IL-KHANIDS i. Dynastic history

The first part of this entry will be a short survey of the reigns of the various Il-khans; treatment that is more detailed will be found in the individual entries for each ruler. The second part will review some of the salient characteristics and institutions of the state they ruled, as well as certain significant developments in it.

Dynastic history. The Mongols had been present in the Islamic world since the invasion of Čengiz Khan in 1219. Even after he withdrew to the steppe in 1223, a relatively small Mongol force remained in Transoxiana and Khorasan. This garrison, reinforced over the years by additional Mongol forces, gradually expanded Mongol control over all of Persia and beyond, even bringing the Saljuq kingdom of Rum under their rule after the victory of Köse Dagh in 1243. The Jazira and Iraq remained beyond the reach of the Mongols at this time, although their forces occasionally raided these countries as well as north Syria in the 1240s. Mongol control over this vast territory was far from complete, and the Nezāri Esmāʿilis (the so-called Assassins) maintained their independence in a series of castles in the mountains south of the Caspian Sea and in Kuhestān in eastern Iran. A rudimentary bureaucracy was also established, staffed by Central Asian Muslims, members of select Central Asian sedentary peoples with administrative traditions (most prominently Uighurs and Khitans), and some Persian bureaucratic families who had decided that it was in their interest, and perhaps the interest of the Muslims at large, to join the Mongol administration. The dismay of the Persians at the influx of the parvenus and their important role in the new administration is seen in the words of ʿAlā’-al-Din al-Jovayni (q.v.; see Jovayni, ed. Qazvini, I, pp. 4-5; tr. Boyle, I, pp. 7-8), scion of a well-established family of bureaucrats, whose father Bahāʾ-al-Din had begun serving the Mongols. His brother Šams-al-Din eventually became Hülegü’s chief minister (ṣāḥeb divān), a post which seems to have merged with the position of vizier at this time. ʿAlāʾ-al-Din was to become the governor of Baghdad and one of the most important historians of Mongol Iran.

This period of relatively loose control and somewhat small Mongol presence, albeit over a large area, ended with the arrival of Hülegü in Transoxiana in 1255. This prince had been dispatched by his brother Möngke in 1251, soon after the latter’s accession to the position of Great Khan or Qa’an (< qaghan), who also sent another brother, Qubilai, to China. Hülegü took with him an enormous army, supposedly two out of every ten Mongol soldiers, who were accompanied by families and herds. This, then, was not just a military campaign but also the mass migration of a large portion of the Mongol nation to Persia and the surrounding countries. According to Rašid- al-Din (q.v.), Hülegü was first to eliminate the Ismāʿilis and to subjugate the rebellious Lurs and Kurds. Subsequently he brought about the submission of the caliph and, finally, enacted the laws of Čengiz Khan in the territory from the Oxus river to the borders of Egypt. This is clear enough, but the author adds that the Qa’an publicly ordered his brother to return to Mongolia upon completing this mission, but secretly told him that he and his descendents were to remain in the country. Rašid-al-Din does not explicitly deal with the question of establishing a dynasty, but perhaps even this fuzzy information should be suspect, as this author was a clear apologist for the dynasty. This, and somewhat contradictory information in other sources such as the Mamluk author Ebn Fażl-Allāh al-ʿOmari (ed. Lech, pp. 2, 17), have led to suggestions that Hülegü may have been exceeding his orders by establishing a full-blown, practically independent dynasty (see Jackson, 1980, pp. 220-22, and the entry on Hülegü Khan for a fuller discussion) in the confusion and civil war that followed Möngke’s death in August 1259. Certainly, as ʿOmari states (ed. Lech, p. 2), the Il-khanid state, as we can now conveniently call it, suffered from a problem of legitimacy vis-à-vis the neighboring Mongol states of the Chaghatayids in Transoxiana and the Jochid Golden Horde to the north, and thus stressed its family connection with the Great Khan.

By 1256, Hülegü had all but eliminated the Ismāʿilis as an independent force in Persia (although individual forts remained independent for some time, even years), and had moved with the bulk of his army to Azerbaijan, which was to become the center of the Il-khanid state. After some desultory negotiations, Hülegü marched his army to Baghdad in early 1258, which surrendered after some spirited fighting by its army and populace. The caliph al-Mostaʿṣem was executed, and the Abbasid caliphate was virtually extinguished. Nevertheless, a scion of the dynasty eventually made his way to Syria, where the Mamluk sultan had him crowned as caliph (1261), thereby inaugurating a puppet caliphate that gave the sultanate a certain legitimacy, at least in its war against the Mongols. At the beginning of 1260, the Il-khans began a campaign into north Syria. Aleppo and its environs were quickly taken, and the general Kitbuqa was dispatched to the south with a division to occupy Damascus and reconnoiter in Palestine and trans-Jordan. In late winter, Hülegü withdrew from Syria to Azerbaijan, perhaps due to a lack of pastureland for the many horses of his large army, or in order to be prepared in case of a Jochid attack via the Caucasus, now that his patron, Möngke, had died. Taking advantage of the presence of only a relatively small Mongol force in Syria, the Mamluk Sultan Qoṭoz moved into the country from Egypt, defeating Kitbuqa on 3 September 1260 at ʿAyn Jālut in northern Palestine, thereby ending the short period of Mongol rule west of the Euphrates. Without a doubt, Hülegü, as well as his successors, desired to avenge this embarrassing and unexpected defeat, but other, more pressing matters usually prevented them from devoting their full attention to it. Hülegü, in any event, had to satisfy himself with raids along and across the frontier. From the Mamluk point of view the war took on clear religious overtones: they were the defenders of Islam against the infidel Mongols. The latter’s religious status did not prevent many Muslim Persian bureaucrats from serving them to the best of their abilities.

