[116] When Alfonso of León-Castilla decided to seize upon the fortuitous murder of the king of Navarra in June 1076 and marched to annex the lion's share of that kingdom, the forces he led had probably been assembled originally for a campaign in the south. Since his father Fernando's death, the kingdom had been poised for the appropriation of the southern half of the Duero Basin. The Muslim taifa kingdoms of the south had been reduced already to the status of client states. To the east and to the west, the flanks of that great region in the upper reaches of the Duero and on the coastal plains of Portugal had been secured. The northern half of the meseta had been largely repopulated. Only the division of the Fernandine kingdom had delayed the outright seizure and resettlement of the sprawling area between the river and the Cordillera Central.

This trans-Duero region, as it was ordinarily styled in the north of the day, was a long triangle whose northern leg was constituted by the Duero itself. Its north-south base was constituted by the mountains of Portugal which were virtually impassable from the Duero down to the Tajo except by the winding route that the railroad follows to this day from Coimbra on the Mondego, along the upper valley of that river, and over the 800-meter-high pass that leads down to modern Guarda. The longest leg ran from southwest to northeast along the watershed formed by the Gredos and the Guadarrama mountains from the Tajo to the Sorian highlands. The distances involved were 320 kilometers, 200 kilometers, and 420 kilometers respectively. Within the legs lay a virtual kingdom of some 50,000 square kilometers. It was more than one and a half times the size of all of Galicia or about six times the area of modern Wales.

Neither the terrain nor the climate of this area offered significant challenges to its acquisition and utilization by the north. In these respects it was of a piece with those lands that Leonese farmers and shepherds had spent the past two and a quarter centuries learning to exploit. In 1076 its essential emptiness was a function of its location. It lay, and had lain, between Islam in the south and Christianity in the north as a [117] kind of "world of war" whose possibilities were beyond more than very partial development by either protagonist. This condition has been brilliantly examined and demonstrated by Claudio Sánchez-Albornoz, and although his thesis, or at least the rhetoric in which it is expressed, has been challenged it has largely been accepted by subsequent scholars.⁽¹⁾ The effective domination of the trans-Duero was to be one of the major problems of Alfonso VI's reign, and even at his death its regular governance was largely supplied from towns around its perimeter.⁽²⁾

The challenge to such a process was constituted, in good measure, by those taifa states that also possessed access to the region. In the west the gap between the Gredos mountains and those of Portugal was controlled by the Muslim kingdom of Badajoz. In the center the kingdom of Toledo possessed practical if more difficult entry by means of a series of passes through the Guadarrama, which opened to the sites of Avila and Segovia and through the rolling country to the east of these mountains and the west of the Sierra de Albarracín. In the east possible, but much more difficult, entry could be obtained by the taifas of Zaragoza and Valencia by means of the old Roman road that followed the meanderings of the upper reaches of the Henares and Jalón rivers.

Though the monarchs of the age did not speak of grand strategic concepts ordinarily, there can be no doubt that they utilized their practical elements quite consciously. The campaigns of the late reign of Fernando I display a clear recognition of the problems mentioned just above. Alfonso himself as king merely of León had attacked Badajoz in 1068 even though it had then been a tributary of his brother, García. As ruler of a reunited León-Castilla, Alfonso VI would demonstrate the same grasp of the great necessities for aggrandizement in the trans-Duero as had his father.

Because of its position, which formed by far the greatest portion of the southern border of the trans-Duero, the kingdom of Toledo was the essential key to its occupation. That taifa must be effectively neutralized or, as events would have it, conquered. How early Alfonso VI's mind turned from the first to the second cannot be precisely determined. As we have seen, in the late summer and fall of 1074, he was already campaigning in the Muslim southeast of the peninsula in concert with his ally and client, al-Mamun of Toledo. Clearly this campaign had two immediate aims both of which were achieved: the exaction [118] of an annual tribute from the king of Granada to Alfonso and the addition of Córdoba to the territories of al-Mamun.

The ambitions of the Leonese monarch in 1075 were perhaps already reaching beyond mere suzerainty and parias. The first recorded presence at court in March of that year of the Mozarabic noble Sisnando Davídez, who held Coimbra, may be significant.⁽³⁾ If an offensive against the taifa of Badajoz were in prospect, both the stability of the Portuguese frontier and the prospect of cooperation from it would be important. However, action on that front seems to have been delayed when the assassination of al-Mamun in Córdoba in June 1075 demanded immediate attention to the affairs of Toledo. The deceased taifa king's son and successor al-Qadir was unable to prevent the declaration of independence in Valencia that followed his father's death, and the entire eastern sector seemed to demand Alfonso's attention. He appears to have inspired raids from Zaragoza on Valencian territory and evidently was prepared to take the field himself in 1076 when the great opportunity in the Rioja presented itself.

The king was not to be totally diverted from his earlier purposes by the Riojan expedition. The settlement of affairs there was virtually uncontested, and he was back in Castilla and just south of Castrojeriz on August 1, 1076. Two private donations of that date were confirmed by the bishops of Burgos and Palencia and one of them also by the bishop of Santiago de Compostela, the latter of whom had been at Calahorra with the king less than a month earlier, and other court figures.⁽⁴⁾ The royal party was moving south and into the lands across the Duero for on August 17, 1076, it was about twelve kilometers north of Sepúlveda. There Alfonso issued a charter authorizing the monastery of Silos to repopulate the area around San Frutos to the southwest of Sepúlveda.⁽⁵⁾[119] The charter was confirmed exclusively by Castilian magnates so that the bulk of the royal court may have proceeded from Castrojeriz to León.

The king himself seems to have returned to León shortly thereafter for a series of private and royal charters place him there from August 27 to September 8, 1076, in the company of the usual members of the court.⁽⁶⁾ However, Alfonso returned in the fall to the area of Sepúlveda where we find him issuing a fuero to the inhabitants of that village on November 17, 1076. The confirmants are again Castilian magnates.⁽⁷⁾

The appropriation of the trans-Duero was beginning, then, with the initiatives of the king in the Sepúlveda area. That village was an obvious place to start for it was a militarily strong position raised above the junction of the Castilla and Duratón rivers, the latter of which then ran down to the Duero at the castle of Peñafiel some forty-five kilometers to the northwest. The highlands beginning to the east of the settlement offered additional security, but to the west the meseta lay open to its surveillance. Fifty kilometers to the southwest lay the ruins of Segovia, and only twenty-five kilometers to the southeast was the 1,440-meter-high pass of Somosierra through the Guadarramas to Madrid and Toledo. Sepúlveda was therefore a spot of great strategic significance on the eastern edge of the meseta and may well have had a small population already to serve as a nucleus for the royal refoundation.⁽⁸⁾ [120] The charters of 1076 and the evidence of language indicates that the new population was drawn largely from Castilla la Vieja.

