(Ar. Banū Ayyūb), a Kurdish family who first became prominent as members of the Zangid military establishment in Syria in the mid-sixth/twelfth century.


AYYUBIDS (Ar. Banū Ayyūb), a Kurdish family who first became prominent as members of the Zangid military establishment in Syria in the mid-sixth/twelfth century. Seizing power in Egypt in 564/1168, they were soon able to extend their dominion to the Yemen, Syria, and much of the Jazīra. However, they lost the bulk of their lands during the crises of the mid-seventh/thirteenth century, though they did continue to hold two minor principalities for some time thereafter (Ḥamā until 740/1341, Ḥeṣn Kayfā until 866/1462). In addition, many members of the family continued to be maintained in Egypt as pensioners of the Mamluk sultanate until the early eighth/fourteenth century.

The Ayyubids traced their ancestry back to Šāḏī, a notable of the Kurdo-Armenian town of Dvin (Ar. Dabīl) in the first quarter of the sixth/twelfth century. Ebn Ḵallekān [Beirut, I, pp. 255-56, tr. de Slane, p. 243] identifies him as belonging “to one of the most eminent and respectable families of Duwîn” (men ahl Dowīn wa men abnāʾ aʿyānehā wa’l-moʿtabarīn behā). Šāḏī’s father is commonly called Marwān, but nothing whatever is known about him. It does seem clear, however, that Šāḏī was a member of the Rawādīya section of the powerful Haḏbānīya tribe, and that the Rawādīya were the dominant Kurdish group in the Dvin district. It is quite evident that the progenitor of the Ayyubids was no simple pastoralist, but a member of the sedentary political-military elite of a marginal but very complex region.

Once the Ayyubids were ensconced in power, some of them sponsored genealogies showing that they were not Kurds, but rather of noble Arab descent, stemming from the Morra b. ʿAwf—or even from the Banū Omayya: On one level, such genealogies are obviously fictions. However, Minorsky (Studies in Caucasian History, pp. 114-16, 123, 128-30) argues that the Rawādīya Kurds should perhaps be connected to the descendants of the Arab general Rawwād Azdī, who was governor of Tabrīz ca. 200/815. These men, having become Kurdicized, emerge in the late fourth/tenth century as the paramount clan among the powerful Haḏbānīya tribe in Azerbaijan, whence one branch moved to take up residence in the district of Dvin at some point in the eleventh century. If Minorsky’s speculations are sound, then the fictitious Arab genealogies of the Ayyubids not only contain a kernel of truth but preserve an authentic folk memory.

The rise of the Ayyubids really begins with the exile of Šāḏī from Dvin, perhaps in 524/1130, when a Turkish general seized the town from its Kurdish prince. Šāḏī migrated to Iraq with his two adult sons, Najm-al-dīn Ayyūb and Asad-al-dīn Šīrkūh, and was appointed commandant of Takrīt by an old friend from Dvin, Mojāhed-al-dīn Behrūz, who had become the šeḥna of Iraq under the Saljuqs. But Šāḏī soon died, and his place was taken by his elder son Ayyūb. Political vicissitudes soon compelled another exile, however, and in 532/1138 Ayyūb and Šīrkūh took service with the powerful atābak of Mosul (Mawṣel), ʿEmād-al-dīn Zangī, and were quickly numbered among his most powerful and influential officers—a status which they retained also under his son Nūr-al-dīn Maḥmūd (r. 541/1146-569/1174).

In 564/1168 Šīrkūh was given command of Nūr-al-dīn’s third expeditionary force to Egypt, and succeeded in compelling the helpless Fatimid caliph al-ʿĀżed to name him vizier. Dying a few months thereafter (Jomādā II, 564/March, 1169), he was succeeded by his nephew Ṣalāḥ-al-dīn Yūsof b. Ayyūb. It is this man, known in the West as Saladin, who was the real founder of his family’s power. With great skill, Ṣalāḥ-al-dīn first suppressed all effective opposition in Egypt, and then, after his master Nūr-al-dīn’s death in 569/1174, began to take control of Syria piece by piece. In 579/1183 he occupied Aleppo (Ḥalab), the last center of Zangid power in Syria, and in 581/1186 he was able to reduce even Mosul to client status. Ṣalāḥ-al-dīn is of course best known for his wars against the Crusaders, and these dominated the rest of this reign until his death (589/1193). Though his heirs would extend their hegemony in the Jazīra and Armenia, it is nevertheless true to say that Ṣalāḥ-al-dīn’s wars marked out the Ayyubid sphere of influence in the Middle East.

