History, Profiles, Reviews, Occasional Meanderings

Category Archives: 861-884

The Samarran Crisis and Abbasid Fragmentation


Note: I wrote this article for a recently concluded course. I’ve made slight modifications before publishing it here, mainly grammatical or spelling changes. I reserve full rights, of course.

Ibrahim Moiz, copyright and full rights reserved

The decade following the dramatic murder of Abbasid caliph Mutawakkil Jafar I in 861 has been recognized as one of the most volatile periods in Islamic history. It featured the murder of at least four – if not five consecutive – caliphs, of whom two had been engaged in a brief but bitter contest for power over Iraq; a fragmentation in the corporate elite; as well as the breakdown of Abbasid authority both near and far in a series of revolts. Though by the mid-870s the Abbasids had, largely through the energetic efforts of Mutawakkil’s son Muwaffaq Talha, managed to salvage the situation, an inexorable process had begun whereby the caliphate’s extremities became effectively independent. Though the logical link between central collapse and peripheral secession has been recognized, less clear is the exact nature of that link. This paper argues that such effectively autonomous rulers as Ahmad b. Tuloun and Ismail b. Ahmad at Egypt and Khurasan respectively based and legitimized their effective autonomy not only on the fact of central erosion but along the same patterns and trajectories as the conflict in Samarra and Baghdad.

This paper proceeds in the following sequence. First, it provides a background to the Abbasid elite structure in the Samarran heyday; next, it chronicles the tumultous events of the 860s, and the trajectory of the various elite networks that dominated the Abbasid centre and competed for its control. It then examines the move towards autonomy in first Egypt-Levant and then Khurasan-Transoxania in the context of the Samarran tumult, drawing a link between the elite networks competing in the centre and their offshoots in the periphery.

Abbasid power in the Samarran heyday

The garrison city Samarra, built by Mutasim Muhammad III in 836, came to epitomize the special core of Turkish crack troops that this caliph had introduced into the caliphal forces on an unprecedented scale . This corresponded with a period of centralization arguably initated by Mutasim’s predecessor, Mamoun Abdullah III, but which really came into its own during Mutasim’s reign. It featured the centralization of Abbasid power on a perhaps unparalleled level in the dynasty’s history; the caliph paid close and personal attention to affairs as far-flung as Egypt, Anatolia, Sind, and Khurasan . To be sure, dynasts already ruled in the caliphs’ name – the Aghlabids, Samanids, and Tahirids (or more accurately Ruzaiqids ) – but the Ruzaiqids were close members of the caliphal cohort, rendering their autonomy in Khurasan at this time largely an extension of caliphal control, while the long-influential Samanids nonetheless only received formal gubernatorial appointments during the ninth century, under the jurisdiction of the Ruzaiqid governor-generals . Aghlabid autonomy can be partly explained by circumstances specific to North Africa, which the Abbasids never managed to decisively overcome ; the appointment of a loyal abna lieutenant was the closest they would ever get to controlling the region. In short, the Abbasid caliphate in the mid-ninth century was more centralized across a larger expanse of territory than had been the case either before or after.

The Turkish slave corps’ role in the subsequent anarchy has led to them assuming an outsize role in much historiography of the period. In fact, though Turkish slave officers came to constitute a defining and dominant part of Mutasim’s regime – especially after he purged the Khurasani factions hitherto influential in the army following an alleged “coup” by his nephew Abbas b. Abdullah III shortly after his accession – even at Samarra, they constituted only one conspicuous bloc in the army, and were by no means themselves homogenous, as subsequent events would show. The Ruzaiqid potentate Ishaq b. Ibrahim b. Hasan, an Abbasid regnal pillar for several generations, is recorded to have gently chided a regretful Mutasim over his promotion of such Turkish potentates as Ashinas Abu Jafar, Wasif Abu Salih, and Aytakh Abu Mansur, likening them to branches without roots – in contrast to the Ruzaiqids and other preceding elites who had strong roots with the Abbasid family . Ashinas passed away during Mutasim’s regime and Mutawakkil eventually managed to corner and execute Aytakh . By 861, the remaining senior Turkish officers from this generation included Wasif, the cupbearer Bugha Abu Faris, and the elderly generallisimo Bugha Abu Musa, whose families would play major roles in the events that followed.

