2. The remarkable career of Jānibak al-Ṣūfī: contemporary and modern understandings

3. Towards an integrative approach: Jānibak al-Ṣūfī's ego network in the Mamluk/Anatolian Frontier zone

What should we make of these differing perspectives on Jānibak? In the third and main part of this article, we will not try to engage in any futile argumentation in favour of one or the other. Rather, we wish to present a different, more nuanced approach, which tries to integrate these different perspectives and understandings into a more comprehensive analytical framework.

Two related assumptions will be the starting point for this re-assessment. The first is the acknowledgement of the reality that throughout most of the fifteenth century, southern and eastern Anatolia remained a dynamic and permeable zone between Ottoman, Egyptian and Timurid sovereignties, where by a kind of a process of imperial stretch the physical enforcement of their trans-regional authorities was limited and where as a result plenty of opportunities and strategies were available for local and regional leaders to manifest themselves against their subordinates, peers, competitors and overlords.Footnote ⁴⁹ As a result, we need to operationalise the fact that between 838/1435 and 841/1438 Jānibak was active in a frontier zone, on the constantly shifting fringe of competing Mamluk, Timurid and Ottoman spaces of power and authority. This means that we have to accept that in this enormous geographical area stretching between Timurid Iran, Ottoman West-Anatolia and Mamluk Syria, political instability, constantly changing sovereignties, and continuous competition for control over, and access to, resources defined the standards of power elites’ social practice, and that this created huge opportunities as well as dangerous pitfalls for local chiefs who wished to try their luck, for regional leaders who wished to pursue their ambitions, and for trans-regional leaders who wished to manifest their authority and expand their reach. For Jānibak, this means that we cannot neglect the fact that he should be also considered from a local Anatolian perspective, without at the same time forgetting that more than anything it was a trans-local high-profile Mamluk background that made him who he was, even in remote areas such as Sinop by the Black Sea.

The second basic assumption which follows from this is that we need to accept the simple fact that the historical realities and perceptions of Jānibak's case may have meant many different things to many different people at many different times and in many different places. It is therefore extremely relevant to take into account who those different people were for whom Jānibak's case may have been meaningful, and to see how these differing perspectives and the different relationships that tied them to Jānibak as a result can further understandings of this story. Most importantly for the purpose of this article, these two basic assumptions mean that conflicting assessments of Jānibak's agency versus his non-agency can be reconciled by an exploration of the varied and interlocking social groups of agents that interacted in meaningful ways with Jānibak between 838/1435 and 841/1437, an interaction that coalesced into a complex and dynamic ego-network that was brokered by Jānibak's agency and identity.Footnote ⁵⁰ More precisely, this ego-network will be reconstructed through the prism of the changing scale of the spatial realities (local, regional, trans-regional) in which it operated, three interrelated socio-spatial network perspectives that help to make more sense of Jānibak's story, as well as of the different relational contexts in which it was constructed: a local or intra-regional Türkmen Anatolian perspective, an inter-regional Mamluk Anatolian perspective, and a trans-regional Anatolian frontier zone perspective.

4. Jānibak as a power broker in Türkmen Anatolia

From his appearance near Ottoman Tokat down to his last combat near Aqquyunlu Çemişgezek, the historiographical representation of Jānibak's Anatolian adventures and connections come across as not necessarily different from those of local chiefs in the area. Just as these chiefs all did, Jānibak al-Ṣūfī basically used the interlocking opportunities offered by this political and military frontier zone — local autonomy amidst fluid trans-local sovereignties, direct access to resources that were beyond the physical reach of trans-local monopolies, and a readily available pool of experienced warriors that were not bound to trans-local loyalties — to carve out a place for himself in this huge and hybrid social space.

All reports agree that Jānibak never operated alone, that there were always people moving around through South-East Anatolia with him. Even in his final hours of freedom, on Friday 25 Rabīʿ II 841/25 October 1437, when he was trying to escape from his Aqquyunlu hosts, Egyptian sources suggest that he only did so after “consultation with his companions (fa-shāwara aṣḥābahu)” and that he fled “with some twenty horsemen from among his companions, so as to save himself (wa-maʿahu ʿishrūn fārisan min aṣḥābihi li-yanjū bi-nafsihi)”.Footnote ⁵¹ It remains impossible to retrieve the actual identity of these “companions”, but the fact that they were reported as having remained with Jānibak in his hour of need, when he had been obliged to seek Aqquyunlu shelter, suggests that there was a strong tie binding them to their leader's cause. Another reference indeed suggests that these “companions” refer to a select personal retinue of “sworn brothers”: two months earlier, on 15 Muḥarram 841/19 July 1437, when Jānibak and his Dulgadirid partner Nāṣir al-Dīn Meḥmed had been camping near Marash, Egyptian sources again explain that “Ibn Dulghādir and Jānibak al-Ṣūfī were with a small group (nafar qalīl), because their main party (jumūʿahumā) had remained with the amir Sulaymān b. Nāṣir al-Dīn Meḥmed, besieging Qaysariyyat al-Rūm (Kayseri)”.Footnote ⁵² A small band was all that had been accompanying them when they were surprised by an overwhelming majority of “[the sultan's] nāʾibs and their peers with about 2,000 horsemen”.Footnote ⁵³ What is important to understand from this is that it suggests that Jānibak had been leading others, next to those close “companions”, on his Dhū l-Ḥijja 839/July 1436 flight with Nāṣir al-Dīn Meḥmed deep into Ottoman lands and then on their Ramaḍān 840/March 1437 march back against Kayseri; these were probably a mixture of Dulgadirid and other troops and followers, put to work in the Kayseri-siege, while their leaders had gone out raiding or hunting in old Dulgadirid territories with a small selection of followers, who can only have been — indeed — their personal retinues.