In the present context, however, Ṣalāḥ-al-dīn is most significant to us as the creator of a political system which would provide the framework of attitudes, expectations, and goals which governed the conduct of his successors for the next half century. This political system, which may be called the family confederation, was closely modeled on Saljuq and Zangid practice, and had affinities as well to the political structures which his grandfather, father, and uncle had known in Armenia. The fundamental (and generally unspoken) principle in this system was that political authority within a state should not be the sole possession of one man, but rather should be shared among the leading male members of the ruling family. This sharing of authority was accomplished by assigning to each suitable prince an appanage of his own, which he would govern as a hereditary and autonomous principality. One member of the ruling family might indeed be recognized as its head, and as such would be accorded a certain degree of deference by the other princes. But in spite of the grandiloquent titles which commonly accompanied such a dignity, there were seldom any regular administrative mechanisms by which the senior ruler could translate his primacy of honor into effective power. Control and coordination within the confederation could be achieved only if the senior ruler commanded a preponderance of military force, or if he enjoyed a high degree of natural familial authority over the other princes. That is, a man who was the father of his “colleagues” could expect to have far more authority than an uncle or older brother. In this kind of system, kinship links among the princes were certainly the primary (though of course not the only) determinant of their political behavior. (The first to identify this system of politics seems to have been Barthold [Turkestan3, pp. 268, 306-07]. Further discussion of this tradition in Humphreys, From Saladin to the Mongols, pp. 66-75.)

If one accepts that Ṣalāḥ-al-dīn’s political outlook was largely shaped by such considerations, his decision (after his conquest of Aleppo in 579/1183) to parcel out his vast conquests as appanage states will come as no surprise. Likewise, the rationale behind his choice of his major heirs will be clear: three of his sons, a nephew, and his younger brother al-Malek al-ʿĀdel Sayf-al-dīn.

The history of Ṣalāḥ-al-dīn’s successors is extremely complex: however, it can be summed up in the observation that his two principal successors as chiefs of the Ayyubid confederation made no significant changes in the structure of politics which he had established, but were rather content to manipulate it for their own ends. Thus his brother al-Malek al-ʿĀdel Sayf al-dīn (r. 596/1200-615/1218) was able to ensure his authority by assigning the major appanages to his own sons. On the other hand, al-Malek al-Kāmel Nāṣer-al-dīn (r. 615/1218-635/1238), who was older brother or cousin to the other princes, was able to achieve a modicum of control over the confederation only through his superb diplomacy and an occasional resort to military force. Like his predecessors, al-Kāmel divided the lands he controlled between two sons, al-Malek al-ʿĀdel II and al-Malek al-Ṣāleḥ Najm-al-dīn Ayyūb. And as before there was a protracted struggle to determine the hierarchy of power and prestige within the confederation. But the issue this time was drastically different, for the series of wars between 635/1238 and 643/1245 led to a fundamental alteration of the Ayyubid constitution.

This change of political structure is linked to the name of al-Ṣāleḥ Ayyūb, the nominal head of the Ayyubid kingdom from 637/1240 to 647/1249. In crude terms, what al-Ṣāleḥ did was to wrench the confederation—or at least those parts of it which he controlled—into a unitary state governed by men who were his functionaries rather than his colleagues, and who were not members of the Ayyubid family. The motive force for al-Ṣāleḥ’s changes may be attributable to his domineering temperament and his deep-seated sense of alienation from his kinsmen and their traditions. On the other hand, this alienation was not merely a question of personality, but must also be seen in the context of family structure. First, al-Ṣāleḥ’s reign represented the fourth generation of Ayyubid dominion, and the ruling princes were no longer close kin, but in effect autonomous dynasts who shared a common eponymous ancestor. Second, al-Ṣāleḥ himself—the nominal head of the Ayyubid house—was only a cousin or nephew to the other princes, and there was no one else of sufficient seniority and prestige to assume his role. The general applicability of these facts can be seen in the similar (though more tentative) policy of administrative centralization being carried out simultaneously in north and central Syria by al-Nāṣer Yūsof of Aleppo (634/1236-658/1260).

The full significance of al-Ṣāleḥ’s centralizing policy emerged only after his death. Just six months later, his elite Baḥrīya regiment (composed of his personal mamlūks) seized power in Egypt and promoted one of its own members, ʿEzz-al-dīn Aybak, to the throne. When the Mongol cataclysm of 658/1260 destroyed the Syrian Ayyubid kingdom of al-Nāṣer, the Mamluk regime was able to occupy his former dominions. These were then incorporated in a highly centralized state controlled from Cairo, in which formal authority was vested in a single autocrat. This policy should occasion no surprise. The new rulers’ training as royal mamlūks had taught them to think of authority as embodied in one man. Likewise, their whole political education and experience had been as instruments of al-Ṣāleḥ’s policy, and they knew no political system other than his.

