The Era of Inter-Mongol Warfare II

The Toluid Civil War was a war of succession fought between Kublai Khan and his younger brother, Ariq Böke, from 1260 to 1264. Möngke Khan died in 1259 with no declared successor, precipitating infighting between members of the Tolui family line for the title of Great Khan that escalated to a civil war. The Toluid Civil War, and the wars that followed it (such as the Berke–Hulagu war and the Kaidu–Kublai war), weakened the authority of the Great Khan over the Mongol Empire and split the empire into autonomous khanates.

Compounding the problems of warfare between different Mongol khanates, and of the curtailment of resources, were contests for the throne of one ulus or another, as a khan or his henchmen tried to confine the succession within the segmentary lineage of his immediate family while a rival party upheld the principle of seniority. The sources highlight discord of this sort within the Ilkhanate in 1282–4 (between uncle and nephew), within the ulus of Orda at the turn of the thirteenth century, over the succession to Qaidu in 1303 (when his throne was claimed by a grandson or great-grandson of Güyüg), and within Chaghadai’s ulus in c. 707–8/c. 1308–9 (when Naliqo’a, a descendant of Büri, displaced Du’a’s line). In 1284 and 1295, Ilkhanid Iran, commonly viewed today as the most stable of the three westernmost Mongol states, was the scene of an armed struggle for the throne.

Admittedly, the Chinggisids had not lost sight of the goal of world-conquest even after 1260. In the very context of Berke’s diplomatic exchanges with Sultan Baybars, a Mamlūk author makes him deprecate the fact that Mongols were now falling to the swords of fellow Mongols and add – somewhat ingenuously – ‘had we remained united, we could have conquered the world’. Indeed, Anne Broadbridge adduces evidence that Berke viewed his ally Baybars as a subject ruler. The Ilkhans, smarting from the reverse at ‛Ayn Jālūt and nursing the conviction that their Mamlūk antagonists were nothing more than the Mongols’ runaway Qipchaq slaves, appear no less committed to expansion. In a letter to Baybars in 667/1268 Abagha employed the traditional expressions il/el (‘peace’) for the Sultan’s submission and yaghi (‘rebellion’) in the event that Baybars continued to reject Mongol imperial authority. Abagha also made the dubious statement that a quriltai of Chinggisid princes (‘all of us older and younger brothers’) had recognized the qaghan’s authority. This seemingly formed part of Ilkhanid diplomatic stock-in-trade. Abagha’s envoys at Lyons in 1274 spoke of a recent peace (otherwise unrecorded) between the Ilkhan and his Mongol neighbours. And his successor, Tegüder Aḥmad, in his second letter requiring the submission of the Mamlūk Qalāwūn in 682/1283, reinforced the message by alleging that the Mongol rulers were once more at peace. Ghazan would make the same claim in a proclamation to the Mamlūk military in Syria in 699/1299–1300. On these latter two occasions, a spurious Mongol consensus was being used as a covert threat that the Ilkhans were now free to turn against their external enemies.

Peace proved, at best, ephemeral. In 699/1299–1300 an observer in Mamlūk Egypt commented that all the monarchs in the east and the west alike were at war; and the Mongol dominions were no exception. In a letter of 1305 to the French King Philip IV, Öljeitü hailed the reconciliation of 1304 as laying to rest quarrels that had raged for forty-five years. He concluded with veiled threats against those who persisted in their hostility towards either the Mongols or the Franks, and the idea of Mongol world-rule was understandably muted. His fellow Chinggisids, when making peace in the previous year, and Özbeg, the khan of the Golden Horde, in relation to Western European monarchs during the 1330s, had fewer inhibitions about using the rhetoric of world-domination. But during the decades following Möngke’s death, that rhetoric more often than not had evoked an ideal to which Mongol princes could only pay lip-service.

