itself in June. In August the king was even in Alpenes, more than a hundred kilometers south of the city, and the Río Ebro in what could only have been a great raid, or razzia, during which he had apparently nothing to fear from the enemy in his rear.⁽⁶⁾

The siege of the [305] capital of al-Mustain continued into the fall although ultimately the taifa weathered this storm to survive yet another seventeen years.

There is not the slightest indication that Alfonso VI reacted in any fashion to these developments although obviously he was well aware of them. The Leonese monarch of necessity must concentrate on the Murâbit threat. Some portion of the early winter was probably occupied by the obsequies of his sister, Urraca. We are informed that she died in 1101 but not when.⁽⁷⁾ The demise of the last of his siblings, the one who had held Zamora for him in the time of his exile by his brother Sancho, and a trusted adviser of his mature years, would have been a blow to him. It likely took place in mid-winter for that infanta appears in none of the court documents of the year.

A happier note was struck when the king received from the Byzantine Emperor Alexius Comnenus a cross, fashioned from the wood of the "true cross," very richly adorned. This imperial gift was presented by the king in turn to the abbey of Sahagún.⁽⁸⁾ Very likely the gift had been conveyed from Constantinople by Court Fernando Díaz of Asturias, who is a second Iberian whom we can identify to have participated in the First Crusade.⁽⁹⁾

On March 19, 1101, Alfonso granted a fuero to the Mozarabs of Toledo, but it was probably issued at Sahagún.⁽¹⁰⁾ Two days later most of the same court figures, including counts Raymond and Henry, figure in a charter of the latter to the abbot of that monastery.⁽¹¹⁾ The court seems to have remained in that district through the celebration of Easter, which that year fell on April 21.⁽¹²⁾ This paschal celebration also [306] marked the end of the long, tangled dispute over the succession to the see of Santiago de Compostela for on that day, according to the "Historia Compostelana," Diego Gelmírez was finally consecrated its bishop.⁽¹³⁾ Our source does not say where the consecration took place or which prelate presided. I believe that the very absence of the mention of the place by those authors who are usually such enthusiastic local patriots argues that it was performed elsewhere, and either León or Sahagún are the most likely locations. Fletcher believes that the celebrant was Paschal II's initial choice, the bishop of Maguelonne in France, but Serrano holds for the pope's second choice, the bishop of Burgos.⁽¹⁴⁾ I myself incline to think that the latter's proximity, the fact that his see was directly dependent on the papacy, and later evidence of special friendship between Gelmírez and García of Burgos make it probable that Serrano is correct.

A private donation to Sahagún and an agnitio executed in León after judgment by the king prove that the court remained in the León area until after May 15, 1101.⁽¹⁵⁾ By June 3, the king had probably moved south to Valladolid.⁽¹⁶⁾ But before then Count Henry for one had left the court, and on June 8 he executed a charter in Portugal.⁽¹⁷⁾ Doubtless these moves were designed to allow a closer surveillance of the endangered frontiers for by late May it would have been known that the Emir Mázdali was mobilizing the Murâbit forces in Andalucía. As it was to develop, that army was to be used against Valencia rather than the center or the west of the Christian frontier. This development left Alfonso free to concentrate on encouraging the repopulation of the trans-Duero.

That objective, which had been in train since 1076, had taken on special urgency with the relentless Murâbit attacks. The earliest efforts of the crown known to us were concerned with the area from Sepúlveda west through Cuéllar, Fuente el Olmo de Iscar, Olmedo, and Medina del Campo, ranging anywhere from twenty to forty-five kilometers in depth south from the Duero. In fact the document of June 3, 1101, mentioned above dealt with an exchange of property in Iscar between the monastery of San Zoil in Carrión and Count Pedro Ansúrez. The monks of San Zoil ceded to Count Pedro a church in Iscar that had been earlier given to them by Count Martín Alfónsez, who had died about [307] 1092 or 1093. The exchange thus speaks to the development of a settlement there which had reached some maturity already by the beginning of that decade. That process may be generally the case along this earliest line of settlement.

In the eastern, or Castillian, trans-Duero, Segovia lay sixty kilometers south of Cuéllar. Snuggling at the foot of the pass of Navacerrada over the Guadarramas, it was at the southern limit of the territory. The process of repopulation had begun in 1088.⁽¹⁸⁾ Here on one of the most direct and shortest routes from the north to the new realm of Toledo growth was fast enough for that settlement to be in violent revolt against control from Toledo in 1114 and to have exacted successfully the restoration of its ancient bishopric by 1120.

But as the Guadarramas run toward the southwest the trans-Duero gains steadily in depth, and we cannot be sure when the first settlers drifted into Coca, twenty-five kilometers southwest of Cuéllar, or into Arévalo, yet another twenty-five kilometers southwest of Coca. Bishop Pelayo only tells us that both were repopulated by Alfonso VI.⁽¹⁹⁾ Early on the development of this sector was the particular responsibility of Count Pedro Ansúrez, operating from his foundation of Valladolid on the Pisuerga just north of that river's junction with the Duero. As early as 1095 he had endowed the collegiate church of Valladolid with a church in Cuéllar.⁽²⁰⁾ Indeed we can quite sensibly envision, from 1076 on, a gradual movement of small homesteaders across the Duero occupying the lands of its south bank to varying degrees of depth. From that date on they had been effectively free of organized campaigns from the Muslim south. For defense against more casual depredations, such lands lay within easy hail of a variety of Leonese strongpoints at Zamora, Toro, Tordesillas, and Valladolid. In addition, the land itself partook of the character of that immediately north of the Duero in soil, climate, and precipitation. It was, then, both easy and safe for the immigrants to work.

How far south this band of gradually advancing settlement had reached after a quarter of a century in 1101 is impossible to measure with entire accuracy. I suspect that it may have reached as far south as Arévalo, where the plain of the river basin begins to yield to the uplands rising to meet the Cordillera Central. Cattle, sheep, and goats probably were run south of that divide, but it seems unlikely that settled, [308] agricultural villages were established in the region. The actions that Alfonso VI was now to set in motion were to have the effect of establishing advance military bases beyond the line of effective Leonese settlement at Avila and Salamanca.

The rationale for this action, I suggest, was that Alfonso was taking into serious consideration the possibility of the loss of Toledo. Such a disaster would have created a far more dangerous situation than that which had existed before 1085 when Toledo had been in the hands of a friendly tributary state. In Murâbit hands the former taifa would have offered a base for extended raids into the now Christian trans-Duero. A strongly fortified military base at Ávila would therefore have made possible a defense of the pass of Arrebatacapas just as Segovia functioned farther east to block that of Navacerrada. Another major military base at Salamanca would protect the exposed western flank of the trans-Duero region. This was, in fact, the military function of that town down until the reconquest of Coria in 1142 by Alfonso VII. In 1101 his grandfather still held that advance post but had to anticipate its probable loss as a concomitant of any loss of Toledo. Salamanca would then become the base for operations against any Muslim army advancing north through the gap between the Cordillera Central and the mountains of Portugal along the old Roman silver road from Mérida to Zamora. The task of establishing these two bastions for the defense of the entire western trans-Duero was entrusted in 1101 to Count Raymond rather than to Pedro Ansúrez.