Elite fragmentation and competition in Samarra’s turbulent decade

Mutawakkil’s murder was preceded by his very public humiliation of his son Muntasir Muhammad IV, whom he superseded with a younger brother, Mutaz Zubair. This decision was taken reportedly on the advice of another court faction with Turkish heritage, albeit freeborn and heavily “Arabicized”: Mutawakkil’s vizier Ubaidullah b. Yahya b. Khaqan and his uncle Fath b. Khaqan, Mutawakkil’s closest friend. Also reportedly in the gunsights for an Aytakh-style demotion was Wasif, who therefore lent his tacit support, and five of his sons, to a conspiracy, in whose formation the cupbearer Bugha Abu Faris played a major role, to murder Mutawakkil and install Muntasir on the throne. The assassination team was led by the cupbearer’s protégé, Baghir Abu Muhammad, and while accounts on its composition vary, the most liberal account includes Muntasir’s Turkish friend and future vizier Utamish b. Khatrakin, the cupbearer Bugha, as well as Wasif’s sons – among whom Salih would emerge as the most prominent character – and Musa, the deputizing son of the elderly Bugha Abu Musa, who was himself away on campaign and apparently innocent of the conspiracy. The implication, however, is that the remaining Turkish proteges of Mutasim and their families stood to lose from Mutawakkil’s decisions, and that they preempted it. Incidentally one of Mutawakkil’s younger sons, Muwaffaq Talha, was also present at the scene; he is reported to have initially attempted to fight off the attackers and then withdrawn to his chambers. Muwaffaq’s subsequent influence among the second and third generation of Turkish officers, particularly Musa b. Bugha, makes his role in these events quite murky.

Mutawakkil and Fath having been murdered while Ubaidullah found out too late to influence events, Muntasir assumed the caliphate and, allegedly under pressure from the Turkish magnates, excluded and imprisoned the next brothers in line, Mutaz Zubair and Muayyad Ibrahim. Muntasir’s own rather murky demise just six months later in the prime of his life is perhaps indirectly explained by a reference from Masoudi, who claims that Bugha Abu Faris – the most unscrupulous villain in the sources – and Baghir had originally intended to murder Muntasir, and then decided first to do away with the father first and split the son against his brethren so as to avoid punishment . This dastardly plot is otherwise uncorroborated, but may be taken as an explanation of Muntasir’s sudden death in the context of the Turkish potentates’, and in particular the cupbearer’s, schemes. Such an impression could have only been reinforced when Bugha and Wasif, blithely ignoring the succession plan, instead installed an obscure cousin, Mustaeen Ahmad I b. Muhammad, to replace Muntasir. Mustaeen’s apparent weakness against the coupmakers was mocked in a poem, recorded by Masoudi, that compares him to a parrot in the cage of Bugha and Wasif, echoing their orders.

The next chapter in the drama, paving the way to civil war, is explained in the sources as the result of an intrigue between the older Turkish potentates, Bugha and Wasif (the elder Bugha, Abu Musa, had passed away by this point), and their younger co-conspirators in Mutawakkil’s murder. Musa b. Bugha had by this time departed on campaign, but the vizier Utamish b. Khatrakin – accused of mass corruption during a period of financial hardship among the rank-and-file – and Baghir Abu Muhammad still remained; it appears, though the sources take care to clarify his role as more passive, that Mustaeen had also tried to cultivate these younger Turks to their elder colleagues’ detriment. Excepting the role of Bugha and Wasif, however, Utamish and Baghir’s downfalls were quite different; a mutiny by rank-and-file troops prompted the vizier’s execution , while Baghir, a popular commander with the Turkish rank-and-file, was murdered by Wasif’s kinsfolk and his execution provoked the troops into riot . Bugha and Wasif had successfully wedged a gap between the caliph and Baghir, who then allegedly plotted to kill Mustaeen. In both episodes, interestingly, the Christian secretary Dulail b. Yaqub, who served first Utamish and then Bugha and was targeted by Baghir, played an important role; he appears to have epitomized the regime’s corruption for the frustrated Turkish troops.

Wasif and Bugha spirited Mustaeen away to Baghdad, where the Turkish contingent was far more diluted by other troops, especially the abna and Khurasani troops who had been sidelined in the Samarran period. The return to influence that Mustaeen’s escape portended for this group was epitomized in the leading role that Muhammad b. Abdullah b. Tahir, the able Ruzaiqid governor, assumed in the ensuing war, where it was he – not Mustaeen, Wasif, or Bugha – who assumed overall command. That the Baghdadis viewed the Samarran forces – mixed though they were, even if the Turks were the largest faction – as representing the unwelcome Turkish hordes is evidenced even in contemporary chronicler Tabari’s description of the subsequent civil war, where he refers to the Samarran forces en masse simply as the “Turks”, and where he clearly favours the Baghdadi side.

We need not delve into the fascinatingly extensive material that Tabari provides on the subsequent 865-66 war other than to note the maneouvres by the various elite factions involved. Wasif and Bugha – who, as Gordon notes, were now perilously cut off from their power base – assumed the role of bystanders in Baghdad. Led primarily by the Ruzaiqids, the Baghdad forces included the Khurasanis and remnants of the old abna families, such as the Banu Abi Khalid; Arab bedouins, primarily employed in securing supply routes, particularly in the tribal Anbar region; urban ayyaroun militias, who shared their name and urchin background with their more famous namesakes in Khurasan; heretical mubayyids, apparently diluted remnants of the ghulat cults first encountered a century earlier in the wake of the Abbasid revolt; Arab tribal notables; and freeborn officers of diverse backgrounds. In short, it comprised those who felt threatened by the Samarran legions.