This new Christian influx would have had to practice some agriculture, largely in garden plots one suspects. From the beginning they must nevertheless have relied primarily on the herding of cattle, sheep, horses, and goats for their livelihood. The unoccupied expanses of the trans-Duero were uniquely adapted to such utilization, and all surviving evidence suggests that they were quickly turned to such purposes.⁽⁹⁾ Wealth in such a form, mobile and based on fortified cities, was much simpler to protect than were sprawling fields against Muslim raiders or pastoralists, themselves entrenched on the northern slopes of the Cordillera Central and pursuing much the same sort of life.⁽¹⁰⁾ Under such circumstances, stockraiding also became a regular part of economic activity as well as an occasional military duty. In both aspects, the ranging of the new settlers over the plains of the south would provide their sovereign with erratic but important new intelligence of events there and further to the south.

That information was not reassuring initially. The ambitious and subtle king of Sevilla, al-Mutamid, had seen in the death of al-Mamun of Toledo in June 1075 an opportunity to reestablish his preeminence in Andalucía. In 1076 he had overrun and annexed the taifa of Denia on the southeastern coast. The following year he reasserted his control over Córdoba and at least some of the southern territories of Toledo in the basin of the Río Guadiana. He may also have already appealed for aid to the growing Murâbit empire in North Africa. Continuing internecine warfare in the Muslim south was to Alfonso's benefit, of course, but only so long as no clear victor appeared to restore effective unity there.

Meanwhile, the Leonese monarch had other concerns which claimed precedence for the moment. Returning from the Sepúlveda region, he had probably spent the Christmas season in León or Sahagún. A private donation to the latter on January 15, 1077, was confirmed by a number of court figures.⁽¹¹⁾ The king issued a charter himself at León on January 30, and a private charter of the same date indicates a larger assemblage [121] than usual.⁽¹²⁾ The documents do not then reveal the royal itinerary until April and May when the king was in Burgos wrestling with the problem of the liturgy and related matters.⁽¹³⁾ June and July were spent in the vicinity of León.⁽¹⁴⁾

The court must have departed for the west in mid-July for a document of the Galician monastery of Celanova seems to record the royal presence in the Orense area on July 31, 1077.⁽¹⁵⁾ It then proceeded northwest to Santiago de Compostela where the king confirmed an agreement between the bishop and the abbot of the monastery of Antealtares, which cleared the way for the beginning of a new cathedral there, on August 17, 1077.⁽¹⁶⁾ It seems hardly possible that Alfonso was still in Galicia so late, but a charter he granted to the monastery of San Antolín of Toques, between Santiago and Lugo near the pilgrim road, is dated October 17, 1077. I incline to believe that the fourteenth-century copyist mistook the date.⁽¹⁷⁾ But, even discounting the latter charter as dated, it would have been impossible for Alfonso already to have been back in León on September 3, 1077, when he appeared as confirmant to a charter of his sisters in León. This charter is a reworking of an earlier one and cannot be relied upon.⁽¹⁸⁾

What does seem to be clear is that Alfonso again traversed the entire breadth of his kingdom in the late fall of 1077. On November 5, 1077, he was in the Rioja where he confirmed a private charter to San Millán de La Cogolla.⁽¹⁹⁾ Five other charters or confirmations of the king, which deal with matters concerning the Rioja but which are dated only by year, are probably of the late fall as well.⁽²⁰⁾ Whether Alfonso was in [122] Castilla on December 17, 1077, when the abbot of Oña and the bishop of Burgos exchanged properties with his consent, is not clear but it is likely.⁽²¹⁾ Before he had left the Rioja the Leonese monarch had perhaps met with Sancho Ramírez of Aragón or his representatives and specified yet further their respective holdings in the former kingdom of Navarra. A charter of the latter, dated only to 1078, cites Alfonso as reigning from Nájera to perhaps as far east as Estella.⁽²²⁾ Such an agreement would have freed the Aragonese king to put to best use the aid he was to receive in 1078 from an expedition of Duke Hugh I of Burgundy.⁽²³⁾

On January 29, 1078, Alfonso VI was back at Sahagún where he confirmed a private donation to that monastery.⁽²⁴⁾ There he seems to have remained through the winter, the Easter season, and even as late as May 14, 1078.⁽²⁵⁾ For the six months that follow, it is impossible to determine the whereabouts of the king and court.

A number of reasons lead me to suspect that he was again engaged in the task of repopulating the trans-Duero. First, a similar dearth of documents in the summer and fall of 1074 accompanied a long absence from the north precisely while he was campaigning in Andalucía. But that campaign was recorded in Muslim sources for it had a major impact upon the taifas. Their silence in 1078 would seem to rule out that sort of major campaign. Second, the almost frenzied mending of fences at home suggested by Alfonso's continual movement in 1077 from Castilla, to Galicia to the Rioja is more readily understandable as the prelude to an anticipated absence of some duration. Third, later tradition will fairly consistently speak of a campaign, which ended with the conquest of Toledo, as being of seven years in length.⁽²⁶⁾

Nevertheless, I do not believe that Alfonso's aims had yet matured [123] to such a point. As we have seen, as late as 1080 he would suggest the creation of an archbishopric to Gregory VII. Surely that would have been too early if he had been thinking of the restoration of Toledo, but at the same time, if he already foresaw that city's recapture, it would have obviated the need for another archbishopric. It is more likely that subsequent chroniclers saw a fairly general, if local, series of initiatives in the region south of the Duero as steps toward a truly remarkable advance in fact achieved subsequently. It has been asserted recently that Alfonso concluded treaties with Sevilla and Zaragoza in 1078 precisely to isolate Toledo, but no authorities have been adduced to bolster the statement.⁽²⁷⁾

Inherently more likely is that Alfonso spent the summer and fall creating advance posts in the trans-Duero from various strongpoints along the line of the Duero itself much as he had if 1076 in the case of Sepúlveda. Although he furnishes no dates, Bishop Pelayo specifically lists Salamanca, Avila, Coca, Arévalo, Olmedo, Medina, Segovia, Iscar, and Cuéllar as places repopulated south of the Duero by Alfonso VI.⁽²⁸⁾ Salamanca, Avila, and Segovia, all lying far south and for that reason too exposed to Muslim attack, would await later attention. We shall not be far from the truth if we envison the king directing his attentions to the remaining places in the summer and fall of 1078.