From about this time we can talk about an independent Il-khanid state, within the context of the breakup of the united Mongol empire. One symbol of this discord was the civil war in Mongolia and northern China between Hülegü’s two brothers, Qubilai and Ariq Böke. The former achieved victory and the title of Qa’an in 1264 and recognized the de facto independence of Hülegü in the west. The dissolution of the united empire is also symbolized by the outbreak of war between the two cousins, Hülegü and the Jochid Khan Berke, in the winter of 1261-62, although sustained fighting only began in the summer of 1262. Initially, the forces of the Golden Horde penetrated into Il-khanid territory, but these were beaten back; and Hülegü’s army, under the command of Abaqa (q.v., the future Il-khan), advanced across the Caucasus into the land of the enemy. Eventually the Il-khanid forces were routed, and an uneasy frontier was established along the Kur river. War between the two Mongol neighbors erupted upon Hülegü’s death in 1265 and continued in a desultory fashion along the river until Berke’s death in 1267. In general, intermittent war with the Golden Horde, or the threat of it, plagued the Il-khanids and was an important factor contributing to their inability to pursue the war against the Mamluk sultans, allies of the khans of the Golden Horde.

Abaqa’s reign (1265-82) was one of consolidation; there were no more conquests, but the state was more institutionalized and enjoyed relative internal stability. The continuing role of Šams-al-Din al-Jovayni as chief minister (ṣāḥeb-divān and vizier) should be mentioned. In the realm of foreign relations, perhaps the most important development was the war with the Chaghatayid Khan Baraq. This saw the latter’s invasion deep into Il-khanid territory in 1270, but ended in his defeat by the Il-khans at the battle of Herat in July of that year. After this, the Il-khanid-Chaghatayid front remained quiet along the Oxus, but this did not prevent Abaqa’s troops from raiding Bukhara in 1272-73. The front with the Golden Horde was also generally quiet, which may have led in the end to the revival of war with the Mamluks. There were at least half a dozen major raids into Mamluk territory during Abaqa’s reign, culminating in the large-scale invasion of Syria in 1281 led by the Il-khan’s brother Möngke Temür. This force was defeated by the Sultan Qalāwun (r. 1279-90) after a difficult battle, and half a generation passed before the Mongols again attempted a major offensive into Syria. The Mamluks themselves waged an active campaign of raiding across the frontier as well as against the Il-khanid ally of Lesser Armenia, reaching a peak with Baybars’ massive incursion in 1277, which included the defeat of the local Mongol force at Albustayn (Elbistān) in April of that year. In order to deal better with the Mamluks, Abaqa dispatched four embassies to the West calling for a joint campaign against the common enemy. Here he continued and strengthened a policy already adopted by Hülegü, but in the long run nothing came of these maneuvers. There had been, however, an attempt to coordinate strategies with Prince Edward of England, who arrived in Acre in 1271, resulting in a fairly large Mongol raid to the north of Syria. This raid, as well as Edward’s activities, led to no changes in the military or political situation in the country.

Abaqa was succeeded by his brother Tegüder, who had converted to Islam as a youth, taking the name Aḥmad. His short reign (1282-84) included a campaign to reinstate Islam as the official religion, negotiations with the Mamluks, and strife with Arḡun, son of Abaqa, which eventually led to Tegüder’s demise. The research of Adel Allouche has shown that the Il-khan’s diplomatic attempts vis-à-vis the Mamluks were still veiled calls for full submission to Mongol rule. While public expressions of Tegüder’s faith were made, and in principle, the laws regarding the ḏemmis were to be carried out, there is evidence for the continued maintenance of traditional Mongol religious tolerance. There is, indeed, little doubt regarding the Il-khan’s general incompetence, especially in his dealing with the challenge represented by Arḡun, but the matter of his religion should not be discounted as a factor that encouraged opposition among the Mongol princes and grandees. If nothing else, his abandonment of traditional Mongol religion may well have served as a rallying cry for political and other opposition. It appears that the Mongol ruling class in Persia was not yet ready to accept wholesale conversion to Islam, either among themselves or among the masses of Mongol troops.

Arḡun’s reign (1284-91), began with the final deposition and execution of Šams-al-Din Jovayni and his replacement by a Mongol officer, Buqa, who in turn served three years before his elimination. Eventually (1289), the office of vizier was given to Saʿd-al-Dowla, whose efficiency in collecting money, and his religion (Jewish), gained him many enemies. He was removed from office and killed during his patron’s fatal illness. Twice, in 1288 and 1290, Arḡun took to the field to meet invasions (or perhaps large-scale raids) from the Golden Horde into the Caucasus; in both cases the invaders were repulsed. On the Central Asian front there were also problems; in early 1288 there had been an attack from Qaidu and his Chaghatayid supporters. Toward the end of Arḡun’s reign, this front erupted again: Nowruz (q.v.), the son of the earlier Mongol viceroy in the area, Arḡun Āqā (q.v.), had long been in rebellion against the Il-khan. He had earlier fled to Central Asia, and now returned with the support of Qaidu. At the time of Arḡun’s death, his son Ḡāzān (q.v.), the governor of Khurasan, was retreating, although eventually he was to be reconciled with Nowruz. Arḡun launched no major campaigns against the Mamluks; perhaps he was preoccupied with other problems, or was hoping for cooperation from the West. The Mamluks, however, were not idle and sent raiders across the frontier in 1285, 1286, and 1289. Arḡun’s lack of military initiatives against the Mamluks belies his diplomatic activities to garner support from the Franks in Europe for a common campaign against this enemy. Arḡun sent four delegations to the West, the most famous being that of the Nestorian prelate Rabbān Ṣauma in 1287, who visited, inter alia, the papal court. (An interesting Syriac account of this mission has come down to us: see Budge, The Monks of Kûblâi Khân in the Bibliography.) The story is recounted by Bar Hebraeus (tr. Budge, p. 486), that 900 Franks came to Iraq in order to build a fleet to harass Muslim shipping, apparently in the Indian Ocean. Nothing came of this project, if indeed it ever existed. Arḡun continued the religious policy of his forefathers—a basic belief in shamanism, combined with an interest in Buddhism, and discernable sympathy for the eastern Christians in his kingdom. He showed perhaps less tolerance for Islam than did other pagan Il-khans.

