12 Bedouin political movements and dynasties

DOI: 10.4324/9780429348129-12

Origins

Despite their Arab origins, the Hamdanids were products of the middle ‘Abbasid political system, relying on ghilmān for their military power and professional bureaucrats to collect their taxes. After the early stages, their role as tribal leaders played no part in their government, and the tribe from which they sprang disappeared into obscurity. In both Mosul and Aleppo, they were succeeded by leaders, the ‘Uqaylids and Mirdasids, whose power was directly dependent on their tribal followings, who employed no ghilmān and remained first and foremost bedouin shaykhs even when they acquired the rights to collect taxes from settled areas and cities. Nor was this process confined to Mosul and Aleppo; all around the Fertile Crescent from southern Iraq to Palestine, we find bedouin tribes and their leaders establishing or attempting to establish small independent states. The Banū Asad (Mazyadids) in the Kūfa area, the Numayrīs in Ḥarrān, the Banū Kalb in the Damascus area and the Banū Ṭayy in southern Palestine are all examples. Some of the tribes which became prominent at this time, notably the Banū Kalb and Banū Asad, had been important in their areas since Umayyad times, but many others, including the ‘Uqayl, Numayr, Kilāb and Ṭayy, seem to have been mostly newcomers to the districts they came to dominate. Equally tribes which had been important in the Umayyad and early ‘Abbasid periods are heard of no more. The example of the Taghlib has already been cited, but other important groupings – the Tamīm and Shaybān in Iraq, the Tanūkh, Lakhm and Judhām in Syria – seem to have declined during the ‘Abbasid period. The rise of the bedouin dynasties of the fourth/tenth century was the result of both an upheaval among the bedouins of the Syrian and north Arabian Deserts and the changing relationships between the settled and pastoral peoples of the area. To understand this process, we must investigate the history of the bedouin in the previous century.

The bedouin tribes of the Arabian and Syrian Deserts had formed a large part of the armies of the Islamic conquests. In Umayyad times, they had continued to be influential, as supporters of the dynasty or of factions within the ruling house or as Kharijite rebels opposing Umayyad authority. After the ‘Abbasid revolution, however, their importance began to decline. The Khurāsāniyya who formed the most important group in the armies of the new dynasty lived not as bedouin but as townsmen settled in cities like Raqqa and Baghdad. Tribal leaders like Yazīd b. Mazyad of Shaybān and Kharijite rebels like al-Walīd b. Ṭarīf (killed in the Jazīra, 179/795) continued to play a reduced part in the politics of the caliphate but the defeat of al-Amīn (198/813) and the rise to power of the military and bureaucratic élite under al-Mu‘taṣim (218–227/833–842) meant an end to this role. From this time, the bedouin were effectively refused access to government and to government patronage and they no longer received any subsidies. From being a dominant Arab race, they became, once again, as they had been before the coming of Islam: the ignorant, uncouth, despised outsiders.

As has already been pointed out, the bedouin life is not economically self-sufficient. The bedouin are dependent on settled people for additional income and, by paying them ‘aṭā’ (salaries), entertaining them at court and allowing them to participate in the military expeditions against the Byzantines, the early Islamic state had subsidized them. The cutting off of these subsidies by al-Mu‘taṣim led to a revival of nomad pressure on the settled people of the area, which was to be the major feature of the fourth/tenth- and fifth/eleventh-century history of the Fertile Crescent. Disturbances among the bedouin were not unknown in the third/ninth century. In 230/845, there were widespread disturbances in the Ḥijāz caused by the Banū Sulaym, who were taxing markets and stopping caravans. The Caliph al-Wāthiq sent one of his Turkish generals, Bughā the Elder, with an army of Turks and Berbers to suppress them and took the opportunity to intimidate other tribes, notably the Banū Kilāb, then found in central Arabia farther to the east. The discontent remained. In 251/865, the Banū ‘Uqayl, complaining about their poverty, blocked the road between Mecca and the port of Jedda, a vital route for supplies for the Holy City. A generation later in 285/898, the Banū Ṭayy were responsible for pillaging the Ḥajj (pilgrimage) caravan as it passed through their territory. Not only was the safeguarding of the Ḥajj a matter of concern to the government for religious reasons but it was also the main commercial activity in Arabia, and the bedouin were signalling that they wanted a share of the profits. The Ṭayy acquired great wealth from this attack, and it is a sign of changing circumstances that they do not seem to have been effectively punished.

These discontents were given a unity and an ideological purpose in the rebellions of the Qarmaṭīs (or Qarāmiṭa, often referred to as Carmathians in Western writings). The Qarmaṭī movement led to fundamental changes in the distribution and relative strengths of the bedouin tribes in the Syrian and Arabian Deserts and this probably marks the most important such movement between the Islamic conquests and the Wahhābī wars of the eighteenth and early nineteenth centuries which largely shaped the modern tribal map of the Syrian and north Arabian Deserts.

The Qarāmiṭa

The Qarāmiṭa were a branch of the Ismā‘īlī movement. It seems that the sect was named after one Ḥamdān Qarmaṭ, who, before 260/873–874, had begun to spread Ismā‘īlī ideas among the agricultural peasantry on the Sawād of Kūfa, preaching and collecting contributions from the faithful. At first, the movement was not particularly militant, but it inevitably came into conflict with the ‘Abbasid authorities and there were a series of small-scale rebellions from 284/897 onwards, all of them crushed by the government of the able and ruthless Caliph al-Mu‘taḍid. The Ismā‘īlī movement also began to suffer from internal divisions. The exact nature of these disagreements is shrouded in mystery, but their consequences were to be far-reaching. It seems that the movement was directed from a base at Salamiyya, on the western edge of the Syrian Desert, probably by ‘Ubayd Allāh, who was to become the first Fatimid ruler in north Africa ten years later.

The original doctrine of the Ismā‘īlīs concerned the identity of the true imām and a division of opinion as to who was the rightful successor of Ja‘far al-Ṣādiq (d. 148/765). His son, Ismā‘īl, who was originally held to be his successor, had died before his father. By the mid-fourth/tenth century, the majority of the Shī‘a of Iraq believed that Mūsā al-Kāẓim, another son, was Ja‘far’s successor and that the line of rightful imāms descended from him. As their name implies, however, the Ismā‘īlīs held that Ismā‘īl, or his son Muḥammad, was the last true imām, that he would come in glory as the mahdī and that until that time there was no visible and acknowledged imām on earth. ‘Ubayd Allāh the Fatimid, however, broke with this, claiming that he was himself a descendant of Muḥammad b. Ismā‘īl and that the Ismā‘īlīs should pay him homage as the true imām. This move provoked strong opposition in the sect and many, among them Ḥamdān Qarmaṭ, refused to accept the new doctrine. A split developed between those who accepted that ‘Ubayd Allāh was indeed the mahdī and those who believed that the Ismā‘īlīs should continue to await the return of Muḥammad b. Ismā‘īl.

The immediate effect of the schism was to weaken the movement in Iraq, and Ḥamdān himself disappears from the record. Other missionaries, however, continued to spread the word in different areas, keeping to the old Ismā‘īlī faith. They and their followers are conventionally known as Qarmaṭīs or Qarāmiṭa to distinguish them from the supporters of the Fatimids. They no longer derived their main support from the Sawād of Kūfa but began to proselytize in other parts of the Muslim world.