What resources Raymond had to employ must have been thin. There was perhaps already some modest, probably seasonal, population at both sites.⁽²¹⁾ Immigration was systematically encouraged but took a very long time to have an appreciable effect. Our clue here is the absence of both a bishop of its own at Ávila until 1121 and the complete absence of cathedral documents, always the first to appear, before the second quarter of the twelfth century.⁽²²⁾ Salamanca had a bishop of its own in 1102, as we shall see, and even a few court documents from the first decade of the century, although private documents begin in 1132. Nevertheless its first bishop seems to have controlled Ávila as well, and the peculiar connections between Salamanca and Zamora that exist during his tenure suggest that he ruled from Zamora.⁽²³⁾ That would [309] have been both safer and more effective if actual settlement in his new diocese at the time was concentrated along the south bank of the Duero. How soon Count Raymond could have begun raising, or repairing, the fortifications of either fortress is an open question, but such work must have been undertaken almost immediately even if of a temporary nature initially.⁽²⁴⁾

While king and count were occupied with building a defense in depth against possible future reverses, in the east the Emir Mazdali had advanced on Valencia and laid siege to it in late August as he would continue to do for some seven months.⁽²⁵⁾ Alfonso seems to have been confident that the city would hold out successfully because he took no known action in 1101. By November 5 at least he had retired to his ordinary winter quarters at Sahagún, and there he remained through December.⁽²⁶⁾

At the very beginning of the new year the court was still there, but [310] the king must already have been making plans to leave.⁽²⁷⁾ On February 13, 1101, he was at Lerma on the Río Arlanza some forty miles south of Burgos "coram magno exercitu militum qui ibant cum rege Valenciam decercare eam de moros." The text as we have it does not list confirmations beyond that of the king, the queen, and the royal notary. The direction from which the king had come is indicated by the fact that this royal fuero was issued to the village of Vallunquera a few miles south of Castrojeriz.⁽²⁸⁾ The army that Alfonso was leading on this occasion must, it seems, have been drawn exclusively from the magnates of the eastern frontier for an agnitio involving the monastery of Sahagún and drawn up on February 15, 1102, shows most of the court still there. Their numbers included Count Raymond, Archbishop Bernard, and the royal alférez, Pedro Alvarez.⁽²⁹⁾ The same situation is shown by a charter of Abbot Diego of Sahagún, dated March 6, which Count Henry also confirmed.⁽³⁰⁾ This extraordinary circumstance may be explained by the king's concern for the continuing defense of the Tajo frontier in his absence.⁽³¹⁾

From Lerma there are too many possible routes to Valencia for us to determine which one the royal host actually took. The distance involved in any case would have made the court's arrival at that destination unlikely before the middle of March, which date would coincide well with the narrative accounts we possess. The Murâbit siege force withdrew thirty kilometers down the coast to Cullera. Alfonso then [311] examined the condition of the city's defenses and reconnoitered the enemy and came to the reluctant decision that it would be impossible to hold Valencia. After about a month he evacuated its Christian population and set fire to the city as he withdrew. As the Leonese king retreated, the Emir Mázdali occupied what was left of Valencia sometime between April 21 and May 20, 1102. "By the reduction of Valencia," said a Muslim chronicler, "Yusuf Ibn Táshufín saw himself master of the whole of Andalus, with the exception of Zaragoza, which remained in the hands of its king, Al-mustain Ibn Húd. The Murâbits dared not molest this prince, on account of the great distance of his kingdom, and of his being in league with the Christians, whose tributary he was."⁽³²⁾

That assessment of affairs in the northeast of the peninsula proved to be too conservative even in the short run. The king of Zaragoza had been able to judge for himself the tide of events and had responded accordingly. In Morocco Yusuf ibn-Tashufin was arranging to have his son, Ali recognized as his successor and al-Mustain despatched his own son, Abd al-Malik, to participate in both the celebration and the recognition. As a result a treaty emerged from the attendant negotiations that was of great utility. When Abd al-Malik returned in the fall of 1102 he found the new Murâbit governor of Valencia, Abd Allah ibn-Fatima, already on the outskirts of Zaragoza. On being informed of that treaty the governor of Valencia withdrew.⁽³³⁾ But we may be sure that the same treaty also ended, at least for the time, the payment of parias to Alfonso VI by the taifa king. Yusuf had always been adamant that Spanish Islam should not finance his enemies by payments to the infidel.

Nor was that significant realignment the full measure of Christian reverses in the northeast. On September 14, 1102, Count Ermengol V of Urgel was killed and his forces destroyed at a battle in the valley of the Segre at Mollerusa, twenty kilometers east of Lérida.⁽³⁴⁾ This event is to be related to the events attributed to the fall of 1102 by the Muslim chronicler al-Kardabus, who says that after the occupation of Valencia the Murâbit governor attacked and ravaged the lands of Barcelona, carrying [312] off church bells, crosses, and precious vessels and dispersing castle garrisons.⁽³⁵⁾ Count Ermengol probably came to grief while attempting a counterstroke. Thus by the beginning of winter Valencia had been lost, Zaragoza had thrown its lot in with the Murâbit juggernaut, Urgel was leaderless, and Barcelona had been bloodied. Only Pedro of Aragón had escaped unscathed. In December that king was in Estella on the edge of Alfonso VI's territories, seeking consultation and cooperation perhaps.⁽³⁶⁾

The king of León-Castilla, after supervising the evacuation of Valencia in early April, may have returned to León. Important elements of the court were there on May 11, 1102.⁽³⁷⁾ I regard it as more probable, however, that he went directly to the Toledan vicinity where he was found granting a fuero to the inhabitants of Aceca twenty kilometers northeast of Toledo on June 5.⁽³⁸⁾ That would have been the logical place to resettle at least some of the Valencian refugees. The most prominent of those refugees was elsewhere though. On June 22 Count Raymond issued a charter to Bishop Jerome, "magistro nostro," granting him two churches in Zamora.⁽³⁹⁾ The confirmations suggest that Zamora was the place of issuance. Another charter of the same date which provides for a formal restoration of the diocesis of Salamanca is a forgery of the period 1120-1135 fathered on the former document.⁽⁴⁰⁾ In the latter all of the churches of the diocese of Salamanca and Zamora are turned over to Bishop Jerome but no mention is made of Ávila.

By the end of July the royal court had returned to Sahagún and Count Raymond and his wife had rejoined it.⁽⁴¹⁾ There is no sign of Bishop Jerome there. More importantly for our purposes, neither is there any evidence of the presence of Count Henry. Al-Kardabus tells us that some Murâbit divisions from Córdoba invaded Castilla, presumably the lands of Toledo, in that year and fought sanguine battles [313] there with "the accursed Errink."⁽⁴²⁾ Again Count Henry is found in command of that sector of the frontier.

The whereabouts of Alfonso VI's court for the remainder of the year is unknown. Count Raymond spent most of the fall in Galicia. On October 6, 1102, he confirmed the rights of its bishop over the city of Lugo and on October 24 granted property to his faithful follower, the Galician magnate Ero Armentáriz.⁽⁴³⁾

In early January 1103 an ecclesiastical council was held in Carrión de los Condes which certainly discussed a territorial dispute between the bishoprics of Santiago de Compostela and Mondoñedo.⁽⁴⁴⁾ Doubtless, after the fashion of the day, it was held in the royal presence and was as much a meeting of the royal curia as it was a church council. It has been asserted that it approved the restoration of the see of Salamanca, the installation of Jerome as its first bishop, and the latter's administration of the diocese of Ávila and Zamora, none of which can be demonstrated but some of which may have happened.⁽⁴⁵⁾ It has also been suggested that Zaida's son, Sancho Alfónsez, was named heir to the kingdom of León-Castilla in that council and that Count Henry of Portugal, under cover of a pretext of a pilgrimage to Jerusalem, repaired to Rome to seek papal assistance in maintaining an independent position in the peninsula.⁽⁴⁶⁾ The latter supposition, like those concerning Salamanca, suffers from the implicit assumption that modern institutional arrangements were foreseen and were the conscious objects of twelfth-century desires.