The years after Arḡun’s death were characterized by political and economic confusion. Gayḵatu, a brother of Arḡun, beat out Ḡāzān, who was preoccupied with Nowruz’s rebellion, to the throne. Ḡāzān had no choice but to acquiesce in this fait accompli. The new Il-khan (r. 1291-95) was known among Mongols and the wider population for his moral turpitude and extravagance. The Mamluks, now under the rule of al-Ašraf Ḵalil (r. 1290-93), attacked and took Qalʿat al-Rum in southeastern Anatolia. This was followed by an exchange of truculent letters between the two rulers, but the Mamluk sultan did not live long enough to attempt to carry out his threat to reconquer Baghdad. The Il-khan, in any event, did not initiate any major offensives into Syria. Likewise, the Central Asian and Caucasian fronts were also relatively quiet. Perhaps the most notable, or rather infamous, development of this reign was the introduction of paper money (chʿao) in the markets of Tabriz in 1294 by the vizier Ṣadr-al-Din Zanjāni. This was a complete failure and contributed to the economic crisis fueled by the large-scale death of livestock (the so-called jut), mismanagement, and unbridled spending by the court. Gayḵatu’s unhappy reign was brought to an end by Baidu, a nephew of Abaqa, who took the throne in the early spring of 1295. Baidu’s regime lasted only a few months; the historian Rašid-al-Din, a noted partisan of Baidu’s nemesis Ḡāzān, does not even bother to mention him as a ruler. Baidu certainly did not have time to make his mark; some six months after seizing power, he was removed and executed by officers of Ḡāzān.

With the accession of Ḡāzān (r. 1295-1304), the Il-khanate enters a new and dynamic era. A harbinger of these changes was Ḡāzān’s conversion to Islam even before he gained the throne. Nowruz, with whom he had been reconciled, was instrumental in bringing about this conversion, as well as the successful struggle with Baidu. With the defeat of the latter and the beginning of Ḡā-zān’s rule, we can now talk of the Islamization of the Mongols in Persia and the surrounding countries, although not all aspects of traditional Mongol belief disappeared immediately. In spite of the patronage of Muslim building and institutions, and the increasing employment of Islamic titles and other forms of legitimization, there is a continued use of Mongol forms of rule and ideology, as well as ongoing connections, perhaps more attenuated than previously, with the Great Khan in the East. We cannot, therefore, speak of a break from Mongol tradition, but a transition to other, more Islamic and perhaps Iranian, forms of government and symbolism.

Ḡāzān’s regime commenced with the removal of several princes who were perceived as undermining his rule. He and his successors continued purging real and imagined royal conspirators, thereby dangerously reducing the pool of potential dynasts in the long run. In 1297, Ḡāzān also executed Nowruz, the strongman of the regime, who was suspected of secret contacts with the Mamluk enemies. In spite of the religious change, there was a remarkable continuity of foreign policy. The conversion of the ruler brought no relaxation of relations with the Mamluks—quite the opposite. Ḡāzān pursued his war with the rulers of Egypt and Syria with vigor, now adding Muslim justification to traditional Mongol ones. An invasion of Syria at the end of 1299 led to the Mongol victory at the Wādi al-Ḵaznadār near Homs; this is the one significant Il-khanid success in a field battle against their Mamluk enemies in the sixty-odd years of the war between the two regional powers. The victory led to a 100-day Mongol occupation of Damascus, but the Mongols withdrew of their own accord, perhaps due to troubles on other frontiers and possibly due to the projected logistical difficulties of maintaining a cavalry army and its numerous mounts in Syria over the summer. A year later, Ḡāzān tried again, demonstrating his determination to gain control over the country and to defeat the Mamluks decisively. This campaign was curtailed while still in north Syria, when unusually wet and cold weather was encountered. Two years later, the Il-khan again sent a large army over the Euphrates under the command of Qutlu-šāh. This force was defeated in April 1303 south of Damascus at Marj al-Ṣoffar. This was the last of the great Mongol invasions across the Euphrates, although warfare along the frontier was to continue for more than a decade. These campaigns had been interspersed with an exchange of missives with the Mamluk sultan, al-Nāṣer Moḥammad. The truculent nature of these letters from both sides, laced with Islamic motifs and each disparaging the religiousness of the other, shows that the possibility of Mamluk-Il-khanid peace was still far in the future.

Ḡāzān also continued his predecessor’s attempts to forge an alliance with the rulers of Europe against the Mamluks. In April 1302, he wrote to Pope Boniface VIII; it appears that an earlier order (yarliḡ) had been issued, which might well have been a detailed plan for a joint Mongol-Frankish campaign (A. Mostaert and F. W. Cleaves, “Trois documents mongoles des Archives secretes vaticanes,” Harvard Journal of Asiatic Studies 15, 1952, p. 469). As in earlier cases, nothing came of these missions, although the Mongol victory of 1299 and the subsequent occupation of Syria had awakened hopes in the West for a return of Jerusalem to the Christians. Again, little of substance occurred except for some raiding along the Syrian coast (see S. Schein, “Gesta Dei per Mongolos 1300: The Genesis of a Non-Event,” English Historical Review 94, 1979, pp. 805-15). Around the beginning of 1300, Ḡāzān received a letter from the Khan of the Golden Horde, Toqta, in which the old Jochid demands for Azerbaijan were rehashed. This claim was dismissed out of hand, but there is no record of actual hostilities breaking out on this front during this reign. On the other hand, there are reports of problems on the Chaghatayid front (Hetʿum, p. 196). These incursions may have been at least part of the reason for Ḡāzān’s withdrawal from Syria in 1300.