In the light of the above outline, is it proper to think of the Ayyubid confederation as a specifically “Kurdish” state? On the level of political structure, the governing attitudes of the Ayyubid confederation can certainly be related to the political institutions of their original homeland. On the other hand, these institutions do not differ significantly from the underlying structures of contemporary Turkish states, and in many ways the Ayyubids can be understood simply as a successor kingdom to the Saljuqs.

On the level of his culture, the matter is equally ambiguous. The Ayyubids ruled a predominantly Arabic-speaking region, and many of their princes became very proficient in Arabic letters and in the religious sciences. However, we see many signs of a continuing connection with their homeland and with Iranian culture generally. Thus, it is clear that al-Malek al-ʿĀdel and his son al-Malek al-Moʿaẓẓam ʿĪsā (d. 624/1227) still spoke Kurdish or even New Persian. And al-Moʿaẓẓam’s particular interest in Iran is seen in his patronage of two works (in Arabic) by Fatḥ b. ʿAlī Bondārī: one, a translation of the Šāh-nāma (ed. ʿA. Aʿẓam, Cairo, 1350/1931); the second, the standard abridgment of ʿEmād-al-dīn Kāteb Eṣfahānī’s history of the Saljuqs (ed. M. Th. Houtsma, Recueil de textes relatifs à l’histoire des Seldjoucides, Leiden, 1886-1902, II). Still, there is no evidence of any widespread translation movement among the Ayyubids, or of any general devotion to the Persian classics.

It is true, however, that the personal influence of Iranian scholars was very much felt in the religious sciences. Dominique Sourdel has shown that almost one-third of the madrasa professors in Aleppo between about A.D. 1150 and 1250 were of Kurdish or Iranian origin. The same figures would not hold for Damascus, let alone Egypt, but their presence in these places was far from negligible. However, most of these men had come to Syria not under the Kurdish Ayyubids but under the Turkish Zangids, particularly Nūr-al-dīn Maḥmūd, in the third quarter of the sixth/twelfth century (D. Sourdel, “Les professeurs de madrasa à Alep aux XIIe-XIIIe siècles d’après Ibn Šaddād,” Bulletin d’études orientales 13, 1949-51, pp. 85-115).

The most visible and distinctive Kurdish presence was in the army. This process too had begun long before the Ayyubid seizure of power, when the atābak ʿEmād-al-dīn Zangī had begun trying to control the Kurdish-dominated mountains north of Mosul in order to ensure regular access to Kurdish recruits. Though the Kurds were viewed with some disdain by Turkish troopers, they were adept at the same mode of warfare (mounted archery) and on a tactical if not social level were easily integrated into the regular Zangid forces. The policy of Kurdish recruitment was doubtless inspired by two problems: 1) That Syria and the Jazīra were unable to obtain many Turkish ḡolāms from Central Asia at this time, and thus could not rely on chiefly mamlūk forces; 2) while Turkman tribesmen were plentiful in this region, their primary loyalties were to their tribes, and they could not be subjected to regular discipline. In this light, Kurdish mercenaries might well seem relatively cheap and reliable.

In spite of the importance of Kurdish recruitment for the Zangid armies, one should not suppose that the Kurds were ever more than a minority of these forces. And though Kurdish troops become more visible than ever before in the reign of Ṣalāḥ-al-dīn, they certainly remained a minority, constituting at the highest possible estimate one-third of his forces. The Kurds in Ṣalāḥ-al-dīn’s armies were sometimes recruited and placed as individuals, but they are more commonly found as members of tribally organized units, of which the sources name four: the Hakkārīya (certainly the largest and most powerful), the Mehrānīya, the Ḥomaydīya, and the Zarzārīya (see, e.g., Abū Šāma, Ketāb al-Rawżatayn II, pp. 144, 179). After Ṣalāḥ-al-dīn’s death, however, such tribal units are rarely recorded, and Kurdish soldiers appear either as individuals or under the collective appellation “al-Akrād.” After Ṣalāḥ-al-dīn, in fact, the Kurds seem to have become a far less prominent part of the Ayyubid military establishment; their amirs are less often members of the political elite, which becomes increasingly Turkish.