What we can see is that the court was at Sahagún on January 21, 1103, and Bishop Pelayo of Astorga was in attendance.⁽⁴⁷⁾ It was probably at Sahagún, then, that Alfonso VI made a grant to the church of Astorga on January 25, 1103. That charter together with another private document [314] of the same date, issued at court, perhaps enables us to see who attended the royal court at Carrión.⁽⁴⁸⁾ In addition to Archbishop Bernard of Toledo the bishops of Astorga, León, Palencia, Oviedo, Santiago, Lugo, and probably Burgos were present. Bishop Jerome seemingly was not. Queen Elizabeth, Count Raymond and Infanta Urraca, and Count Henry and Infanta Teresa were present. But more significantly, Sancho Alfónsez confirmed both documents, his fathers charter as "Sanctius infans quod pater fecit confirmo." While he had probably not been formally proclaimed heir, he now had become a public figure. He would have been just about ten years of age.

In February the court continued at Sahagún. Both the young Sancho and Count Henry figured in the documents of February 10 and 25.⁽⁴⁹⁾ In March the court may have traveled east to Burgos where Alfonso extended the fuero of Burgos to villages of its alfoz on March 19, 1103.⁽⁵⁰⁾ The confirmation list to this document has not survived. But on March 23, the king issued a charter to the monastery of Oña whose witness list suggests that the court was in Castilla.⁽⁵¹⁾ Here Infans Sancho confirmed but Count Henry seems not to have been present although Count Raymond and his wife were. In late April and early May the court had returned to the León-Sahagún vicinity but Count Henry still appears to be absent.⁽⁵²⁾

In fact a Portugese document, dated May 1103, shows the magnate Soeiro Mendes exercising jurisdiction in Coimbra together with Infanta [315] Teresa "usque ad venitam comitis de Iherusalem." Now clearly if Count Henry had been in León as late as February 25 and would be in Portugal, as we shall see, in mid-July there was hardly time for him to have gone to the Holy Land and precious little time for even a flying visit to Rome. For now the document presents an enigma.⁽⁵³⁾

Alfonso VI had departed the Sahagún area in late May. Muslim sources tell us that he began a siege of Medinaceli in 1103. That reaction was characteristic of the Leonese monarch. If Valencia had been lost and Zaragoza had embraced the Murâbits then what was essential was to prevent their easy communication and coordination at all costs. So long as Muslim forces held Medinaceli it was impossible for Alfonso to block either at the easiest point, the narrow, winding gorges of the upper Jalón. For an active defense, such as the king almost always favored, the hilltop fortress at Medinaceli became the prime objective.

The king could not have reached that objective himself before about the end of the first week in June, but perhaps advance forces had been sent ahead to initiate the siege. Or perhaps the Murâbit, by this time, could gauge his likely responses or were quite as well aware of the strategic necessities of the new situation as were the Leonese. In any event, the Muslim response was swift. The combined forces of the governors of Valencia and of Granada advanced to the relief of Medinaceli. Our source says from Calatayud but there is clearly some confusion here. That town on the lower Río Jalón was a logical staging and supply area for an army that planned to advance up the river to the relief of Medinaceli. But such an advance was the most difficult way in which to try to approach the enemy.

Perhaps what the brevity of the account obscures is that the governor of Valencia made such an approach alone not to force a passage so much as to immobilize Alfonso by the threat of doing so. Then the governor of Granada could make an unimpeded advance across the meseta of southern Castilla la Nueva and ravage the territories of Toledo. Such a strategy might well have worked. But somehow troops were found to meet this second threat without giving up the siege of Medinaceli. Sometime in June the Muslim army was brought to battle and defeated [316] near Talavera de la Reina, seventy kilometers west of Toledo. The governor of Granada was himself killed in the battle.⁽⁵⁴⁾

It is possible that the charter given to the church of Toledo by Alfonso on June 22, 1103, was an act of thanksgiving for this victory.⁽⁵⁵⁾ The king was accompanied by the queen, Sancho Alfónsez, Count Raymond and Urraca, the archbishop of Toledo, four bishops, three counts, and a variety of Toledan figures. Count Henry was in the north of Portugal at the time near Guimãraes.⁽⁵⁶⁾ This victory must have been crushing enough to relieve the military threat for the remainder of the summer for, while the siege of Medinaceli was to continue, some members of the court who confirmed the diploma of June 22 confirmed a private donation to Sahagún on July 10, 1103.⁽⁵⁷⁾ Some also figured in another private charter later in the month.⁽⁵⁸⁾

[317] The king himself may have traveled instead to Burgos to whose good men he granted a charter on July 25, 1103, whose list of confirmations is lost.⁽⁵⁹⁾ There is a somewhat obscure reference to the Emir Yusuf, who was in Córdoba in September 1103 and Algeciras in November, having to order 1,500 cavalry north from Valencia to protect al-Mustain from the forces of Alfonso.⁽⁶⁰⁾ It would have been in keeping with the unflagging energy of that monarch to have immediately followed up a victory with military pressure on Zaragoza to resume payment of the parias.

The next notice of the court comes in October when the king, his wife, and his son confirm an exchange between Count Raymond and the bishop Oviedo.⁽⁶¹⁾ Count Henry and Infanta Teresa had rejoined the court by then. Where the document was executed is not evident. On November 7, 1103, major court figures including counts Raymond and Henry confirmed a sweeping donation made by Count Pedro Ansúrez to Bishop Raymond of Palencia.⁽⁶²⁾ Again the place of issuance is undetermined. The supposed donation by infanta Urraca, the sister of Alfonso VI, dated to November 13, 1103, cannot help us for it obviously is not of that date.⁽⁶³⁾ Urraca had died in 1101. The charter of Alfonso VI to the Castillian monastery of Oña, dated December 12, 1103, may indicate that the court was again in eastern Castilla.⁽⁶⁴⁾ If so it may have been for negotiations on common concerns as to the taifa of Zaragoza with the king of Aragón. On December 11, Pedro I was at Estella on the Castillian border.⁽⁶⁵⁾

In 1104 Alfonso VI maintained the initiative he had secured in the preceding year. Nevertheless military operations began slowly. The court itself seems to have been in its ordinary wintering spot until well [318] after Easter, which fell on April 17.⁽⁶⁶⁾ The only royal charter of this period, granted to the bishop of Oviedo on March 16,1104, shows Alfonso accompanied by his queen and his two daughters by her, the infantas Sancha and Elvira. This is their first known citation. The king's son Zaida and his sons-in-law and their wives also confirmed.⁽⁶⁷⁾ The two counts appear to have been at court all winter.

Some portion of the court seems to have remained at Sahagún all through that summer.⁽⁶⁸⁾ But through the winter and spring the siege of Meinaceli had gone on, and that key fortress surrendered in July 1104.⁽⁶⁹⁾ Alfonso may have been there for the capitulation or he may have left that agreeable formality to a lieutenant. We know from Muslim sources that he was again pressing his advantage in that summer with a great raid through the territories of Sevilla. The Murâbit governors of both Sevilla and Granada were forced to join forces to contain it.⁽⁷⁰⁾

That punitive expedition may have been over well before September 14, 1104, when a private charter shows many of the ordinary court figures, including Count Henry, at Sahagún.⁽⁷¹⁾ It was certainly over prior to October 11, when Alfonso himself seems to have been at Burgos, confirming the possessions of a monastery there. Almost all those who confirm the document are local figures.⁽⁷²⁾ Count Raymond was at the other end of the realm where he confirmed a local privilege of the Galician monastery at Samos on October 25, 1104.⁽⁷³⁾

By the end of that year there were some momentarily threatening developments in the east of the peninsula that cast a shadow over the limited initiative Alfonso had been able to reclaim on the meseta. In September Pedro I of Aragón had died without a son to succeed him. His brother Alfonso I el Batallador inherited the Aragonese throne and, as his sobriquet implies, was to prove himself a great warrior and the ultimate conqueror of Zaragoza in 1118. But for the moment at least in León the king must have wondered how successful the Batalador would be in making good his claim to his deceased brother's throne and whether or not, if successful, he would continue Pedro's policy of deference toward León-Castilla. Of necessity the Leonese monarch must work for a strong ruler in Aragón so as to contain the great advantage the Murâbit had secured with the seizure of Valencia.

