Muslim Spain and Portugal

A Political History of al-Andalus

Hugh Kennedy

• Publisher: Routledge
• Year: 2014
• ISBN:
  9781317870401
• Edition: 1
• Topic: Geschichte
• Subtopic: Weltgeschichte

This is the first study in English of the political history of Muslim Spain and Portugal, based on Arab sources. It provides comprehensive coverage of events across the whole of the region from 711 to the fall of Granada in 1492. Up till now the history of this region has been badly neglected in comparison with studies of other states in medieval Europe. When considered at all, it has been largely written from Christian sources and seen in terms of the Christian Reconquest. Hugh Kennedy raises the profile of this important area, bringing the subject alive with vivid translations from Arab sources. This will be fascinating reading for historians of medieval Europe and for historians of the middle east drawing out the similarities and contrasts with other areas of the Muslim world.

Chapter One
The Conquest and the Age of the Amirs, 711-56

The Iberian background

The Iberian peninsula, divided into the great provinces of Tarraconensis, Carthaginensis, Baetica, Lusitania and Gallaecia, had been one of the richest and most developed areas of the western Roman Empire, but for three centuries before the coming of the Muslims it had been dominated by warrior aristocracies of Germanic origin. The most successful of these were the Visigoths who had first entered the peninsula in the early fifth century. With the accession of King Leovigild in 569 the Visigothic monarchy entered on a century and a half of stability during which kings, based in Toledo, exercised effective power.

The kingdom shared many of the characteristics of the post-Roman kingdoms in France and Italy. In general the Visigothic monarchy was a reasonably strong and effective instrument of government: it did not suffer the internal divisions and progressive debility of the contemporary Merovingian monarchy in France and, right up until the Muslim invasions, the kings maintained their control over most of the Iberian peninsula. In theory the monarchy was elective and successive church councils of Toledo in the midseventh century had laid down the rules: the king was to be elected by the bishops and nobles. He was to be a catholic Christian, a Goth by descent and of free birth. He was to be elected either in Toledo or on the site of the previous king's death, and before his accession he had to swear to uphold the laws of the realm. In practice, the choice of monarchs was confined to the most important lineages and there was a natural tendency for fathers to wish to pass their crowns to their sons, as Leovigild did to Recared in 586 and Egica did to Witiza in 702. At the same time, there seems to have been a strong feeling among the nobility against the establishment of a purely hereditary succession and perhaps that a lineage which had held the crown too long should be replaced.

Under the king, the chief men of the state were the nobles, mostly of Gothic origin, and the bishops, mostly recruited from the Hispano-Roman landowning class, although the distinctions between these two groups must have largely disappeared by the beginning of the eighth century. Besides providing spiritual leadership, the bishops were also among the largest landowners and most powerful political figures in the land. The nobles, who sometimes bore the title of dux (duke) or comes (count), were also owners of large, often underexploited estates cultivated by semi-free peasants. It was the nobles too who provided the army: apart from a royal guard, there was no standing army and the nobles brought their followers in response to the royal summons and the king might reward them with gifts of gold or silver. In general this simple military system seems to have functioned fairly successfully, but it probably meant that the bulk of the troops owed their first loyalties to their lords, rather than to the monarchy.

It is impossible to make any precise assessment of the population or economy of the peninsula. It has been plausibly suggested¹ that the population had been about six million in the early Roman period but had been reduced by plague and war to four million by the later Visigothic period. Archaeological evidence shows that the large open cities of the earlier period had shrunk into small fortified settlements. Country estates and their buildings were certainly more primitive than the great latifundia and villas of the imperial Roman period.

Economic life was almost entirely localised: there is little evidence of long-distance trade and both the small towns and the large estates were effectively economically self-sufficient. The circulation of coinage was extremely limited and most transactions were conducted by barter.

Later Visigothic Spain and Portugal was a fairly stable society and, apart from a limited Byzantine incursion around Cartagena in the south-east, there had been no outside invasion for a couple of centuries. On the other hand, we can picture a very empty landscape, where settlements were few, far between, poor and primitive. Agricultural resources were in many cases neglected or underexploited. There were areas, too, notably in the northern mountains where the Basques and the Asturians lived, where the people were totally independent of any form of royal control and where a primitive mountain society vigorously resisted outside control. Such was the land the Muslims invaded.

