[3] In the second half of the eleventh century a reconstituted kingdom of León-Castilla was poised on the northern bank of the Duero River. Before it to the south lay the almost deserted other half of the most extensive river basin in the Spanish peninsula. From its origin in the eastern highlands of Soria to its mouth on the Atlantic at Oporto the Duero stretches for more than 900 kilometers across the meseta of Castilla la Vieja and drains some 98,000 square kilometers.

For two centuries now, since Ordoño I of Asturias (850-866) had re-populated the city of León, the shallow, meandering river had served as a moat and wall protecting the growing Christian communities to its north against all but the best organized expeditions of Islam. The river itself is never deep during spring, summer, or fall, the campaigning seasons, but from the Portuguese highlands in the west to the Sorian highlands in the east the north bank of the river generally rises steeply to form a natural wall. This natural fortification was never impenetrable, of course, but from very early it was improved by a line of fortresses that made it an even more formidable obstacle. From Zamora, to Toro, to Tordesillas, to Simancas, and beyond through Peñafiel, Aranda de Duero, and Peñaranda de Duero, major fortifications guarded the river crossings at the average of one every thirty kilometers. To reduce any one of them was difficult; to leave them, unreduced, behind an invading army was perilous.

North of this frontier lay the kingdom of León-Castilla. Even then the realm comprised three major and geographically distinct parts. On the far side of the Cantabrian Mountains, along the Bay of Biscay, lay Asturias, whose major passes communicating with the south averaged 1,500 meters in altitude. To the west, fronting on the Atlantic, Galicia stretched south to what is now northern Portugal, also separated from León by a series of rugged ranges whose passes averaged 1,000 meters in height. By the beginning of the tenth century these two well watered valley regions were more heavily populated than ever before in their history.

From these two sheltered redoubts, a gradually increasing flow of [5] settlers moved south onto the Leonese and Castilian meseta, into the country of valleys and streams between them called the Bierzo, and south across the Miño River into northern Portugal. In these latter two areas they had found a hilly, wet, and sheltered environment which was essentially an extension of their ancestral lands. But when they had begun to emerge onto the Leonese-Castilian meseta they faced instead a flat, dry, open world that furnished a fundamentally different arena for the testing of both agricultural and political skills.

The history of this two-century-long process is most incompletely written or researched, but enough is known about it, and the monuments of art, architecture, and demography it produced, to allow one to extrapolate a general picture.⁽¹⁾ The repopulators should be envisioned as hardly more than small groups of extended families inching south along the tributaries of the Duero: the Balimbre, the Tuerto, the Órbigo, the Luna, the Bernesga, the Esla, the Cea, the Carrión, the Pisuerga, and as many more. At first they would have cultivated the lands immediate to the rivers where the agricultural challenge would not have been too severe. As these river-bottom settlements leap-frogged south in a series of advancing settlements paralleling one another across the meseta from east to west strong points or castles seem to have protected them at the distance of not more than five or six kilometers along each watercourse. Gradually then, not only did the line of settlements along the tributaries reach the Duero, but a continuous line of minor fortifications stretched back from the major fortresses of the Duero into the foothills of the northern and western mountains. Like Castilla, early León as well was a "tierra de castillos."

As the pressure of population increased in these river valleys, the wooded land bordering each of them was gradually attacked, making east to west communication and association more regular. The average distance between the tributaries rarely exceeded fifteen to twenty kilometers after all. But the lateral movement away from the river bottoms created an agricultural world in which dry farming was usually a necessity and wheat a more and more important crop. In adjusting to the [6] changing conditions of agriculture so created, the skills provided by yet another current of immigration must have been of the greatest importance. The bearers of these techniques were the Mozárabs, Christians of Arabic culture, who moved northward in large numbers from the last half of the ninth century. These newcomers settled widely in the western and northern portion of the meseta north of the Duero even up to the foothills of the Cantabrians.⁽²⁾

By the beginning of the eleventh century 150 years of such gradual repopulation had created a new center of gravity in the Christian world of northern Spain. The older realms of Asturias and Galicia were increasingly becoming peripheral even if never unimportant. Constantly reinforced by new streams of immigrants from the older lands to its north and south, León-Castilla had become a sturdy society of agricultural hamlets, small monasteries, and scarcely larger cities. The resilience of this society was dramatically demonstrated between 985 and 1002 when the dictator and general of the world of Muslim Spain, Almanzor, took and sacked Burgos, León, and Zamora in turn. Nevertheless, he was unable to occupy it permanently, and such damage as he was able to do was gradually made good by Alfonso V of León (999-1028).