The first of these was the desert to the west of Kūfa on the road to Tadmur (Palmyra) and Damascus. This area was dominated by the powerful Kalb tribe, who it would seem had already been attacking traffic on this route. A Qarmaṭī missionary from the Sawād called Zikrawayh sent his son al-Ḥusayn to make contact with these bedouin, and he made enough conversions to be able to mobilize a considerable military force. He was joined by Yaḥyā, who was probably his brother. They took code names to conceal their identities – Yaḥyā being Ṣāḥib al-Nāqa (Master of the She-camel), while al-Ḥusayn was Ṣāḥib al-Shāma (the Man with the Mole). In 290/903, they besieged Damascus, but the city was stoutly defended by Ṭughj, father of Muḥammad b. Ṭughj the ikhshīd (ruler of Egypt, 323–334/935–946), the attack was repulsed and the Ṣāḥib al-Nāqa killed. This was not the end of the affair, however. The Ṣāḥib al-Shāma took over leadership and the Qarmaṭī army pillaged much of northern Syria, including the towns of Homs, Ḥamāh, Ma‘arrat al-Nu‘mān and Ba‘albak. A contemporary source describes the army as “most of the people of the desert” and there could be no clearer illustration of the resurgent power of the bedouin. It was at this time too that the Qarāmiṭa took Salamiyya and destroyed the residence of ‘Ubayd Allāh the Fatimid, who had prudently left for north Africa shortly before.

This full-scale rebellion produced a major ‘Abbasid reaction. A large army under the command of Muḥammad b. Sulaymān was sent to Syria, and in 291/904 the Qarāmiṭa were severely defeated east of Ḥamāh and the Ṣāḥib al-Shāma killed. Despite this setback, the movement continued to attract support. In 293/906, the Kalb and other Qarmaṭī tribes sacked the Ḥawrān area and Tiberias as well as making another unsuccessful attempt on Damascus. In the same year, Zikrawayh himself left his hiding place near Kūfa and took the field in person for the first time. The next year (294/907), he was killed leading an attack on the Ḥajj caravan, and with his death the Qarmaṭī movement among the tribesmen of the Syrian Desert effectively came to an end. It is noteworthy that the leadership of the movement came from the settled people of the Sawād. The bedouin were obviously restive before the arrival of the Qarmaṭī missionaries, but it was the preaching of the sect which acted as a catalyst and gave purpose to their raids. The leaders of the armies were the two sons of Zikrawayh, both of whom claimed religious authority as descendants of Muḥammad b. Ismā‘īl, not shaykhs of the Kalb or any other tribe. The aims of the military activity, however, reflected the priorities of the bedouin: to secure tribute or booty from the settled communities. No effort was made to occupy conquered cities or establish an administration to control them. The effect of the rebellion was to weaken the settled communities to the advantage of the bedouin, a process which was to continue throughout the next century and a half.

Another area where Qarmaṭī missionaries met with success was in Baḥrayn. (In the early Islamic period, this term applied to the mainland areas of the Ḥasā’ province of Saudi Arabia as well as the island nowadays called Baḥrayn, which was then known as Uwāl.) This area was inhabited by a heterogeneous population of farmers in the oases, merchants travelling the Gulf to such ports as Baṣra and Sīrāf, and the bedouin of the desert areas. It was among this varied population that an Ismā‘īlī missionary of Persian origin, Abū Sa‘īd al-Jannābī, managed to recruit a large number of followers. By 286/899, the year of the split in the Ismā‘īlī movement, he had converted much of the population and married into one of the leading commercial and landowning families, that of al-Ḥasan b. Sanbar. He also looked for allies among the bedouin of the desert and found them in the Banū Kilāb and the Banū ‘Uqayl, who now came to occupy the leading position among the tribes of central and eastern Arabia, as the Tamīm and ‘Abd al-Qays had done in early Islamic times. This partnership between the Qarmaṭī leadership, their followers among the settled population, and the bedouin was to prove extremely formidable. Hajar, the capital of the province and seat of the ‘Abbasid governor, was taken, and in 287/900, an ‘Abbasid army was decisively defeated. Then, the Qarāmiṭa began to threaten the city of Baṣra itself. This first militant stage was brought to an end in 300/913 when Abū Sa‘īd was assassinated in his new capital at Ḥasā’ near Hajar. He was succeeded by his son Sa‘īd, who entered into negotiations with the Baghdad authorities. He stressed the poverty of his followers and seems to have sought a subsidy from the government and trading opportunities, an interesting insight into the motivations of these early Qarāmiṭa.

While ‘Alī b. ‘Īsā was influential in Baghdad, this policy proved effective, and in 303/915–916 the wazīr sent presents and granted the right of free trade in Sīrāf, on the eastern shore of the Gulf. In 310/922–923, however, the position changed when Sa‘īd was replaced by his much more militant younger brother Abū Ṭāhir, while the next year ‘Alī b. ‘īsā lost power to Ibn al-Furāt, who was less inclined to take a conciliatory attitude. From this point onwards, it was war. In 311/923, Baṣra was taken and sacked by the rebels, with great destruction of life and property. Since 303/916, the Ḥajj had been allowed to go in peace, but in 311/924, it was attacked by a force of Qarāmiṭa, 800 horse and 1,000 foot, who pillaged it. The same happened again the next year, while the Ḥajj of 313/926 was allowed to go in peace only after the payment of a heavy subsidy. By 315/927–928, Baghdad itself was threatened, and the wazīr, al-Khaṣībī, sent for Ibn Abī’l-Sāj of Āzarbayjān for more troops. The effectiveness of the Qarāmiṭa contrasted markedly with the uselessness of the Baghdad army, a fact commented on by contemporaries. Interestingly, ‘Alī b. ‘Īsā, observing, rightly, that Ibn Abī’l-Sāj would not be successful, suggested that it would be better to employ the bedouin Banū Asad to guard the Ḥajj and the Banū Shaybān to attack the Qarāmiṭa; neither of these tribes were involved in the rebel movement and both could be expected to oppose the Kilābīs and ‘Uqaylīs who fought for the Qarāmiṭa. His advice was not accepted and the Baghdad forces were again defeated.

In 317/930, not only was the Ḥajj attacked by the Qarāmiṭa and most of the pilgrims killed, but the Black Stone of the Ka‘ba was stolen and taken to Baḥrayn.. This act caused great outrage, but the Baghdad government was powerless and the stone remained in Qarmaṭī hands in Ḥasā’ until 339/951. The reasons for the removal of the stone are not entirely clear. There is no reason to suppose that it was a protest against idolatry; the Ismā‘īlī Fatimids were to be notable patrons of the Ḥajj in later years and the Qarāmiṭa do not seem to have destroyed other shrines. They appear to have had no objection to the Ḥajj as long as they, and especially their bedouin supporters, were properly rewarded for allowing its safe passage. They must certainly have hoped for concessions from the government in exchange for its return and may even have hoped to divert the Ḥajj, with the trading opportunities it presented, to their own capital at Ḥasā’.