Medina del Campo, Olmedo, Coca, Fuente el Olmo de Iscar, and Cuéllar, on the other hand, all form a rough salient along the bank of the Duero stretching from twenty to forty kilometers south from that [124] river and a rough fifty kilometers from east to west. Any point in that area could have been easily and quickly succoured or reinforced from the fortresses at Tordesillas, Valladolid, or Peñafiel securely ensconced along the northern bank of that river. If Arévalo was repopulated at this same time, they would together have formed a wedge thrust south into the trans-Duero for a total distance of fifty kilometers at its deepest point. They would all have been mutually supporting and represent a practical, strategic bite into that area which would complement what had been achieved in the Sepúlveda area fifty kilometers to the east two years before in 1076. It may be significant that when next we catch sight of Alfonso VI he is granting a fuero to Santa María de Dueñas, midway between Valladolid and Palencia, on November 5, 1078.⁽²⁹⁾ Possibly he was returning from just such a labor.

In all probability the court spent the late fall and the Christmas season in the proximity of Sahagún and León as was customary. In any event the court was clearly at Sahagún on February 5, 1079.⁽³⁰⁾ By March 12, 1079, it had moved north into Asturias, which was the only major province except Portugal not visited during the preceding year, where the king granted a charter to the monastery of San Vicente in Oviedo.⁽³¹⁾ The business of the crown there must have been of high import for the trip over the Cantabrians would have been grueling at that time of year. Doubtless it had to do with preparing for an assemblage of the royal host for by April 7, 1079, the king was already on campaign not far from Toledo in the center of the peninsula. The extraordinary urgency at work becomes even more impressive when we reflect that to travel from Oviedo to Toledo in little more than three weeks would have forced Alfonso to ignore the solemnities of the Easter season since the feast fell that year on March 24.

Nevertheless the facts can hardly be controverted. The charter of April 7, 1079, is an original. It bears a note, added in another but contemporary hand in my judgment, that tells us that it was issued by the Río Guadarrama, which flows into the Tajo just twelve kilometers west of Toledo, while on campaign.⁽³²⁾ What extraordinary circumstances had called the Leonese monarch into the field with such haste?

Although the chronology of events in the Muslim half of the peninsula [125] is anything but clear it is known that al-Mutamid of Sevilla had secured at least temporary control of the taifa of Murcia in 1078, probably while Alfonso was busy repopulating the trans-Duero. Added to Córdoba, retaken in 1076, and Denia, seized in the same year, the Muslim king's realm stretched from the Atlantic to the Mediterranean. But even worse had transpired, at least from the standpoint of Alfonso. Sometime in early 1079 his ally and client al-Qadir had lost control of the taifa of Toledo. A series of risings in his chief city had resulted in the installation there by the conspirators of al-Mutawakkil, the taifa king of Badajoz in his stead. So the situation still stood in June 1079.⁽³³⁾ Al-Qadir had himself fled first to Huete and then farther east to Cuenca.

The position of Alfonso VI's army a dozen kilometers west of the city of Toledo in early April, then, is best understood as constituting a blocking force that would disrupt communication and reinforcement between al-Mutawakkil in Toledo and the latter's permanent seat of power in Badajoz. Information is simply lacking to follow the development of the campaign in any detail, but two ultimately successful initiatives of Alfonso seem clear. First, he extended his military campaign to the hereditary domains of the king of Badajoz and had succeeded by September in conquering the town of Coria.⁽³⁴⁾ The fall of that fortress gave the Leonese king a base from which to stage an attack on Badajoz itself some 125 kilometers to the southwest.

Second, Alfonso also undertook diplomatic measures to arraign al-Mutamid, king of Sevilla, against al-Mutawakkil. This episode is recorded in the Historia Roderici as an embassy by El Cid to collect the parias, or tribute due to León, from al-Mutamid, but it also clearly involves an alliance between the Christian and Muslim kings. As such it resulted in Alfonso's ambassador fighting a pitched battle against the taifa king of Granada, another Leonese client, who was trying to take advantage of the confused situation in the Muslim south to attack the Sevillan.⁽³⁵⁾

[126] Both plans to contain and to threaten al-Mutawakkil worked sufficiently well to allow Alfonso himself to leave the culminating military actions against Coria to his subordinates. By the end of July the king was again in the Rioja where he granted a charter to the monastery of San Millán.⁽³⁶⁾ To judge by those who confirmed it his entourage was made up entirely of Castilians. A month later, on September 3, 1079, he had been joined by the full royal court when he issued a charter making the monastery of Santa María of Nájera a dependent of Cluny.⁽³⁷⁾ Certainly one of the reasons for which Alfonso was in the Rioja was to receive his new fiancée, Constance of Burgundy, as has been developed in the preceding chapter. Yet that was not the entire reason.

That rich territory had only been annexed in 1076, and then under the most peculiar circumstances. Alfonso had returned there in 1077 and now did so again. It would be ingenuous not to realize that he was experiencing real difficulty in working out a satisfactory political settlement there. Indeed, as others have pointed out, the very granting of that old royal monastery of the kings of Navarra to Cluny was designed to strengthen royal power in the province as well as to reward a steadfast friend.⁽³⁸⁾ And the charter shows the brother and sisters of the late Sancho García IV of Navarra clustered around Alfonso, associated with him by their acts of confirmation and lending legitimacy to his power. But even better provision would have to be made for the control of the province if the king was to be able to concentrate on the opportunities of the south. For the present, however, Alfonso seems to have returned to the center of his realm with his bride-to-be for there is evidence that he was at Sahagún on December 6, 1079.⁽³⁹⁾ In that area, probably León, he was married perhaps before Christmastide as suggested by a charter of Christmas day.⁽⁴⁰⁾

During the following year the documents provide only slim guidance as to the whereabouts of the king. The court seems to have been at Sahagún in late January.⁽⁴¹⁾ Apparently it remained there until late April.⁽⁴²⁾ The winter season was occupied with the delicate negotiations over the liturgical question and the royal marriage detailed in the preceding [127] chapter. It was climaxed by a settlement perhaps reached in council in Burgos but in any event fully in effect by the time of the great meeting of the royal curia reflected in the document of May 8, 1080. After that date no documents give trustworthy evidence as to the royal whereabouts until the late fall.⁽⁴³⁾

This interval Alfonso spent in the south reaping the fruits of his efforts there the previous summer. Al-Mutawakkil was forced to abandon Toledo and al-Qadir was restored to power in that taifa.⁽⁴⁴⁾ Nevertheless Coria was not restored to the king of Badajoz because its acquisition constituted a major advance in what was still, I am convinced, the Leonese monarch's primary goal, the repopulation of the trans-Duero. Situated on the Río Alagón some forty-five kilometers northeast of the point at which that tributary flowed into the Tajo, that fortress town sat squarely in the gap between the western end of the Cordillera Central and the mountains of Portugal. In Leonese hands it guaranteed the relative safety of new settlements in the whole of the western trans-Duero region from Zamora south to Salamanca and Alba de Tormes.