The geographical diversity of the new realm about to emerge was matched by its linguistic and political heritage. The vernacular of Galicia was surely incomprehensible to the Asturians, and the speech of the latter must have been almost as difficult to understand for the inhabitants of Old Castile. The largely north to south, parallel course of the repopulation carried these linguistic properties southward and, in León in particular, brought them into contact with yet more archaic forms carried north by the Mozárabs.⁽³⁾ The relative isolation, one from the other, of these linguistic groups served to perpetuate their differences into the eleventh century.

Within an institution like the church, and to a lesser extent the royal government, a simple Latin served to provide a medium of communication. Among the greater nobility, whose family relations and estates seem to have been widely extended from an early date, some type of lingua franca must have already emerged for the conduct of their affairs. But for the great mass of the population, the peasantry, the incomprehensibility of their neighbors speech reinforced the particularism inherent in their geographical separation. In the course of their [7] everyday lives, scarcely touched by commerce or travel, this would have meant little. Only when they were marshaled into a war band or army, perhaps the one, true general endeavor of the age, would it have mattered. Then, however, they would have been ungovernable and unleadable except by the nobility of their own district or province. Such, one expects, was a principal root of the political particularism that found all too dramatic expression in the events of the age.

The recovery of the kingdom of León from the great devastations it had suffered in the time of Almanzor did not leave its king, Alfonso V, in anything like uncontested control of the meseta north of the Duero. In the tenth century an independent county of Castilla had arisen under Fernán González (930-970) in its eastern reaches and whose territories fluctuated from the Duero in the south to the Basque country of Alava and to Asturias de Santillana in the north. Its fortunes had waned, however during the minority of its last count, García Sánchez (1017-1029). The question of the day was just by whom the county of Castilla would be absorbed.⁽⁴⁾

The death in battle of Alfonso V of León in 1028 opened a decade of kaleidoscopic changes on the meseta. He left behind him a kingdom nominally ruled by his eleven-year-old son, Vermudo III (1028-1037), but actually in the hands of the boys mother, Urraca, sister of the king of the mountain realm of Navarra. Sancho el Mayor of Navarra (1000-1035) by this time not only held that tiny realm but had extended his sway over the minuscule counties of Aragón, Sobrarbe, and Ribagorza, the wild Basque country, and the upper Ebro River district of Rioja. In addition, Sanchos influence was powerful in the extreme east, where Count Berenguer Ramón of Barcelona (1018-1035) was almost his client, and in Castilla, for the sister of its young count was his wife.⁽⁵⁾

In 1029 Count García Sánchez of Castilla was about to be married to Sancha of León, the sister of Vermudo III, an arrangement apparently sanctioned by the king of Navarra, when the count was murdered in the city of León. Sancho el Mayor then claimed the county of Castilla in his wifes name and installed in it their son, Fernando, as the new count of Castilla. The Navarrese kings policy now became a mixture of aggrandizement sometimes ostensibly on behalf of his son and at other times simply in his own behalf. He seized the linguistic borderlands between the Cea and the Pisuerga rivers, long a bone of contention [8] between León and Castilla. After he had forced the marriage between Fernando and Sancha, Vermudo III's sister, in 1032, those lands went to Castilla as part of her dowry. In 1034 he wrested the city of León itself from Vermudo, who retreated into Galicia, and began to style himself "Emperor" on his coinage. By the time Sancho died in 1035 the meseta north of the Duero was dominated by the mountain kingdom of Navarra.