The Muslim conquest, 711-16

The Muslim invasion of Spain and Portugal was in many ways the logical and necessary extension of the conquest of North Africa.² Before the coming of the Muslims the area of the modern states of Libya, Tunisia, Algeria and Morocco was occupied by two groups, the Byzantines and the Berbers. The Byzantines held a number of strongholds on the coast, notably Tripoli and Carthage, which they kept supplied by sea. When they reconquered the land from the Vandals in the reign of Justinian, they had established an elaborate system of defences on the southern frontiers of the settled areas, but these seem to have been abandoned by the mid-seventh century when the Muslims began to attack. The remaining Byzantine garrisons defended their coastal strongholds stubbornly and they held out much longer than the garrisons in Syria and Palestine had done a generation before, but they could easily be bypassed by overland invaders and were only a real threat when allied to the Berber tribes of the area.

The Berbers were the real power in the land. They were, and still are, the indigenous inhabitants of North Africa, with their own, unwritten, language, quite distinct from either Latin or Arabic. Their social structure was tribal and they seem to have owed their first loyalties to their kin. Apart from this, there were wide variations of lifestyle. Some Berbers were acculturated to the Byzantine world and many were Christians. Others seem to have lived a much more separate existence and some at least were still pagans. There were Berbers who lived in the cities, many more who lived as farmers in mountain villages, some who kept sheep and goats in the steppe lands and yet others who wandered as the Touareg do today in the remote and awesome wastes of the Sahara. Berber genealogies are much less fully recorded than Arabic ones and it is difficult to gauge whether the scattered references we have reflect a static relationship or whether some groups were expanding at the expense of others.

According to Arab sources, the Berber tribes were divided into two groups, called Butr and Barānis, just as Arab tribes were divided into Qays/Muḍar and Yemen. Most of the Berbers who joined the Muslim conquest and settled in al-Andalus came from the Butr group. They seem to have retained their tribal identities and probably their pagan religion. In the sixth century these tribes were moving west from Tripolitania and putting pressure on the Byzantine settlements.³ The Barānis, by contrast, were older-established tribes who had entered into closer relations with the Byzantines and had in many cases converted to Christianity. This suggests that the Muslims assumed the leadership and gave extra momentum to tin existing movement of populations among the Berbers, and this goes some way to explaining the success and completeness of their conquests. In practice, the difference between Butr and Barānis seems to have had little effect on the politics of al-Andalus, unlike the murderous disputes between Qays/Muḍar and Yemen among the Arabs, and divisions among the Berber tribes were based on smaller units of individual tribes and extended families and their relationship to Arab groups.

The conquest of North Africa⁴ had begun as early as 22/642 when the conqueror of Egypt, 'Amr b. al-'Ās, led an expedition to Barqa in Cyrenaica. From there he dispatched an army to Zawīla, an oasis settlement to the south, led by 'Uqba b. Nāfi'al-Fihrī. 'Uqba came from a branch of Quraysh, the Prophet Muḥammad's tribe, and his father Nāfi' had been one of the first Muslims to settle in Egypt. He came from the elite of early Islamic society and he used his position to make contacts among the Berber people of the area, alliances that were to make his family the most powerful in North Africa and al-Andalus before the coming of the Umayyads in the mid-eighth century. When 'Amr returned to Egypt, he left 'Uqba in charge at Barqa.

The conquest of North Africa was difficult, partly because of political disputes among the Muslims but more because of the vigorous resistance put up both by the Berber tribes of the interior and the garrisons of Byzantine cities like Tripoli and Carthage on the coast. More than the other Arab commanders, 'Uqba seems to have understood that the key to subduing North Africa was to enrol the support of Berber tribes. In 50/670 he founded the Muslim settlement of Qayrawān, away from the coast, in the central plain of Tunisia. Like earlier Arab garrison cities at Kūfa, Baṣra and Fusṭāṭ, this was designed to be a settlement where the Muslims could preserve their identity and from which they could dominate the surrounding country.