The seemingly unstoppable advance of the movement was brought to an end by internal feuds. In 319/931, Abū Ṭāhir was persuaded that a Persian adventurer of obscure origin was indeed the promised mahdī, and yielded authority to him. The adventurer’s behaviour soon made it clear that the Qarāmiṭa had been duped. He was deposed and killed, but the event seems to have damaged the morale of the sect. Aggressive activities did not cease, but the Qarāmiṭa seem to have been anxious for peace. As early as 320/932, we find Qarmaṭīs serving as soldiers in the army of the ‘Abbasid general Mu’nis. In 327/939, Abū Ṭāhir made his peace with the Baghdad authorities, promising to protect the Ḥajj in exchange for an annual subsidy and a payment by the pilgrims. The death of Abū Ṭāhir in 332/944 and the return of the Black Stone to Mecca in 339/951 mark the assimilation of the Qarmaṭī state into the Muslim political order. Ibn Ḥawqal gives an interesting account of the community as it existed in the 350s/960s. The ruler was advised by a council, called the ‘Iqdāniyya, whose opinions he was obliged to take into account. The administration included al-Ḥasan b. Sanbar’s son, himself called Sanbar, whose family were effective partners in government with the descendants of Abū Saīd al-Jannābī. There was a ṣāḥib al-shurṭa (chief of police), a kātib (secretary), an army commander (who was a member of the Kilābī bedouin tribe) and a ṣāḥib al-barīd (chief of the post, who was also a Kilābī). The bedouin seem to have played an important part in military affairs. The revenues of the state were drawn from the land tax, tolls on the Baṣra and Mecca roads, taxation of the pilgrimage and tolls from shipping in the Gulf; commerce was vital to the financial well-being of the Qarmaṭī state. They maintained representatives in Kūfa, near Baṣra and eventually in Baghdad itself to organize their affairs. There was a marked contrast between the peace and prosperity of the Qarmaṭī state and the tyranny and chaos which prevailed elsewhere in the Fertile Crescent.

The last outbreak of Qarāmiṭa militancy was connected with the Fatimid conquest of Egypt. In the main, they kept on good terms with rulers of the Fertile Crescent, the Ikhshidids and the Hamdanids who sent them gifts, but relations with the Fatimids were complicated by the fact that the Qarāmiṭa did not acknowledge the Fatimid claims to be the true imāms. From 353/964, the Qarāmiṭa began to take an aggressive line under the leadership of al-Ḥasan al-A‘ṣam. An expedition was launched against Syria, perhaps to help the Fatimids in their plans to take Egypt but possibly to secure increased tribute from the Ikhshidid government. Tiberias was sacked. In 357/968, the adventure was repeated and both Damascus and Ramla were taken by the Qarmaṭī armies, and again in 358/969 after the Fatimid takeover of Egypt, when al-Ḥasan b. ‘Ubayd Allāh the Ikhshidid agreed to pay tribute for Ramla and Damascus. The Qarāmiṭa seem to have felt that Syria was now their protectorate, and this led to a dramatic breakdown in relations with the Fatimids. Open warfare soon flared. The Qarmaṭī effort was hampered by internal strife at Ḥasā’ but in 360/971 Damascus and Ramla were taken and Egypt invaded. To demonstrate their disenchantment with the Fatimid regime, the Qarāmiṭa proclaimed ‘Abbasid authority in the areas they captured. Thereafter, Qarmaṭī attacks on Egypt and Syria continued until they were defeated by the Caliph al-‘Azīz in 368/978. They retreated to Ḥasā’ in exchange for the promise of a subsidy. Apart from a brief late occupation of Kūfa, this marked the end of the Qarmaṭī threat to the neighbouring lands.

The movement was now confined to Baḥrayn. In accepting this, it lost the allegiance of the bedouin tribesmen who had supported the sect when there was the opportunity for booty and tribute. In the end, it was a tribal leader from the Muntafiq branch of the ‘Uqayl who turned against them, capturing and pillaging Ḥasā’ in 378/988. After this, the Qarāmiṭa could no longer tax the commercial routes of the north Arabian and Syrian Deserts, and it was nomad tribal leaders who took over this role.

The bedouin dynasties

The rise and decline of the Qarāmiṭa were the prelude to the establishment of the bedouin dynasties of the late fourth/tenth and fifth/eleventh centuries. The Qarāmiṭa had profoundly affected the balance of power among the tribes in the north Arabian and Syrian Deserts, and in the main it was tribes which had been involved in the movement which came to dominate the area – not, however, in the name of a religious ideal but for their own tribal interests; and they were now led not by missionaries from the settled peoples but by their own tribal shaykhs. Kilāb and ‘Uqayl had been the leading supporters of the Qarāmiṭa of Baḥrayn and it seems that in the second half of the fourth/tenth century, many of them, disillusioned with the decline of the power of the movement, drifted north to join fellow-tribesmen already established in the hinterlands of Mosul and Aleppo. The influence of the Kalb in central Syria had probably been consolidated by their leading role in the rebellions of the Syrian Qarāmiṭa, while the Ṭayy seem to have moved from north Arabia to Palestine during the last waves of Qarmaṭī attacks on the area. Of the tribes which founded bedouin states at this period, only the Asadīs in the Kūfa area had played no part in the movement.

The establishment of the bedouin states was, however, more than just the result of a new wave of Arab immigrants from Arabia. The founders of these bedouin states all owed something to the patronage of rulers of settled states. The Kilābīs owed much of their predominance in northern Syria to the support given them by Sayf al-Dawla the Hamdanid. The primacy of the ruling clan of the ‘Uqayl was greatly aided by ‘Aḍud al-Dawla the Buyid, who made them responsible for the disciplining of their fellow-tribesmen, thus giving their authority the support of his government. It was the attempts of the last Hamdanids to counter the influence of the Kurd Bādh in the Mosul area by granting lands to the ‘Uqaylids which ensured their control of northern Iraq. Equally, the Mazyadid leaders of the Asad tribe owed much to the patronage of Bakhtiyār and later contenders for power in Buyid Baghdad who sought their support, while Ibn Ustādh-hurmuz (Buyid governor of Baghdad, 392–401/1002–1010) was to rely on them to discipline the bedouin of the area. None of the dynasties would have achieved power without the patronage of settled rulers.

The dynasties which prospered were only able to do so because they controlled the revenues of towns and settled areas. The power of ‘Uqaylid rulers lay in their tribal following, but it was control of Mosul which brought them the wealth these followers needed. Ṣālih. b. Mirdās was simply one of a number of Kilābī chiefs until he took possession of Aleppo, a position which assured his primacy in the tribe. In the desert, a nomad state was impossible since no chief could command absolute authority. Contact with settled powers and peoples enabled some ruling clans to establish themselves as effective dynasts, even while retaining a nomad lifestyle. The possession of the revenues of settled lands also helped the chiefs to increase their tribal following. At first, the dominant position of the ‘Uqayl in northern Iraq looks to be the product of a vast influx of tribesmen from Arabia, but this may be something of an illusion. There certainly were new immigrants from northern Arabia at this time, but it may also be that many of the members of these newly dominant tribes in fact came from other groups, had attached themselves to the successful tribe and were henceforward counted among the ‘Uqayl or Kilāb. Such a process would account in part for the greatly expanded numbers of these tribes and for the disappearance from historical sources of other bedouin tribes in the area.