A notable event in Ḡāzān’s reign was the elevation of Rašid-al-Din Hamadāni (q.v) as co-vizier (with Saʿd-al-Din Sāvaji) around 1298. Rašid-al-Din first has importance as an administrator, lasting in this job until 1318. No less important, he was entrusted by Ḡāzān with writing the history of the Mongols and their steppe predecessors. This was the Tāriḵ-e ḡāzāni-e mobārak, which was expanded during the reign of Öljeytü (1304-16) into a history of humanity as then understood, with the title Jāmeʿ al-tawāriḵ. Besides being a monument to the wide intellectual horizons occasioned by the period of Mongol rule, it is also a major (and according to many scholars the most important) source for the history of the Mongols in general and the Il-khanids in particular. It is not without its biases, particularly towards the house of Toluy (q.v.), i.e., Hülegü’s father, and therefore the ancestor of his own patron. Among the significant information transmitted by Rašid al-Din is extensive evidence, including the original texts of orders from the Il-khans, regarding the widespread economic and administrative reforms that Ḡāzān saw fit to enact towards the end of his reign. The reports on the poor state of the economy that Ḡāzān met upon his rise to power, and the extent of these reforms (and their execution), should be taken with some reservations, since Rašid al-Din is virtually our only source for these matters and he himself was heavily involved in their implementation. This subject will be discussed in more detail below.

Ḡāzān died in 1304 at the age of 33. Possessing wide intellectual horizons, he was in many respects the greatest of the Il-khans. He was succeeded by his brother Öljeytü, whose early reign was marked by an attempt at unity and an end to conflict among the various Mongols rulers. This rapprochement was highlighted in the Il-khan’s letter to Phillip the Fair of France in 1305. Even though there was a call for concerted action against the Mamluks, there is also perhaps a veiled threat towards the Franks if they did not join the campaign (D. Sinor, “The Mysterious ‘Talu Sea’ in Öljeitü’s letter to Philip the Fair of France,” in D. Sinor, Inner Asia and its Contacts with Medieval Europe, London, 1977, article no. XIV). At the same time, Öljeytü also instigated an exchange of embassies with the Mamluks, calling for peace, although it would appear that this was either a subtle demand for surrender or an attempt to bide time until a major campaign could be launched. After several years of waiting for a Frankish response to his call for joint action, Öljeytü decided to launch the offensive with only his own forces. He was probably encouraged by Mamluk deserters who arrived in the summer of 1312 under Qarasunqur, former governor of Aleppo. This campaign, in spite of the large Mongol forces that set out, was a fiasco. The Mongols, led by the Il-khan himself, wasted their strength against the Mamluk border fortress of al-Raḥba along the Euphrates. After a siege lasting several weeks, during which the Mongols suffered from disease and other problems, the Mongols withdrew at the end of January 1313. Although there would be further fighting along the border in the following years, this was the final attempt of the Il-khans to invade Syria in force.

Öljeytü also had little success in 1307 in bringing the troublesome Gilān region under firmer control. The Mongol campaign, led by the Il-khan himself, met with stiff resistance; the senior general Qutlu-šāh was killed, and the Mongols suffered an embarrassing withdrawal. More success was achieved in the East, when in 1314 interference from both the Chaghatayids in Transoxiana and the Negüderis (Qauranas) in Afghanistan was brought to an end.

Early in his reign, Öljeytü dedicated the city of Solṭāniya in Azerbaijan, a project that had been initiated by Arḡun. Öljeytü built his mausoleum there, one of the pinnacles of Il-khanid architecture in Persia. In the winter of 1307-8 he declared his allegiance to the Shiʿites, and during the following years he made efforts to enforce this belief among the Muslims of his kingdom. This “conversion” certainly fueled the enmity of the Mamluks, and perhaps Öljeytü’s adoption of this faith was indirectly influenced by this ongoing conflict. Some Mamluk sources report that Öljeytü renounced his Shiʿite faith towards his death, but this is probably no more than wishful thinking on their (and their patrons’) part.

The last of the Il-khans was Abu Saʿid, the son of Öljeytü, who was twelve years old at the time of his accession to the throne. Real power was clearly in the hands of Čobān (q.v.), the leading Mongol grandee. Early in his reign, the venerable vizier and historian Rašid-al-Din was tried and executed, leaving his enemy ʿAli Šāh as the sole vizier. The latter maintained this position until his natural death in 1324, a unique occurrence in the annals of the Il-khanate. The early years of Abu Saʿid’s reign were punctuated by invasions and disorder. In 1319, Yasaδur, a Chaghatayid prince based in what is today Afghanistan, who earlier had submitted to the Il-khans, now rose in revolt. This was soon put down, but at the same time there had been an invasion via the Caucasus by Özbek, khan of the Golden Horde. This was repulsed by Čobān, although the young Il-khan’s personal bravery can be noted. The forces of the Golden Horde attacked a second time in 1325, and again Čobān repulsed them, even penetrating enemy territory in the aftermath of his success. Perhaps more significant was a revolt of Mongol notables led by Irenjin. It appears that Abu Saʿid himself may have initially encouraged these officers against his guardian. Subsequently, however, the Il-khan changed sides and together with Čobān defeated the rebels. Due to his own courage, Abu Saʿid received the epithet Bahādor “brave” (< Mongol baghatur “brave”), which henceforth remained attached to his name.