This concentration on the development of the region between the Río Duero and the Cordillera Central was, assuredly, complemented by the continuance of the older policy of maintaining a general suzerainty over the taifa states. The former could only proceed so long as they remained divided among themselves and their wealth was exploited by means of the parias to provide extraordinary resources for the endeavors of the king of León-Castilla. In this light, the self-restraint of Alfonso in regard to the continued if dependent existence of the taifas constituted a fundamentally enlightened self-interest, and his restoration of al-Qadir to Toledo becomes fully comprehensible.

The statement of such a policy is, in fact, given to us by a contemporary source, the Muslim king of Granada. Abd Allah puts in the mouth of Alfonso himself, in the course of their negotiations in 1074, the simple and direct explication of that policy. "The costs of conquest," [128] Alfonso is made to say, "in both men and money would be more than I could hope to gain from it even if I were successful. In the event that I did conquer, because of their religion I could not count on the fidelity of its population and it would be impossible to kill them all and to repopulate with men of my own faith. It is better for me then to set one against the other and exact tribute from all. Eventually they will become so weakened as to surrender to me of their own will and I will gain my ends without effort."⁽⁴⁵⁾ One may well suspect that this little speech was a literary device of the Granadan king. Still the ideas expressed in it must surely have been the commonplaces of political counsel in the Leonese court of the late eleventh century.

But notwithstanding, the seeming incapacity of al-Qadir to control the taifa of Toledo had already in 1079 begun to carry Alfonso VI beyond the strict adherence to such policy of relative moderation. Toledo friendly and strong was a key necessity for the peaceful penetration of the trans-Duero, and the inability of its king to control even his domestic enemies had made open intervention there absolutely necessary to the Christian monarch. As conditions for al-Qadir's restoration, the so-called Pact of Cuenca, Alfonso extracted from his dependent the surrender of two key military bases. These were Zorita and Canturias.⁽⁴⁶⁾ The first, 110 kilometers to the northeast of Toledo, seems to have assured Alfonso himself of easy access to the realm of Toledo while denying it to enemies from Valencia or Zaragoza. The second, 75 kilometers west of Toledo, would represent added assurance against future incursions from Badajoz. For the support of these permanent garrisons, at least in part, Alfonso also exacted additional concessions of the sort related to the imposts of yantar or hospedaje within his own realm.⁽⁴⁷⁾ Doubtless such precautions seemed sensible and necessary to the Leonese king, but they were bound to exacerbate relations between al-Qadir and his own subjects and make further intervention by Alfonso necessary.

On the other hand, the very restraints implicit in the policy of supporting al-Qadir could and did involve the Leonese monarch in serious trouble with his subjects. The events of late 1080 or early 1081 furnish a celebrated case in point. The earliest source for the incident, the Historia Roderici, tells of a Muslim raid on the Castilian stronghold of San Esteban de Gormaz on the upper reaches of the Duero.⁽⁴⁸⁾ Whence the [129] raiders had come we are not told, but both geography and subsequent events make it likely that they were natives of one of the nearby Muslim hill towns such as Atienza, Sigüenza, or Medinaceli. As such they were technically subjects of al-Qadir. In retaliation El Cid, acting apparently on his own initiative, launched an extensive and destructive raid into the eastern domains of the taifa of Toledo. His action presented his king with major diplomatic problems. On the one hand it further damaged al-Qadir's prestige with his own subjects. On the other, it was bound to raise the question in the minds of the taifa kings as to the practical immunity the parias actually bought. At the same time, it set the worst possible precedent for others of Alfonso's own nobility to whom a quick border raid was an attractive opportunity for enrichment.

The Leonese monarch reacted in the only possible fashion. He exiled El Cid from his domains. The subsequent epic literature of course was to malign the king and to glorify Rodrigo Díaz de Vivar. Deriving from Castilla in a period when León and Castilla were separate and frequently hostile kingdoms, that was to be expected. In addition, however, that literature cast the whole of their subsequent relationship into the largely personal terms of envy and jealousy on the part of the king, rivalries at court, and forbearance and self-abnegation on the part of the injured noble. Precisely since it was literature, and this was the stuff of a good story, that is hardly surprising. What is surprising is that subsequent serious historians to the present day have accepted the complete substitution of personal passions for serious political concerns at virtually its face value. Doubtless many personal motives were involved in the resolution of the problem El Cid had created by his own action, but we are utterly without a historical basis on which to discern or describe them.

Fancy aside, we do not know whether Alfonso VI exiled Rodrigo Díaz in anger or in sorrow. We can see that it was the least measure he could have taken that would safeguard the policy he was pursuing. Exile, apart from the dramatic elements that have been added to it ever since, at least had the appearance of a chastisement for one who had broken the kings peace. But, when we consider the sequel and results of the banishment, we cannot even be sure that it was indeed a real penalty in the eyes of either of the protagonists. For practical purposes, El Cid was to lead a military force into the east of the peninsula where he would have a completely free hand. Ultimately that journey was to make of him a great, territorial prince. Assuredly both great skill and great good fortune contributed to that result, but the opportunity was there in 1080 and both king and noble must have known it.

One also must reflect on the surprising extent to which the exile and [130] Rodrigo's activities complemented the interests of Alfonso. El Cid was first to attack the taifa of Valencia which had broken away from Toledo in 1075 and had since been the enemy of both al-Qadir and his overlord. This presence of the Castilian also placed a check on the initiatives of the Count of Barcelona and of al-Muqtadir of Zaragoza, both of whom had already displayed ambitions of their own for aggrandizement at the expense of the inept Abu Bakr of Valencia. Yet since these effects were the work of an official exile, Alfonso himself could be held responsible for none of them nor was his own power or prestige contingent on Rodrigos success. His were simply the benefits.