In economic terms, the changing relationships between the nomads and the settled population can be seen in the laments of numerous sources about the decline of settled agriculture and the occupation by the bedouin of farmland. The areas worst affected were probably Trans-Jordan, where almost all urban and much village life seems to have come to an end at this time, the Ḥawrān south of Damascus, and the northern Jazīra, where Ibn Ḥawqal describes the process most clearly. But there is evidence from the immediate neighbourhood of Damascus, Homs and in central Iraq (in the once fertile Sawād) which all points in the same direction. There can be no doubt that the century 950–1050 saw a vast increase in the area used for nomad pasture and the collapse of the agricultural economy in districts which had once been the granaries of the ‘Abbasid caliphate. It would be wrong to present a purely negative view of these changes. There were cities which benefited from the bedouin ascendancy. Aleppo seems to have been more prosperous under the Kilābī Mirdasids than it had been under the Hamdanids. It may have been that tax levels were lower under bedouin rule than under the domination of regimes like the Buyids or Hamdanids with expensive ghilmān to pay. Certainly, the city which suffered the best-documented urban crisis in this period, Baghdad, was never under bedouin control, while in nearby Ḥilla the camp of the Mazyadid clan of the Banū Asad developed into a flourishing city.

The Mazyadids

The Banū Asad owed their rise to power in the Kūfa area to Buyid weaknesses and the endemic tensions between Turks and Daylamites in Iraq. From their modest beginnings as bedouin shaykhs, the Mazyadid clan used the power of the tribe to found an enduring and powerful state whose influence was to last for a century and a half in the lands that bordered the Euphrates to the south of Baghdad and whose most lasting achievement was the foundation of the city of Ḥilla, which replaced Kūfa as the leading town of the area and is today the local provincial capital. The shaykhs of Asad first began to play an important role in politics in 364/974–975, when the Buyid Bakhtiyār was attempting to dislodge the Turkish commander Alptakīn from Baghdad and invited one Ḍabba b. Muḥammad al-Asadī to raid the area around the city; it was a sinister portent for the future as, for the first time, nomad depredations were given government approval. Two years later, in 366/977, when Bakhtiyār was facing the army of his cousin ‘Aḍud al-Dawla, he called on the support of another Asadī leader, Dubays b. ‘Afīf. Like so many of the things Bakhtiyār did, this turned out to be an ill-judged move; not only did the Asadīs desert him in the conflict, precipitating the breakup of his army, but they also pillaged his baggage.

‘Aḍud al-Dawla was strong enough to take firm measures against the bedouin, and in 369/979–980 he sent an army to the desert stronghold of the Banū Asad at ‘Ayn al-Tamr. The tribe was scattered and their leaders only just escaped. The death of ‘Aḍud al-Dawla in 372/983 and the rivalry between his sons shifted the balance once more in favour of the bedouin. As soon as ‘Aḍud was dead, Dubays b. ‘Afīf was employed by a Buyid rival to challenge the power of ‘Aḍud successor, Sharaf al-Dawla. In 387/997–998, ‘Alī b. Mazyad, now apparently the chief of the Asadīs, refused to pay tribute to Bahā’ al-Dawla in Baghdad and declared instead for Ṣamṣām al-Dawla, then ruling in distant Shīrāz. Bahā’ al-Dawla’s forces marched against ‘Ayn al-Tamr, but unable to secure their supplies in the desert, they were forced to make peace. This treaty confirmed ‘Alī’s power and status, and from this time onwards, the Mazyadids enjoyed virtual independence. The failure of the Buyids to prevent this was largely a product of their own internal rivalries. The nomads were able to provide a force of cavalry which made them sought-after allies in times of conflict, while on those occasions when the government did try to reassert its power, they could take advantage of desert retreats where only ‘Aḍud al-Dawla was able to reach them.

The Asadīs became increasingly bold in their raids into settled areas, sometimes in alliance with the Banū ‘Uqayl. In 389/999, they pillaged Dayr ‘Āqūl on the Tigris, and in 392/1002 Asadīs and ‘Uqaylids joined forces to prevent the Buyid commander in Baghdad from retaking al-Madā’in, then under ‘Uqaylid “protection”. The Buyid ruler, Bahā’ al-Dawla, was by now permanently resident in Shīrāz, and his commander in Baghdad, Abū Ja‘far al-Ḥajjāj, found himself almost entirely surrounded by bedouin or Kurdish nomad forces, the city an island of Buyid control in a countryside dominated by pastoral peoples. For a time it seemed as if Baghdad too would be lost, but in 392/1002, a decisive battle was fought near Ḥilla between a joint army of ‘Uqaylid and Asadī bedouin on the one side and al-Ḥajjāj, aided by the bedouin Banū Khafāja and the Kurdish ‘Annazids, on the other. Baghdad was saved for the Buyids but the incident shows how powerful the Arab tribes had become and how they could roam at will through the Sawād; Asadīs from west of the Euphrates could sack Dayr ‘Āqūl on the east bank of the Tigris. Not since the time of the Arab conquest had nomad bands had such freedom, and their activities must have been both a cause and a consequence of the impoverishment and depopulation of rural, agricultural communities.

The replacement of al-Ḥajjāj by Ibn Ustādh-hurmuz in Dhu’l-Ḥijja 392/October 1002 gave the Mazyadids further opportunities for meddling, and for the next five years, they supported their old enemy al-Ḥajjāj in opposition to his successor. In 397/1007, an agreement was reached whereby ‘Alī b. Mazyad was given the honorific title Sanad al-Dawla (Support of the State) as a recognition of his status as an independent ruler under Buyid overlordship. Thereafter, Ibn Ustādh-hurmuz and his successor in Baghdad, Fakhr al-Mulk (401–407/1011–1016), usually kept on good terms with ‘Alī and even used him as a sort of policeman to maintain some Buyid influence in their area. In 404/1011, for example, when the lawless Banū Khafāja pillaged the Ḥajj caravan, the authorities in Baghdad turned to ‘Alī, who was able to recover much of the booty and bring the marauders under some sort of control.

When ‘Alī b. Mazyad died in 408/1018, he left his tribe as the leading group in the areas to the south and southwest of Baghdad and his own family as the leading clan and the most reliable allies of the Baghdad authorities. Both tribe and family, however, still had rivals. The Banū Asad were opposed by the ‘Uqaylids from the north. In 401/1010, the ‘Uqaylid leader, Qirwāsh, made his ill-fated attempt to transfer his allegiance from the ‘Abbasid caliphs to the Fatimids and the collapse of his bid for total independence severely weakened ‘Uqaylid power in the Baghdad area. Their place was taken to some extent by the Banū Khafāja, who now became the Asadīs’ main bedouin rivals in the lands to the west of the Euphrates. Within the tribe, ‘Alī had also faced opposition, and the Buyid governors of Baghdad had, on occasion, sided with his opponents to prevent him from becoming too powerful. On his death, his son and successor, Dubays, who had already been invested with the title Nūr al-Dawla by the Buyids, was opposed by his brother al-Muqallad, who sought the alliance of the Turkish soldiers of Baghdad who pillaged his brother’s camp. But Dubays, too, found allies in the city, and al-Muqallad was eventually driven out and forced to find refuge with the ‘Uqaylids in northern Iraq.