These troops and personal retinues however, were only a tiny remainder of the much larger hosts that had been following the leaderships of Nāṣir al-Dīn Meḥmed and Jānibak al-Ṣūfī before the two devastating Syro-Egyptian campaigns in the regions of Elbistan and Gaziantep, in the summer of 839/1436. Jānibak's total defeat near Gaziantep, and his narrow escape from the arrest which befell most of his followers, clearly was a turning point in his Anatolian career, as was remarked by the contemporary Egyptian historian Ibn Taghrī Birdī:

Much more information is indeed known about the wide range of supporters that had been rallying to Jānibak's cause before Dhū l-Ḥijja 839/July 1436. Already when he had been residing in Ottoman Tokat, many surely had been attracted by the appeal of a new and promising leadership, confirmed by the support from the Ottoman sultan Murād Beg's representative in Tokat, and by the terror which was rumoured to be spreading among sultan Barsbāy's agents in northern Syria and Anatolia.Footnote ⁵⁵ Others were drawn in by a very pro-active campaign of official correspondence from Tokat; the substance of one type of such correspondence is suggested by the summary of the first contact that was established between Jānibak and Nāṣir al-Dīn Meḥmed in Elbistan, in Shawwāl 838/May 1435:

Other types of such correspondence to lesser regional figures included more direct appeals for support, alliance or even subordination to Jānibak's authority.Footnote ⁵⁷ Whatever its actual nature and faraway sources’ reconstructions of it, this soon turned out to be a very successful campaign, feeding many rumours in Syro-Egyptian lands about who was or was not part of Jānibak's growing Anatolian network of allies, partners, and subordinates;Footnote ⁵⁸ most importantly, it resulted in the build up of a substantial military force around Jānibak, eventually — by the time of these troops’ fatal attack on the sultan's army near Gaziantep in Dhū l-Ḥijja 839/July 1436 — reaching a substantial size to Anatolian standards, assessed by Egyptian sources at “about 2,000 horsemen”.Footnote ⁵⁹ The wide variety of local participants in this new leadership project is best illustrated in a summarising report by the contemporary court historian Ibn Taghrī Birdī, who tried to capture this remarkable emergence and deployment of a highly diverse new Anatolian power network in the following sentences:

From this and similar reports that provide at least a few more details about the individuals and groups that joined Jānibak's ranks, there actually emerge three types of Türkmen participants in this new network, each more or less integrated into it in different ways and for different reasons.

A first type concerned those that had accepted Jānibak's leadership in unequivocal fashion, Ibn Taghrī Birdī's “large group” of supporters, those who had rallied to Jānibak's cause when he was in Tokat, or perhaps even already before, as well as those who had come to Tokat to also physically strengthen his ranks. Source reports have next to nothing to say about these rank-and-file whose chief Jānibak became, but most of them probably were of very mixed Anatolian Türkmen origins, surely also including at least some bands of Türkmen brigands and freebooters hoping to benefit from their new leader's actions. It may also be assumed that this first circle of subordinate anonymous supporters made up a majority among the great host of horsemen that rode with Jānibak in Dhū l-Ḥijja 839/July 1436. At the core of this first, subordinate and anonymous Türkmen type of supporters in Jānibak's Anatolian network there undoubtedly stood his afore-mentioned personal retinue: two or more dozens of personal bodyguards and sworn brothers who would stay with Jānibak up to the dramatic end of his adventures. “His dawādār” or master of the pen-box, mentioned above as representing Jānibak in the first negotiations with Nāṣir al-Dīn Meḥmed, must have been a member of this close retinue, as would have been other holders of similar household-positions of prestige in Jānibak's entourage. By analogy with that dawādār, also the other representative in this important mission to Nāṣir al-Dīn Meḥmed — referred to as Meḥmed b. Kundughdī b. Ramaḍān al-Turkumānī — must have been a member of that private retinue; nothing further is known about this Meḥmed, but his name suggests that it concerned a member of the Türkmen Ramadanid dynasty of Adana and Tarsus who, for unknown reasons, was trying his luck way beyond his family's homeland, as a trusted member of Jānibak's close entourage.Footnote ⁶¹ This mainly Türkmen identity of Jānibak's personal retinue and military household at the centre of the growing number of his Anatolian followers is finally indicated by the fact that another envoy from Jānibak — sent in Shawwāl 838/May 1435 on a similarly important mission to the sultan's territories in northern Syria but intercepted and arrested near Aleppo — was described in the sources as “a Türkmen known as Meḥmed (rajul turkumānī yuqāl lahu Muḥammad)”.Footnote ⁶²

Whereas these Türkmen followers were in a clear subordinate position to Jānibak's leadership, a very different kind of relationships was established with Türkmen leaders and their followings in the region of South-Eastern Anatolia. For relationships established before Shawwāl 838/May 1435 with local rulers such as Isfandiyar, the Türkmen ruler of Sinop, and Yörgüç Pasha, the Ottoman governor of Amasya, information about what exactly connected them to Jānibak's cause — apart from a general attitude of hospitality for a refugee Egyptian grandee — remains wanting.Footnote ⁶³ We are better informed, however, about the ties that were created after the Shawwāl 838/May 1435 promotional campaign of envoys and letters sent from Tokat to Türkmen amirs such as — in Ibn Taghrī Birdī's words — “Nāṣir al-Dīn Meḥmed b. Dulghādir, nāʾib of Abulustayn, Aslamās b. Kubak, Meḥmed b. Quṭbakī, and Qarā Yuluk”. Almost all of these has an interesting story to tell about the dynamic contours of Jānibak's Türkmen network.