Unfortunately the precise time when this dramatic decision was taken is impossible to establish. The last time that Rodrigo Díaz de Vivar confirmed a document of his king was at the great court of May 8, 1080.⁽⁴⁹⁾ Shortly thereafter the king left for the campaign that was to reinstall al-Qadir in Toledo, and the exact whereabouts of the court are unknown until it reappears at Sahagún on November 28, 1080, where it probably remained through the Christmas season.⁽⁵⁰⁾ It is quite possible, then, that as the success of Alfonso carried him to the west of Toledo in the summer of 1080, some of the Muslim of the Medinaceli area risked a raid on San Esteban de Gormaz. The same absence of the king was required for the practical vacuum of authority that permitted El Cid to respond in kind. The likelihood is, then, that Alfonso was faced with the decision as to how the resultant problem could be resolved when he returned to winter at Sahagún in the late fall of 1080.

Such decisions would hardly have been made except as a complex of adjustments to all the pressing concerns of the past few years. Rodrigo Díaz himself was not a problem except as he represented the unruly ambitions of the minor nobility. He had played a modest part in the court of Sancho II, when fiction is segregated from fact, as we saw in a preceding chapter. His role in the court of Alfonso VI was even more modest. Menéndez Pidal would make much of the royal favor that bestowed on El Cid a niece of Alfonso VI as his wife in 1074.⁽⁵¹⁾ But the [131] carta de arras of that date cannot indeed be of that date as we have seen. Nor is the trail that leads from Doña Jimena to Alfonso VI any more satisfactory. The Historia Roderici names as Jimena's father a Count Diego of Oviedo.⁽⁵²⁾ Unfortunately no count of Oviedo of that name can be described for the eleventh century. Menéndez Pidal based his reconstruction of the genealogy of Rodrigo's wife on a document of Oviedo dated August 13, 1083, which is clearly false.⁽⁵³⁾ El Cid's own family was of minor importance.

The major problem was the stability and protection of the eastern frontier of the realm. The exile of El Cid would provide a safeguard or a buffer in the southeast. In the northeast the Rioja remained a concern and one that had called Alfonso from his southern campaigns twice since 1076. Though most historians have followed Menéndez Pidal in his assertion that Alfonso VI had made García Ordóñez count in Nájera in 1076, I do not feel that such an appointment was likely. Neither the fuero of Nájera nor that of Calahorra in that summer contained any mention of García Ordóñez. Instead, those persons mentioned as influential in the king's counsels are Diego Alvarez of Oca and the latters brother-in-law, Lop Jiménez, son of Jimeno López. The latter was count of Vizcaya and of Nájera. The collaboration of these three magnates helped win the Rioja for Alfonso. If, then, two documents of 1077 refer to a Count García in Nájera, is it not at least likely that the reference is to García Jiménez, brother of Lop Jiménez, as to García Ordóñez?⁽⁵⁴⁾

In fact, after his last reliable notice as royal alférez in mid-1074, García Ordóñez vanishes from the royal entourage for six years.⁽⁵⁵⁾ In the interim we hear of him as leading an attack on the taifa king of Sevilla in 1079 in conjunction with two Navarrese magnates who had been prominent in the service of that kingdom and then of Sancho II of Castilla, Fortun Sánchez and Lop Sánchez. That attack was contrary to the policy of Alfonso VI at the time and was repulsed by Rodrigo Díaz de Vivar, just then performing an embassy to al-Mutamid.⁽⁵⁶⁾ This context [132] strongly suggests that García Ordóñez was then himself a refugee in the Muslim south. But royal policy changed and his fortunes improved markedly, for in 1080 a document of Nájera itself cites him as count there.⁽⁵⁷⁾ His first reappearance at court and in the capacity of count is in the same charter of May 8, 1080, which is the last record of El Cid at court.⁽⁵⁸⁾ Finally, on April 18, 1081, García Ordóñez will confirm a charter, not only in the company of his king, but also in concert with his own wife. She is of course Urraca, sister of the late king of Navarra, Sancho García IV.⁽⁵⁹⁾ By this latter date, the elements of a new settlement for the eastern frontier of the realm had been largely determined.

Alfonso VI seems to have devoted most of 1081 to its design. The court was apparently at Sahagún on February 1 of that year and may have continued there through March 31.⁽⁶⁰⁾ Certainly by April 18, 1081, the court was in Nájera where the king confirmed the charter of Ramiro García, brother of Sancho García of Navarra, to the monastery there. Probably sometime during this journey, the king also confirmed an old charter of San Millán though his action is dated merely by year.⁽⁶¹⁾ On May 14, l08l, probably in or near Burgos to judge by those who confirmed, Alfonso granted yet another charter to Cluny.⁽⁶²⁾ Back in León by June 9, l081, the king issued another charter to a tiny Leonese church.⁽⁶³⁾ July 15, l08l, found the court at Sahagún again.⁽⁶⁴⁾

The known itinerary of the crown in the spring and early summer of l08l precludes placing Alfonso's campaign against al-Mutawakkil in that year.⁽⁶⁵⁾ He may have made some military demonstration in the fall [133] for on September 9, 1081, one of his court nobles, Diego Ansúrez, made a will at León setting out the conditions to be observed if he were killed in battle and his body were not found, on the one hand, and if it were recovered, on the other.⁽⁶⁶⁾ If something of a military excursion did take place it must have been quite brief, however, for the court was again at León on December 3, 1081, and Christmas was probably celebrated in Burgos.⁽⁶⁷⁾

All surviving evidence points, then, to Alfonso VI having concentrated on solving the major problems of the eastern frontier in 1081. The decision to exile Rodrigo Díaz de Vivar may have been the catalyst but it was far from constituting the entire royal policy. The recall of García Ordóñez from Andalucía, his marriage into the royal dynasty of Navarra, and his appointment as count in the Rioja would at the same time balance the activities of El Cid. The mutual hostility between the two is the secondary theme of all of the epic literature, and there is no reason why we should reject it as unhistorical.

If the two nobles were already enemies, the new situation of each was bound to enhance their rivalry. The fortunes of both were of necessity intimately entwined with influencing the actions of the taifa of Zaragoza. That greatest of the Muslim kingdoms of the east bordered on the Rioja in the south and Valencia in the north. Though nominally a client state of León-Castilla it pursued a largely independent policy within its own region, and its cooperation or at least neutrality would be crucial to the ambitions of either García Ordóñez or Rodrigo Díaz.

Alfonso VI's actions reflect an intention to play upon this and other rivalries. In July 1081 García Ordóñez's brother, Rodrigo, had already become royal alférez, a position García himself had held briefly some seven years previous. At the same time, Alfonso had promoted or allowed the consolidation of the three Basque provinces of Vizcaya, Alava, and now Guipúzcoa under his old ally, Lop Jiménez, by 1081.⁽⁶⁸⁾ That barony constituted a more effective deterrent to any western ambitions of Sancho Ramírez of Aragón, but it also necessitated a stronger rule in the Rioja to moderate the ambitions of the Basque magnate in turn.