It should be noted that before hostilities commenced, “second-rank” Turkish officers linked to Bugha and Wasif, including Bugha’s former bodyguard Bayakbak, followed the escaped caliph Mustaeen to Baghdad, where they humbled themselves and assured him that he could safely return to Samarra. Mustaeen refused, and the officers’ indignation was compounded by the patronizing attitude displayed by the Ruzaiqi governor Muhammad b. Abdullah b. Tahir. Returning to Samarra, they and the troops freed Muntasir’s imprisoned brothers and declared Mutaz Zubair their caliph. Mutaz’s brother Muwaffaq Talha, the closest Abbasid prince to the Turks, commanded the army that marched on Baghdad. While both sides were clearly multiethnic in composition, it is clear that a configuration of Samarran “up-and-comers” versus older, primarily freeborn Abbasid officers was taking place. Thus Attab b. Attab, a Khurasani veteran, quietly escaped to Baghdad, while Bugha’s own son Abdullah, having originally returned to Baghdad, quickly collected what information he could and then defected to the attackers.

The war lasted about a year. If Tabari’s richly detailed account is to be believed, the Samarran force, despite their military specialization, initially floundered in their assaults against a large, well-defended city. It was rather an eventual shrink in supplies, as well as clever diplomacy by the Samarrans, that turned the tide against Mustaeen. Mutaz and Muwaffaq’s priorities can be seen in the fact that they never made a serious attempt to encircle Baghdad, which was able to dispatch expeditions against proto-Shia rebels in Kufa at the same time as the campaign was ongoing. Once it was clear that the city would not be stormed after several attempts, Muwaffaq resorted to diplomacy. He managed to win over a steady trickle of increasingly important defectors: these included Abbasid amirs, the staff who had accompanied Mustaeen, and the Banu Khaqan, led by Ubaidullah b. Yahya and his uncle Muzahim b. Khaqan – after all, it had been this family that originally persuaded Mutawakkil to choose Mutaz as his successor, and it can have been no great task for Mutaz to win them back to their stomping grounds in Samarra.

Even Muhammad b. Abdullah b. Tahir, suffering a slow-burning ulcer that would claim his life a few years later and engaged in disputes with other Ruzaiqid notables, finally deserted Mustaeen. Interestingly, Baghdad’s populace seems to have maintained the strongest sympathy for their beleaguered caliph; when they suspected the governor as preparing to defect, they nearly mobbed him and had to be calmed by Mustaeen himself. With the various elites won over by Mutaz’s generous terms, Mustaeen abdicated and withdrew to exile; far less generous to his cousin, Mutaz had him murdered by chamberlain Saeed b. Salih just months later.

Mustaeen’s murder came in the context of a new factional struggle, between Mutaz and the senior Turkish commanders. Bugha Abu Faris and Wasif Abu Salih had been restored to some dignity after having lost the war, but the suspicious young caliph could hardly ignore his experience. He originally tried to have them murdered by the Ruzaiqids, but the pair quickly found out and obtained protection from Mutaz’s half-brothers Muayyad and Muwaffaq, who had close links to the Turkish corps at large. Silently fuming, Mutaz soon had the brothers imprisoned on sedition charges that are hard to verify; after a Turkish contingent led by Musa b. Bugha attempted to break them out and was only stopped by a Maghribi contingent, Mutaz had Muayyad executed, though the particularly strong ties between Musa and Muwaffaq appear to have prevented the latter’s murder . Seeing potential in the hostility between the Maghribis and Turks, Mutaz appeared to have tried to weaponize the former against the latter.

A more successful stratagem occurred in Mutaz’s attempts to play the Turkish potentates off one another. During 867, he first isolated Wasif by showering favours on Bugha Abu Faris while giving Musa b. Bugha control over Jibal, the Iranian highlands currently held by Wasif; Musa’s field commander, Muflih Abu Abdurrahman, routed Wasif’s vassal, the Arab Dulafi chieftain Abdul-Aziz b. Qasim b. Isa in a major military campaign whose roots lay definitively in Samarra’s politics; Musa spent the majority of the next few years on campaign. Meanwhile, Wasif was soon lynched in a mass mutiny over pay that was probably manipulated by Mutaz or Bugha, the latter conveniently slipping away shortly before events escalated and leaving Wasif to his fate.

Nonetheless it was not long before Bugha’s turn came. Too shrewd to miss the signs, he tried to precede events by arranging a marriage between his daughter Juma and Wasif’s senior son Salih. Mutaz, however, availed of a dispute between Bugha and Bayakbak in order to levy the latter against Bugha. Panicking at a sudden attack by Bayakbak during winter 868, Bugha fled but was soon abandoned by his troops; trying to escape to Salih’s residence, he was caught en route and beheaded at Mutaz’s orders. Thus in the space of a few months the young caliph had outmaneouvred Samarra’s most seasoned intriguers.