However, for all of Sancho el Mayors diplomatic and military skill, one thinks that such a situation was essentially a fortuitous result of the youth of his opponents. Certainly the brute facts of geography and demography militated against its continuance. After his death, Vermudo III was immediately received back into León and soon began a campaign to recover from Castilla and his brother-in-law the disputed territory between the Cea and Pisuerga. But Count Fernando defeated and killed Vermudo III at the battle of Tamarón on September 4, 1037. Since the latter died without an heir, the kingdom of León now recognized Sancha and her husband as its rulers, and Fernando was anointed king in the royal city on June 22, 1038. The dynasty of Navarra had triumphed perhaps, but the united realms of León-Castilla with its rimlands of Asturias and Galicia had become the political center of the north Spanish, Christian society.

The internal consolidation of the realm of León-Castilla under Fernando el Magno and Sancha (1037-1065) is a history that remains to be researched and written. For our present purposes it is sufficient to say that within little more than a decade and a half Fernando was strong enough to begin a policy of expansion that ended only with his death.

The first gains came at the expense of his elder brother, García Sánchez III of Navarra (1035-1054), whom he defeated and killed at Atapuerca on September 15, 1054. García had been the heir not only of that kingdom and the traditionally Castilian lands of Alava, the Rioja, and the Bureba at his fathers death but also of his fathers ambitious southern policy. The victory at Atapuerca reduced the kingdom of Navarra to a vassal state under the late kings son, Sancho García IV (1054-1076), but Fernando was generous in his triumph and required the cession only of the Bureba district on the west bank of the upper Ebro River around the royal monastery of Oña.⁽⁶⁾The other heirs of Sancho el Mayor were not then a problem for Fernando. Gonzalo had died in 1045 and his counties of Sobrarbe and Ribagorza had been acquired by Ramiro. The illegitimate scion of Sancho el Mayor, Ramiro I of [9] Aragón (1035-1063) thus enlarged his tiny realm and would begin to style himself king, but no one could have anticipated as yet the spectacular future that awaited that remote, landlocked territory.⁽⁷⁾

The great enemy, however, was Spanish Islam. Although the caliphate of Córdoba had disintegrated after the death of Abd al-Malik, Almanzors son, in 1008, a series of powerful taifa states still rimmed the southern half of the Duero basin, inhibiting its repopulation from León-Castilla. These major taifa kingdoms were those of Zaragoza, Valencia, Toledo, and Badajoz.⁽⁸⁾So long as these Muslim realms were unshaken, Christian shepherds might utilize the trans-Duero, seasonally and at their peril, to graze their herds, but the expansion there of a permanent agricultural population from the north simply could not occur. The Muslim use of that vast expanse was of the same character and for largely the same reason.⁽⁹⁾The area was an open frontier exposed to destructive raids from both north and south as opportunity offered and when cupidity found a suitable objective. Very gradually, Fernando I set about modifying that situation, creating the grounds of permanent settlement.

Already in 1055 he may have begun his attacks on the Portuguese territories of the taifa of Badajoz.⁽¹⁰⁾ Portugal was walled off from the meseta of León-Castilla by a continuous mountain chain, impassable for practical purposes by an army of the day. Access to it from León was simplest by the route that led through southern Galicia and emerged into the valley of the Miño River, part of the present-day boundary of Spain and Portugal, orjust south of it into the valley of the Limia River. What then presented itself south of the Miño was a land composed of a series of river basins made up chiefly of those of the Duero, the Mondego, and finally the Tajo. These rivers all drop down sharply from the mountains in the east, cross a shallow plain rarely more than thirty kilometers wide, and empty into the Atlantic. Only the last of these rivers provided a possible invasion route from Spain and, at that time, from Muslim Spain.