The decision to stabilize the eastern frontier by allowing or promoting the emergence of three strong nobles with virtually viceregal authority in extensive provinces created dangers of its own as was subsequently to be demonstrated. Nevertheless the experiment worked well [134] enough to be repeated a decade later in the western territories of the kingdom. In both cases, the stimuli seem largely to have been the same. The first of these was the sheer size of the kingdom as it had been reunited under Alfonso and had continued to swell into first the Rioja and then the trans-Duero. The difficulties of communication and travel, hence effective control, multiplied with every sizeable accession of territory. The second was a pressing necessity to concentrate on the affairs of the Muslim half of the peninsula.

Perhaps as early as the end of 1080 and almost certainly by the middle of 1081, Alfonso had come to the realization that nothing short of the outright annexation of the taifa of Toledo was going to protect his essential ends in that region. He had restored al-Qadir to power but that king proved unable once again to administer his own affairs effectively. After his restoration the Muslim had found it impossible to raise the indemnity which had been the price of Alfonso's support. As was his preferred practice, Alfonso expected his allies to defray the costs of his campaign.

As a result of the default in the indemnity, the Leonese monarch took control of yet another fortress, identified as "Canales," and garrisoned it.⁽⁶⁹⁾ If this location can be identified with the present remains of a castle just thirty kilometers north of Toledo, then its cession provided almost the last necessity for an outright siege of the city of Toledo itself.⁽⁷⁰⁾ Alfonso's garrisons already occupied the fortresses of Zorita to the east of the city and Canturias to the west. While the southern approaches of the city remained clear thus far, these bases were quite adequate to conduct a continuous harrying operation against the city itself and the surrounding countryside that nourished and supported it. Unless major military assistance could be secured from Muslim Andalucía, the ruler and inhabitants of Toledo had no choice but to come to terms with an opponent so situated.

As events were to demonstrate, no pitched battle would be fought. A gradual campaign of attrition and a slowly increasing military pressure against the city itself would finally result in the peaceful capitulation of Toledo in the spring of 1085. But even when Alfonso VI began to think in terms of the ultimate occupation of the city and taifa of Toledo, he understood that the real problem would not be its immediate conquest. Implicit in the remarks attributed to him by Abd Allah in 1074 was the recognition that the real task would be the fashioning of a [135] political settlement and the marshaling of resources for repopulation there which would allow him to retain what he had overrun.

Even if, as subsequently it would develop, Alfonso was able to hold only the northern half of the basin of the Tajo, the effort required would be enormous. The area so defined constituted better than forty thousand square kilometers, and it was separated from the territories of León-Castilla not only by the massive Cordillera Central but also by the as yet virtually empty plains of the trans-Duero, averaging a hundred kilometers north to south across. Its population was composed of Muslim, Christians, and Jews in undetermined proportions but all of a culture quite alien from that of the north. The difficulties of the conqueror might well be symbolized by the city of Toledo itself. If Josiah Cox Russell is correct, the urban population of twenty-eight thousand was itself greater than the combined populations of every major town in the kingdom of León-Castilla.⁽⁷¹⁾ Unlike the latter, it was also a true commercial center that battened on the trade that ran along the old Roman road from Cadiz, Sevilla, and Córdoba in the south to Zaragoza in the north.

Once the outright conquest of the taifa of Toledo was undertaken, Alfonso VI would have another world to reorient and restructure as it was absorbed into a Christian northern Spain which had been developing along other pathways for almost four hundred years. Furthermore, the moment that conquest was effected, the kingdom of Toledo ceased to be a client state whose surplus wealth could be drained away to secure other ends. Immediately it became a province to be nurtured, developed, governed, and protected. The scarcest of all resources of León-Castilla, human lives, must be invested there. Small wonder then if Alfonso VI hesitated long before committing himself irreversibly to such a course.

1. Sánchez-Albornoz, Despoblación y repoblación.

2. Reilly, León-Castilla under Queen Urraca, pp. 296-300.

3. Sisnando is usually thought to have been installed as the governor of Coimbra by Fernando I after the city's capture in 1064. Emilio García Gomez and Ramón Menéndez Pidal, "El conde mozárabe Sisnando," Al-Andalus 12 (1947): 35. That may be, but the first documentary record of his rule there dates to May 1, 1070, and even that document is not beyond suspicion. See chapter 2, note 63. The charters of Alfonso VI of 1075 may actually record his initial appointment as well as his presence. See chapter 5, note 63.

4. August 1, 1076. Serrano, ed., Becerro gótico, pp. 239-40. It appears to have been issued in Hinestrosa about three kilometers south of Castrojeriz for all the men of that village are recorded as present. Aug. 1, 1076. Bruel, ed., Recueil des chartes 4:604-607. It is a donation to Cluny. An apparent reworking of the charter is dated to Jan. 29, 1077. Ibid., pp. 622-25.

5. Emilio Sáez, ed., Colección deplomática de Sepulveda (Segovia, 1956), pp. 3-7; and Férotin, ed., Recueil des chartes, pp. 23-26, both of whom date it to Aug. 20, 1076. The problems surrounding the issuance of this charter have been studied intensively by Antonio Linage Conde, "La donación de Alfonso VI a Silos del futuro priorato de San Frutos y el problema de la despoblación," AHDE 41 (1971): 973-1011; and María de la Sotterana Martín Postigo, "Donación del lugar de San Frutos por Alfonso VI a Silos: Reconstitución del privilegio por las fuentes. Estudio diplomático," Estudios Segovianos 22 (1970): 333-96. The latter redates the original text to Aug. 17, 1076. The charter was issued at Navares and there are three villages with some variation of that name north of Sepúlveda.

6. Aug. 27, 1076. Acad. Hist., Colección Salazar, 0-17, fol. 775r-v; and BN, Manucritos, 720, fol. 224r-v; the latter dated to Sept. 1. A royal charter given to Cluny. Aug. 1076. García Larragueta, ed., Documentos de Oviedo, pp. 232-33. A private charter probably issued at this time. Sept. 5, 1076. AHN, Códices, 989B, fol. 82r. Again a private charter. Sept. 8, 1076. AC León, Códice 11, ff. 53v-54r; Acad. Hist., Catedrales de España, 9-25-1-C-4, ff. 70r-72r; pub. ES 36:65-67 append. This last is a charter of the infantas Urraca and Elvira to the church of León.