Dubays was to remain a leader of the Mazyadid clan for the next sixty-three years, until 474/1081, a period which allowed the consolidation of their power in the area. It is worth remarking on the great longevity of many of the nomad chiefs of this period. Dubays reigned longer than any other ruler in the medieval Islamic world. Some of his contemporaries were equally fortunate; the Kurd Naṣr al-Dawla the Marwanid ruled for fifty-two years and Qirwāsh b. al-Muqallad the ‘Uqaylid chief for forty-nine. Among sovereigns of more established, urban dynasties, only al-Mustanṣir the Fatimid at fifty-eight years could rival the nomad dynasts and that only because he, unlike them, succeeded as a small child. By contrast, the Buyids tended to die young after short reigns. Sharaf al-Dawla, to name only one, had taken possession of the lands of his father, ‘Aḍud al-Dawla, and had ruled in Fars and Iraq. He might well have been able to restore the fortunes of the dynasty, but he died in 380/990 after just two years in charge, aged only twenty-eight. This longevity had a political importance, since it enabled these dynasties to establish themselves without the ceaseless succession disputes and minorities which damaged other regimes. After the deposition of Fakhr al-Mulk in 407/1016, no Buyid ruler or governor in Baghdad was powerful enough to threaten his independence, while the ‘Uqaylids were increasingly preoccupied by affairs in the north. In 435/1044, Dubays allied with his old rival, the ‘Uqaylid Qirwāsh, against their new common enemy, the Ghuzz Turkmen invaders, and defeated them. With the establishment of Seljuk power in Baghdad, Dubays was faced with a new and much more formidable government in the city, but the Mazyadids were able to maintain, and even increase, their power.

The rise of the Mazyadids illustrates the complex and changing relations between the Buyid government and the bedouin. The Mazyadids needed and exploited the Buyids for the granting of titles and subsidies, while the Buyids, in turn, needed them, intermittently, for military support against their rivals and as policemen on the fringes of the desert. The rivalries between the Buyids and the tensions between Daylamites and Turks on the one hand, and the rivalries within the Banū Asad, and between the Asad and other bedouin tribes on the other, meant that the position was constantly changing, and the political history of this period is often tortuous.

In many ways, the early Mazyadids ruled as bedouin shaykhs. At first, their priorities were pillage and tribute. As they became more powerful in the later years of ‘Alī’s leadership and in the time of Dubays, they began to make permanent Ḥimāya agreements with settled communities. At least until the second half of the fifth/eleventh century, the Mazyadids lived a bedouin, nomad life rather than settling in cities. They employed no expensive ghilmān, relying instead on their tribal following of up to 2,000 horsemen, a substantial force by the standards of the time. And since they had no ghilmān to pay they needed only the most basic administration and seem to have felt no need for wazīrs and bureaucrats. This did not mean that they were hostile to urban life; it is fascinating to observe how the Ḥilla (encampment) of the Mazyadids was transformed during their rule, into Ḥilla, the walled permanent city which became their capital.

The ‘Uqaylids

In the Jazīra and northern Iraq, power passed into the hands of the leaders of the Banū ‘Uqayl tribe. Of all the dynasties which appeared at this time, the shaykhs of the Banū ‘Uqayl were perhaps the most truly bedouin in character, never settling in towns and remaining entirely dependent on their tribal followers for military support. Leadership among the members of the ruling clan was decided on traditional bedouin lines, the whole kin sharing in the exercise of authority, and no one man achieved the absolute status typical of more settled dynasties.

The ‘Uqaylīs are found in the Jazīra in Umayyad times and consolidated their hold on the pastoral lands of the area in the Hamdanid period, when they were the chief beneficiaries of the decline and emigration of the Banū Taghlib. Along with the Numayrīs, they are found as irregular troops in the armies of the Hamdanids. This period also shows the increasing incursions of the nomad tribes on settled areas, and the geographer Ibn Ḥawqal presents a vivid impression of the decline of agriculture around such Jaziran towns as Ra’s al-‘Ayn and Ḥarrān. As with the Asadīs in the Sawād, the rise of the bedouin dynasties was both a cause and a consequence of the decline of agriculture in favour of pastoral economy. The takeover of the Jazīra and Mosul by ‘Aḍud al-Dawla in 369/979 led to a firm government policy with regard to the tribes, and ‘Aḍud al-Dawla insisted that ‘Uqaylid chiefs, among them al-Musayyib b. Rāfi’, be responsible for the actions of their followers – a policy which caused some of them to take refuge across the Euphrates out of his reach. After ‘Aḍud al-Dawla’s death in 372/983, the situation became disturbed again and the ‘Uqaylīs and Numayrīs were obliged to defend their lands from the Kurds under Bādh to the north. When, in 379/989, Bahā’ al-Dawla sent the Ḥamdānī brothers Ibrāhīm and al-Ḥusayn to Mosul to rally support against Bādh, they turned to the ‘Uqaylīs for support, offering in exchange control of the old Taghlibī centres of Jazīrat Ibn ‘Umar, Balad and Niṣībīn. The Kurdish threat was contained, but the ‘Uqaylī chief Muḥammad b. al-Musayyib b. Rāfi’ took advantage of the confusion to execute the last Hamdanid in the area and the rival chief of the Numayrīs and make himself master of Mosul itself, the first member of the tribe to acquire an interest in the city.

Muḥammad pledged his allegiance to Bahā’ al-Dawla and was obliged to accept a Buyid presence in Mosul, while enjoying a share of the revenue. Shortly after his death in 386/996, however, the last Buyid governor was forced to withdraw. Mosul came under ‘Uqaylid rule and was to remain so for the next century. After Muḥammad’s death, a power struggle developed between his brothers. It was ‘Uqaylid custom that leadership pass to the eldest, ‘Alī, who was supported by the majority of the tribe but opposed by another brother, al-Muqallad, who had acquired interests in revenue collecting in a number of districts on the Euphrates between Anbār and Kūfa, and had developed contacts with the Buyids and the Baghdad military. He therefore offered to take over the revenue farming of Mosul and began recruiting Turks and Daylamites to his cause. At first, ‘Alī was prepared to allow his brother Mosul while he himself continued to live a bedouin existence as the chief of the tribe, but his secretary persuaded him that this was impossible and that if al-Muqallad was the master of Mosul, he would certainly use his position and wealth to subvert ‘Alī’s authority. In 386–387/996–997, there was much confused fighting, which resulted in a compromise and the division of the revenues between ‘Alī and al-Muqallad. The failure of al-Muqallad’s bid determined the whole nature of the ‘Uqaylid state. ‘Alī stood for traditional bedouin leadership and relied on the military power of the tribe, al-Muqallad, for a government based on Turkish ghilmān and Daylamite troops, very much in the Hamdanid mould. Both brothers were given honorifics by the Buyids, Janāh. al-Dawla for ‘Alī and Ḥusām al-Dawla for al-Muqallad, so marking their effective independence. Despite lavish gifts, al-Muqallad was unable to win the support of more than 2,000 of the ‘Uqaylis, while ‘Alī could count on 10,000. There were strong pressures exerted by the tribe on the two brothers to avoid war, and al-Muqallad was obliged to agree to a compromise after his sister, in a very traditional gesture, had threatened to shame herself in front of the whole tribe if he did not. Al-Muqallad, we are told, was “like wax in her hand”. ‘Alī’s victory meant that the ‘Uqaylid shaykhs would not follow in the footsteps of the Hamdanids and replace their tribal following with salaried ghilmān but would continue to rely on the Banū ‘Uqayl for military support. The dynasty based its power on the bedouin, and its policies reflected their needs.

The division of authority continued after ‘Alī’s death when his brother, al-Ḥasan, took over his position with the tribe, while al-Muqallad devoted most of his energy to his interests in central Iraq. He even tried to make himself the master of Baghdad and was negotiating with army leaders there when he was murdered by his Turkish retainers in 391/1001. His position was inherited, not without difficulty, by his son, Qirwāsh, the greatest of the early ‘Uqaylid chiefs, who united the tribe under his leadership. It was he who led 7,000 ‘Uqayl, in alliance with the Banū Asad, when they fought the Baghdad commander Abū Ja‘far al-Ḥajjāj and his allies from the Banū Khafāja near Ḥilla in 392/1002. The battle was a humiliating defeat for the ‘Uqaylid leader and the women of Khafāja raided and pillaged the ‘Uqaylid camp: it was like the jāhiliyya all over again.