At the end of 1103 Alfonso VI had responded to the emergency created by the death of Count Ermengol V of Urgel at the battle of Mollerusa in late 1102. In a momentous but exceedingly obscure set of circumstances the great magnate Pedro Ansúrez was exiled from the realm and took up residence for the next six years in the country of Urgel. The domestic implications of this event will be treated in the following chapter. The net effect in the east was to secure there an able figure who was at the same time unfailingly loyal to the larger purposes of the king of León-Castilla. In a very curious fashion the exile of Pedro Ansúrez between 1104 and 1109 strengthened the strategic position of Alfonso VI in the east just as he had the exile of Rodrigo Díaz de Vivar between 1080 and 1099. The modern mind puzzles at the willingness of medieval subjects to forgive their king his necessities.

Sometime in 1104 Count Pedro Ansúrez had taken up his duties as regent for Ermengol VI of Urgel who was his grandson. The knowledge of Pedro's new charge had reached Pope Paschal II already in May of 1104, so it must have occured quite early in that year.⁽⁷⁴⁾ By 1105 the count had begun to add his direction to the constant attrition against the northern borders of Islam in the east of the peninsula, which had been the preeminent role of Aragón in recent years. Before November [320] 3, 1105, he had taken the fortress town of Balaguer, twenty-five kilometers north of Lérida on the Río Segre. It was a major victory, and to achieve it he had drawn Count Ramón Berenguer of Barcelona into an alliance.⁽⁷⁵⁾

As the year began in León-Castilla, Count Raymond and his wife were in Galicia where they issued a charter on January 16, 1105, to the monastery of San Juan de Poyo.⁽⁷⁶⁾ The couple were accompanied not only by the bishops of Compostela, Orense, and Túy, but also by their daughter Sancha, of whom this is our first notice.

Count Henry and Infanta Teresa, on the other hand were at court where they granted a charter jointly to Cluny and its dependency of San Isidro de Dueñas of some Portuguese possessions. Bishko, who first discovered this charter, made it the linchpin of his argument that Henry was seeking a coalition against his cousin and the young Sancho Alfónsez early in that year.⁽⁷⁷⁾ His argument, though it suggests that the charter might have been issued at Sahagún, presumes the absence of Alfonso VI and indeed Henry's absence from the royal court until mid-May. All that one need concede to this careful argument designed to place the Pact of Succession between Raymond and Henry in the spring of 1105 is that there was indeed an impressive gathering of Portuguese magnates at Alfonso's court on January 5, 1105. Why that should have been so we cannot say.

However, the royal notary, Pelayo Eríguez, confirmed Henry's charter. When on February 6, 1105, Alfonso VI granted a charter to Bishop Jerome the name of the notary is unfortunately missing, but the bishops of both León and Palencia, who days before had confirmed Henry's charter, now confirmed the king's even if Count Henry did not.⁽⁷⁸⁾ Less than a week later, both counts, Henry and Raymond, joined Alfonso in confirming a private charter to Sahagún.⁽⁷⁹⁾ It is also worth noticing that although the king's charter to Jerome entitles the latter "bishop of Salamanca" the church granted to that prelate was again a [321] church in Zamora. Salamanca continued to be an isolated outpost whose resettlement and fortification had hardly begun.

The royal court was still in the Sahagún area on March 31, 1105, when Alfonso VI gave a charter to the canons of Astorga, confirmed by his son Sancho and Count Raymond.⁽⁸⁰⁾ It was from court that the latter may have granted a charter to the church of Orense on March 18. Two of the bishops of Galicia were there as was Archbishop Gerald of Braga also.⁽⁸¹⁾ The court remained in that district until June 1, 1105, and on the latter date counts Raymond and Henry were present.⁽⁸²⁾

In May of the year pressure was maintained south of the line of the Tajo by forces of León-Castilla, but we do not know who commanded. Our only information is that such an expedition was defeated in the mountains well south of Toledo.⁽⁸³⁾ Of further fighting during the year we hear nothing, so Pedro Ansúrez's capture of Balaguer in late fall marked the major success of that campaigning season.

When the royal court again becomes visible on September 9, 1105, it is again at Sahagún and counts Henry and Raymond are in attendance.⁽⁸⁴⁾ The king had likely been attending to the strengthening of the trans-Duero during the summer. Later in September Alfonso issued a charter to a Castillian supporter, possibly at Burgos.⁽⁸⁵⁾ In December the court had returned to the Sahagún vicinity and Jerome of Salamanca was there as was Gerald of Braga, and the bishop of Compostela as well as the king's Burgundian sons-in-law. It was there that Count Raymond may have issued his fuero in favor of the men of Compostela,⁽⁸⁶⁾[322] Alfonso's court continued at the center of the realm into the middle of March of the new year, but in January the documents evidence a much larger than ordinary curia being held with no fewer than nine archbishops and bishops attending as well as his sons-in-law.⁽⁸⁷⁾

On March 19 the king issued a charter to the church of Oviedo confirmed by his son among others.⁽⁸⁸⁾ On April 9 Count Raymond issued one charter to the canons of Lugo and another to its merchants.⁽⁸⁹⁾ All three were produced at court which remained in the Sahagún area through the first weeks of June.⁽⁹⁰⁾

During the month of August Count Henry and Infanta Teresa were in Portugal.⁽⁹¹⁾ Where Alfonso spent the summer or fall we cannot say with certainty. Probably he himself was the leader in a great raid which not only penetrated to the extreme south into the territories of Málaga but whose superiority was so assured against reprisal that it was able to escort north large numbers of Mozarabs for resettlement in Christian territories.⁽⁹²⁾ On the evidence the Leonese seemed to have regained the [323] initiative in a surprising fashion since the dark years between 1097 and 1102. If they were not sufficiently strong to make major advances still a most aggressive defense continued to keep the Murâbit off balance and contributed substantially to the capacity of the Christian frontiers to withstand future attack.

For the moment at least the Murâbit's empire was fully occupied with the changes consequent upon the death of Yusuf ibn-Tashufin. Alfonso VI's great opponent died on September 2, 1106. He was succeeded by one of his sons, Ali ibn-Yusuf. The latter had himself been born of a Christian slave, probably from the peninsula.⁽⁹³⁾ Under Ali the Murâbit empire in Spain would reach its greatest extent, but it would also begin its gradual decline.