Jochid ambitions and Ilkhanid territory

The khans of Jochi’s line never relinquished their claims on the rich grazing-grounds in Azerbaijan and Arrān, which had been appropriated by Hülegü. Not that the Jochids were consistently aggressive. Rashīd al-Dīn assures us that after Mengü Temür made peace with Abagha (doubtless in 669/1270, when he sent envoys to congratulate the Ilkhan on his victory over the Chaghadayid khan Baraq), there were no hostilities until 687/1288; he is mistaken, since we know from other sources of a large-scale attack in 678/1279–80. Again, the same author tells us that since the invasion of 687/1288 peace had reigned down to the time that he wrote, though he ascribes it to the Jochids’ weakness and speaks of their friendship as merely an outward display; and elsewhere, in any case, he reports yet another attack in 689/1290. Toqto’a sent envoys to make a truce with Gaikhatu in 693/1294, but since they also made ‘all sorts of requests’ it is possible that the purpose was merely to renew Jochid demands for the return of territory south of the Caucasus. On the other hand, Waṣṣāf evidently regarded Gaikhatu’s reign as something of a watershed, since he associates it with a resumption of diplomatic contacts between the two powers. The lack of incursions south of the Caucasus and the absence of any recorded embassies from Toqto’a to the Mamlūk Sultan prior to 1304 are perhaps due to the conflict with Noghai and hence a need to avoid provoking the Ilkhan. During the latter stages of the struggle, both sides courted Ghazan, but – so Rashīd al-Dīn smugly informs us – the Ilkhan nevertheless resolutely refrained from intervening for his own advantage. A Mamlūk author claims that Toqto’a had made peace with Ghazan, presumably to cover his flank.

Charles Halperin argued vigorously that conquests beyond the Caucasus were the primary concern of the khans of the Golden Horde, far outstripping, that is, the attractions of the economically less desirable Rus´ principalities. If anything, Jochid diplomacy and military activity would grow more menacing in the fourteenth century. Toqto’a reiterated the habitual Jochid demands in 702/1302–3, and they were revived on Özbeg’s accession in 712/1312.85 The latter would personally head two major incursions into Azerbaijan, in 719/1319, during Abū Sa‛īd’s minority, and in 736/1336, following that monarch’s death; in 722/1322 we find him allied with the Chaghadayid khan Köpek, who was then attacking the Ilkhanate. In 758/1357, some years after the demise of the Ilkhanate, Özbeg’s son Janibeg invaded Azerbaijan, and as late as the 1380s the khan Toqtamish was still staking the time-honoured Jochid claim to this region.

In an incomplete list of those borderlands that were wasted and whose populations were killed or fled elsewhere, Rashīd al-Dīn mentions the region between Darband and Shīrwān. We are told very little, in fact, about the damage that Jochid-Ilkhanid warfare inflicted on this territory, though it is worth noting that, during the first few decades of the Ilkhanate, command of the incursions from Darband tended to be entrusted to Jochid princes whose fathers’ deaths at the hands of Hülegü’s forces gave them a particularly strong personal motive for vengeance: Noghai, the son of Tutar, in 660–1/1262 and 663/1265, and Tama Toqta, a son of Balagha (Balaqan), in 687/1288. On the other hand, Abagha’s construction of a barrier (sibe, sübe) and deep ditch north of the Kur river in 663/1265 possibly limited the Jochid forces’ access to Azerbaijan and Arrān, further south, and reduced their capacity to inflict damage.