7. Emilio Sáez, Rafael Gibert, Manuel Alvar, and Atiliano González Ruiz-Zorilla, Los fueros de Sepúlveda (Segovia, 1953), pp. 44-51. The document raises many problems already discussed in my León-Castilla under Queen Urraca, pp. 323-24, but I remain convinced of the reality of a royal charter of that date. See also my "Chancery of Alfonso VI," pp. 10-11. A thirteenth-century copy unknown to the editors exists in AHN, Port., Mosteiro de Lorvão, Caixa 60, Maço 1, no. 19.

8. In addition to the sources cited in notes 5 and 7, see the discussion in Salvador de Moxó, Repoblación y sociedad en la España cristiana medieval (Madrid, 1979), pp. 44-45, and Antonio Linage Conde, "En torno a la Sepúlveda de Fray Justo," Homenaje a Fray JustoPérez de Urbel, vol. 1 (Silos, 1976), pp. 60-61.

9. Charles J. Bishko, "El Castellano, hombre de llanura." Homenaje a Jaime Vicens Vives, vol. 1 (Barcelona, 1965), p. 208, and also his "The Peninsular Background of Latin American Cattle Ranching," Studies in Medieval Spanish Frontier History (London, 1980), pp. 494-95.

10. Sánchez-Albornoz, Despoblación y repoblación, p. 375, and Angel Barrios García, "Toponomástica e Historia. Notas sobre la despoblación en la zona meridional del Duero." In En la España Medieval, vol. 2 (Madrid, 1982), pp. 131-32.

11. AHN, Códices, 989B, fol. 232v; pub. Hinojosa, ed., Documentos de León y de Castilla (Madrid, 1919), pp. 32-33.

12. For Alfonso's charter, see chapter 6, note 7. The private charter, AC León, Códice 11, fol. 60r-v, reveals the bishops of Modoñedo and Oviedo at court.

13. See chapter 6, notes 33 and 37.

14. See those who confirm the following documents. June 27, 1077. AHN, Códices, 989B, fol. 146r. July 10, 1077. Bruel, ed., Recueil des chartes 4:627-29. July 12, 1077. AHN, Clero, Carpeta 884, no. 1.

15. Serrano y Sanz, ed., "Documentos de Celanova," pp. 35-37.

16. López Ferreiro, Historia de Santiago de Compostela 3:3-7 append., and Jesús Carro García, "La escritura de concordia entre Don Diego Peláez, obispo de Santiago, y San Fagildo, abad del monasterio de Antealtares," CEG 4 (1949): 112-18.

17. AHN, Clero, Carpeta 557, no. 15. This charter slightly enlarges the privileges contained in the one his brother García had granted some ten years before. See chapter 2, note 42. Although some reasons would have existed for simply reworking this earlier charter and attributing it to Alfonso, the diplomatic of the Alfonsine charter is too good to discount it.

18. AC León, Códice 11, ff. 52v-53r. But see note 6. The same property is conveyed in both but the later one lacks the confirmation of Queen Inés, which ommission would have been politic if that lady had been forced into retirement.

19. Serrano, ed., Cartulario de San Millán, p. 238.

20. Ibid., pp. 238-39, 239-40, and 240-41. Antonio de Yepes, Corónica general de la Orden de San Benito, vol. 1 (Irache and Villadolid, 1609-1621), pp. 33-34 append. Loperráez Corvalán, Descripción históric de Osma 3:6. AHN, Clero, Carpeta 701, no. 18

21. See chapter 6, note 39.

22. Ubieto Arteta, ed., Cartulario de Santa Cruz, pp. 23-24. The charter reads "usque un Ilia." A notice of yet another charter of Alfonso to San Millán, dated only to 1078, may in fact be of the preceding year. Acad. Hist., Colección Salazar, 0-16, fol. 88v

23. Boissonnade, Du nouveau sur la Chanson de Roland, pp. 30-31.

24. AHN, Clero, Carpeta 884, no. 5, and original; and a copy in Códices, 989B, ff. 68v-69r.

25. Feb. 28, 1078. AHN, Códices, 1.043B, fol. 37r-38v, and 363B, ff. 45v-46v and 67r-68v; pub. ES 40:417-22, and Antolín López Peláez, El señorio temporal de los obispos de Lugo vol. 2 (Coruña, 1897), pp. 111-14. Mar. 1, 1078. AHN, Códices, 988B, fol. 19r-v; 989B, fol. 5r-v; Clero, Carpeta 884, no. 7; Acad. Hist., Colección Salazar, 0-22, ff. 7r-8v; pub. Fita, "Concilio nacional de Burgos," pp. 338-41. Mar. 20, 1078. AHN, Clero, Carpeta 884, no. 8; pub. Escalona, Historia de Sahagún, pp. 474-75. Apr. 2, 1078. AHN, Códices, 989B, fol. 108r-v. May 14, 1078. Ibid., fol. 70r-v.

26. Lucas de Túy, "Chronicon Mundi," p. 100. Al-Maqqarí, The History of the Mohammedan Dynasties in Spain, vol. 2, trans. Pascual de Gayangos (London, 1840), p. 263.

27. José Miranda Calvo, La reconquista de Toledo por Alfonso VI (Toledo, 1980), p. 56.

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

29. Julio González, "Aportación de fueros castellano-leoneses," AHDE 16 (1945): 627-29; and Justiniano Rodríguez Fernández, Palencia: Panorámica foral de la provincia (Palencia, 1981), pp. 219-21.

30. AHN, Clero, Carpeta 884, no. 10; and Códices, 989B, fol. 163v. A private charter.

31. Luciano Serrano, ed., Cartulario de San Vicente de Oviedo, 781-1200 (Madrid, 1929), pp. 87-90.

32. AD León, Gradefes, no. 5. "Et ista karta fuit facta et roborata hic in ipso flubis que discurrit justa de toledo. Id est guadarrama in fussato."

33. For the confused events on the Muslim side, I follow the chronology of Huici Miranda, Historia musulmana de Valencia 1:200, who seems to me to have best tracked that jungle of rhetoric.

34. Turk, El reino de Zaragoza, pp. 146-47. The only notice in the Christian sources places the capture of that town in September 1077. David, Études historiques, p. 299. But see Antonio Prieto y Vives, Los reyes de taifas (Madrid, 1926), p. 67, n. 2. Like virtually all the other historians of the question, I follow the Muslim chronology here, which seems best to comport with what is known of Alfonso's movements. For the strategic position of that fortress, see Manuel Terrón Albarrán, El sólar de los Aftásides (Badajoz, 1971), pp. 125-27.