This defeat and the reassertion of Buyid control under Ibn Ustādh-hurmuz soured relations between Qirwāsh and the Buyids. In an effort to regain control in the areas around Baghdad which his father had dominated – not only Anbār to the west and the Euphrates lands but also al-Madā’in to the southeast – Qirwāsh resorted to raiding and kidnapping. His priorities also changed after the death of the last of his uncles, Muṣ‘ab b. al-Musayyib in 397/1006–1007, left him the most senior of the ‘Uqaylid shaykhs and gave him further responsibilities in the Jazīra. Here, the position of the tribe was threatened by the Fatimids and their protégés, the marauding Khafāja.

It was against this background of worsening relations with the Buyids and Fatimid advances from the west into the Euphrates valley that Qirwāsh, in 401/1010, transferred his allegiance to the Fatimid caliphs and had the khuṭba pronounced in the name of al-Ḥākim in the mosques of all the towns under his control, including Mosul, Anbār, Kūfa and al-Madā’in. This was a challenge which could not be ignored and Bahā’ al-Dawla the Buyid ordered his governor in Baghdad, Ibn Ustādh-hurmuz, to take action, sending him 100,000 dīnārs for the purpose. Qirwāsh recognized that he had overstepped the mark and, in default of any support from the Fatimids, he returned to formal allegiance to the ‘Abbasid caliphate. Like the Hamdanids and the Mazyadids, the ‘Uqaylids are usually said to have been a Shi‘ite dynasty, but there is little evidence to support this. That Qirwāsh had no great regard for the ‘Alid family is suggested by the fact that he kidnapped one of its leading members in Kūfa in 395/1004–1005 and held him to ransom for 100,000 dīnārs. Apart from his brief flirtation with the Fatimids, Qirwāsh was content that the name of the ‘Abbasid caliph should be mentioned in the khuṭba in his domains. The truth is probably that he was committed to neither Sunnī nor Shī‘ī party and that he was comparatively indifferent to the claims of formal religion.

The failure of the bid for independence in 401/1010 marks the beginning of the decline of ‘Uqaylid influence in the lands al-Muqallad had acquired in the Sawād. The ‘Uqaylids were faced with the alliance of the Buyid governors in Baghdad and the Mazyadids, while their position was attacked by the rival Khafāja ‘Uqaylids in the desert west of Anbār and Kūfa. The main reason for the continuing conflict was the need for pasture. It would seem that the ‘Uqaylīs used the same grazing lands occupied by the eastern Shammar in the nineteenth century. Their summer pastures were in the northern Jazīra, while their winter pastures were in the Anbār–Kūfa area. If Qirwāsh were to be acceptable as the leader of the ‘Uqayl, he had to ensure that his followers had access to these grazing areas. The priorities of the ‘Uqaylī leaders remained the priorities of a pastoral society. They had no difficulty in maintaining amicable relations with the Marwanid Kurds, for example, because there was no such competition, but they were deadly enemies of the Khafāja, who threatened their grazing. Qirwāsh’s own mother was a Hadhbānī Kurd and relations with them seem to have been friendly. In 411/1020–1021, Qirwāsh was defeated near Sāmarrā by a coalition of Mazyadids, Baghdad troops and dissident members of his own clan, a defeat which led to the establishment of a branch of the family in Takrīt. In 417/1026, the ‘Uqaylids, allied with the Mazyadids, suffered a humiliating defeat when 10,000 men in their army were defeated by a much smaller force of 1,000 Khafāja. After this, they continued to hold Anbār but seem to have been driven out of the more southerly areas around Kūfa.

In 417/1026, Qirwāsh’s leadership was challenged in the north at exactly the same moment that the Sawād possessions were being lost to the Mazyadids and the Khafāja. The opposition was led by his brother, Badrān, who seems to have wanted his own emirate in Niṣībīn. Both sides prepared for battle, Qirwāsh raising 13,000 men with help from the Marwanids of Mayyāfāriqīn. But in the end, peace was arranged and Badrān was allowed to keep Niṣībīn. Thereafter, Qirwāsh’s position seems to have been unchallenged until the whole tribe was threatened by the arrival of the Ghuzz Turkmen. The Ghuzz were an existential threat to the ‘Uqayl, since they too were pastoral nomads and sought to take the same area and role as the Arabs. There was no room for compromise. When they first attacked in 433/1041–1042, Qirwāsh and his Kurdish allies were defeated, and he was forced to take refuge in Mosul for fear of them. Here, both bedouin and townsmen tried to defend the city against the common enemy but to no avail. The city was taken and Qirwāsh fled by ship, leaving his treasure in the hands of the enemy. A popular rebellion in the city against the Turks in 435/1043–1044 was brutally suppressed. Meanwhile, Qirwāsh was seeking allies against the intruders. He had written to Jalāl al-Dawla, the useless Buyid ruler of Baghdad, who was in no position to help, as well as to the Mazyadid leader Dubays and several Kurdish chiefs, notably Abū’l-Shawk the ‘Annazid. Faced with the common peril, Arabs and Kurds joined under Qirwāsh’s leadership to defend their pastures and their territories. In Ramaḍān 435/April 1044, the two armies met to the northwest of Mosul. At first, the Turkmen were victorious and drove the Arabs right back to their tents. Then, the fortunes of battle changed and the Arabs defeated their enemies with great loss of life. It was Qirwāsh’s greatest moment, and his victory was celebrated by poets in the traditional manner. He had certainly saved not just the ‘Uqayl but other pastoral peoples in the area from being overwhelmed by the newcomers. Qirwāsh must have been an old man by this stage and it may have been his increasing feebleness which led to his being put under house arrest by his brother, Baraka, in 441/1049 after yet another conflict within the ruling clan. For the last three years of his life, he was deprived of all power and died a captive in Rajab 444/October 1052.

Qirwāsh dominated the politics of the Jazīra for half a century. All his life he remained very much a bedouin shaykh, glorying in a genealogy which went back to the fathers of the Arab race. In his poetry, he despises inherited wealth and boasts of the swiftness of his horse, the sharpness of his sword and the generosity with which he gives. As far as we can tell, he usually lived in the Ḥilla, the nomad encampment, only visiting the towns under his control when they were threatened, like Anbār in 417/1026 or Mosul in 435/1043–1044. His recorded utterances suggest that he had a very low opinion of townsmen but that does not mean that ‘Uqaylid rule was particularly oppressive; certainly, the people of Mosul were prepared to take up arms alongside their ‘Uqaylī masters against the Ghuzz Turkmen. The centre of power had shifted from the city to the camp, and ‘Uqaylid “government” was remote from the affairs of the town. Unfortunately, we are very badly informed about the history of Mosul or the other ‘Uqaylid towns in this period. In 411/1020–1021, Qirwāsh had a dispute with his representatives in Mosul, who were his wazīr, Ibn al-Maghribī (later to move to the service of Naṣr al-Dawla the Marwanid), and a local landowner called Sulaymān b. Fahd, which suggests that he employed non-bedouin agents in the cities; and we should probably be right to assume that in Mosul, as in other towns of the period, local urban élites played an important part in day-to-day administration. The pattern of rule was not that of direct government of the cities and taxation of the countryside, as practised under the ‘Abbasids, but rather the Ḥimāya or protection agreement where, as in the modern khuwa agreement, the settled people pay the bedouin for their protection but otherwise manage their own affairs.