The two characters that look least familiar in this list, and about whom almost no further information has been preserved, are Aslamās b. Kubak and Meḥmed b. Quṭbakī. For the latter Meḥmed, no further data could be retrieved, apart from the fact that Ibn Taghrī Birdī and his Egyptian colleagues indeed all linked him to a position of local Türkmen leadership in southeastern Anatolia.Footnote ⁶⁴ Some better idea of what this type of leadership consisted of, may be retrieved from the datum that has been preserved for Aslamās b. Kubak. Although no biographical information has been preserved for Aslamās himself, there has for another member of his family, providing a much better insight into the position and status of this family in southeast Anatolia. This concerns the amir al-Ḥusayn b. Kubak al-Turkumānī, identified by Ibn Taghrī Birdī as “amīr al-Turkumān al-Kubakiyya”, leader of the ‘Kubak-Türkmen’ group.Footnote ⁶⁵ Several Egyptian source reports detail how, for several years and often leading a mixed host of Türkmen and Kurdish troops, this Ḥusayn b. Kubak and his brother Sūlaw had been raiding and causing havoc in the region of Cairo-controlled Malatya, even looting the town of Malatya itself on several occasions.Footnote ⁶⁶ Eventually, Ḥusayn b. Kubak was murdered in 821/1418, by order of sultan al-Muʾayyad Shaykh, as part of his Anatolian campaigns to restore Mamluk hegemony in the region.Footnote ⁶⁷ Although they had been subdued by the assassination of Ḥusayn, the Kubak family's leadership position over a local gathering of Türkmen groups in or on the fringes of the Malatya region seems to have continued. This is suggested by the fact that — at least in one version of the first phase of Jānibak's story by al-Ṣayrafī — Jānibak and his following only joined forces with Aslamās (and with Meḥmed b. Quṭbakī) when they were moving south from Divriği to Malatya, that is, when they were marching through Kubakiyya homelands.Footnote ⁶⁸ This local leadership is also suggested by Ibn Ḥajar's phrase that it were “the amirs of Aslamās b. Kubak and of Meḥmed b. Quṭbakī” who had joined Jānibak's ranks.Footnote ⁶⁹ It is, finally, most clearly visible from the fact that Aslamās suddenly appeared in Cairo in late Ramaḍān 839/April 1436, where he was duly rewarded by sultan Barsbāy for abandoning Jānibak at this crucial moment, just before the Gaziantep confrontation, and received gifts and honours, after which — in Ibn Ḥajar's words — ‘the sultan [. . .] sent him back to his lands’”.Footnote ⁷⁰ This Aslamās clearly then represents another type of participants in Jānibak's network, a type that also encompassed his peer Meḥmed b. Quṭbakī and — according to Ibn Taghrī Birdī — “other Türkmen amirs like them”Footnote ⁷¹ : they were some of the many small-time southeast-Anatolian Türkmen leaders that were active on the fringes of Mamluk suzerainty in an endless competition for opportunities, loyalties, and resources, who from Shawwāl 838/May 1435 onwards rallied with their personal retinues and troops around Jānibak; when the large-scale Syro-Egyptian campaigning of Ramaḍān-Dhū l-Ḥijja 839/March-July 1436 erased any apparent prospect for new opportunities, Aslamās, and probably also many of the other small-time chiefs like him — of whom, indeed, nothing further is heard —, proved to be of little use to Jānibak, and most of them returned to their previous activities in search for new opportunities in the area.

Ibn Taghrī Birdī's list includes two more names of Türkmen amirs in Jānibak's network: “Nāṣir al-Dīn Meḥmed b. Dulghādir, nāʾib of Abulustayn” and “Qarā Yuluk”. The Dulgadirid leader of Elbistan and the Aqquyunlu leader ʿUthmān Beg Qarā Yülük have been introduced above as the dynastic leaders of regionally powerful Türkmen tribal confederations — the one emanating from the regions of Elbistan and Marash, the other from the Upper Euphrates region —, and throughout their long careers each of Meḥmed and ʿUthmān in their own ways enjoyed very mixed relations with attempts to extend Cairo's suzerainty over the lands they controlled. In terms of resources, manpower, and regional authority, these two leaders clearly were of a very different standing from that of small-time chiefs such as Aslamās. The ties that bound these regional leaders and their families to Jānibak's cause also were of a far more complex nature than was the case for Türkmen amirs like Aslamās. They represent, therefore, a third type of Türkmen partners in Jānibak's network.

Relations with Nāṣir al-Dīn Meḥmed in particular developed in peculiar and complex ways. As illustrated in a quote above, various sources agree that at first these relations were of a rather formalised and nominal nature, Meḥmed simply guaranteeing Jānibak's agents “that he would not hand over nor abandon Jānibak al-Ṣūfī in case he would come to him”.Footnote ⁷² Soon thereafter, in Rabīʿ I 839/October 1435, this formal acceptance of Jānibak's Anatolian position by the Dulgadirid leadership seemed to materialise in real support, when near Malatya Jānibak's group was strengthened by Nāṣir al-Dīn's son Sulaymān and an alleged number of “150 horsemen”.Footnote ⁷³ At first however, this turned out to be a ploy designed to win Jānibak's trust, isolate him from his followers, capture him, and take him to Elbistan.Footnote ⁷⁴ Later, when soon after their arrival in Elbistan Nāṣir al-Dīn undid his son's actions and restored Jānibak to his leadership, this Dulgadirid double-dealing enabled the unexpected creation of a remarkably close partnership between Nāṣir al-Dīn Meḥmed and Jānibak, that was to last until Muḥarram 841/July 1437. Jānibak's cause and that of Nāṣir al-Dīn Meḥmed's became inextricably connected, and wherever Nāṣir al-Dīn turned up throughout these twenty odd months — whether between Elbistan and Marash, in Ottoman lands, or in Karamanid territory — “his partner” (nazīluhu)Footnote ⁷⁵ Jānibak and his followers were by his side. Actually, this partnership was cemented at some point between Rabī I and Rajab 839/October 1435 and February 1436 by the marriage between Jānibak and Nāṣir al-Dīn Meḥmed's daughter Nafīsa.Footnote ⁷⁶ By the time of the Syro-Egyptian campaigning of Ramaḍān-Dhū l-Ḥijja 839/March-July 1436, the powerful reality of this relationship was demonstrated by the presence of “all the sons of Nāṣir al-Dīn Beg Ibn Dulghādir — except for Sulaym ān —” in Jānibak's army that marched on Gaziantep.Footnote ⁷⁷ Its strength was then illustrated by the fact that even despite the disastrous outcome of the confrontation near Gaziantep, the partnership between the two leaders — now on the run — was continued beyond Dhū l-Ḥijja 839/July 1436. Only in Muḥarram 841/July 1437, when their encampment in the Marash region was attacked by the governor of Divriği, did their relationship end. Upon this new calamity, Ibn Taghrī Birdī explains, “Ibn Dulghādir started to hate him, and they separated from that day onwards”.Footnote ⁷⁸