In the northeast of the peninsula the Murâbit would finally manage to annex the taifa of Zaragoza outright in 1110, but in the meanwhile such help as they could furnish to their coreligionists there could not prevent the continuing attrition of its frontier lands at the hands of Aragón. Alfonso I on his accession in 1104 had immediately taken up the process begun by his father and so brilliantly continued by his brother Pedro. Sometime in the summer of 1106 he had overrun Ejea de los Caballeros fifty-five kilometers northeast of Zaragoza and but forty kilometers east of Tudela, the third city of the taifa. On another front he also siezed Tamarite, thirty-five kilometers north of Lérida, its second city.⁽⁹⁴⁾

Tamarite was also, however, only the same distance west of Balaguer, the fortress town taken by Count Pedro Ansúrez at the end of the preceding year. It is not surprising, then, that in 1106, probably after the fall of Tamarite, Count Pedro ceded to the bishop of Huesca a church in Balaguer.⁽⁹⁵⁾ The donation by the count of one-third of this city to Alfonso I of Aragón is undated but is also likely of the same period.⁽⁹⁶⁾

The military cooperation that the latter document mentions was eminently sensible for two warriors both concerned with harassing if not conquering the Muslim of Lérida. That Count Pedro had also sworn fidelity to the Aragonese monarch and, that, at some future point, the [324] young Ermengol would be expected to do the same is also made clear. Though such close relationships may seem surprising, Sancho Ramírez, Alfonso I's father, had been married to a daughter of Ermengol III of Urgel and Pedro I, Alfonso I's brother, had once seen the possibility of inheriting the country itself.⁽⁹⁷⁾ Pedro Ansúrez was continuing the traditional policy of Urgel in behalf of his grandson.

Late in the fall of 1106 it may be that the king of León-Castilla was at Burgos or some other location in eastern Castilla. Two documents, one a charter of the king to the monastery of Oña and the other an agnitio concerning the monastery of San Millán, both record his actions though they are dated only by year and did not specifically place him there.⁽⁹⁸⁾ By December 17, 1106, the court would seem to have been at León.⁽⁹⁹⁾

No significant military events seem to have taken place in 1107. As best we can tell from the documents, the year was largely untroubled and on January 18 the court was where it usually was to be found, that is, at Sahagún.⁽¹⁰⁰⁾ Infans Sancho Alfónsez confirmed a private charter there as did counts Raymond and Henry. Shortly thereafter Raymond and Urraca may have journeyed to Galicia for on March 17, 1107, they granted a charter to the church of Compostela confirmed by largely provincial figures.⁽¹⁰¹⁾ The royal court continued in the Sahagún area into April.⁽¹⁰²⁾ Then in León at the very beginning of May Alfonso held a great curia in which his son Sancho Alfónsez was declared his heir. This was a step of major importance in the politics of the realm and will be examined as such subsequently. The evidence for the meeting is scattered and partial but, for the moment, we can be satisfied with the documents.

On May 8, 1107, Alfonso VI granted a charter to the church of Toledo whose dating formula concludes "Roborato vero in castro de monzon coram omni sue expedicionis multitudine dum attenderet ad aragon post celebratum concilium apud legione."⁽¹⁰³⁾ The confirmation list reveals that the Burgundian cousins and their royal wives must have [325] been present at the curia. The castle of Monzón was a royal stronghold some twelve kilometers north of Palencia.

Six days later, on May 14, 1107, Alfonso granted the privilege of coinage to the bishop of Santiago de Compostela "quando rex de burgis agressus cum sola castellanorum expeditione super vascones et aragonenses iter direxit." Again all the major figures of the dynasty are present, and this time Sancho Alfónsez confirms as "regnum electus patri factum."⁽¹⁰⁴⁾

What this expedition directed against the Basques and Aragonese effected, if anything, we just do not know. There are no other notices of it but these two. Perhaps it was to bolster the position of León-Castilla in the Rioja or to shore up the northwest frontier of Zaragoza. In any event it was soon over for the court had returned to Sahagún before the middle of June and remained there for the rest of the summer.⁽¹⁰⁵⁾

Events during the remainder of the year 1107 were of the sort that would affect primarily the domestic affairs of León-Castilla and the imminent crisis of succession: the death of Count Raymond, the death of Queen Elizabeth, and perhaps the growing infirmity of the old king himself. The examination of these developments is best left for a more detailed consideration of the progression of that crisis. For now it may be said that these new factors tended to undermine the considerable progress of the past four years. As so often since the battle of Zalaca in 1086 and the emergence of the North African Murâbit empire as a major factor in peninsular affairs, from 1102 to 1107 Alfonso VI had managed [326] to stabilize the frontiers of his swollen kingdom. Though Valencia had had to be abandoned, the capture of Medinaceli in 1104 had effectively sealed the corridor of the Jalón. Thus the enemy was forced again to rely on a direct assault along the line of the Tajo as its one avenue of advance.

But the difficulties experienced by the Murâbit in North Africa, consequent on the designation of Ali as heir of Yusuf ibn-Tashufin, had prevented the use of any large African contingent in that effort. As ever, the Leonese kingdom proved itself able to more than manage any force fielded by Spanish Islam. Indeed if the armies of Alfonso VI were not uniformly successful in those four years, it is yet clear that they managed to retain the initiative. Most usually they are found to be operating well south of the Tajo and on more than one occasion deep within Andalucía. Though the precious city of Toledo continued to mark the southernmost extension of the realm, no siege army had appeared before it.

More important still, behind this most aggressive, if essentially defensive, campaigning crucial contributions were being made to the repopulation of the lands north of the Tajo and even north of the Cordillera Central. Mozarabs from Valencia and Málaga were resettled in those territories to swell the Christian population. Encouragement and assistance were given as well to the continuing movement into the trans-Duero of settlers from the north. To be sure, little could be effected in a mere four years to help the kingdom grow into the enormous reaches of territory aggrandized since 1076. But there are times that the smallest increments of strength can prove decisive in a contest. We cannot be sure that such was the case in the early twelfth century, but we can see that such was the policy of Alfonso VI in this penultimate effort of his last years.

2. Antonio Durán Gudiol, La iglesia Aragón durante los reinados de Sancho Ramírez y Pedro I: 1062-1104 (Rome, 1962), pp. 83-85. Guerard, ed., Cartulaire de labbaye de Saint-Victor de Marseille 1:450-51.

3. Turk, Reino de Zaragoza, pp. 180-81.

4. AHN, Clero, Carpeta 890, no. 14; a copy in AHN, Códices, 989B, fol. 112r-v.

5. Antonio Durán Gudiol, "La Santa Sede y los obispados de Huesca y Roda en la primera mitad del siglo XII," AA 13 (1965): 47.

6. This document was discovered by Santos García Larragueta, "Un documento original e inédito de Pedro I," in Homenaje a Don José Esteban Uranga (Pamplona, 1971), pp. 55-56, who dated it to 1099. The citation of Pons as bishop of Barbastro makes that impossible. For this and other treatment of the campaign, see Ubieto Arteta, Historia de Aragón, pp. 131-33.

7. As usual there is some dispute. But see BN, Manuscritos, 1.358, fol. 3v, and Enrique Flórez, Memorias de las reinas católicas de España (1761; Madrid, 1964), pp. 234-35.

8. Puyol y Alonso, "Las crónicas anónimas," pp. 116-17.

9. See note 4, and chapter 14, note 85.

10. The best current study of the complex history of all the privileges of Toledo is Alfonso García-Gallo, "Los fueros de Toledo," AHDE 45 (1975): 459-61, for this text. Copies of the text unknown to this editor exist in Acad. Hist., Colección Salazar, 0-22, ff. 63r-v, and 253r-254r; AGS, Mercedes y Privilegios, legajo 377, no. 9; and BN, Manuscritos, 714, fol. 9r-v.

11. AHN, Códices, 989B, fol. 14r; pub. DMP 1-1:11-12.

12. Apr. 13, 1101. AHN, Clero, Carpeta 890, no. 15; and a copy is bound into the manuscript of the "Historia Compostelana" in the Biblioteca del Palacio Real de Oriente, Madrid, Manuscritos, II-534; pub. Escalona, Historia de Sahagún, pp. 501-502. Apr. 20, 1101,. AHN, Códices, 989B, fol. 146v. A forgery purporting to be an Alfonsine confirmation of previous gifts to Oviedo, pub. García Larragueta, ed., Documentos de Oviedo, pp. 312-19, was dated by the editor to about 1100. See also Fernández Conde, El Libro de Testamentos, pp. 333-42.