In the last two decades of Ayyubid history, however, the Kurds re-emerged as a major political and military force. This was due chiefly to two new Kurdish migrations into Syria (in which both groups appear to carry names of regional origin rather than tribal names). The first of these were the Qaymarīya, who arrived in the train of the Khwarazmian marauders who swept through Syria in 642-43/1244-46; from them were drawn several of the most prominent amirs under al-Nāṣer Yūsof, and they remained a significant force under the Mamluk sultan Rokn-al-dīn Baybars (r. 658/1260-676/1277). The second group was the Šahrazūrīya, fleeing before the Mongols in 657/1259. Though numerous (3,000 mounted warriors), the Šahrazūrīya could not be integrated into the regular forces of al-Nāṣer; they were simply an element in the vast human debris thrown up by the Mongol invasion. In the final analysis, then, the Ayyubids did make a considerable use of Kurdish troops, as had their Turkish predecessors and rivals (the Zangids and Artuqids), but we cannot say that theirs was a Kurdish army.



The narrative sources for the history of the Ayyubids are very numerous, but the essential texts—more or less in order of composition—are as follows. ʿEmād-al-dīn Moḥammad Kāteb Eṣfahānī, al-Barq al-šāmī (Bodleian, Bruce 11 and Marsh 425; see H. A. R. Gibb, “Al-Barq al-Shāmī; The History of Saladin by the Kātib ʿImād-al-Dīn al-Iṣfahānī,” WZKM 52, 1953, pp. 93-115 for a resumé of the contents).

Idem, al-Fatḥ al-qossī fi’l-fatḥ al-qodsī, ed. Landberg, Leiden, 1888; tr. H. Massé, Paris, 1972.

Bahāʾ-al-dīn b. Šaddād, al-Nawāder al-solṭānīya wa’l-maḥāsen al-yūsofīya or Sīrat Salāḥ-al-dīn, ed. J. Šayyāl, Cairo, 1964; tr. Wilson and C. R. Conder, The Life of Saladin, London, 1897.

Ebn al-Aṯīr, vols. XI-XII. Ebn Ḵallekān [Beirut], I, pp. 255-57; VII, pp. 139-45 (the most important statements concerning the family’s origin).

Abū Šāma, Ketāb al-Rawżatayn fī aḵbār al-dawlatayn, 2 vols., Būlāq, 1287-92/1871-75.

Idem, al-Ḏayl ʿala ’l-Rawżatayn, ed. Kawṯarī, Cairo, 1336/1947.

Sebṭ b. Jawzī, Merʾāt al-zamān fī taʾrīḵ al-aʿyān, ed. Jewett, Chicago, 1907.

Ebn Wāṣel, Mofarrej al-korūb fī aḵbār Banī Ayyūb, ed. J. Šayyāl, 5 vols. (in progress), Cairo 1953-.

R. S. Humphreys, From Saladin to the Mongols: the Ayyubids of Damascus, 1193-1260, Albany, 1977, so far the only full-length attempt to deal with the whole period of Ayyubid rule. C. Cahen, “Ayyubids,” in EI2 I, pp. 796-807 (a superb short account with a genealogical table, to be supplemented by H. A. R. Gibb, “The Aiyubids,” in K. M. Setton, ed., History of the Crusades II, Madison, 1969, pp. 693-714.

On the origins of the Ayyubids, see V. Minorsky, Studies in Caucasian History, London, 1953, pp. 107-57.

On the careers of Ayyūb and Šīrkūh, see N. Elisséeff, Nūr al-dīn, II, Damascus, 1967, passim.

On Ṣalāḥ-al-dīn, the most useful studies are recent ones: H. A. R. Gibb, The Life of Saladin, Oxford, 1973, A. S. Ehrenkreutz, Saladin, Albany, 1972 (an interpretation diametrically opposed to Gibb’s), and M. C. Lyons and D. E. P. Jackson, Saladin: the Politics of the Holy War, Cambridge, 1982.

On the later rulers, see F.-J. Dahlmanns, al-Malik al-ʿĀdil, Giessen, 1975.

H. L. Gottschalk, al-Malik al-Kāmil und seine Zeit, Wiesbaden, 1958 (very important).

G. Schregle, Die Sultanin von Ägypten, Wiesbaden, 1961.

Finally, on the Kurds and the Ayyubid army, see D. Ayalon, “Aspects of the Mamluk Phenomenon. II,” Islam 54, 1977, pp. 1-32.

R. S. Humphreys, “The Emergence of the Mamluk Army,” Stud. Isl. 45, 1977, pp. 89-93, 46, 1977, pp. 150-52, 155-57.

(R. S. Humphreys)

Originally Published: December 15, 1987

Last Updated: August 18, 2011

This article is available in print.
Vol. III, Fasc. 2, pp. 164-167