As seen above, at that particular moment, Jānibak sought refuge with the sons of that other regional leader of repute and status, the Aq Qoyunly ʿUthmān Beg Qarā Yuluk. However, by Muḥarram 841/July 1437, ʿUthmān Beg was already dead, and his sons Meḥmed and Maḥmūd proved unreliable hosts, to say the least. Clearly, at this particular moment, there was no longer reason for them to ally with Jānibak, and just as Nāṣir al-Dīn Meḥmed had done, and as Aslamās and his peers had done in the spring of 839/1436, they forsook him.Footnote ⁷⁹ This act of betrayal was however not unanimously welcomed among the new Aqquyunlu leadership, the brothers of Meḥmed and Maḥmūd.Footnote ⁸⁰ Most importantly, it was not entirely consistent with previous Aqquyunlu policy vis-à-vis Jānibak's cause. In fact, the previous policy mirrored the positive engagements of the Dulgadirid leadership's far more than this negative outcome would suggest. Although no direct information has survived, it is highly likely that the above mentioned Tokat correspondence to “Qarā Yuluk” concerned an initiative parallel to that to Nāṣir al-Dīn Meḥmed. In this case too, the promotional campaign seems to have resulted in a formal expression of acceptance and nominal support for Jānibak by ʿUthmān Beg. This is at least suggested by the fact that shortly after Shawwāl 838/May 1435, Jānibak joined forces with ʿUthmān Beg's son Meḥmed — who would betray him some two years later — who “received them honourably and strengthened their [forces]”.Footnote ⁸¹ This active involvement of the Aqquyunlu in Jānibak's Anatolian network, however, came to a premature end when the Aqquyunlu were summoned north by Shāh Rukh and when ʿUthmān Beg died in the course of that campaign. In fact, it was only after the sudden disruption of this Aqquyunlu association that Jānibak entered into more active contact with the Dulgadirids.

Nāṣir al-Dīn Meḥmed and ʿUthmān Beg Qarā Yuluk, as representatives of the third type of members in Jānibak's network, display indeed a very particular sort of interaction with Jānibak's leadership, which is very different from that of the other two types. Unlike Jānibak's entourage of Türkmen followers and of small-time local chiefs, these regional leaders —including also the ruler of Sinop and the Ottoman governor of Tokat — acted as patrons, or at best as partners, rather than as subordinates for Jānibak's leadership project. As a result of this, they limited themselves to a symbolically empowering recognition of Jānibak's status, followed only in second instance by active support, which was then however rather delegated to sons and other agents. In the Aqquyunlu case, the relationship was then cut short by other significant occurences. In the Dulgadirid case, the opposite happened and the relationship was maximised by a very different turn of events, which from Ramaḍān 839/March-April 1436 onwards resulted in all but the annihilation of the Dulgadirid enterprise itself.

Before that moment however, as Ibn Taghrī Birdī rightly remarked, “rulers and common people (mulūk wa-khalāʾiq) had been rallying to [Jānibak]”;Footnote ⁸² more precisely, between Shawwāl 838/May 1435 and Ramaḍān 839/March-April 1436 an intricate three-tiered web of vertical and horizontal ties was knitted around the newly emerging leadership of Jānibak, consisting of a dynamic following of Türkmen warriors and small-time South-East-Anatolian chiefs, and of regional partnerships with Türkmen dynastic rulers, with the Dulgadirid Nāṣir al-Dīn Meḥmed in particular. Whatever the historical dynamics of this ego network of Jānibak al-Ṣūfī beyond Dhū l-Ḥijja 839/July 1436, its combination of horizontal ties with regional Türkmen rulers and of vertical relations of power with Türkmen followers and local chiefs, spread out over an area that stretched from Sinop in the North to Gaziantep in the South, demonstrates first and foremost how this Egyptian amir of Circassian origins was all but an outsider to Türkmen Anatolian realities.Footnote ⁸³ Through this network, Jānibak's leadership and the ups and downs of its performance were deeply integrated into the Anatolian framework in which they were developing, and from this perspective, Jānibak's full agency on the Türkmen Anatolian stage stands out from his active brokerage at the centre of that ego network.

This full integration of Jānibak al-Ṣūfī and his leadership in their Türkmen Anatolian context emerges not just from the different types of Türkmen groups and actors that rallied around his cause. It also transpires from the type of actions that generated the transformation of these disparate groups and actors into one powerful cohort. In the best of Türkmen chieftainship traditions, between Shawwāl 838/May 1435 and Dhū l-Ḥijja 839/July 1436 in particular, ranks were closed behind Jānibak's leadership, and income and resources were generated for his followers and their leaders, by their joint marching against Anatolian towns and their joint raiding of villages and local communities. At first, with Aqquyunlu support, “raids” (ghārāt) were organised against the fortified town of Divriği, “the people of which were gravely affected and the surroundings of which were looted (ḍāyaqū ahlahā wa-nahabū ḍawāḥīhā)”.Footnote ⁸⁴ Then, they moved further south and laid siege to the town of Malatya (nazalū ʿalā Malaṭiya . . . wa-ḥaṣarū Malaṭiya), where their ranks were briefly strengthened by Dulgadirid warriors.Footnote ⁸⁵ Upon the settlement of subsequent Dulgadirid differences, Jānibak “fell upon Marash (nazala ʿalā Marʿash)”.Footnote ⁸⁶ Finally, in Dhū l-Ḥijja 839/July 1436, it was said that “Jānibak and his followers agreed to make a raid on (ijtamaʿa Jānī Bak wa-man maʿahu an yakbasūhu)” Gaziantep,Footnote ⁸⁷ after which they were defeated and Jānibak only barely managed to escape north. Throughout this first and second phase of Jānibak's Anatolian whereabouts, and to some extent also thereafter,Footnote ⁸⁸ the reality of his Türkmen network's activities as much as the representation thereof in Egyptian sources were clearly framed in violent actions and phrases that do not differ in any way from how any other local Türkmen chief was thought to perform and establish local authority: securing revenue for followers and supporters by raiding, looting, besieging and using violence. Jānibak al-Ṣūfī himself therefore was acting just as such a Türkmen chief, generating his authority from a full-scope local agency that emanated from and resulted in a complex ego-network of Türkmen partners and supporters, who closed ranks behind him, but only as long as his leadership appeared to guarantee them benefit.

5. Jānibak as a power broker in Mamluk Anatolia

As suggested above, this Türkmen context of supporters, partners and local violence, however, was not all that made for Jānibak's case. In many ways, the above mentioned Egyptian source representations of Jānibak's Anatolian activities could well just be interpreted as representing a framing in negative terms of the renewed public activities of someone whom dominant Cairene circles had been made to think of as Barsbāy's dormant nemesis ever since this sultan's accession in 825/1422. In this respect, the contemporary court historian al-ʿAynī's representation of these events’ reception in Cairo is tellingly clear, when he referred to “messages that had come about the attack of the enemy and the destruction of the lands (al-akhbār allatī taʾtī min hujūm al-ʿadūw wa-fasād al-bilād)”.Footnote ⁸⁹ The framing of these Anatolian activities with labels of social disruption and violence reveal how Anatolia was indeed perceived in Cairo as a region beyond the sultan's direct control but within his sphere of moral responsibility, a distant frontier zone that at the very same time was in constant need of the preservation and restoration of social order by the sultan's agents. More importantly, the framing of Jānibak's activities from this perspective of the challenge to a Cairo-guarded regional social order also reveals how — at least in the minds and thoughts of Cairo's political and cultural elites — Jānibak operated not just in a context of Türkmen friends, followers and forays, but also on a fringe of the Mamluk sultanate. From this perspective of a specifically Mamluk liminal space and frontier zone, it was Jānibak's Mamluk identity and especially the memory thereof, rather than his local chieftainship, that were considered meaningful and that added to his actions a layer of subversive meanings, in direct opposition to his former opponent for the sultanate, al-Ashraf Barsbāy.