The attack on the Muslim in Portugal then would proceed, as it had [10] for centuries, from Galicia. The basin of the Duero in its lower reaches and Oporto at its mouth together with the country north to the Miño had long been held by León-Castilla. But a hundred kilometers upriver the Muslim stronghold of Lamego was a nuisance if not a threat. On November 29, 1057, Fernando I's army captured it and the upland valleys it had protected.⁽¹¹⁾

With the basin of the Duero cleared, Fernando now moved toward the reconquest of that of the Mondego. The first step was the reduction of Viseu, which overlooked the middle reaches of that river from the north, some seventy kilometers northeast of Coimbra. That was accomplished on July 25, 1058, but a long and grueling battle for the remainder of that basin was to follow.⁽¹²⁾ It was not until six years later that the hilltop, fortress city of Coimbra fell, after a six-month siege, on July 25, 1064.⁽¹³⁾ With that victory, the basin of the Mondego was secured for León-Castilla, and in central Portugal only the basin of the Tajo remained to Islam.

The arduous Portuguese campaigns by no means monopolized all the time and energies of Fernando I, and much of the fighting must have been carried out by essentially local troops and in his absence. Contemporaneously the Leonese monarch was more than active on his eastern frontier as well. His gradual reclamation of the traditional Castilian territory held by Sancho García IV of Navarra was apparently a political and diplomatic process rather than an overtly military one for it has left no record in the chronicles and must be followed in the tortuous record of the charters. By 1065, however, it had carried his domains up to the borders of Rioja itself.⁽¹⁴⁾

Fernando's pressure against the great taifa of Zaragoza has left a more prominent record but one with a difficult chronology. The primary chronicle of his reign tells of a campaign up the narrow valley of the eastern Duero into the Sorian highlands which took the fortresses of San Esteban de Gormaz, Berlanga, and Vadorrey, and then swung south across the rolling country to Santiuste, Huermeces, and Santamara. This drive put him in a position to threaten the Roman road from [11] Toledo to Zaragoza, which was the great avenue of communication and reinforcement from Muslim Andalucía to the latter, or to proceed down that road against the taifa kingdom of Toledo on whose borders he now stood. The chronicles editor dates this campaign about 1060.⁽¹⁵⁾ Whatever its precise date the offensive certainly took advantage of the preoccupations of al-Muqtadir, king of Zaragoza. The Muslim leader was in the process of absorbing the neighboring taifa of Tortosa⁽¹⁶⁾ and at the same time defending his northeastern frontier against the coordinated attacks of Ramiro I of Aragón and Ramón Berenguer I of Barcelona.

The combination of these problems seems to have led him to drop his alliance with the relatively ineffective Sancho García IV and to become a tributary state of León-Castilla.⁽¹⁷⁾ However much the taifa king may have considered this step a temporary expedient, circumstances were to make it a permanent arrangement down to 1065. The pressure of Aragón in the north proved to be unrelenting, and in 1064 a crusade of papal inspiration brought a host of foreign warriors into the northeast of Spain where they scored an impressive triumph, capturing the fortress of Barbastro in 1064.⁽¹⁸⁾

The same circumstances, by effectively nullifying the power of Zaragoza, allowed Fernando to turn his attention to the central taifa, Toledo. Although it is difficult to know their precise nature, relations between the Leonese king and al-Mamun (1043-1075), the king of Toledo, were already close. The last known Mozarabic bishop of Toledo, Pascual, was consecrated in León in 1058, and he certainly would have been at least acceptable to Fernando.⁽¹⁹⁾ In 1062, the latter swept around the eastern end of the Guadarrama mountains into the eastern reaches of al-Mamuns realm. He took Talamanca north of Madrid and [12] pressed on to lay siege to Alcalá east of that same city. Wasting the countryside extensively, he demanded tribute from the Toledan king, who agreed to pay it to secure the formers withdrawal.⁽²⁰⁾

Fernando I now had forced two of the greatest of the taifa kingdoms to render annual parias, or tribute, and the establishment of these exactions provided him with even greater resources which could be turned to military purposes. He was not slow to utilize them in just that fashion for in 1063 the armies of León-Castilla struck even deeper into Andalucía. They ravaged the lands of the taifas of Sevilla and Badajoz, forcing the former to ransom itself and probably the latter as well.⁽²¹⁾ In addition to generalizing yet more the system of parias, the great raid, or razzia, probably aimed at a weakening of Andalusian Islam, which would disable or disincline it from reacting to the final offensive against Coimbra in the following year.