In 681 'Uqba led a spectacular raid to the west in which he reached Tangier and the Atlantic coast, although there were no Muslim settlements beyond moḍern Tunisia at this stage. This was his last and greatest achievement, but his memory lingered on and his sons continued to play a very important role in the Muslim politics of North Africa. There followed a period when the Arabs were almost driven out and Qayrawān itself fell to the Berber leader Kusayla. The Muslims did not recover the initiative until 74/694 when the Caliph 'Abd al-Malik sent an army of Syrians led by Ḥassān b. al-Nu'mān al-Ghassānī. He captured the last Byzantine outpost at Carthage and defeated the Berber leader, the priestess Kāhina, and in 82/701 established himself firmly in Qayrawān. He was able to do this, not only because of his Syrian troops, but because of his policy of working with the Berbers. Some tribes, like the Luwāta, seem to have remained allies of the Arabs throughout; many others came over after the defeat of Kāhina, including her own sons. They converted to Islam and were enrolled in the Muslim dīwān, receiving a share of the spoils like the Arabs.

Ḥassān was dismissed by the governor of Egypt, who supervised the western provinces in 704, probably because he was too successful, and was replaced by Mūsā b. Nuṣayr, a man of obscure origins who had risen in the financial administration of the Umayyad empire. He continued Ḥassān's policy of recruiting converted Berbers into the Muslim armies and using this new force to extend his control further to the west until, in about 90/708, he took Tangier and appointed a Berber supporter of his, Ṭāriq b. Ziyād, as governor.

The conquest of North Africa had been achieved by an alliance of Arabs and Berbers in the name of Islam. As the conquest proceeded, so the importance of the Berber contribution increased. By the time the Muslims were conquering the area of modern Morocco, it is probable that the great majority of the troops in their army were Berbers, newly converted to Islam. These Berber troops received a share of the booty but, apart from Ṭāriq b. Ziyād in Tangier, they do not seem to have occupied positions of political importance. Many Berbers entered into wālā' agreements with important Arab groups or individuals. They were then described as mawld (pl. mawālī) of so and so (Ṭāriq b. Ziyād, the Berber governor of Tangier and probably a man of considerable importance in his own community, for example, is described as mawlā of Mūsā b. Nuṣayr, the Arab governor of Qayrawān). This relationship can be described as a sort of clientage, by which members of the conquered peoples were converted (you could not be a non-Muslim mawlā) and given a position in the Muslim community in exchange for their loyalty and support.⁵ These networks were very important in the fluid politics of early Muslim North Africa and al-Andalus and were often more useful than tribal followings in building up a power base: both the family of 'Uqba b. Nāfi' and the Umayyads depended heavily on their mawālī to support their political ambitions.

The governors (wālī or 'āmil are the two Arabic terms used to describe this office), by contrast, were dependent for their authority on the governor of Egypt, and a change of command in Fusṭāṭ (Old Cairo) almost certainly meant a change in Qayrawān. This pattern became even more pronounced in al-Andalus, where the position of the governor was constantly threatened by changes of policies or personnel in Qayrawān or Fusṭaṭ. In these circumstances, it was difficult for a commanding personality to establish himself for long and the governors were often transient figures who made little impact on the country.

The conquerors fed off further conquests. It is true that subject Christian and pagan Berbers are said to have been obliged to pay jizya or poll-tax, but there is little indication of any formal taxgathering machinery. Most of the soldiers must have served in the hope of booty and new lands rather than for a salary and Muslim dominion in North Africa had to expand to survive. If the booty dried up and no new opportunities appeared, then the groups and tribes would turn in on each other and disintegration would inevitably follow. The conquest of Tangier effectively meant the end of westward expansion; now only Spain could offer the sort of opportunities the state needed to be able to survive.

Our understanding of the Muslim conquest of al-Andalus and the establishment of Arab rule is hampered by the nature of the sources. No contemporary Arabic accounts of the conquest survive and the earliest major sources which have been passed down to us are collections of historical anecdotes (akhbār) preserved in a number of works dating from the tenth century onwards, notably the anonymous Akhbār al-Majmū'a (Collection of Anecdotes) from possibly c. 940⁶ and the Ta'rīkh Iftitāḥ al-Andalus (History of the Conquest of al-Andalus) of Ibn al-Qūṭiya (d. 977).⁷ Both these collections arrange their materials more or less in chronological order but they are not annals and are more concerned with vivid and interesting stories than the careful ordering of events. The Akhbār is particularly important for the pre-Umayyad period, while Ibn al-Qūṭiya gives vivid and gossipy accounts of the courts of the Umayyad amirs.

In the tenth century these accounts were edited and systematised using the cri...