This Mamluk royal-centre-versus-subversive-periphery understanding of Jānibak's Anatolian activities was not however, just a matter of Egyptian historiographic rhetoric and political discourse. Underlying facts suggest that this operating of Jānibak in a specifically Mamluk frontier zone reflects another dimension of the historical reality of Jānibak's story, which ran parallel to the Türkmen context and which was equally crucial for its proper understanding. As in the Türkmen case, this reality of the Mamluk frontier reveals itself best via the further analysis of Jānibak's brokerage of an ego network, with an additional Mamluk type of members and an additional Mamluk layer of meanings equally meriting full attention.

First, there is the unmistakable and simple fact of the north-south axis along which Jānibak's actions between Shawwāl 838/May 1435 and Ramaḍān 839/March-April 1436 proceeded, from Sinop by the Black Sea, over Tokat in Central Anatolia, and Divriği and Malatya to the southeast, to eventually Gaziantep some 100 kilometres north of Aleppo (see Figure 1: map). Whether real or not, an intention to target the sultan's authority in Syria was certainly suggested by this physical course of the Anatolian whereabouts of Jānibak's network, and in Aleppo and Cairo subsequent reports surely triggered their being perceived as such.Footnote ⁹⁰ This perception of a direct threat to the Cairo court's regional interests was certainly also the result of the fact that the different Anatolian towns and regions that appeared (or were perceived) as the specific targets of Jānibak's raids and sieges in this period — Divriği, Malatya, Gaziantep — all happened to be directly controlled by Barsbāy's agents, whereas lands under local or regional Türkmen control all seemed to be left alone. The growing threat posed by Jānibak and his troops to Barsbāy's authority and suzerainty in this frontier zone in general, and to his direct control over and strategic interests in the wedge of towns, routes, basins, plateaux and mountain ridges between Dulgadirid and Aqquyunlu lands — the “Mesopotamian Marches (thughūr al-Jazīra)”Footnote ⁹¹ — in particular, was thus sourly felt by the sultan's local agents. To them, and to their overlords in Cairo, it may well have looked as though Jānibak was pursuing a full takeover of Cairo's control over this strategic Anatolian zone, before moving further south.

Map 1. Main towns, cities, routes and geographical markers in 9^th-/15^th-century southeastern Anatolia and Syria (important towns in Jānibak's story are marked by cartouches, blue cartouches indicating that these were towns under direct Egyptian control)

Figure 1: Jānibak's Türkmen Anatolian network

This Mamluk dimension to Jānibak's actions was certainly not just a matter of geographical strategies, centre-periphery perceptions, or personal ambitions. It was also deeply connected to the composition of Jānibak's ego network, which in fact included apart from its Türkmen participants a remarkable Syro-Egyptian element that suggested that at least by Ramaḍān 839/March-April 1436, Jānibak's Anatolian campaign had successfully merged with, or even transformed into, a full-fledged anti-Barsbāy rebellion.

In Ibn Taghrī Birdī's description of key members of Jānibak's ego network, quoted above, one element has so far indeed not yet been discussed. “The amīr Qurmush al-Aʿwar, one of the former muqaddams ʾalf in Egypt [. . .] showed up with him”, Ibn Taghī Birdī explained, “and with all who had joined him — including the amir Qurmush — he travelled from Tokat”.Footnote ⁹² In this list — as well as in the reality of the network taking shape around Jānibak— the amir Qurmush al-Aʿwar was awarded a central role as a key protagonist, who operated as a sort of right hand to Jānibak.Footnote ⁹³ This amir Qurmush was claimed to having been very close to Jānibak — in his biographical notice, Ibn Taghrī Birdī illustrates this bond by making him say that “I have carried Jānibak al-Ṣūfī on my shoulders in the Circassian homelands and I have raised him as though he was a son”;Footnote ⁹⁴ Qurmush had joined Jānibak at Tokat in Shawwāl 838/May 1435 after many years of Syro-Egyptian experiences that paralleled those of his would-be son: a senior amir in Egypt, the one-eyed (al-aʿwar) Qurmush sided with Jānibak in 824/1421 and after Barsbāy's victory he was removed to a position in Damascus, where, in 826/1423, he joined another unsuccessful revolt against Barsbāy's leadership and then disappeared from the sultan's radar — “it seems that he had been hiding in these regions”— for more than ten years, until he appeared again in Tokat to join his lost foster-son Jānibak.Footnote ⁹⁵

Qurmush, however, was not the only Syro-Egyptian veteran to join Jānibak. There actually was a fourth type of members in Jānibak's network, consisting of another group that accepted his leadership in ways similar to how his Anatolian Türkmen accepted it. These members, however, all shared a very different, non-Anatolian background that was directly related to Jānibak's Egyptian origins. Reporting about the fatal battle near Gaziantep in Dhū l-Ḥijja 839/July 1436, the contemporary chronicler al-ʿAynī (as his name indicates originally from Gaziantep!) consistently refers to a much wider Syro-Egyptian involvement:

This passage suggests not only that Qurmush was perceived in Cairo as a leading figure in Jānibak's entourage, but also that there were good reasons to believe that there were many dozens like him in that entourage: al-ʿAynī's use of the common identifier turk, in clear contradistinction to Jānibak's Türkmen followers, suggests that these indeed were renegades from the political and military elites of the Dawlat al-Atrāk, the so-called Mamluk polity.Footnote ⁹⁷ Most prominent among these turks in Jānibak's camp — and most suggestive for their renegade Syro-Egyptian originsFootnote ⁹⁸ — was the ‘deserter’ from the milieu of amirs in Aleppo, Kumushbughā, identified by Ibn Taghrī Birdī as “the amir Kumushbughā, known as Amīr ʿAshara, one of the amirs of Aleppo, who had gone from Aleppo and rallied to Jānibak al-Ṣūfī long time before this date (Dhū l-Ḥijja 839/July 1436)”.Footnote ⁹⁹ Having been — just as Jānibak and Qurmush — a mamlūk apprentice of sultan al-Ẓāhir Barqūq, this Kumushbughā al-Ẓahirī was said to have had a low ranking amir's default career in the anonymous mass of “military ranks and governorates (imriyyāt wa-wilāyāt)” before going over to Jānibak's side.Footnote ¹⁰⁰ Apart from Qurmush and Kumushbugha, none of the other turks in Jānibak's entourage are further identified in any of the available sources, suggesting that they all were of lower rank and status than Kumushbughā. What is clear, however, is that they were by Jānibak's side when he attacked the sultan's forces near Gaziantep in numbers sufficiently large to be noticed and reported, and that the majority of them shared Kumushbughā's renegade status, for they included, as Ibn Taghrī Birdī remarked in the biography of Jānibak, “a group from the amirs of Aleppo and elsewhere [that] had fled to him (wa-kāna qad haraba ilayhi jumāʿa min umarāʾ Ḥalab wa-ghayrihā)” just before the Gaziantep confrontation.Footnote ¹⁰¹

Whatever their origins and reasons for joining Jānibak, what all these Syro-Egyptian members of Jānibak's ego network clearly had in common was their antagonistic attitude towards Barsbāy's authority.Footnote ¹⁰² Apparently, Barsbāy's state was perceived as quite inhospitable by all of them, and from the moment Jānibak's project presented itself as a viable alternative that gradually came within reach due to its successful southward progress through ‘the Mesopotamian marches’, these outcasts from Barsbāy's authority switched sides —“they all agreed to fight against al-Malik al-Ashraf”, Ibn Taghrī Birdī remarkedFootnote ¹⁰³ —, from Qurmush who went to Tokat already in Shawwāl 838/May 1435, over to Kumushbughā who joined some time later, and to other amirs who came from northern Syria when Jānibak was camping near Gaziantep, in Ramaḍān 839/March-April 1436. In fact, in the original letter writing campaign from Ottoman Tokat, also other members of Barsbāy's political and military elites had been targeted, such as the afore-mentioned nāʾib of Daranda and even the nāʾib of Aleppo, the amir Qurqumās al-Shaʿbānī.Footnote ¹⁰⁴ These attempts allegedly came to naught, with the interception of these messages and the execution of Jānibak's envoys.Footnote ¹⁰⁵ But there are also source reports that suggest that the anti-Barsbāy element in Jānibak's network may have been — or was at least perceived as having been — larger and more threatening than what materialised near Gaziantep in Dhū l-Ḥijja 839/July1436.Footnote ¹⁰⁶

Figure 2: Jānibak's Mamluk Anatolian network

Therefore, the combination of an apparent spatial strategy developing in this Mamluk frontier zone, with a coalition of old and new outcasts from Barsbāy's polity gaining strength on the fringe of his authority, ensured that Jānibak's activities were understood in Cairo in the clearest of anti-Barsbāy terms. This perceived threat would certainly explain why a reaction was not left — as was more usual for local disturbances in this Mamluk frontier zone — for local agents of the sultan (the governor of Aleppo in particular) to solve. Between Jumādā II and Shaʿbān 839/January and March 1436, substantial Egyptian forces, involving seven of Barsbāy's most trusted amirs with their personal troops, “1,000 royal mamluks, and 1,000 soldiers from the ḥalqa”, were equipped in Cairo and sent up north.Footnote ¹⁰⁷ Between late Ramaḍān and Dhū l-Ḥijja 839/April and July 1436, this full-scale deployment of Barsbāy's force in and near the northern Mamluk frontier zone indeed was successful in restoring the sultan's social order in the area, by defeating and breaking up Jānibak's network, laying waste to Dulgadirid Elbistan, and then pursuing Jānibak and his Dulgadirid partner Nāṣir al-Dīn Meḥmed up to Sivas in central Anatolia. Most revealing for this anti-Barsbāy dimension of Jānibak's ego-network — at this Gaziantep confrontation in particular, and within the reality of the Mamluk frontier zone as much as in its perception at Barsbāy's court — is the subsequent punishment of those from Jānibak's Syro-Egyptian followers who were captured near Gaziantep. Once brought to Aleppo, they were all put to death, but the remains of Qurmush and Kumushbugha were subsequently given a treatment that is only reserved for — as al-ʿAynī indeed already suggested above — ‘rebels’ and ‘deserters’ who had been perceived as a true and realistic threat for Barsbāy's royal authority, and whose defeat — as all source reports agree — had to be advertised in the clearest and most public of terms, in the periphery as much as in the centre of the sultan's power. Their heads were cut off, sent to Cairo, paraded in the city's streets, and then publicly disposed of in the most humiliating of ways.Footnote ¹⁰⁸ This Mamluk dimension and the shared agency in it for Jānibak and his Syro-Egyptian supporters is surely most dramatically summarised in Ibn Taghri Birdī's report of this Gaziantep confrontation and its aftermath:

6. Jānibak as a pawn in frontier Anatolia

Jānibak's activities however, were not just fully integrated into their Türkmen contexts and meaningful, even acute, within a Mamluk context. The staging of Jānibak's story also needs to be considered from a third, interlocking perspective: that of a multi-directional Anatolian frontier zone, where traditional Egyptian suzerainty was constantly challenged from, and undermined by, various and competing regional and trans-regional forces. Surely southern and eastern Anatolia were fully integrated as a dynamic frontier zone into the performance of Barsbāy's sovereignty, in both discursive and coercive ways, but Jānibak's story also demonstrates how different political centres in Anatolia, and beyond, simultaneously engaged with or attempted to draw closer to this frontier zone, in this particular case by using Jānibak's position to challenge Barsbāy's authority, on a regional as well as on a trans-regional scale.

In the first instance, the different regional partners in Jānibak's Türkmen ego network — both Aqquyunlu and Dulgadirid — did not just have mere local reasons to act. It was rather Jānibak's Egyptian background and his very concrete subversive meaning within the context of Barsbāy's authority that made him useful, for allying with him equalled the public performance of a powerful act against Barsbāy's suzerainty. Regional rulers such as Nāṣir al-Dīn Meḥmed and surely also Qarā Yülük ʿUthmān Beg were very much aware of that dimension, as was Barsbāy. Jānibak certainly also tried to put this symbolic dimension of his role in Anatolia to best use, but this proved, more often than not, beyond his control.