This cannot have escaped the emergent Turkish elites, and rather than wait for Mutaz to pick them off they pounced first. Barracks unrest over salaries was the catalyst, and Salih flexed his military resources to arrest Mutaz’s vizier Ahmad b. Israil as well as two other influential secretaries to the Turkish elite, Isa b. Ibrahim b. Nuh and Hasan b. Makhlad b. Jarrah. After a public humiliation they yielded some money, but this only delayed the inevitable. There was, in fact, a potential lifeline when lower-ranked Turkish officers – who could scarcely have been unaware of the potentates’ vast wealth – offered Mutaz that they would murder Salih if he only paid them; the caliph pled his fabulously wealthy mother Qabiha to lend some money, but to no avail. The lifeline missed, Salih, Bayakbak, and Musa b. Bugha’s brother Muhammad flung Mutaz in prison, where he starved to death.

Before Mutaz passed away, he was prevailed upon to sign a letter that confirmed his abdication in favour of his cousin, Muhtadi Muhammad V, son of former caliph Wathiq. The new caliph had a reputation for frugality, pietism, and gravity that must have helped his case; he used this reputation, however, to shore up his position, often making moralistic public recantments against the overbearing potentates. The targets included not only civilians but also the lower-ranked soldiers, to whom Muhtadi personally appealed and also sent his kinsmen, notably his brother Abdullah b. Haroun II, to negotiate and address grievances. The caliph appears to have realized that there was potential in allying with junior officers against senior officers.

In the meantime, however, Muhtadi had little choice but to cooperate with the commander who had installed him, Salih b. Wasif. For a brief period in 869, Salih ruthlessly dominated Samarran affairs and attempted to obtain money by hook or crook; he seized Qabiha’s hidden wealth and, against the caliph’s wishes but apparently in league with the new vizier, Abdullah b. Muhammad b. Yazdaid, again brutally extracted funds from the disgraced bureaucrats– arranging the public torture of the hapless Ahmad b. Israil, Isa b. Ibrahim, and Hasan b. Makhlad, of whom only Hasan survived the ordeal. The brutality drove Muhtadi into near-hysterics, but in the meantime a more imposing threat approached, for the bereaved Qabiha had escaped and reached out to Musa b. Bugha, whom her son had promoted to command the front, calling him to return. Refusing polite requests by Muhtadi to remain at the front, Musa moved inexorably on Baghdad with a large, proven force. The returning commander’s reputation was such that Salih lost the vast majority of his guard to desertion; he fled underground, but not before fatefully having stoked mistrust between the caliph and Musa . Drawing on the bitter experience of the past decade, Musa suspected the caliph of secret collusion with Salih, and was in any case reluctant to let his guard down until Salih’s discovery and elimination; Muhtadi, rather oddly – since most chroniclers paint a picture of Musa as refreshingly honest and dutiful, in spite of his regicide – suspected him of corruption. At one point the commanders discussed murdering the caliph; however, Bayakbak’s brother Tughtaya reportedly dissuaded them, citing Muhtadi’s upright character and the repercussions associated with such an act.

Nor were the commanders, again, the only Samarran Turk protagonists; junior officers and troops, still awaiting pay, conducted negotiations with the caliph; one condition they raised was to replace their variegated commanders with a trustworthy Abbasid scion, a role that would later fall to Muwaffaq . The Samarran slave troops in particular had been indoctrinated into reverence for the caliphal family. When Salih was discovered and executed, Musa, Bayakbak, and Muflih departed on an overdue campaign against one of the many revolts that had broken out. A short-term compromise was agreed between the various factions; such potentates as Musa and Bayakbak retained their rank and pay, while Musa’s brother Muhammad b. Bugha was promoted to command the Samarran army; in view of the following events, this appears to have been a ploy by Muhtadi to put the onus of payment onto Musa’s network. When in summer 870 another mutiny over pay arose Muhtadi had Musa’s brothers thrown in prison, where Muhammad was murdered, and recalled the commander from the front.

The succession of ensuing events was swift but complex . Muhtadi simultaneously mobilized a large army; along with the embittered families of Bugha Abu Faris and Salih b. Wasif as well as younger Turks, this included Khurasani officers such as Attab b. Attab and Masrur Abu Hashim. It was a coalition of variant interests, united only in their alarm at Musa’s influence. The caliph meanwhile tried to play off Bayakbak, who had longtime links with the Banu Khaqan and Banu Wasif, against Musa. But Bayakbak, apparently playing a double role, both informed Musa – who swiftly dispatched Muflih with a force outside Samarra – and presented himself to the caliph, who suspecting treachery had his disembodied head flung into the attackers’ ranks. Rather than demoralize them, this only whipped Bayakbak’s contingent into a rage and prompted several Turkish units to mutiny while others deserted. The deserters included Yarjukh, a Turkish commander whose daughter had married Bayakbak’s stepson Ahmad b. Tuloun, and who had hitherto backed Muhtadi. The battle was fierce, showing how close Muhtadi had come to overcoming the Samarran command; eventually, however, he was forced into flight, pursued by a Turkish contingent – led, ironically, by Bayakbak’s brother Tughtaya, who had so staunchly defended the caliph the previous year – and cruelly done to death.