An important development in the realm of foreign relations was the end of hostilities with the Mamluks. This began with feelers put out, evidently first by the Mongols, through the good offices of the international slave trader, Majd-al-Sallāmi, around 1320. This led to negotiations that in turn resulted in the signing of a peace treaty in 1223. Thus ended the sixty-odd year war with the chief enemy, certainly the main non-Mongol one, of the Il-khans. In 1327 began the crisis that led to Čobān’s downfall and death. Henceforth Abu Saʾid ruled fully in fact and not just in name. The following years were relatively tranquil. A contemporary observer would probably not have discerned that the Il-khanate was nearing its end, but the rebellion of Irenjin and other tensions hint that the ties between the Hülegüid dynasty and the Mongol ruling class were weakening. With the unexpected death in 1335 of Abu Saʿid, who had set out to meet the reported advances of an invasion from the Golden Horde, the ruling house basically came to an end. The purges and rebellion of the previous generation or so had left their mark; there were virtually no Hülegüid scions to serve as ruler (Abu Saʿid had no sons), and in any event the senior Mongol grandees fell out among themselves, each supporting different pretenders. This led to the disintegration of the kingdom, the story of which is beyond the scope of this article (see ČOBĀN, CHOBANIDS)

Institutional, social, and cultural aspects of Il-khanid rule. The most important institution in all the Mongol states, including the Il-khanate, was the army, the basis of conquest and power. All Mongol grandees and princes were senior officers, and the Il-khan, even at the end of the dynasty, would often go out at the head of his troops on campaign. The original Mongol army that Hülegü brought with him (supplemented by the Mongol garrisons already in the Middle East) was composed overwhelmingly of light cavalry: Mongol troops riding steppe ponies and armed with bows and hand weapons. Troops went on campaign with a string of ponies that they would ride in turn, both on long marches and during battle. The principal tactics were the massed cavalry charge, accompanied by a barrage of arrows, and attempts to encircle the enemy. If the enemy were broken, then the Mongols would close in; if it maintained its position, then the Mongols would launch successive attacks until their objective was achieved. From the beginning, the Mongols were aware of the importance of siege warfare. Hülegü brought with him several hundred households (or perhaps squads) of siege engineers, who were probably both sappers and artillery experts (cf. Allsen, 1987, p. 202). The question whether the Mongols employed gunpowder-based explosives (or even artillery) in sieges in the Islamic world remains open, at least to this author’s mind. In any event, there is no evidence of them in the war against the Mamluks. Early on, the Mongols employed auxiliary troops from local rulers, be they Muslims (Saljuqs from Rum, Mosul, etc.) or Christians (Greater and Lesser Armenia, and Georgia). In the former case, many of these soldiers were probably mamluk-type troops, i.e., slave soldiers, of local rulers; the Georgian and Armenian contingents probably included some infantry. In any event, the evidence points to the continuation into the fourteenth century of traditional Mongol tactics, using massed light cavalry. There may have been some heavier-armed cavalry units, perhaps the guard division of the ruler himself (the bahādoriya “braves”). There is clear evidence that the Il-khanid army was organized into divisions, each called tümen. These were, in theory, composed of 10,000 men, although in reality they probably were undermanned.

After the initial period of conquest and consolidation, the army found employment against different types of enemies: (1) other Mongol states and forces (the Golden Horde, Chaghatayids and Qaraunas), (2) local rebels (Kurds, Lurs, local Muslim kings), (3) rebellions of Mongol princes and senior officers (who were supported by part of the army), and (4) the Mamluks. In general, the Mamluks had the upper hand in this sixty-year war along the frontier, in the use of espionage and in the major field battles. The one Mongol victory in the field in a major confrontation was at Wādā al-Ḵaznadār in central Syria in 1299. This led to a temporary Mongol occupation of Syria but to no lasting change of the status quo along the border. It has been suggested by J. M. Smith, Jr. and David O. Morgan that the logistical limitations of Syria prevented the Mongols from concentrating a large number of troops in Syria for any length of time; there simply was not enough pasturage in the country to support the many mounts that the army brought with it. Without anyone contesting the importance of this factor, others may have been just as significant. One was the inability of the Il-khans to absent themselves for any length of time from the center of their kingdom, and another was the dangers (perceived or real) from other Mongol states. In general, it appears that the Mamluk sultans attributed more importance to this war than did the Il-khanids. For the former, it was seen as a matter of life and death. The resources and attention that the Mamluks devoted to the war may be one of the main reasons for their continuing success. It was recognition of their inability to defeat the Mamluks that apparently impelled the Mongol ruling group in Iran to initiate the negotiations that led to the peace treaty of 1323.

The desire to defeat the Mamluks also led to ongoing negotiations with Western powers such as the pope and the kings of France, England, and Aragon. While the Il-khans had some contacts with the Franks in the Levant (particularly with the Prince of Antioch, who officially submitted to the Mongols even before the campaign in Syria in 1260), it became clear that these were of minor importance and that a deal would have to be struck with at least one of the Western potentates. Hülegü’s initial maneuver, directed to Louis IX of France and apparently to the Pope as well, still bore traces of the traditional haughty Mongol attitude toward non-Mongol rulers, albeit more subtly expressed in this case. Later missives were more restrained, as the Mongol rulers saw the strategic importance of gaining allies against the Mamluks. It appears that the impression frequently conveyed by the Mongol envoys, who were often Eastern Christians, was that the Mongols either had converted to Christianity or were about to do so. This may have been due to a deliberate Mongol ploy to encourage cooperation, an exaggeration on the part of the envoys, or simply a misreading of the religious tolerance of the Mongols, along with their sympathy for the Eastern Christians and the fact that some Mongol queens were Christians, mainly Nestorian. The last recorded Il-khanid embassy to the West was dispatched to Philip the Fair of France by Öljeytü in 1305.

Overall, in spite of some warm words and general expressions of good will and a desire to cooperate against the common enemy, little came of these exchanges of letters and oral messages. The closest that the Mongols came to actively cooperating with the Franks was during the crusade of Prince Edward of England, who arrived in Acre in 1271. After an exchange of letters, Abaqa sent a significant force into northern Syria. Edward, however, dissipated his strength on pointless raids in Palestine; and the Mongol force withdrew. So ended the one real attempt at Il-khanid-Frankish military cooperation. (See R. Amitai, “Edward of England and Abagha Il-khan: A Reexamination of a Failed Attempt at Mongol-Frankish Cooperation,” in M. Gervers and J. M. Powell eds., Tolerance and Intolerance, Syracuse, N.Y., 2001, pp. 75-82.) At the same time, one might note the evidence cited above about the Frankish (probably Italian) sailors in Iraq at the end of the thirteenth century, who supposedly were planning attacks on Muslims (i.e., enjoying Mamluk patronage) in the Indian Ocean.