The borderlands between the Central Asian Mongols and the Yuan

For twenty years or so Qaidu and Du’a proved capable of stemming the advance of the Yuan. Almaligh itself, lost in 1271 to the army commanded by the Qaghan’s son Nomoghan, was recovered a few years later. The Uighur iduq-qut, who owed allegiance to Qubilai, was forced to abandon Beshbaligh and take up residence first at Qaraqocho, at Qāmul (Qomul; Hami) and then at Yongchang in Gansu; he was only briefly restored at Qaraqocho in 1313. Qaidu and the Chaghadayids appear to have appointed their own clients as iduq-quts. In passages that seemingly relate to his overland journey to China (c. 1274), Marco Polo describes Kāshghar and Khotan as still subject to the qaghan (whereas Yārkand was already in Qaidu’s power). At any rate, Qubilai’s garrisons evacuated both towns in the late 1280s. Qaidu did not, it seems, go so far as to occupy them (although Waṣṣāf includes Kāshghar among his territories), but these regions were in Chaghadayid possession by the second decade of the fourteenth century. Qubilai had thus forfeited control over important Muslim settlements in Central Asia. The seemingly inaccessible Muslim kingdom of Badakhshān, too, which recognized the qaghan’s suzerainty, was being repeatedly harassed by Qaidu’s forces by the turn of the century. In 716/1316, when we find a contingent from Badakhshān collaborating with a Chaghadayid invasion of Khurāsān, it was evidently under Chaghadayid overlordship.

Uighūristān and other lands between Qaidu’s dominions and those of the qaghan are the second region that Rashīd al-Dīn singles out as subject to devastation in the inter-Mongol wars. The Muslim populations of these territories undoubtedly suffered in the course of repeated fighting. In 1266 Baraq plundered Khotan, then governed by one of Qubilai’s lieutenants, and Du’a maintained the pressure. The anonymous life of Mar Yahballāhā claims that Hoqu (a younger son of Güyüg) had slaughtered thousands of people in Khotan just before the Catholicos passed through the region in c. 1274 and that the caravan routes had been interrupted; Yahballāhā and Rabban Ṣawma found Kāshghar, which had been recently sacked by Qubilai’s enemies (doubtless Hoqu again), bereft of inhabitants. The Yuan government took measures in 1274 to provide comfort and assistance to Khotan, Yārkand and Kāshghar. Hoqu had been pushed into the camp of Qubilai’s enemies by an unprovoked attack on the part of Nomoghan’s colleague, the general Hantum Noyan.

The labile frontiers in eastern Iran

Rashīd al-Dīn’s list of devastated territories omits the long frontier between Iran and the Chaghadayids. Here there was no sibe to hamper invading forces, which may well be the reason why one of al-‛Umarī’s informants assured him that a single Chaghadayid trooper was worth a hundred from the Qipchaq steppe and that attacks from beyond the Caucasus provoked less alarm in the Ilkhanate than did those by the Chaghadayids in the east. As on their eastern frontier, so also did Qaidu and his allies make notable advances to the south, chiefly at the expense of the Ilkhans.

In Hülegü’s time, the Oxus had divided the Chaghadayid dominions from the Ilkhanate. Although Baraq’s invasion of Khurāsān in 668/1270 failed, the Ilkhan Abagha was dissuaded from sending a force to apprehend the recalcitrant malik of Herat in 674/1275–6, on the grounds that Khurāsān was still a wasteland as a consequence of Baraq’s attack and was in no state to support such a campaign. The Ilkhanate’s eastern territories were also open to attacks by the Negüderis or Qara’unas, based in present-day Afghanistan. The Persian sources testify to their turbulence; for Sayfī their depredations were proverbial. Abagha attempted to exert indirect authority over them through refugee Chaghadayids such as the former khan Mubārak Shāh. The policy failed: Mubārak Shāh proved a fickle subordinate, meeting his death in an attack on Kirmān in 674/1275–6; and when Böjei’s son ‛Abd-Allāh, who commanded the Negüderi bands in the Ghazna region a few years later, responded to an overture from the rebel Sultan Ḥajjāj of Kirmān, he was acting on behalf of Qaidu and the Chaghadayids. An audacious Negüderi raid on Fārs and Kirmān in the winter of 677/1278–9, doubtless provoked by this appeal, in turn prompted Abagha’s expedition to Khurāsān and Sīstān in the following year. Abagha drafted certain Qara’unas bands into Ilkhanid service and transported them to central and western Iran. But their confrères in eastern Iran and Afghanistan continued to pose a threat to the territories of the Ilkhan’s clients, raiding Kirmān again three years later; Waṣṣāf says that the people of Shīrāz lived in fear of their attacks every winter until the end of Arghun’s reign (690/1291). When one of Abū l-Fidā’s informants told him that Hurmuz was ruined as a result of Tatar incursions, he must have been referring to the Negüderis.