It was Yarjukh who produced Muhtadi’s cousin, another of Mutawakkil’s sons called Mutamid Ahmad II b. Jafar I; he was duly proclaimed the caliph with the further cooperation of the Banu Khaqan: Ubaidullah b. Yahya and his brother Abdullah both served as vizier in the next few years. The unexpected stability that ensued in the caliphal centre was in large part attributable to the amicable relations between the Turks, particularly Musa b. Bugha, and Mutamid’s brother Muwaffaq Talha. Now, at last, the caliphate proceeded to set its provinces in order; a myriad conflicts had festered both near and far, and now Muwaffaq set about addressing them one by one. It is notable that contrary to the pattern of the past decade and even his own conduct with Muhtadi, Musa and his lieutenants –notably Muflih and his son Abdurrahman, Ibrahim b. Sima, and Ishaq b. Kundachik – showed complete deference to Muwaffaq, showing a discipline and willingness to fight that had been rare over the past decade . With the exception of Ishaq, who we shall encounter below, these lieutenants and others were killed in the battlefield against various rebels such as the Gulf plantations’ Zanj slaves, the Farsi adventurer Muhammad b. Wasil b. Ibrahim, and the Sistani commander Yaqub Saffar b. Laith b. Muaddal; Musa himself eventually stepped down in favour of Muwaffaq’s Khurasani preference, Masrur Abu Hashim, and passed away quietly a few years later, earning a public eulogy at court by the caliph Mutamid.

It was no mean feat for Muwaffaq to have restored some semblance of order to the caliphate’s centre within a few years. This would appear to lend itself to a resurgence of central control. Yet by the 880s, a steady decentralization of caliphal authority had ensued: Ahmad b. Tuloun became effectively autonomous in the west, and Samanid prince Ismail b. Ahmad b. Asad in the east. To understand how this happened, we must look at the direct links between these far-flung regions and the networks that fought out the 860s conflict.

Centre and periphery in the Samarran period

Despite the Abbasid breakdown, the caliphate was still a somewhat compact neighbourhood during this period – insofar as events in one part, particularly the centre, were directly linked to events in another. To take one example, we may note that shortly after Mutawakkil’s murder, the legendary frontier commanders Ali b. Yahya and Aqta Umar b. Ubaidullah, were slain in their campaigns against the Byzantines. This news quickly reached Iraq, further stoking unrest at the role of the Turkish praetorians, who were perceived as collectively abandoning their roles and, in contrast to the veteran freeborn commanders at the front, concentrating their energies on intrigue and murder . The effect worked the opposite way, too, with central politics affecting the periphery. Another frontier commander, Tarsous’ Turkish governor Balkajur, competed with its notables for influence in the city. When Mutaz took power at Samarra after Mustaeen’s flight, these notables arrived at Baghdad and tried to persuade Mustaeen that Balkajur had treacherously thrown in his lot with Mutaz. The governor may well have been sacked had Wasif Abu Salih not been at hand to retort that he had probably been unaware of Mustaeen’s survival; sure enough, Balkajur soon pledged loyalty to Mustaeen and so retained his position.

The most powerful example of the interplay between the centre and periphery comes from the Ruzaiqid family, that veteran interface between caliphal Iraq and Khurasan. Disputes within the Ruzaiqid family, which especially sharpened after their senior statesman Muhammad b. Abdullah b. Tahir passed away in 867, directly impacted and were impacted by events in Khurasan, where competition among the family flourished in an atmosphere of increased corruption. Sulaiman b. Abdullah b. Tahir’s troops in Khurasan provoked a revolt in Daylam that soon rallied behind an Alid contender for the caliphate, Hasan b. Zaid b. Muhammad, and expelled the Ruzaiqid governor. Escaping to Iraq with a large, ravenous army, Sulaiman assumed the Iraqi governorate from his brother Ubaidullah, but soon found its revenue insufficient to pay his troops and descended into a tortuous feud with his cousins. Meanwhile the troops proceeded to plunder the countryside; among their acts was the release of a Mosul notable called Musawir b. Abdul-Hameed from prison, where he had been sent for Kharijite activities . Once released, Musawir resumed his revolt; it would be left to Musa b. Bugha to mop up both revolts years later. In these episodes, we can see a clear link between separate events in far-flung parts of the realm.