Egyptian sources provide most detailed information about how Nāṣir al-Dīn Meḥmed tried to instrumentalise Jānibak in his strained relation with Barsbāy. Some time before the arrival of Jānibak on the scene, Barsbāy's agents had intervened in a conflict within the Dulgadirid family for control over the town of Marash, and they had captured one of Nāṣir al-Dīn Meḥmed's sons and taken him to Cairo as a prisoner of the sultan; soon thereafter, another of Mehmed's sons, Sulaymān, lost the town of Kayseri to the Cairo-backed Karamanids.Footnote ¹¹⁰ Cairo-Dulgadirid antagonism was thus on the rise, and with one of his sons in an Egyptian prison, Nāṣir al-Dīn Meḥmed had little room for manoeuvre. As a sort of very last resort, he sent his wife on a diplomatic mission to Cairo to try and obtain the release of his son and the return of Kayseri. When Jānibak's envoys presented their master's case to Nāṣir al-Dīn in Elbistan however, he saw an opportunity to create some more leverage in his negotiations with Barsbāy, and he therefore schemed with Sulaymān to capture Jānibak and to try and use him to strike a better deal with the sultan.Footnote ¹¹¹ “[Sulaymān] caught [Jānibak near Malatya]”, the contemporary scholar Ibn Ḥajar summarised the continuation of these events,

Once it became clear to Nāṣir al-Dīn that his wife and son had safely returned from Cairo — his son even in the capacity of Cairo-appointed governor of Marash — and that Barsbāy's support for Karamanid control over Kayseri was to remain unchanged, he almost immediately changed policy. Jānibak al-Ṣūfī was released and restored to his local leadership position, and bonds between them were — as mentioned above — strengthened through marriage. Nāṣir al-Dīn Meḥmed thus openly joined ranks with an Egyptian refugee, whom Barsbāy's court had continued to consider a rebel and an enemy ever since 824/1422. His anti-Barsbāy intentions were publicly confirmed when he sent back empty-handed the envoy Barsbāy had sent to Elbistan to take Jānibak to Cairo.

It was against this background of crisscrossing opportunities offered by the Anatolian frontier zone that Nāṣir al-Dīn became an active partner in Jānibak's ego network, participating with his sons in this local campaign that was gradually transforming into a full-fledged anti-Barsbāy rebellion. As much as this may have suited the interests of Jānibak himself and those of his Syro-Egyptian partners, this movement certainly also agreed with the intentions of the Dulgadirid leadership to wrest themselves from Barsbāy's frustrating suzerainty, “to abandon treating the sultan with flattery and to start thinking of the change of fortune (wa-taraka mudārāt al-sulṭān wa-ashghala fikr al-dawla)”, al-Maqrīzī explained, “because the rebellion of Jānibak had brought about something [that looked] wiser and more opportune [to pursue] (li-annahu qad jāʾa min khurūj Jānibak mā huwa adhā wa amarr)”.Footnote ¹¹⁴

Despite the paucity of further information, it is very likely that the same was actually true for other regional partners in Jānibak's Anatolian ego network, and that the Türkmen ruler of Sinop, the Ottoman governor of Tokat, and the Aqquyunlu leader ʿUthmān Beg Qarā Yülük all decided to invest one way or another in the anti-Barsbāy symbolism that Jānibak happened to bring into their Anatolian midst. This is at least suggested from the only extant information on Jānibak's dealings with one of these latter regional leaderships, in particular at the very end of his Anatolian adventures. As detailed above, by Rabīʿ II 841/October 1437 the Aqquyunlu Meḥmed b. ʿUthmān and his brother Maḥmūd succumbed to the pressures of Barsbāy's governor of Aleppo and exchanged Jānibak — or rather his Anatolian frontier value as a challenge to Barsbāy's authority — for the neat sum of 5,000 dinar.Footnote ¹¹⁵ However, their brother Ḥamza was reported by al-Maqrīzī to have had very different plans.

Just as Nāṣir al-Dīn Meḥmed two years before, Ḥamza b. Qarā Yülük (d. 848/1444), ruler of Mardin and Erzincan and the ascending successor to Aqquyunlu general leadership, saw an opportunity in Jānibak's sudden appearance in Aqquyunlu lands to boost his campaign to re-negotiate the nature of Barsbāy's suzerainty. This did not materialise, however, due to his brothers’ more pragmatic concerns for resources, nor did Ḥamza's wider anti-Barsbāy campaign bear any real fruits for him. It rather resulted in a last deployment of Barsbāy's military apparatus in the Anatolian frontier zone between Shaʿbān 841 and Muḥarram 842/February and July 1438, in intrusions deep into Aqquyunlu lands up to Erzincan, and in the ritual submission of several Anatolian local and regional leaders — including Ḥamza himself — to Barsbāy's suzerainty.Footnote ¹¹⁷

Barsbāy's troubled relationship with the Anatolian regional leaderships of the Dulgadirids and the Aqquyunlu and the way that Jānibak's anti-Barsbāy meanings were — or were at least aspired to be — operationalised within that relationship thus ran very parallel courses, even in their outcomes of another reconfirmation of Barsbāy's suzerainty in this Anatolian frontier zone, by his successful deployment of violence and symbolic communication. As the Aqquyunlu case suggests, however, this build up of tension in the Anatolian frontier zone did not just revolve around Jānibak al-Ṣūfī's case. Transcending this by far was the continuous negotiation over the physical and symbolic frontiers of the sultan's authority in Anatolia between these regional leaders and Barsbāy's court, and in this negotiation Jānibak al-Ṣūfī was just one of those instruments that could offer the former considerable leverage.

This symbolic value of Jānibak al-Ṣūfī in the Anatolian frontier zone as Barsbāy's accessible nemesis was not just picked up by regional leaders such as Nāṣir al-Dīn Meḥmed and Ḥamza. It was also operationalised by trans-regional leaders, especially when they appeared to pursue an extension of their authority in that frontier zone and attempted to challenge Barsbāy's traditional suzerainty over it. As mentioned before, this concerned the actions of the Ottoman ruler Murād Beg and, in particular, of the Timurid leader Shāh Rukh.