That six-month siege of the hilltop city on the Mondego seems to have fully engaged the military energies of Fernando during 1065, but in the final year of his reign he launched yet another offensive against Spanish Islam. This time the main target was the taifa of Valencia on the eastern coast. A strong Muslim reaction had occurred in the east, and the combined forces of Zaragoza and Valencia had struck into the valley of the upper Duero in 1064.⁽²²⁾ There had been a massacre of Christians in Zaragoza in January 1065.⁽²³⁾ Doubtless al-Muqtadir had broken off payment of the parias as well. Then in April a combined army of Zaragoza and Sevilla had retaken the fortress of Barbastro, crushing the French crusaders who had remained there.⁽²⁴⁾

The sources do not reveal much about the Christian campaign but there seems to be no doubt that it was a success. Fernando penetrated to the very environs of the city of Valencia itself and there defeated Abd al-Malik in the late fall of 1065. The defense of the latter seems to have been so inept that it moved al-Mamun of Toledo to assume the real direction of the affairs of that taifa, whose king was already his son-in-law.⁽²⁵⁾ Zaragoza returned to its tributary status and the Leonese monarch, having fallen ill, ended the campaign in November. This was to be Fernando's last triumph, however, for he never recovered from his [13] illness. On December 27, 1065, he died in the city of León and was buried there in the church of San Isidoro.⁽²⁶⁾

When Fernando el Magno died the kingdom of León-Castilla was arguably the greatest power in the entire peninsula. The three strong taifa kingdoms of Zaragoza, Toledo, and Badajoz had all been forced to recognize his overlordship, to pay parias regularly, and had yielded up portions of their territories of greater or lesser extent. He had made his strength felt, as well, in the rather more distant Muslim realms of Valencia and Sevilla. Among the major kingdoms of Spanish Islam only Córdoba and Granada had gone untouched. In the Christian north, Navarra had become a vassal state, given up territory, and was pinned against the Pyrenees by León-Castilla and Zaragoza. Aragón was still a petty principality of the northeastern foothills and remote Barcelona no more than a struggling if ambitious county.

For twenty-eight years Fernando and Sancha had ruled León and Castilla on the plain north of the Duero together with Asturias and Galicia in the mountains of the north and west. Their efforts had made them now masters of the northern two-fifths of modern Portugal as well and of a strategic salient extending along the upper Duero into the highlands of Soria. The Muslim taifas of the northern tier of Spanish Islam had been humbled and shaken. All appeared to be in readiness for the acquisition and repopulation of the 50,000 square kilometers of the trans-Duero, the southern portion of the meseta which lay between the Duero River and the Cordillera Central of the Guadarrama and Gredos mountains and the Sorian highlands and the mountains of Portugal to the east and west. The beginning of that recovery was to be retarded nonetheless because before his death Fernando had provided for the division of his own realm between his three sons.

1. For the history of the kingdom and society of León-Castilla up to the period here considered the most useful volumes are by Justo Pérez de Urbel and Ricardo Del Arco y Garay, Historia de España, vol. 6, España cristiana, comienzo de la reconquista (711-I038), 2d ed. (Madrid, 1964), and Alfonso Sánchez Candeira, El "Regnum Imperium" leonés hasta 1037 (Madrid, 1951). The relevant portions of Luis García de Valdeavellano, Historia de España de los orígenes a la baja Edad Media, 2d ed., vol. 1 (Madrid, 1955), are always useful. Recent Spanish scholarship has produced two powerful surveys in José Angel García de Cortazar, Historia de España Alfaguara, vol. 2, La época medieval (Madrid, 1973), and in José Luis Martín, La península en la Edad Media (Barcelona, 1976). The best survey in English is that of Joseph F, OCallaghan, A History of Medieval Spain (Ithaca, 1975).

2. The indispensable work for all of this development is Claudio Sánchez-Albornoz, Despoblación y repoblación del valle del Duero (Buenos Aires, 1966).