Qirwāsh’s obligations were to the Banū ‘Uqayl who followed him; his major concern was to provide them pasture and the fruits of taxation. Within his own clan, he was obliged to allow the formation of sub-amirates centred on towns such as Takrīt and Niṣībīn to satisfy the demands of his kin for their own Ḥimāyas. The appearance of these secondary states should not be seen as a sign of disintegration but rather as a sign that traditional patterns of clan rule were being followed.

The Mirdasids

In Aleppo, the position of the Hamdanids was taken over by the shaykhs of another bedouin tribe, the Banū Kilāb. Like the ‘Uqaylīs, the Kilāb had appeared in northern Syria in the aftermath of the Islamic conquest but, again like their neighbours to the east, they seem to have been strengthened by a new wave of immigration at the beginning of the fourth/tenth centuries and taken over lands along the fringes of the desert previously inhabited by tribes such as the Tanūkh. They had helped Sayf al-Dawla, the first of the Hamdanids, to power in the city and under Hamdanid rule their power had prospered at the expense of their neighbours to the south, the Kalbīs. Byzantine attacks on Aleppo are said to have diverted trade farther east through the Kilābī dominated town of Khunāṣira. By the end of the fourth/tenth century, the tribe controlled most of the area to the east and south of Aleppo itself, along with the Euphrates towns of Bālis and sometimes Raḥba. After the failure of the Hamdanid dynasty, Aleppo was the scene of a struggle between the last of the Hamdanid ghilmān, led by Lu’lu’ and his son Manṣūr, with Byzantine support, the Fatimids and the Kilābīs, in which the Kilābī leader, Ṣālih. b. Mirdās, finally emerged victorious.

Aleppo was different from Mosul in a number of ways. While the ‘Uqaylids in Mosul could feel comparatively secure from attempts by the later Buyids to assert their power, Aleppo lay on the frontier between the Byzantine and Fatimid spheres of influence. The area controlled by the tribe was comparatively small, certainly smaller than the Hamdanid state of Sayf al-Dawla. Antioch to the west was in Byzantine hands, while Homs and Ḥamāh to the south were usually under Fatimid domination. This meant that successful rulers of Aleppo had to be diplomats to manoeuvre among the great powers as well as being able to secure the support of their followers. The sources for Aleppo in this period are much fuller than those for Mosul and the ‘Uqaylid areas, and this allows us to see more clearly the interaction of nomad and settled elements in the state. The Kilābī tribesmen remained the backbone of the army, and without their support no member of the Mirdasid family could exercise effective control. While the bedouin were a fairly effective military force, even defeating a large Byzantine army on one occasion, they demanded a say in the running of the state in return. The Mirdasid family were not old-established leaders of the tribe. It was the fact that Ṣālih. seized Aleppo which gave him paramountcy among the shaykhs of the tribe. There were other clans who felt that the Mirdasid family had no more rights than their own, and some of them were prepared to make alliances with the Fatimid general Anūshtakīn al-Dizbarī when he drove the Mirdasids from the city in 429/1038. The Kilābīs did not settle down but maintained their nomad life, often camping close to Aleppo so that they could exert political pressure if necessary.

The other important element was the people of Aleppo, who, led by the ra’īs al-balad or shaykh al-dawla chosen from a leading local family, often played an important role in politics. The city had its own militia, the aḥdāth, which was not dependent on the Mirdasids but usually cooperated with them. This uneasy partnership was the natural result of bedouin rule, and both sides needed each other. It would be a mistake to imagine that this bedouin rule was resented by the townsmen. Such evidence as we have suggests that the city was more prosperous under the Mirdasids than it had been under the Hamdanids. Among the citizens themselves, there were divisions between rich and poor, and this often prevented them from maintaining a common front against an external foe. Nonetheless, the people of Aleppo, mostly fighting as infantry, were a force to be reckoned with. An additional complication was provided by the citadel, then as now standing high above the centre of the city, which could often maintain itself as a separate political unit and withstand blockades of up to a year. This sometimes led to a divided governorate, one authority for the town and another for the castle, and it was control of the citadel which enabled the Fatimids at various times to maintain a presence in Aleppo when they had been ejected from the city.

The rule of the Mirdasid family in Aleppo, which lasted, with interruptions, from 414 to 472 (1023–1079), was always precarious and dependent on balancing the interests of the ruling family, the Kilābī tribesmen, the citizens of the town and the surrounding great powers. Ṣālih. b. Mirdās himself gained possession of the city from the Fatimid garrison in 414/1023, with the aid of some of the citizens, including Sālim b. Mustafād, who became ra’īs al-balad, but he himself never seems to have lived there. No sooner had he appointed deputies to manage the city in his absence than he left for Palestine to join the great coalition of bedouin tribes, the Kilāb, Kalb and ṭayy, against the Fatimids and the general Anūshtakīn al-Dizbarī. He does not appear to have returned to the city before his death in 420/1029 and the battle of Uqḥuwāna near Lake Tiberias, where the bedouin were decisively defeated by the Fatimids. After his death, his inheritance was divided – one son, Thimāl, getting the citadel, while another, Naṣr, took the town. But the arrangement broke down the next year when Naṣr, with Byzantine support, took over the citadel as well, while Thimāl retired to the east to Bālis and Raḥba, where he enjoyed the support of the kilābī tribesmen. Naṣr also had to face the opposition of Sālim b. Mustafād and the aḥdāth, but they were crushed when they rebelled. On the international side, Naṣr attempted to maintain good relations with Byzantines and Fatimids alike; in 422/1031, he joined the Byzantines to suppress a Druze disturbance in the Jabal Summāq which threatened both their interests, while the Fatimids accepted him as their representative in Aleppo.

This dual control, Naṣr in the city, Thimāl with the tribe, lasted until 429/1038 when the aggressive Fatimid commander in Syria, Anūshtakīn al-Dizbarī, launched a major expedition against the city, supported not just by the Turkish and Berber troops of the Fatimid army but by the bedouin Ṭayy, Kalb and even some Kilāb. Naṣr gathered his army and his clan, including his brother Thimāl, to face the enemy but he was no match for superior forces and was killed fighting bravely. Thimāl took refuge in Aleppo. Here, the shaykhs of the city promised to support him against the enemy but his own advisers urged him to flee, saying that the townspeople would hamper his efforts and that if he remained in the city, the Kilābī tribesmen would desert him. It was the classic dilemma of the nomad ruler: did the town or the tribe have priority? Thimāl decided for the tribe and, leaving his nephew al-Muqallad b. Kāmil in the city, fled east across the Euphrates to take refuge in the Jazīra. Al-Muqallad was soon forced to surrender, Anūshtakīn’s men entered the city and mamlūks (slave soldiers) were appointed to run it. It seemed as if Mirdasid rule had come to a premature end.