According to Waṣṣāf, Qaidu stationed his son Sarban south of the upper Oxus. No date is given, but already in c. 1290 we find Sarban based near Shabūrghān and Du’a’s noyan Yasa’ur in the regions of Balkh and Bādghīs. A bid by the Central Asian Mongols to control the Negüderis of the Ghazna region through the renegade Ilkhanid noyan Nawrūz failed when he turned against Qaidu and then in 694/1295 submitted to Ghazan. They succeeded, even so, in bringing the majority of the Negüderis within their orbit. In the mid-to-late 1290s Du’a summoned the Negüderi leader ‛Abd-Allāh, himself a Chaghadayid prince, and replaced him with his own eldest son Qutlugh Qocha. Rashīd al-Dīn speaks of Qutlugh Qocha in terms that suggest he virtually ruled the ulus jointly with his father. He enjoyed overall command of various princes at the head of forces totalling five tümens and was master of a considerable tract between the Oxus and the Arghandāb, including, if Waṣṣāf is to be believed, even Merv.

From their advance bases, the Central Asian Mongols launched frequent attacks on the Ilkhan’s territories and inflicted greater damage. Waṣṣāf hints that Qutlugh Qocha benefited from the defection of Ilkhanid troops, possibly attracted by the prospect of rich plunder from his raids on India. When the forces of Du’a and Qaidu, guided by the rebel noyan Nawrūz, entered the province in 690/1291, says Rashīd al-Dīn, the killing, pillage and destruction they perpetrated defied description. Although Nīshāpūr successfully resisted a siege, the villages were raided and many captives carried off; the shrine at Ṭūs (Mashhad) was also plundered. In 695/1295–6 Du’a and Sarban profited from Ghazan’s departure from Khurāsān to subject that province and Māzandarān to a campaign of devastation lasting for eight months. Malik Fakhr al-Dīn of Herat offered the depredations of Qutlugh Qocha’s forces as a reason for failing to send the Ilkhan the stipulated tribute. During Ghazan’s absence on campaign in Syria in 700/1301, they ravaged Kirmān and Fārs, penetrating as far west as Tustar (Shustar). The winter of 702/1302–3 witnessed an attack on Khurāsān by Sarban and his lieutenants, whose attempt to effect a junction with Qutlugh Qocha’s army, however, miscarried badly. Such incursions must have been responsible for slowing any recovery from the campaigns of Chinggis Khan. According to Ḥamd-Allāh Mustawfī, the town of Jurjān was still ruined and sparsely populated in his own day; and it was surely Chaghadayid attacks that Marco Polo had in mind when he observed that Balkh had been ravaged many times by ‘the Tartars and other peoples’ and that its fine mansions still lay in ruins. Significant recovery may have been delayed until the early fourteenth century, when the wazir ‛Alā’ al-Dīn Hindū Faryūmadī is said to have restored Khurāsān to its former splendour.