35. Menéndez Pidal, España del Cid 2:921-22.

36. July 20-Aug. 1, 1079. The date is uncertain. AHN, Clero, Carpeta, 1.048, no. 16, has the latter date but it is a copy of 1485. Pub. Ildefonso Rodríguez de Lama, ed., "Colección diplomática riojana," Berceo 16 (1961): 366-68. Serrano, Cartulario de San Millán, pp. 246-47, published an earlier copy dated July 20, 1079.

37. See chapter 6, note 57.

38. Peter Segl, Königtum und Klosterreform in Spanien (Kallmünz, 1974), p. 211.

39. AHN, Códices, 989B, fol. 66v.

40. See chapter 6, note 58.

41. Ibid., note 59.

42. AHN, Códice, 989B, fol. 128v.

43. See chapter 6, note 66. There is also a document of Sahagún dated May 29, 1080, which seems to have been confirmed by the royal alférez and the bishop of León, but that is slim evidence to place the court there so late. AHN. Códices, 989B, fol. 90r.

44. Huici Miranda, Historia de Valencia 1:200. Menéndez Pidal, España del Cid 1:264-65; González, Repoblación de Castilla la Nueva 1:71-72; and others have dated the entire al-Mutawakkil episode in Toledo rather to 1080-81. The works of the later Muslim chroniclers on whom all parties rely are obviously difficult to reconcile precisely. I have chosen to follow Huici Miranda's reading of them for two reasons. First, as we shall see, the documents of Alfonso VI in the spring and summer of 1081 put him far from Toledo. Second, the capture of Coria in 1079 becomes a logical part of a campaign against a king of Badajoz ensconced in Toledo rather than appearing as an isolated venture.

45. Abd Allah, Memorias, pp. 157-59.

46. Lévi-Provençal, Islam dOccident, p. 127; González, Repoblación de Castilla la Nueva 2:72-73.

47. Abd Allah, Memorias, p. 163.

48. Menéndez Pidal, España del Cid 2:923.

49. See chapter 6, note 66.

50. AHN, Códices, 989B, fol. 138r. A second document of December 3, 1080, is untrustworthy as to contents but may accurately reflect the continuing royal presence there. AHN, Clero, Carpeta 890, no. 6; Códices, 989B, ff. 22v-23r; Acad. Hist., Colección Salazar, 0-22, fol. 13v. See Ana María Barrero García, "Los fueros de Sahagún," AHDE 42 (1972): 399-401. A document dated Dec. 4, 1080, puts the court near Benavente some seventy-five kilometers southwest of Sahagún, AHN, Clero, Carpeta 1.239, no. 18; copy in BN Manuscritos, 18.387, ff. 295v-297r. García Alvarez, "Catalógo de documentos," Compostellanum 11 (1966): 312-13, called the latter an original but the caroline script marks it as a copy.

51. España del Cid 1:210-21.

52. Ibid., 2:921.

53. Ibid., pp. 719-24. For the document, see García Larragueta, ed., Colección de Oviedo, pp. 255-56. For the critique, Fernández Conde, Libro de Testamentos, pp. 288-90.

54. Serrano, ed., Cartulario de San Millán, pp. 234-36 for the relationship of the brothers; pp. 23 9-40 for one reference to this Count García; and Loperráez Corvalán, Obispado de Osma 3:6, for the other.

55. July 1, 1074. AHN, Códices, 989B, fol. 137v; pub. Escalona, Historia de Sahagún, p. 473, with a date of June 24, 1074. He does appear in the problematic carta de arras of El Cid and in the false diploma of Alfonso to Burgos in 1075.

56. See note 35. All three appear together in the charter of Sancho II of January 18, 1070, for instance. See chapter 3, note 18.

57. AHN, Códices, 105B, fol. 95r-v. It is dated only by year.

58. See chapter 6, note 66.

59. AHN, Códices, 105B, ff. 99r-100v; pub. Rodríguez de Lama, ed., Colección de la Rioja 2:91-92.

60. Feb. 1, 1081. AHN, Códices, 989B, fol. 134r. Mar. 31, 1081. Ibid., fol. 193v. The latter document may not be entirely trustworthy since it was confirmed by Bishop Pedro of Astorga who had been forced into retirement by this time. On the other hand, a charter of Alfonso VI himself, dated Jan. 6, 1081, to the Castilian monastery of San Pedro de Arlanza, which would seem to place the king there, is suspect by reason of the confirmation of the later Bishop Raymond of Palencia Serrano, ed., Cartulario de San Pedro, pp. 155-56.

61. Serrano, ed., Cartulario de San Millán, pp. 94-95.

62. Bruel, Recucil des chartes 4:719-22, published one copy. Another, AHN, Clero, Carpeta 1.030, no. 5, was published by Rodríguez de Lama, Colección de la Rioja 2:93-94. Yet another copy exists in Acad. Hist., Colección Salazar, 0-17, ff. 788v-790r.

63. AC León, Códice 11, ff. 96r-97r; another copy in Acad. Hist., Catedrales de España, 9-25-1-C-4, ff. 76r-78v. Pub. Luis Sánchez Belda, Cartulario de Santo Toribio de Liébana (Madrid, 1948), pp. 118-20.

64. AHN, Códices, 989B, ff. 20v-21r. Again the confirmation of Bishop Pedro of Astorga raises a question.

65. Cf. Miranda Calvo, Reconquista de Toledo, pp. 80-81, and González, Repoblación de Castilla la Nueva 1:72-73.

66. AC León, Códice 11, ff. 29v-30r.

67. Dec. 3, 1081. AHN, Clero, Carpeta 959, no. 9; pub. Vignau, Cartulario de Eslonza, pp. 74-75. Dec. 25, 1081. AC Burgos, vol. 2, pt. 1, fol. 50r-v; pub. Serrano, Obispado de Burgos 3:61-62. This original charter was not confirmed by the bishop of León.

68. Gonzalo Martínez Díez, Alava medieval, vol. 1 (Vitoria, 1974), p. 95.

69. Ibn al-Kardabus, "Kitab al-Iqtifa," in al-Maqqari, The History of the Mohammedan Dynasties in Spain, vol. 2, ed. and trans. Pascual de Gayangos (1843; London, 1964), p. xxx append.

70. Miranda Calvo, Reconquista de Toledo, p. 81.

71. Russell, Medieval Regions and Their Cities (Bloomington, 1972), p. 178.