The overall failure of diplomacy to achieve a common military strategy with the Frankish rulers in Europe belies a well-developed trade with Europe, particularly via Italian merchants. Especially important were the Venetians, although their Genoese competitors also were active in the Il-khanate in spite of their close relations with the Golden Horde and the Mamluks. The Italians maintained colonies in Azerbaijan: not the least in Solṭāniya, where eventually a bishopric was established. Much of this trade was transient, i.e., Mongol Iran was a conduit for merchandise moving between the Mediterranean and points further east such as China and South Asia, and there are examples of Italian merchants (most famously the Polos) using Mongol Iran as a springboard for activities in other Mongol-controlled areas of Asia. The importance of these Italian merchants and the trust placed in them by the Mongol authorities is shown by their occasional use as envoys to the West by the Il-khans.

The checkered and often difficult relations between the Il-khans and their royal cousins to the north and northeast have already been mentioned. Relations with the Great Khan or Qa’an (Qāʾān in Persian and Arabic texts) were much warmer, due both to the close familial ties (all were descendents of Toluy), as well as the fact that they were far away and therefore there was no opportunity for border disputes. The importance of these ties for the legitimacy of the Hülegüid house is noted by the Mamluk author ʿOmari (ed. Lech, p. 1). Hülegü’s immediate successors enjoyed official investiture from the Great Khan Qubilai, although it came some time after their initial gaining of rule. Under Ḡāzān and his successors, some of these ties were loosened. The title Il-khan is used less often, and the official expression of subservience by Ḡāzān to the Great Khan on his coins all but disappears. On the other hand, there are several examples of delegations going back and forth between the two rulers, even more so than in the past; the largest number of missions was exchanged under Abu Saʿid. More importantly, the Great Khan had a high commissioner in the Il-khanid court, Bolad Ching-Sang (d. 1313), giving clear evidence of the official preeminence of the Great Khan. It is clear, then, that the Il-khans never gave up their de jure recognition of the Great Khan’s preeminence.

Bolad also played a certain role in the cultural exchanges between the two extremes of the Mongol world, Iran and China. As Thomas Allsen has recently shown, the movement of people, commodities, and ideas across Asia was mutually beneficial to these countries, contributing to influences in both directions. Areas of influence included cartography and astronomy, medicine, agriculture, textiles, food, and printing. For example, on one hand, various crops were introduced from the Islamic world into China, while on the other hand, culinary influences went both ways. The Mongols were not merely passive conveyors of cultural artifacts but actively shaped the nature of this exchange by their own tastes and preferences.

The administration of the Il-khanate was an interesting mixture of institutions from Muslim Iran and the Mongol empire (itself combining several traditions), along with some original features and ongoing stopgap measures. Large numbers of Muslim bureaucrats, mostly Persian-speaking, quickly made their peace with the new non-Muslim rulers and joined their service. Mention has already been made of the Jovayni family; but Jean Aubin has brought to our attention the existence of the so-called Qazvini group of officials, and there were surely others. This scholar has also made it clear that the distinction between Mongol officers and Persian-speaking officials was not as clear-cut as has been thought. Certain Mongols (or other Central Asians) functioned successfully as senior bureaucrats (e.g., Arḡun āqā, q.v., and Buqa), and many of the Persians were well acquainted with Mongol ways and language. On occasion they would even command military campaigns, such as Šams-al-Din Jovayni’s in Anatolia in 1277. There is evidence that the tax system was both more onerous and more capricious than before the Mongols, causing much difficulty for the agricultural sector and for economic life in general. This is related to the question of the economic decline of the Il-khanate. A recent study by G. Lane permits us to see the economy of Mongol Iran and the surrounding countries in a more nuanced way. In general, we should be wary of attributing the effects of the initial Mongol conquest (1219-23) in northeast Iran, and the subsequent three decades of minimal Mongol control, to the initial period of Il-khanid rule. It can be stated that Hülegü and Abaqa’s reigns led to a certain stabilization and some attempt to rebuild the economy, at least in certain areas. At the same time, we should be aware that areas such as Iraq and the Jazira suffered from Hülegü’s conquest, and the latter region, as a frontier area, subsequently declined economically and demographically (see the evidence in Rašid-al-Din, Baku, III, pp. 557-58). Other areas, particularly Azerbaijan, prospered, due, inter alia,to Mongol patronage. Southern Iran (Fārs and Kermān) also maintained a level of prosperity, due apparently to the indirect nature of Mongol rule. It appears that Khorasan never fully recovered from the effects of the original conquest, as well as the fighting with the Chaghatayids and internal disorders such as Nowruz’s revolt. In any event, the misrule of Gayḵatu seems to have plunged the kingdom as a whole into an economic crisis, perhaps exacerbating previous problems of over-taxation and the abandonment of agricultural land in certain areas.

Our understanding of the problem is certainly colored by the account of Rašid-al-Din. As mentioned above, since he was in the service of Ḡāzān and was writing at his behest, these reports should be treated with some caution, not the least since Rašid-al-Din himself was co-vizier and therefore responsible for carrying out the Il-khan’s efforts at reform meant to rectify this situation. These administrative and economic reforms were recorded in detail by Rašid-al-Din himself in his history, with the texts of many orders (yarliḡs) given verbatim. The only independent, and quite general, confirmation of these reforms is by the Mamluk author al-Ṣafadi (al-Wāfi be’l-wafayāt, MS. Topkapı Sarayı Ahmet III 2920/25, fol. 61a), but there is no denying that a great many reforms were conceived and at least partially executed. It is also clear that the economic situation that Ḡāzān inherited was not auspicious, and strong administrative measures were needed to stabilize and improve matters. It appears that these measures, enacted towards the end of Ḡāzān’s reign, had some effect. A certain increase in state revenues is reported, and the impression from the accounts of his successors’ reigns is not that of endemic economic crisis.