13. ES 20:27-31.

14. Fletcher, Saint James' Catapult, p. 113. Serrano, Obispado de Burgos 1:368.

15. May 14, 1101. AHN, Códices, 989B, fol. 138r-v. May 15, 1101. Floriano Cumbreño, ed., El Libro Registro de Corias 1:168-70.

16. Mañueco Villalobos and Zurita Nieto, eds., Documentos de Valladolid 1:67-71.

17. DMP 1-1:12-13.

18. Julio González, "La Extremadura castellana," p. 299. On the argument that the site had never been entirely deserted, which I am inclined to grant, see Moxó, Repoblación y sociedad, pp. 43 and 206.

19. Sánchez Alonso, ed., Crónica del Obispo Don Pelayo, p. 81.

20. See chapter 12, note 80.

21. Angel Barrios García, Estructuras agrarias y de poder en Castilla: El ejemplo de Avila, 1085-1320(Salamanca, 1983), pp. 128-71, is a fascinating attempt to reconstruct the circumstances of the age of settlement. For his chronology, however, the author is far more trusting of later chronicles than I should care to be.

22. Reilly, León-Castilla under Queen Urraca, pp. 244-319.

23. Ibid., pp. 34 and 321-22. Also Julio González, "Repoblación de la 'Extremadura leonesa," Hispania 3 (1943):195-273. An original private document of June 17, 1104, even cites one "Iheronimus camorense sedis." AD León, Santa Maria de Otero de las Deuñnas, no. 216.

24. On the walls see Manuel González García, Salamanca: La repoblación y la ciudad en la Baja Edad Media (Salamanca, 1973), pp. 41-42.

25. Huici Miranda, Historia musulmana de Valencia 1:164-65.

26. Nov. 5, 1101. AHN, Códices, 989B, fol. 132v. Dec. 7, 1101. Ibid., fol. 78r.

27. Jan. 2, 1102. AHN, Códices, 989B, fol. 88r. A royal charter dated Jan. 12, 1102, pub. Múñoz y Romero, ed., Colección de fueros municipales, pp. 398-99, has been interpolated. The text names a Doña Sancha as sister of Alfonso VI but she was sister rather of Alfonso VII. Otherwise those who confirm are proper for the period. Among them is "Sancius filius Imperator," and this would be the first public appearance of the young heir apparent if it could be relied upon. The date that actually appears in the text is 1110, also not given in the Spanish era.

28. Julio González, "Aportación de fueros castellano-leoneses," pp. 629-31. Alfonso's issuance of a fuero to Castrojeriz may be of this date. See Reilly, "The Chancery of Alfonso VI," p. 17. A fragment of an Alfonsine charter which entirely lacks a date may be associated with this journey as well. Pub. Rodríguez de Lama, ed., Colección diplomática medieval de la Rioja 2:101-102.

29. AHN, Clero, Carpeta, 890, no. 18; copy in Códices, 989B, ff. 56v-57r.

30. AHN, Codices, 989B, fol. 172v; pub. Escalona, Historia de Sahagún, p. 502.

31. On the other hand, there may be a dating error in the two Sahagún documents. An altogether peculiar charter of Count Raymond, AHN, Clero, Carpeta 1.749, no. I bis, of about this time seems to show the count in command of a royal army "exercitatus imperatorum _____" (illegible). The charter, a fifteenth-century copy, is dated to April 1, 1101, but bears the episcopal rota of Gelmírez of Compostela, which imitation of contemporary papal practice even he would not have adopted before his episcopal consecration on April 21 of that year.

32. "Historia Roderici," in Menéndez Pidal, España del Cid 2:969. Al-Kardabus, "Kitab al-Iqtifa" ed., 2:xlii append. C. Also see Huici Miranda, Historia musulmana de Valencia, pp. 164-65.

33. Huici Miranda, Historia musulmana de Valencia, 3:17-20, who corrects the dating of Turk, Reino de Zaragoza, 159-63.

34. Próspero de Bofarull y Mascaró, Historia de los condes de Urgel (Barcelona, 1853), p. 361. Losses on the Christian side were said to number 300 knights, which would indicate a combined force probably including troops of the country of Barcelona.

35. "Kitab al-Iqtifa," pp. xlii-xliii append.

36. Luis Rubio, ed., Documentos del Pilar (Saragossa, 1971), pp. 9-10.

37. AHN, Códices, 989B, fol. 76v.

38. García-Gallo, "Fueros de Toledo," pp. 462-63.

39. AC Zamora, Tumbo Negro, fol. 22r-v.

40. June 22, 1102. AC Salamanca, Cajon 16, legajo 1, no. 5. Its most recent editors, José Luis Martín, Luis Miguel Villar García, Florencio Marcos Rodríguez, and Marciano Sánchez Rodríguez, Documentos de los archivos catedralicio y diocesano de Salmanca, siglos XII- XIII (Salamanca, 1977), pp. 83-85, call it an original which it most certainly is not, although the subsequent confirmation of it by Alfonso VII in 1136 probably is both original and genuine. A later copy is in BN, Manuscritos, 3.546, ff. 142v-143r.

41. July 26, 1102. AHN, Clero, Carpeta 890, no. 20; copy in Códices, 989B, fol. 33r. July 31, 1102. Ibid., fol. 91r.

42. "Kitab al-Iqtifa," p. xliii append.

43. AHN, Códices, I.043B, fol. 16r-v; another copy in 363B, fol. 124r; and BN, Manuscritos, 18.387, fol. 303r-v, respectively.

44. It is known from a letter of Archbishop Bernard of Toledo to the bishop of Mondonedo, who had declined to attend, dated Feb. 4, 1103. ES 20:74-76.

45. Fidel Fita, "Concilios nacionales de Carrión en 1103 y León en 1107," BRAH 24 (1894): 292-316, remains the basic treatment. Jerome is supposed to have confirmed a donation of some of the men of Ávila to the Castilian monastery of San Millán in 1103, calling himself bishop of Ávila in the text. Serrano, ed., Cartulario de San Millán, pp. 294-95.The document is at least suspect since the abbot, Blasio, who receives the gift is of the period 1074-1089. DHEE 3:1653.

46. Torquato de Sousa Soares, "O governo de Portugal pelo Conde Henrique de Borgonha: Sus relaçóes com as monarquias Leonesa-Castelhana e Aragonesa," RPH 14 (1974): 378.

47. AHN, Códices, 989B, fol. 91v.

48. Jan. 25, 1103. AD Astorga, Carpeta 1, but up until 1981 actually in the Museo of the cathedral. It had disappeared by the spring of that date. It may have been an original. Copies in AHN, Códices, 1.195B, ff. 385r-387v, and 1.197B, ff. 230v-232v. BN, Manuscritos, 712, fol. 148r-v. Acad. Hist., Collección Salazar, 0-22, fol. 18r. Pub. in translation in Pedro Rodríguez López, Episcopologio asturicense 2:536-38. Jan. 25, 1103. AHN, Códices, 1.195B, fol. 387r-v; and 1.197B, ff. 233r-238r.

49. Feb. 10, 1103. AC Santiago, Tumbo A, fol. 27r-v; Chartularium, fol. 71r-v; pub. López Ferreiro, Historia de Santiago de Compostela 3:54-55 append. Feb. 13, 1103. AHN, Códices, 989B, ff. 122v-123r. Feb. 25, 1103. AHN, Clero, Carpeta 891, nos. 1, 2, 3, and 4, with some variants. Pub. Escalona, Historia de Sahagún, pp. 503-505.