Ahmad b. Tuloun and Samarran network competition

I propose here that the progressive decentralization of the caliphate can be viewed not only as a natural corollary to mayhem in its centre, but similarly linked directly to the competition of elite networks in Samarra. The case of Ahmad b. Tuloun is especially so, because he had actually been involved, tangentially but not insignificantly, with the 860s contest in Iraq. While Ahmad, like Bugha Abu Musa and Musa b. Bugha, is often viewed as an exception to the carnivorously corrupt Turkish potentate, and indeed according to a possibly fictitious account had little respect for his peers, he was nonetheless closely tied to the Samarran Turks and their affairs. His father, Tuloun, had been a slave praetorian commander imported by the Samanids to Samarra; Ahmad himself was familially related to two notable actors in Muhtadi’s downfall, his stepfather Bayakbak and his father-in-law Yarjukh; Yarjukh and Ahmad were especially close. His exploits had also won the personal attention of the caliph Mustaeen, after whose abdication he served as an agreeable escort to exile. When Bayakbak subsequently shot to real prominence during Mutaz’s period, he dispatched Ahmad to deputize for him at his newly acquired province in Egypt. Finally, Ahmad was the link between Bayakbak and Yarjukh, which prompted Yarjukh to desert Muhtadi after Bayakbak’s execution.

If, following the patterns of familial and patrimonial networks among Samarran Turkish commanders, Ahmad was part of the Bayakbak-Yarjukh network, it follows that he would align with Yarjukh’s backing for caliph Mutamid Ahmad II. And though Ahmad mounted his revolt, making Egypt effectively autonomous, during the reign of Mutamid, he consistently pled his case as the caliph’s protector and warden. In this competition Ahmad’s major rival was Muwaffaq Talha, Mutamid’s overbearing if doubtless competent brother, whose own son Mutadid Ahmad III was in line to succeed. Consider the rapid succession and interrelation of events in Egypt and Iraq.

In 868, Bayakbak dispatched Ahmad to Egypt as his deputy. By all accounts Ahmad helped reorder the country’s economy, so that agriculture soared at the same time as Iraq’s economy was caving under conflict . In 870, Bayakbak was slain in the conflict with Muhtadi, and the brothers Mutamid and Muwaffaq came to power with the assistance of Yarjukh and Musa b. Bugha; Yarjukh, linked to Bayakbak via Ahmad, inherited the Egyptian province and maintained Ahmad as deputy there. 873, Yarjukh’s expiry and his replacement with Mutamid’s son and original heir, Mufawwad Jafar b. Ahmad II, gave Ahmad an opportunity to forge links with the caliph . Mutamid appears to have increasingly chafed under Muwaffaq’s dominance, and it is entirely likely that Ahmad was aware of the tension between them owing to the potential conflict between Mufawwad and Muwaffaq’s powerful son, Mutadid, for succession. In the middle 870s, at the height of Muwaffaq’s conflict with the various revolts, Ahmad refused to acknowledge the Abbasid regent, only recognizing Mutamid as caliph.

Having narrowly defeated Saffarid commander Yaqub b. Laith’s invasion of Iraq in 876, Muwaffaq was momentarily free to address this upstart, and tried to send his reliable lieutenant Musa to take over; however, Musa’s shortage of funds prematurely ended his expedition. It was only sustained Byzantine pressure on Egypt that forced Muwaffaq to recognize Ahmad’s position – that too after an appeal from Mutamid, suggesting some success on Ahmad’s part in winning the caliph’s confidence against his regent. Regionally, he faced a challenge from the Aghlabid ruler to his west, Ibrahim b. Ahmad b. Muhammad, which was apparently approved from Samarra but could only make limited headway. Ahmad’s star continued to rise during the late 870s: he removed his main competitor, the bureaucrat Ahmad ibn Mudabbir, who hailed from an important but rapidly declining bureaucratic family whose Iraqi wing had meanwhile been scuttled by the Zanj revolt . This gave Ahmad full control over Egypt, and he advanced into the Levant, coopting or defeating the mostly Turkish area governors up to the Byzantine frontier. His purported aim always having been to consolidate the front for jihad against Byzantium, Ahmad’s efforts did repel Byzantine pressure in the eastern Mediterranean, even if some of his numerous competitors – such as Tarsus’ Turkish commandant, Yazaman Khadim – could claim similar exploits. More to the point, Ahmad so impressed the caliph that during 882 he secretly slipped out of Samarra and entered Syria, hoping to enter Ahmad’s protection.

The implications of Mutamid’s escape, had it succeeded, were profound. It would have reduced the significance of Iraq in the Abbasid order and returned it to Syria; decades earlier, Mutawakkil himself had contemplated such a step. It would also have anticipated the Mamluk sultanate’s protectionism over the caliphate by some four hundred years in the same region. It was, of course, clearly a threat to the long-embattled Samarran elite who had fought so hard to save Iraq in the past and who had rallied behind Muwaffaq. It is therefore no coincibtldence that, as he had once dispatched the Samarran veteran Musa b. Bugha againt Ahmad, Muwaffaq now dispatched the deceased commander’s lieutenant, Ishaq b. Kundachik, to take over Ahmad’s domains. Ishaq could not budge the firmly consolidated Tulunid realm, but he did achieve the significant service of intercepting Mutamid and returning him to Samarra’s “safety”. In Mutamid’s adventure, we can see the final competition between two Samarran networks that had briefly coalesced a decade earlier to remove Muhtadi: the networks of Bayakbak and Musa b. Bugha. In the west, it was the former that won out, for Ahmad’s failure to protect Mutamid did not preclude the effective, and sustained, economy of the Egyptian-Levant realm he carved out.