The Mirdasids were saved by rivalries within the Fatimid administration between the wazīr al-Jarjarā’ī and the military commander, Anūshtakīn. Anūshtakīn was deprived of office and al-Jarjarā’ī was happy to have Thimāl and his nephew al-Muqallad appointed as Fatimid governors in Aleppo, and Kilābī forces succeeded in ejecting the Fatimid garrison. Thimāl’s reign from 433 to 449 (1041–1057) was the high point of Mirdasid fortunes. Until the last years, he succeeded in balancing all the different elements superbly. Fatimids and Byzantines, Kilābīs and citizens were all prepared, or forced, to accept his authority. With the Byzantines, who had no aggressive designs on Aleppo and whose main interest lay in keeping the city free from direct Fatimid rule, relations were uniformly good. The Mirdasids paid tribute but received gifts and Byzantine titles in exchange. Thimāl himself was made magistros in 433/1041–1042, and other members of his family, including his wife ‘Alawiyya, received the titles of strategos or patrikios. Even his agreements with the Fatimids did not interrupt these friendly relations.

The Fatimids made three determined attempts to destroy the independence of Aleppo. In 439/1047–1048, Thimāl had a dispute with Cairo about some money the Fatimid forces had left in the citadel when they left the city at the beginning of his reign. The Caliph al-Mustanṣir sent a member of the Hamdanid family, al-Ḥasan b. al-Ḥusayn, against the city, a reminder that although the Hamdanids had disappeared from Mosul and Aleppo, they still remained influential in the Fatimid Empire. The people of Aleppo, however, showed no enthusiasm for this scion of the old dynasty and went out to fight him but were defeated and Aleppo was only saved by a flood which destroyed the enemy’s camp and equipment. The next year, the assault was renewed by the Berber governor of Homs, Abū Shujā‘ b. Kulayd, but he was defeated at Kafarṭāb well to the south of the city itself, by an army of Kilābī bedouin citizens of Aleppo and the local fallāḥīn (peasants: a very rare example of the use of peasants in military activity at this time) under the command of al-Muqallad b. Kāmil al-Kilābī, Thimāl’s nephew and second-in-command. The third attack was led by a ghulām of the Fatimids called Rifq. He had a large army and actually entered the city in 441/1050, but dissensions within his army led to the breakup of the expedition, while he himself died of wounds.

After this third attempt, Thimāl sent a peace mission to Cairo which included his wife, the redoubtable ‘Alawiyya, known as al-Sayyida (the Lady) and the shaykh al-dawla, the leader of the citizens. They secured a treaty which confirmed Thimāl in his possessions and which remained in force for the rest of his rule. Despite some stresses, his relations with the people of Aleppo seem to have been friendly. His rule was remembered in the city as a period of prosperity, and the citizens seem to have preferred Mirdasid to Fatimid rule. There is no mention of any very developed administrative system, although each ruler had a secretary with the rather grand title of wazīr. Under Naṣr, two of these were Christians, but Thimāl seems to have only employed Muslims. The only one who achieved more than local fame was Muḥammad b. Jahīr, an experienced administrator who took service with Thimāl in 446/1054–1055. He doubled the revenues and reorganized the finances, but in doing so made himself unpopular and he was eventually forced to leave.

Ironically, Thimāl’s reign was brought to an end not by outside invasion, but by the opposition of his fellow Kilābī tribesmen. He was very aware of the dangers of alienating these men, who provided the military backing for his regime. He recruited no ghilmān to rival them in the exercise of power and he was careful to distribute money and favours among them. Despite this, however, their leaders remained resentful, feeling that they had as much right as he to wealth and power. Once again, we see the conflict between the ideas of shared leadership in a nomad society and the rule of a single man demanded by an urban state. In the end, in 449/1057, pressure from the Kilāb became intolerable and he resigned his post to the Fatimids in exchange for Beirut, Acre and Jubayl, not because they were richer or carried more prestige but simply because they were farther away from the Banū Kilāb.

Although Mirdasid rule was restored three years later, and Thimāl himself made a short-lived return, the dynasty never again enjoyed the same power. The international scene had changed. Byzantine power, which had been essential for the continuing independence of Aleppo, was greatly weakened. The rising power of the Seljuk Turks was a serious threat. The Byzantines had been pleased to have Aleppo as a buffer state, but the Seljuks sought to deal the Fatimids a mortal blow and needed to use northern Syria as a base. In addition, their Turkmen followers coveted the pastures of the Banū Kilāb. The days of the bedouin dynasts were numbered.

Other bedouin tribes

The Mazyadids, ‘Uqaylids and Mirdasids were the most successful of the bedouin leaders who attempted to establish their autonomy in the Fertile Crescent, and it is their names and genealogies which appear in the handbooks of Islamic history, but they were by no means the only ones. Sandwiched between the Banū ‘Uqayl to the east and the Banū Kilāb to the west, the shaykhs of Banū Numayr were able to establish a transitory authority over Ḥarrān and Edessa until its conquest by the Byzantines in 422/1031. Farther south in Syria and Palestine, the nomads were unable to secure control over the settled lands for a long enough period to establish states. Although they were numerous and wide-ranging, the Banū Kalb never established themselves in Damascus; while the Jarrahids, the leading family of the Banū Ṭayy, occupied Ramla on several occasions and did great damage to the town and the surrounding countryside, they never held on to it for long enough to set up any sort of state. Equally in Iraq, the Banū Khafāja, despite some sporadic successes, were never able to dominate any settled area permanently.

The reasons for these failures lay not so much in the weakness of the tribes as in the strength of the forces opposed to them. While the Mazyadids and the ‘Uqaylids were faced with the enfeebled Buyid regime of Baghdad, anxious to acquire bedouin support and prepared to make concessions to get it, the tribes of southern Syria and Palestine were opposed by a government which could field large armies of Berber troops and Turkish ghilmān. Under the capable leadership of Anūshtakīn al-Dizbarī in the early fifth/eleventh century, the Fatimid cause prospered and, in a great trial of strength, the combined forces of the bedouin tribes were defeated at the battle of Uqḥuwāna. While the Fatimids were prepared to tolerate an independent state in distant Aleppo, at least at certain times, they could not afford to do so in Damascus or on their own doorstep in Ramla and they used all the resources of Egypt to prevent this happening. The case of the Khafāja was somewhat different – here, it was more powerful bedouin rivals, the Banū Asad and the Banū ‘Uqayl, who prevented them acquiring rights over settled communities, and their attempts to ally with the Buyids or Fatimids against their rivals achieved nothing.

The possession of revenues from settled areas affected not just the wealth of the tribe as a whole but its internal structure. Groups like the Kalb which lacked this sort of revenue seem to have had a weak decentralized structure of the sort that is typical of pastoral tribes. When revenue rights were acquired, however, the money was channelled through the chief to the tribesmen; just as the payment of ‘aṭā‘ through tribal chiefs in the early Islamic period had strengthened their position against the rank and file of the tribe, so in the fifth and sixth (tenth and eleventh) centuries, the acquisition of revenues put the ruling clans in a much stronger position. They were then in a position to found dynasties with real authority, rather than be permanently dependent on the consent of the tribe.

The prosperity of the bedouin regimes began to decline after the middle of the fifth/eleventh century. While the Mazyadids were able to survive for another century, the ‘Uqaylids and Mirdasids were undermined by the advance of the Turkmen tribes. The Kilāb soon disappeared entirely as a bedouin tribe, while the mighty ‘Uqayl dwindled to a small remnant in the Ḥarrān area, losing their tribal identity and name in the process. The century 340–440 (approximately 950–1050) was the high-water mark of Arab nomad influence in the Fertile Crescent. In the long term, the century can perhaps be seen as the culmination of half a millennium of penetration by Arabic-speaking pastoral peoples of the settled areas, a process which began in late Byzantine and Sasanian times and continued through the Islamic conquests and the advances of the Qarāmiṭa until it reached its furthest extent in this period.