50. Pub. Múñoz y Romero, ed., Collecion de fueros municipales, pp. 236-38, from a copy then in the municipal archive of Burgos and dated to 1073. Repub. García Sáinz de Baranda, La ciudad de Burgos 2:403-405. There are late copies of the document in Acad. Hist., Colección Salazar, 0-13, ff. 28v-30v and 0-22, ff. 247-248. Full citation to existing copies of this document in Juan Antonio Bonachia Hernando and Julio Antonio Pardos Martínez, eds., Catalogo documental del Archivo Municipal de Burgos: Sección histórica, 931- 1515 (Salamanca, 1983), p. 49.

51. Alamo, ed., Colección diplomática de San Salvador de Oña 1:149-52. A late copy unknown to this editor exists in Acad. Hist., Colección Salazar, 0-7, fol. 32r-v.

52. Apr. 27, 1103. AHN, Códices, 989B, ff. 39v-40r. May 6, 1103. ASI, Reales, no. 134, pub. Amparo Valcarce, El dominio de S. Isidoro de León, pp. 95-97. May 11, 1103. AHN, Clero, Carpeta 891, no. 5. May 15, 1103. AHN, Códices, 989B, fol. 114r.

53. DMP 2:96-97. Sousa Soares, see note 46, imagined the pilgrimage story as a fiction published to conceal a trip to Rome from Alfonso VI. That must remain conjecture. The same author, "Governo pelo Conde Henrique." pp. 374-75, redates the Portugese defeat at Vatalandi to this year, following the analysis of Henrique Barrilaro Ruas, "A data do desastre de Vatalandi," RPH 4 (1949): 361-73. The "Chronica Gothorum," Pierre David, Études historiques, pp. 301-302, dated it to 1110. In the absence of other testimony the battle remains impossible to date securely.

54. My reconstruction of events depends on the source reported in Ambrosio Huici Miranda, "Los Banu Hud de Zaragoza, Alfonso I el Batallador y los Almorávides," EEMCA 7 (1962): 11.

55. García Luján, ed., Privilegios reales de Toledo 2:20-22.

56. July 11, 1103. DMP 2:109.

57. AHN, Clero, Carpeta 891, no. 6, an original; a copy in AHN, Códices, 989B, fol. 164V.

58. July 23, 1103. José María Fernández Catón, ed., Catálogo del Archivo Histórico Diocesano de León, vol. 1, (Leon, 1978), p. 17.

59. Pub. García Sáinz de Baranda, La ciudad de Burgos 2:405-406. Full citation to existing copies in Bonachia Hernando and Pardos Martínez, eds, Catalogo del Archivo Municipal de Burgos 1:50.

60. Julio González, Repoblación de Castilla la Nueva 1:95, n. 53.

61. García Larragueta, ed., Colección de Oviedo, p. 331; a late copy in translation and dated to 1073. The royal confirmation of a private document of Oviedo, dated only by year, may be of this period. Floriano Llorente, Colección diplomática de Oviedo, pp. 207-10.

62. AHN, Microfilmas, Palencia, rollo 1.728, fol. 21r-22r; another copy in Acad. Hist., Catedrales de España, 9-25-1-C-4, ff. 12r-14r.

63. AC Burgos, Vol. 69, Pt. 1, fol. 109r; pub. Luciano Serrano, ed., Cartulario del infantado de Covarrubias (Madrid, 1907), pp. 50-51.

64. AHN, Clero, Carpeta 271, no. 12 pub. Alamo, ed., Colección de Oña 1:155-57. The date in the text is 1105 but the confirmation of Diego Fernández as royal majordomo makes the redating necessary.

65. Lacarra, Colección diplomática de lrache, pp. 104-106.

66. Jan. 2, 1104. AC León, Códice 11, fol. 78r-v. Jan 19, 1104. AHN, Códices, 989B, ff. 126v-127r. Feb. 9, 1104. AHN, Clero, Carpeta 891, no. 9; pub. Escalona, Historia de Sahagún, p. 505. Mar. 2, 1104. AHN, Clero, Carpeta, 891, no. 10; also AHN, Códices, 989B, fol. 162r, dated to mar. 4. Mar. 14, 1104. AHN, Clero, Carpeta 891, no. 11; also AHN, Códices, 989B, fol. 57r-v. Mar. 28, 1104. AHN, Clero, Carpeta 960, no. 22; pub. Vignau, ed., Cartulario de Eslonza,pp. 88-89. Apr. 10, 1104. BN, Manuscrtos, 18.387, fol. 249r-v; pub. Yepes, Corónia de San Benito 2:211-12, under date of 1074; ES 40:226-27; Manuel Murguía, Galicia (Barcelona, 1888), pp. 1032-33; and Antonio López Ferreiro, "Ojeada sobre el estado de los monasterios de Galicia a fines del siglo XI y principios del siguiente," Galicia histórica, vol.1 (1901-1903), pp. 56-57. Apr. 18, 1104. Pub. Hilda Grassoti, "Apostillas a 'el prestimonio' de Valdeavellano," CHE 29-30 (1959): 194-95, n. 86. May 9, 1104. AHN, Clero, Carpeta 891, no. 14; also AHN, Códices, 989B, fol. 152r-v.

67. Pub. Vázquez de Parga, Lacarra, and Uría Ríu, Las peregrinaciones a Santiago de Compostela 3:49-50. The date in text is 1103 but the confirmation of Pelayo Rodríguez as majordomo suggests a date between 1104 and 1107.

68. June 13, 1104. AHN, Códices, 989B, fol. 219v. July 8, 1104. Ibid., fol. 87r-v. Aug. 12, 1104. AHN, Clero, Carpeta 891, no. 18. Aug. 26, 1104. Ibid., no. 20. May 26, 1104. Pub. García Larraqueta, Colección de Oviedo, pp. 333-35, was probably drawn up at Sahagún.

69. "Anales Toledanos," Huici Miranda, ed., Las Crónicas latinas de la Reconquista 1:344.

70. Huici Miranda, "Los Banu Hud de Zaragoza," p. 19.

71. AHN, Clero, Carpeta 891, no. 21; another copy in AHN, Códices, 989B, ff. 32v-33r.

72. Pub. Andrés, "Monasterio de San Juan de Burgos," pp. 122-23; Peña Pérez, ed., Documentación de San Juan de Burgos, pp. 11-12. Another copy in BN, Manuscritos, 5.790, fol. 140v.

73. AHN, Clero, Carpeta 1.240, no. 3; copy in BN, Manuscritos, 18.387, fol. 303v, pub. Maximino Arías, "El monasterio de Samos," pp.66-68.

74. Rodríquez Fernández, Pedro Ansúrez, pp. 69-70, and note 87.

75. See the agreement of Nov. 3, 1105, pub. Francisco M. Rosell, ed., Liber Feudorum Maior (Barcelona, 1945), pp. 165-66.

76. The charter itself is lost and our information about it comes from Prudencio de Sandoval, Antigüedades de Túy, fol. 109r-v; and Historia de los reyes de Castilla y de León, fol. 95r-v.

77. Bishko, "Count Henrique of Portugal," pp. 155-88. The entire article would supply a context for the document. An agnitio that would seem to place Henry in Portugal, dated only to 1105, may have been executed at the same time. See DMP 2:158.

78. AC Zamora, Tumbo Negro, fol. 9r; copies in BN, Manuscritos, 712, fol. 1.64V, and 3.546, fol. 142r.

79. AHN, Códices, 989B, fol. 26v.

80. BN, Manuscritos, 712, fol. 83r-v, and 9.194, fol. 102r; another copy in Acad. Hist., Colección Salazar, 0-22, fol. 19r.

81. AC Orense, Reales, no. 9. Only a fragment remains so that one cannot be sure that it was a comital charter. The form of Raymond's confirmation suggests that it may have been rather a royal one.