Autonomy in the Islamic east

The linkage between the Islamic east – Khurasan and Transoxania – and the Samarran conflict is subtler but nonetheless clear. The Ruzaiqids who governed Khurasan were always closely tied into the caliphal elite, as noted above. This network was an active participant in the Iraqi conflict of the 860s, and on balance lost its former glory. With the exception of Baghdad’s very capable governor Muhammad b. Abdullah b. Tahir, the family fell into a complex series of squabbles. It is not necessary to detail them comprehensively here, except to note that this saw the steady displacement of caliphal control in the east. We have already encountered the misadventures of Sulaiman b. Abdullah b. Tahir above. His nephew, Muhammad b. Tahir b. Abdullah, proved a particularly ineffectual governor-general whose failures provoked differing responses from the Saffarids and Samanids.

The more spectacular response was the military campaign of the coppersmith Yaqub b. Laith b. Muaddal, who exploded from the Sistan backwater to conquer Khurasan and as far as Baghdad’s outskirts before Muwaffaq Talha finally checked his charge. Traditionally historians have tended to classify Yaqub, with his famous disdain for the moribund Abbasids and his militant lower-class origins, as a militaristic and unprincipled adventurer . Tor, more recently, has shown that in fact the Saffarids fit, and certainly tried to fit, the ideal of a frontier ghazi, regularly raiding into the pagan frontiers in the Sind region and espousing an uncompromisingly militant Islam that helped win over some defections from the Kharijite defections . If Yaqub showed ruthless militarism, it only reflected the environment where he operated.

At the outset of his adventures Yaqub captured Herat from its longtime Kharijite pretender, Mutawakkil Abdurrahman, whose head he sent to Iraq with a request for governorship, as well as routed the Alid rebel leader Hasan b. Zaid in northern Iran. This occurred in conjunction with, and may have been intended to legitimate, Yaqub’s entry into Nishapur during 873. Only when Muwaffaq and vizier Ubaidullah b. Yahya, recognizing the upset in the Ruzaiqid-dominated hierarchy of Abbasid vassals in the province that this represented, excluded Yaqub from this hierarchy did the Saffarid commander sweep decisively westwards towards Iraq. Indeed his centrally approved opponent in western Iran, sometime rebel Ali b. Husain b. Quraish, in fact echoed Yaqub’s hostility towards the Ruzaiqids with a complaint to Samarra . The point is not that Yaqub was keen to maintain the Abbasid structure in Khurasan – his personal disdain and mistrust towards the dynasty was well-known – but rather than that he was not opposed to entrenching himself within this structure if he could. It was only when this failed that Yaqub swept across Khurasan and made his famous attack on Iraq. Moreover, this was a common tactic in a period where the Ruzaiqids had been thrown into dysfunction; Yaqub was merely its most successful exponent.

Western Iran was the site of several longtime adventurers whose relations with the caliphate had varied between submission and revolt; during the early 870s, Muwaffaq had dispatched Samarran Turks such as Harith b. Sima, Abdurrahman b. Muflih, and Kayghalagh Abu Ahmad to confront them with only mixed success. These included the tribal Banu Dulaf, sometime lieutenants to the Ruzaiqids, and the notoriously temperamental Muhammad b. Wasil b. Ibrahim . Ibn Wasil’s longevity over a thirty-year period owed much to his purely transactional alliances that alternated between groups as varied as the caliphate and the local Kharijites. Having just broken off his latest alliance with Samarra by killing Harith, Ibn Wasil failed to make a deal with Yaqub and thereby tried to attack him on the caliphate’s behalf; he was duly captured and his long career brought to an abrupt end. Yaqub’s refusal to ally with Ibn Wasil and, even after his defeat against Muwaffaq, the nascent Zanj forces in southern Iraq, do not fit the actions of a purely opportunistic adventurer, no matter his disdain for the Abbasids. It appears that the difference between him and his brother Amr, whom the Abbasids successfully coopted and then played off against other regional forces, is not as stark as is often portrayed; both were willing to work with the Abbasids if the Abbasids would have them at little cost to their autonomy.

Rather the politics of Khurasan in this period should be viewed along the same continuum as the elite struggles in Iraq, predominant among them the Ruzaiqids. Such adventurers as the Banu Dulaf, Saffarids, and others operated in an unstable environment exacerbated by the breakdown of this hierarchy and the dysfunction of the once-solid Ruzaiqid control in the province. Occupying an eminent role in the eroding hierarchy were the Samanids. While the family’s influence in the region predated the Abbasids, we have noted that it was after helping suppress Khurasani revolts that they received official governorates in 820; that is, their official governorate remained subject to the Ruzaiqids, and thus indirectly the caliphate. Fifty years later, even an ineffectual Ruzaiqid scion such as Muhammad b. Tahir was able to overrule a senior Samanid scion such as Herat governor Ibrahim b. Ilyas, whose cautions he ignored and whom he dispatched on a vain mission to check Yaqub’s progress.