From the beginning of their regime, the Mongols were patrons of high culture. Hülegü supported the Shiʿite savant Naṣir-al-Din Ṭusi (q.v.), most famously in the construction of an astronomical observatory at Marāḡa. This ruler surrounded himself with scholars from different religions and cultures (Rašid-al-Din, Jāmeʿ al-tawāriḵ, Baku, III, 91). The massive palace complex at Taḵt-e Solaymān in Azerbaijan (near Marāḡa), built during the time of Abaqa, is another example of a relatively early Mongol construction; the luxurious adornment shows the Il-khans as patrons of fine arts at their best (see Tomoko Masuya, “Il-khanid Courtly Life,” in Komaroff and Carboni, 2002, pp. 84-103). With the final conversion of the Mongols, represented by Ḡāzān’s adoption of Islam, we can see a definite change: patronage is devoted primarily to projects of an Islamic nature. Some of these projects are on a grand scale, such as Öljeytü’s mausoleum complex, or, in the realm of fine arts, the gigantic Koran that he commissioned. Henceforth, in their search for legitimization, the Il-khanids combine Islamic and Mongol motifs on coins, inscriptions, building projects, letters, and orders. Upon their arrival in the Middle East, the Mongols disestablished Islam as the state religion, a matter of some concern for subject Muslims. This was, however, a great propaganda boon for the Mamluks, who constantly harped in public pronouncements on this insufferable situation, on the killing of the caliph, and on the fact that the Mongols were infidels and polytheists. There is no indication that this religiously inspired, anti-Mongol polemic particularly disturbed the Mongols, although it may have caused some unease among the Muslim bureaucrats who served them. Within the Mongol ruling class, particularly among the women, there were those who professed Nestorian Christianity. The Mongols saw no contradiction between this system and their traditional sha-manistic religion. Likewise, Buddhism, evidently of the Tibetan variety, flourished in Iran under their patronage. Hülegü himself was sympathetic to Buddhism as well as to Christianity, surely influenced by his wife, Dokuz Ḵātun, herself a Nestorian. Abaqa and Arḡun had even stronger affinities for this religion, although without disagreeing with traditional Mongol beliefs. Tegüder, Ḡāzān, and Öljeytü all went through Christian and Buddhist phases as children. This religious tolerance or, as some scholars would have it, indifference to religion created a propitious environment for non-Muslim communities, particularly the Christians in Iran, Iraq, and other countries. The early Il-khanid period was indeed the Indian summer of Eastern Christianity in the Middle East; even after the Mongol conversion to Islam, the Christians in their realm enjoyed a greater degree of freedom than in the pre-Mongol period.

The matter of the conversion to Islam of the Mongol rulers, notables, and commoners is one of interest and importance. Some scholars have suggested that this was a search for legitimization among the sedentary, mostly Persian-speaking, subjects. There is, however, no actual proof for this assertion, and in fact the impression is that the Mongol ruling class was not overly concerned with their image in the eyes of their Muslim or non-Muslim subjects, as opposed to the great importance that they attributed to legitimization among the peoples of steppe origin. Only with their conversion to Islam, as noted above, did the Il-khans and their officials look for Islamic motifs to justify their rule, among the ruling class as well as the larger population. Mongol conversion seems to have been a double process: individuals among the elite, and a growing movement among the common tribesmen. The role of Sufis can be noted, although whether this was due to some affinity to shamans or shamanism, another old chestnut in the research literature, may be open to question. It may be due more to the charismatic nature of some Sufis, the perceived success of their miracles and other powers, and the appeal of their down-to-earth, but also spiritually rich, version of Islam. In any event, other forces for Islamization can be ascertained, such as the Muslim officials who surrounded them and the Turks and Mongols who had already converted. Finally, there was the whole matter of cultural diffusion between a strong majority culture and a minority culture which had no proselytizing aspirations but rather had always been open to religious influences and which had created a multi-cultural milieu.

Another perspective is offered by looking backwards from the eventual fate of many of the Mongols in Iran and the surrounding areas. They assimilated into the Turkish-speaking population, some of whom had long been in the region and had long been Muslims, while others had come with the Mongols or in their aftermath; many of this latter group were also Muslims. From this point of view, one can say that the Islamization of the Mongols was a step on their way to becoming Turks or else, perhaps, that their assimilation into the Turkish population was facilitated by their conversion to Islam.

There is no overwhelming indication that the masses of Mongols settled down and became sedentary during the Il-khanid period. Since there is clear evidence that the Il-khanid army of the early fourteenth century was similar in tactics and character to that which came with Hülegü in the mid-thirteenth, we have an indication that the later tribesmen maintained the same pastoral nomadic lifestyle as their forefathers. The famous evidence given by Rašid-al-Din (Jāmeʿ al-tawāriḵ, ed. Jahn, 1940, p. 302) notwithstanding, it appears that Ḡāzān’s distribution of land allocations (eqṭāʿāt) was not executed to any real degree, and even if it was, this does not mean that the Mongols were settling down and beginning to practice agriculture themselves. Quite the contrary ʿOmari (ed. Lech, p. 95) clearly states: “Every tribe has land to reside in and the descendent inherits it from the forefathers since Hülegü conquered this country. Their abodes are in it. They have in it crops for their substance, but they do not live by tilling and sowing.” Until late in the Il-khanate period there is evidence that the royal family and the court in general maintained a peripatetic lifestyle, moving between winter and summer camps (see C. Melville, “The Itineraries of Sultan Öljeitü,” Iran 28, 1990, pp. 55-70).