82. May 6, 1105. AHN, Clero, Carpeta 960, no. 23; pub. Vignau, ed., Cartulario de Eslonza, pp. 90-91. May 15, 1105. AHN, Clero, Carpeta 891, no. 24.June 1, 1105. AHN, Códices, 989B, f. 115r-v.

83. "Anales Toledanos," Huici Miranda, ed., Crónicas latinas de la Reconquista, p. 358. See González, Repoblación de Castilla la Nueva 1:97, n. 61, for the location of the battle.

84. Sept. 9, 1105. AHN, Clero, Carpeta 892, no. 2 a copy in AHN, Códices, 989B, ff. 25v-26r. For the counts, Sept. 11, 1105. AHN, Clero, Carpeta 892, no. 3.

85. Sept. 22, 1105. AC Burgos, Vol. 71, no. 147; pub. Serrano, Obispado de Burgos 3:123-25.

86. Dec. 13, 1105. AHN, Clero, Carpeta 892, nos. 4 and 5, the latter possibly the original; and a copy in AHN, Códices, 989B, fol. 74r-v. Dec. 16, 1105. AC Santiago de Compostela, Tumbo A, fol. 29r; another copy in Chartularum, ff. 71v-72v; also AD Santiago de Compostela, legajo 90, ff. 6v-8r; pub. López Ferreiro, Historia de Compostela 3:61-63. This latter charter reads that it was given "apud tumbam bmi. iacobi apli. causa orationis," which may be a pious, later interpolation or may indicate something more serious. Certainly its author was in León.

87. Jan. 8, 1106. AHN, Códices, 989B, fol. 113r-v. Jan. 10, 1106. Ibid., fol. 112v. Jan. 10, 1106. AHN, Clero, Carpeta 892, no.6; copy in AHN, Códices, 989B, fol.153r. Jan. 12, 1106. Ibid., ff. 113v-114r. Jan. 16, 1106. Ibid., fol. 113r.Jan. 22, 1106. AHN, Clero, Carpeta 892, no. 7. Feb. 13, 1106. AHN, Códices, 989B, ff. 44v-45r. Feb. 19, 1106. AHN, Clero, Carpeta 892, no. 8; copy in AHN, Códices, 989B, ff. 26v-27r. Feb. 26, 1106. Ibid., fol. 131r. March 4, 1106. AHN, Clero, Carpeta 892, no. 9; copy in AHN, Códices, 989B, fol. 136r-v. Mar. 14, 1106. Ibid., ff. 78v-79r.

88. García Larragueta, ed., Colección de Oviedo, p. 336-37, and other copies unknown to the editor in BN, Manuscritos, fol. 111r, dated to Apr. 18, 1071, and Acad. Hist., Colección Salazar, 0-14, ff. 42v-43r, dated Mar. 19, 1080.

89. AHN, Códices, 1043B, fol. 17v; and another copy in 363B, fol. 125r. The second of the same date in AHN, Códices, 1043 B, fol. 16v; and another copy in 363B, ff. 124v-125r.

90. May ,6, 1106. AHN, Códices. 989B, fol. 79r. May 25, 1106. AC Túy, legajo 5, no. 1; a copy in Primero Libro Becerro, ff. 41r-v; pub. in part in María del Carmen Pallares Méndez and Ermelindo Portela Silva, El bajo valle del Miño en los siglos XII y XIII (Santiago de Compostela, 1971), p. 119. This is another charter of Count Raymond. June 10, 1106. AHN, Clero, Carpeta 892, no. to; copy in AHN, Códices, 989B, fol. 40r-v.

91. Aug. 1, 1106. DMP 1-1:14-15. Aug. 25, 1106. Ibid., p. 15. A number of other diplomas ascribed to either Henry or Teresa in this year have to be rejected on one ground or another. Feb. 1, 1106. Ibid., pp. 13-14 and 1-2:555-60. Mar. 19, 1106. Ibid., 1-1:56-57. The confirmations are strictly incredible. May 30, 1106. AHN, Lisbon, Corporaçhes Religiosas, Suplemento, Sala 16, San Pedro de Pedroso, Caixa 1, no. 8 bis. Teresa styles herself "queen," which at the very least would require a date of it 1116-1117. Under these circumstances July 28, 1106, AHN, Clero, Carpeta 892, no. 11, which would place Henry still at Sahagún, must be suspect.

92. "Anales Toledanos," Huici Miranda, ed., Crónicas latinas de la Reconquista, p. 344. Although in a much later publication, "Los Banu Hud de Zaragoza," p. 28, n. 7, the same author argues that the "Anales" has the date wrong, that it should be 1126, and that the reference is to Alfonso I of Aragón. His argument is from the silence of the Muslim sources for 1106.

93. Francisco Codera, Decadencia y desaparición de los Almorávides en España (Zaragoza, 1899), pp.5-6.

94. Turk, Reino de Zaragoza, pp. 184-85.

95. 1100. Pub. Durán Gudiol, ed., Colección diplomática de la catedral de Huesca, p. 123.

96. Rosell, Liber Feudorum Maior, pp. 166-67.

97. Ubieto Arteta, ed., Colección de Pedro I, pp. 25-27.

98. Alamo, ed., Colección de Oña 1:121-22. The document is dated in the text to 1086, which is impossible given those who confirm. Serrano, ed., Cartulario de San Millán, p. 295.

99. AC León, Particulares, no. 290.

100. AHN, Clero, Carpeta 892, no. 13 a copy in AHN, Códices, 989B, ff. 140v-141r.

101. AC Santiago, Tumbo A, fol. 29v; other copies in Chartularum, ff. 72v-73r; and in an unpublished manuscript of Acad. Hist., 9-27-2-E-50, Estefanía, "Memorias," p. 235.

102. March 29, 1107. AHN, Clero, Carpeta 892, no. 14. Apr. 21, 1107. AHN, Códices, 989B, fol. 100v.

103. Pub. García Luján, Privilegios de Toledo 2:22-25, but dated erroneously to March 8.

104. AC Santiago de Compostela. Tumbo A, fol. 27v; and Tumbo C, fol. 218v-219r, with the date as 1077; and Chartularum, ff. 73r-74v; pub. López Ferreiro, Historia de Compostela 3:70-73 append.; and Estefanía, "Memorias," pp. 237-38. It is the second of these copies that, although intermediate in time, gives us the critical but variant reading. Nevertheless it is impossible to imagine a thirteenth-century scribe simply inventing such a reading while the "Sancius filius regis conf." of the earliest copy could easily be abbreviated.
    Claudio Sánchez-Albornoz, "La primitiva organización moneteria de León y Castilla," in Estudios sobre las instituciones medievales españolas (Mexico City, 1965), pp. 441-82, argued on the basis of the "Historia Compostelana." ES 20:63-69, that the charter should be dated rather to 1105. Bishko, "Pacto Sucessório," pp. 186-87, followed Sánchez-Albornoz, and the charter thus became a critical part of the evidence for dating the pact to that year. However, as we have seen, the dating formulas of the two charters would support the common and traditional dating. I do not believe it can be altered on the basis of literary evidence. For the methods of composition followed by the authors of the gesta see Reilly, "Historia Compostelana," pp. 78-85.

105. June 12, 1107. AHN, Códices, 989B, ff. 73v-73(bis)r. July 27, 1107. AHN, Clero, Carpeta 892, no. 15, Aug. 7, 1107. AHN, Códices, 989B, fol. 78v. Aug. 16, 1107. AHN, Codices, 1.439B, ff. 3v-4r, which would place Alfonso in Toledo is a forgery. Aug. 27, 1107. AHN, Códices, 989B, fol. 86r.