The northern, and eventually dominant, section of the Samanid family eventually availed of the Ruzaiqid decline to assert itself with proper autonomy: it was the remarkable Ismail b. Ahmad b. Asad who eventually asserted Samanid dominance. Though the Samanids had traditionally been loyal lieutenants, the Abbasid caliph Mutadid Ahmad III – having learnt, as we saw repeatedly at Samarra, to play off powerful commanders against each other – decided to neutralize the ambitious Saffarid commander Amr b. Laith b. Muaddal by giving him their domain . The Samanids, however, bested their rivals and assumed dominance over the region. Much as with Ahmad b. Tuloun in Egypt, however, Ismail continued to defer to legitimist authority: as Ahmad had guarded the Byzantine frontier, so did the Samanids guard the frontier against the pagan Turks in central Asia, as well as finally wiping out the Alid revolt in northern Iran that had been brought on by Ruzaiqid misrule. The Samanids were not as directly connected with Iraq’s factional intrigues as the Tulunids, yet indirectly the breakdown of Ruzaiqid authority fuelled by those intrigues set in motion a cycle that resulted in their, and Transoxania’s, autonomy.


The decentralization of the Abbasid caliphate in the third quarter of the ninth century was not simply the logical result of central collapse, but tied in closely with the dynamics of the conflict that caused the centre to collapse. The rise of the Tulunids in the west and the Saffarids and Samanids in the east proceeded along patterns and trajectories of elite competition in the centre; both the Tulunids and Samanids, and to an extent even the Saffarids, tried to make themselves autonomous agents of the caliphate in a way that was strikingly similar to various factions – such as the Samarran Turk potentates and the Ruzaiqids – in the Iraqi centre. The outcome of the Samarran conflict was therefore the permanent decentralization of the caliphate into several geographic blocs – the Aghlabid and Tulunid west, the Abbasid centre, and the Saffarid and Samarran east – that theoretically adhered to central rule but in fact constituted the autonomous proto-sultanates that continued right up to the modern period.

Abun-Nasr, Jamil. A History of the Maghrib in the Islamic Period. Oakleigh, Melbourne: Cambridge University Press, 1993.
Adamec, Ludwig. Historical dictionary of Islam. Lanham, MD: Rowman and Littlefield, 2017.
Bianquis, Thierry. “Autonomous Egypt from Ibn Tulun to Kafur, 868-969.” In The Cambridge History of Egypt: Volume One, Islamic Egypt 640-1517, edited by Carl Petry, 86-119. Cambridge: Cambridge University Press, 1998.
Bosworth, Clifford. “The Armies of the Saffarids.” BSOAS 31 no. 3 (1968), pp. 534-54.
Daftary, Farhad. “Sectarian and National Movements in Iran, Khurasan and Transoxania during Umayyad and early Abbasid times.” In History of Civilizations of Central Asia, Volume IV: The Age of Achievement, A.D. 750 to the end of the fifteenth century; Part One: The historical, social, and economic setting, edited by Muhammad Asimov and Clifford Bosworth, 41-60. Paris: UNESCO, 1998.
Gordon, Matthew. The Breaking of a Thousand Swords: A history of the Turkish military of Samarra (A.H. 200-275/815-889 C.E.). Albany, NY: State University of New York Press, 2001.
Ismail, Osman. “Mu’tasim and the Turks.” BSOAS 29, no. 1 (1966), pp. 12-24.
Ismail, Osman. “The Founding of a New Capital at Samarra.” BSOAS 31, no 1 (1968), pp. 1-13.
Al-Masudi, Ali b. Husain, Meadows of Gold, tr. Paul Lunde and Caroline Stone. New York, NY: Routledge, 2010.
Al-Tabari, Muhammad b. Jarir, History of the Messengers and Rulers:
– “Volume 34: Incipient Decline: The Caliphates of al-Wathiq, al-Mutawakkil, and al-Muntasir A.D. 841-863/A.H. 227-248”, tr. Joel Kraemer, State University of New York Press.
– “Volume 35: The Crisis of the Abbasid Caliphate: The caliphates of al-Musta’in and al-Mu’tazz A.D. 862-869/A.H. 248-255”, tr. George Saliba, State University of New York Press.
– “Volume 36: The Revolt of the Zanj,” tr. David Waines, State University of New York Press.
– “Volume 38: The Return of the Caliphate to Baghdad,” tr. Franz Rosenthal, State University of New York Press.
Tor, Deborah, Violent Order: Religious warfare, chivalry, and the ‘Ayyarun phenomenon in the medieval Islamic world. Wurzburg: Ergon, 2007.