Although Qutlugh Qocha was mortally wounded during his return from an invasion of India in 699/1299–1300, his command remained in Chaghadayid hands: the Negüderis were ruled first by his lieutenant Taraghai until 705/1305 and then in turn by two other sons of Du’a, Esen Buqa and (after the latter’s accession as khan of Chaghadai’s ulus in 709/1309) It-qul. It-qul was apparently succeeded by Qutlugh Qocha’s son Dā’ūd Qocha, who was expelled by Öljeitü’s forces in 712/1312 at the instigation of some Negüderi chiefs. Whether the retaliatory expedition sent by Esen Buqa reinstated the prince, we never learn. But in 724/1324 Du’a’s youngest son, the future khan Tarmashirin, crossed the Oxus and seized the Ghazna-Kābul region, only to be defeated in 726/1326 by the Ilkhan Abū Sa‘īd’s forces under Ḥasan b. Choban. Ḥamd-Allāh Mustawfī paints a grim picture of the outrages committed in the city of Ghazna by these Ilkhanid troops, acting though they did on behalf of a Muslim monarch: they destroyed the tomb of Maḥmūd of Ghazna and made no distinction between the citizens and the military. According to Ibn Baṭṭūṭa, Ghazna was again in the hands of Tarmashirin, now himself khan and a Muslim, in the 1330s, but was mostly in ruins. It may be that the devastated condition of the territory extending for twenty days’ journey in the vicinity of Ghazna was in part the work of this Ilkhanid campaign, although al-‛Umarī attributes it to the conflict between the Chaghadayids and the Delhi Sultan.

Transoxiana, Turkestan and Khwārazm

Nor had the Chaghadayids’ other lands been immune from savage attack. Writing in (or at least of) the period prior to the death of the Qaghan Möngke, Juwaynī had contrasted the restoration of Transoxiana to a flourishing condition with the dismal fate of Khurāsān. This happy state of affairs did not last. Rashīd al-Dīn says that the territory of ‘Turkestan’ was devastated successively by Alughu, by his sons Chübei and Qaban, by Baraq (in all probability, Baraq’s sons are intended) and, most recently, by Bayan of the Blue Horde. In the course of the civil wars of 1260–4, as we have seen, Alughu’s troops slaughtered elements representing Berke’s interests in Samarqand and Bukhārā; among other victims was a son of the celebrated Shaykh Sayf al-Dīn Bākharzī. Then in 671/1272–3, on the advice of his chief minister, the Ṣāḥib-dīwān Shams al-Dīn Juwaynī, and in order to profit from the upheavals north of the Oxus as Qaidu tried to assert his control over the sons of Baraq and of Alughu, the Ilkhan Abagha’s forces invaded Transoxiana. They made several attacks on Kish and Nakhshab. When the people of Bukhārā rejected the option of accompanying them back to Khurāsān, the city was subjected to a slaughter, beginning on 7 Rajab/28 January 1273 and lasting for seven days; the number of those killed is given as 10,000, and the prisoners – boys and girls – carried off are put at 50,000. In reprisal for Mas‛ūd Beg’s disdainful attitude towards the Ṣāḥib-dīwān during his embassy to Iran some years before, the college he had built was burned down. Chübei and Qaban, in pursuit of the Ilkhan’s retreating troops, recovered half the captives, though whether they were returned to their homes we are not told. Only three years later, in 674/1275–6, Chübei and Qaban, possibly acting on Qubilai’s behalf, themselves subjected the region of Bukhārā and Samarqand to an orgy of killing and looting, with the result that it lay desolate for seven years.

According to Mīrzā Ḥaydar Dughlāt, the nomads of Mughalistān were known to their more sedentarized western neighbours as Jata/Chete – usually translated as ‘bandits’ and perhaps bearing the connotation of ‘freebooters’, like the word qazaq (whence ‘Kazakh’), which surfaces as the name of a new and powerful grouping around Ḥaydar’s own time. We first encounter the term Jata in Jamāl al-Qarshī’s account of a raid on Kāshghar by ‘the accursed Jatā’iyya’. This episode is undated but must have occurred at a time when the town was in theory under Qaidu’s protection; it may have been the above-mentioned attack by Hoqu. At any rate, the marauders killed a good many people, including members of the ‛ālim class, and drove off 5,000 more as slaves. These Mughal nomads were presumably the ‘infidels’ whose repeated attacks on the villages of Transoxiana, says Ḥamd-Allāh Mustawfī, obliged the local inhabitants to go about armed.