3. The standard volume is that of Ramón Menéndez Pidal, Orígenes del Español, 5th ed. (Madrid, 1964).

4. For its independent history see Justo Pérez de Urbel, El condado de Castilla, 2d ed., 3 vols. (Madrid, 1969).

5. Justo Pérez de Urbel, Sancho el Mayor de Navarra (Madrid, 1950), is the standard biography. The surest guide to the history of the counts of Barcelona is Ferran Soldevila, ed., Historia dels Catalans, 3 vols. (Barcelona, 1964).

6. For the history of the kingdom of Navarra see José María Lacarra, Historia del reino de Navarra en la Edad Media (Pamplona, 1975).

7. José María Ramos y Loscertales, El reino de Aragón bajo la dinastía pamplonesa (Salamanca, 1961), is still the best general account but it may be supplemented in some particulars by Antonio Ubieto Arteta, Historia de Aragón (Zaragoza, 1981).

8. On the defensive frontier of Muslim Spain see Julio González, Repoblación de Castilla la Nueva, vol. I (Madrid, 1975-76), pp. 28-63.

9. ^{Sánchez-Albornoz, Despoblación y repoblación, p. 375.}

^10. The only narrative account of the Portuguese campaigns is confused in its chronology. Pérez de Urbel and González Ruiz-Zorrilla, eds., Historia Silense, pp. 188-94. Joaquim Veríssimo Serrão, História de Portugal, vol. I, 3d ed. (Lisbon, 1979), is a reliable survey.

11. Pierre David, Etudes historiques sur la Galice et le Portugal du VI^e au XII^e siècle (Paris, 1947), p. 296. All references to the old Portuguese annals will be to the editions in this volume. Serrão, História de Portugal 1:66.

12. David, Etudes historiques, pp. 296-97.

13. Ibid., pp. 311-12.

14. José Angel García de Cortázar. El dominio del monasterio de San Millán de la Cogolla, siglos X a XIII (Salamanca, 1969), pp. 169-70. Félix Sagredo Fernández, Briviesca antigua y medieval, 2d ed. (Madrid, 1979), p. 99. The latter authors "Los condes de Bureba en la documentación de la segunda mitad del siglo XI," CH 6 (1975): 91-119, is incorporated literally and completely into the book.

15. Historia Silense, pp. 194-95.

16. Afif Turk, El reino de Zaragoza en el siglo XI de Cristo, V de la hégira (Madrid, 1978), p. 81. Originally pub. in Revista del Instituto Egipcio de Estudios Islámicos 17 (1972-73): 7-122.

17. Charles Julian Bishko, "Fernando I y los orígenes de la alianza castellano leonesa con Cluny," CHE 47-48 (1968): 115-116, who rather dates it to 1058-59. The original English of this study has been printed since in Studies in Medieval Spanish Frontier History (London, 1980), pp. 1-136.

18. Antonio Ubieto Arteta, Historia de Aragón, pp. 54-62. The same author would redate the battle of Graus and the death of Ramiro I of Aragón in battle against the joint forces of León-Castilla and Zaragoza, ibid., pp. 68-76. We need not take up the controversy until later.

19. AC León, Códice 11, fol. 264r-v, pub. ES 36:51-52 app. A private document confirmed, among others, by "Pascualis eps. Toletanus ibi fui tunc ordinatus simul cf." See also Juan Francisco Rivera Recio, Los arzobispos de Toledo desde sus orígenes hasta fines del siglo XI (Toledo, 1973), pp. 205-206.

20. Historia Silense, pp. 196-97.

21. Ibid., p. 198.

22. Ibid., p. 194.

23. Antonio Ubieto Arteta, Historia de Aragón, p. 62.

24. Turk, Reino de Zaragoza, pp. 96-99.

25. Historia Silense, pp. 206-207. Ambrosio Huici Miranda, Historia musulmana de Valencia y su región, vol. 1 (Valencia, 1969), p. 186.

26. Historia Silense, pp. 208-209.