In the Ottoman case, any allusions to Murād Beg's appreciation of Jānibak's frontier value remain very indirect and speculative. As seen above, it was mainly the Ottoman governor in the city of Tokat whose support and promotional campaign enabled the re-appearance of Jānibak as an Anatolian chief,Footnote ¹¹⁸ and after his defeat at Gaziantep in Dhū l-Ḥijja 839/July 1436, it was in Ottoman lands that Jānibak found shelter against his pursuers as well as renewed support to return to the Anatolian frontier zone.Footnote ¹¹⁹ As a result, Egyptian historians’ representations of the reception at Barsbāy's court of this news indicate that Ottoman actions involving Jānibak were considered as aimed against the heart of Barsbāy's authority, generating concrete plans to make the sultan himself march up north and face the enemy.Footnote ¹²⁰ However, whereas in the Tokat case it is highly likely that Murād Beg was involved one way or another in his governor's engagement with Jānibak, in the other case there is nothing in the sources to suggest the same. Stripped of most of his Anatolian and Syro-Egyptian supporters, Jānibak simply had no option but to follow his Dulgadirid partners as long as they endured him, and the fact that Murād Beg did not operationalise his symbolic value during the joint Ottoman-Dulgadirid military operations against the Karamanid leader Ibrāhīm Beg only suggests that this value was not really picked up in the Ottoman camp, or at least that it was not considered worth their trouble. This lack of Ottoman interest soon also became clear in Cairo, where any plans for the sultan's personal engagement in military operations in Ramaḍān 840/March 1437 were similarly and quickly forsaken. The peace treaty concluded shortly afterwards between Murād Beg and Ibrāhīm Beg and the subsequent Ottoman retreat northwards confirmed that, whatever the original Ottoman interests in Jānibak, this had not been a campaign against Barsbāy.Footnote ¹²¹

Things were somewhat different with Shāh Rukh, who used the opportunity of his Azerbaijan campaign to try and extend his authority in truly Timurid universal rule style over the entire eastern Mediterranean. This plan became first known in Cairo when, “in Ṣafar 839/September 1435, the news arrived with the sultan”, Ibn Taghrī Birdī reported, “that Shāh Rukh, the son of Tīmūrlank, had sent robes of honour to sultan Murād Beg b. ʿUthmān, the ruler of Rūm, to the amir Ṣārim al-Dīn Ibrāhīm b. Qaramān [. . .], to Qarā Yülük and his sons, and to Nāṣir al-Dīn Beg b. Dulghādir, making them his governors in their realms”.Footnote ¹²² This was perceived as a direct challenge to the sultanate's suzerainty in the Anatolian frontier zone, a perception that proved true when messages arrived that Shāh Rukh had taken Tabriz. The challenge turned out to be even greater when in early Rajab 839/January 1436, an envoy from Shāh Rukh is reported to have appeared in Cairo with another robe of honour, this time for Barsbāy himself, to become the Timurid ruler's “governor of Egypt”. Enraged and humiliated by this bold diplomatic gesture, Barsbāy is said to have sought to set up an alliance with Murād Beg against Shāh Rukh, but this was cut short by Shāh Rukh's sudden return eastwards in Shawwāl 839/April 1436, and the subsequent disappearance of his threat to Barsbāy's authority and suzerainty.Footnote ¹²³ It was in fact in the midst of this build up of trans-regional tension with Shāh Rukh that Jānibak re-appeared in the Anatolian frontier zone, and that, in Rabīʿ al-Ākhir 839/October 1435, it became known to Barsbāy that Jānibak too had been the object of correspondence from the Timurid ruler. At that time, a most disconcerting intercepted letter was forwarded from Aleppo to Cairo, in which Shāh Rukh “incited Jānibak al-Ṣūfī to take the Syrian lands, and [he promised him] that he [ = Shāh Rukh] would send to him his son Aḥmad Jūkī and [his leading military commander] Bābā Ḥājjī to assist him fighting the sultan of Egypt”.Footnote ¹²⁴ If this letter, and its representation in Egyptian sources, were at all genuine, it seems that Shāh Rukh hoped for much more than merely operationalising Jānibak's Mamluk values and meanings in the Anatolian frontier zone. In this case, the Timurid ruler intended to also operationalise Jānibak's ego-network of Türkmen and Syro-Egyptian partners and followers, and to generate or at least to enhance its gradual transformation into a full-fledged anti-Barsbāy revolution, aiming for Jānibak's full take over of Barsbāy's political project in Shāh Rukh's name, first in Syria, and then, also with Timurid assistance, in Egypt. Among Barsbāy's entourage at least, the coinciding of Shāh Rukh's emergence from the east with Jānibak al-Ṣūfī's in the Anatolian frontier zone was perceived from such a hugely threatening perspective.

For Jānibak al-Ṣūfī himself, this perspective of the Anatolian frontier zone shows how he also got caught up in events that transcended him by far, and how regional and especially trans-regional rulers also attempted and even succeeded in patronising him and his ego-network in their negotiations with and challenges to Mamluk suzerainty. In many cases, this certainly enhanced Jānibak's cause and enabled him to maximise his ego-network, so that in the course of its existence it consisted not just of a variety of partners and followers, but also of a succession of powerful patrons. At the same time, the nature of these patrons’ interests never really — or at best only temporarily — coincided with Jānibak's, so that in the end he and his supporters were more than once victimised rather than supported by their policies. In many ways, even the meanings attached by Barsbāy's entourage to Jānibak's Anatolian whereabouts and the resulting extremely forceful Syro-Egyptian military reactions, especially in Ramaḍān-Dhū l-Ḥijja 839/March-July 1436, can also be considered an unwelcome function of this type of patronage, as they were certainly partly also brought about by trans-regional interests such as those that were imposed upon Jānibak's campaigns.

Figure 3: Jānibak's frontier network

In the Anatolian frontier zone, where interests and authorities of regional and trans-regional rulers were in a continuous flux, their negotiation, challenge and establishment took on a variety of forms and applied a plethora of strategies and tools, often difficult to evade and even more difficult to control for local leaders such as Jānibak. The only constant factor in this process at the time of Jānibak's emergence, however, remained the imperial reach of Barsbāy's suzerainty, established through symbolic means when possible, and through violent force when necessary. Jānibak's mixed adventures with Ottoman, Aqquyunlu, Dulgadirid, and Timurid patrons, and with Syro-Egyptian opponents, are certainly also a token of that.