To sum up: The Il-khanid period was a time of tremendous change in Iran; some areas were adversely affected economically, while others enjoyed prosperity. This was a time of cultural activity, partially inspired by the continentally wide perspectives of the rulers. Mongol rule had an effect on political institutions in the long run in Persia and contributed indirectly the reconstruction of a unified Persia. The borders of the state established by the Safavids are quite similar to those created by the Il-khanids (with the major exception of Anatolia), and under the Il-khans the term Irān-zamin enjoyed wide use. Perhaps most importantly, the demography of the country and the surrounding areas was forever changed. The Eurasian Steppe element, what we can today call Turkic, was greatly augmented under the Mongols, eventually giving us the belt of speakers of Turkic languages from southeastern Europe through much of northern Iran into Central Asia.

IL-KHANIDS ii. Architecture

The architecture produced during the period of Il-khanid rule in Persia and Iraq is notable for its mammoth size, soaring height, sparkling color, and ingenious methods of covering space. Size and scale were seen as representative of power and authority, and the wider palette and increased surface of colored decoration served to further distinguish these buildings from the landscape. Builders in the Il-khanid period continued to use traditional materials such as brick, stone, wood, and plaster in standard forms and building types, which were often combined into larger complexes; but they incorporated new Chinese motifs into the standard decorative repertory. Both the buildings themselves and contemporary descriptions of them show that the new rulers deliberately manipulated architecture to connect themselves to the Persian past, both pre-Islamic and Islamic. The message resonated, for many of the Il-khanid structures served in turn as models for those built by the Timurids in Khorasan and even the Mughals in India.

The Mongol campaigns of the early 13th century devastated the countryside, and the first years of Il-khanid rule were devoted to reconstructing the land and economy. Only two major buildings date from the second half of the century (and both of them are known mainly through archeology), but their individuality already reflects the concerns of the new rulers. The first known act of architectural patronage by the Il-khanids after their conquest of Baghdad in 1258 was the construction of an observatory on a hill five hundred meters north of the summer capital at Marāḡa (Wilber, no. 9). Amply funded by Hulāgu /Hülegü (q.v.), the institution was set up in 1259 under the renowned astronomer Naṣir-al-Din Ṭusi and supplied with instruments by the Damascene Moʾayyad-al-Din Ordi. Excavations of the site by Parviz Varjāvand in the 1960s and 1970s (Varjāvand, pp. 527-36) revealed sixteen units, including a central tower containing a quadrant 45 meters in diameter, a foundry for the manufacture of astronomical instruments, five round towers, and several several large buildings (Plate I). The observatory’s importance is evident from both the quantity and quality of the materials found there (which included baked brick and glazed and luster tiles; Plate II) and the building’s enduring legacy (it seems to have inspired the Timurid ruler Uluḡ Beg to build another in Samarkand nearly two centuries later).

PLATE I. Marāḡa, a panoramic view of the observatory complex. After Varjāvand, p. 401.View full image in a new tab

PLATE II. Marāḡa, reconstructed tower of the observatory. After Varjāvand, facing p. 322.View full image in a new tab

The Il-Khanids maintained the nomadic practice of winter (qešlāq) and summer (yeylāq) camps with elaborate tents, wintering in the warm lands of Mesopotamia around Baghdad and summering in the highlands of Azerbaijan. There, as part of their sedentarization, Hülegü’s son Abaqa (q.v.) began construction of a summer palace southeast of Lake Urmia (Huff, III, pp. 1-31). Completed under his son Arḡun (r. 1284-91, q.v.), the site, known as Taḵt-e Solaymān, stands on the foundations of the Zoroastrian sanctuary of Šiz, which had been favored by the Sasanian kings (see ĀDUR GUŠNASP and also INVESTITURE iii.). German excavations in the 1970s uncovered massive oval walls protected by towers and accessed by a new gate on the south (Plate III). The palace itself had a huge central courtyard (125 x 150 m) that was oriented north-south and encompassed an artificial lake. A portico with ayvāns (q.v.) on the four sides surrounded the courtyard. The north ayvān led to a domed room, which occupied the site of the Sasanian fire temple and probably served as Abaqa’s audience hall. The west ayvān led to a transverse hall flanked by two octagonal pavilions, and the many fragments found there suggest that this area, which had been the throne room of the Sasanian king Ḵosrow II Parvēz (r. 591-628), served as the living quarters of the Il-khanid ruler. The halls and pavilions were lavishly decorated in stucco, including elaborate moqarnas vaults, and tile, including six types ranging from underglaze painting and monochrome glazing to overglaze luster-painting and lājvardina (Qučāni, pp. 74-103). Along with traditional geometric, floral, and figural designs, the tiles were also decorated with new motifs such as dragons and phoenixes as well as scenes and verses from the Šāh-nāma. Both the choice of site and the themes of decoration suggest that the Il-khanid rulers were trying to link themselves to the glorious Persian past.

PLATE III. Azerbaijan, plan of Taḵt-e Solaymān archeological site. After Huff, p. 213.View full image in a new tab

Ḡāzān’s official conversion to Islam in 695/1295 and the concomitant economic reforms carried out under his vizier Rašid-al-Din Fażl-Allāh led to an upsurge in architectural patronage. The prosperity of the Il-khanid state was based on long-distance east-west trade, and to support it caravansaries were built across the realm. The few that survive at Marand (ca. 1330), Sin (730/1330), and Sarčam (733/1333) show a typical rectangular plan protected by bastions and entered through a single monumental portal that led to a large open courtyard (Wilber, nos. 85, 89-90). This plan, which had already been standard in Saljuq times (e.g., Rebāt Šaraf, begun 508/1114-15), set the norm through the Safavid period.

Caravansaries were also built within cities. The best example to survive is the superb Khan Mirjān (760/1359), ordered by the Jalayerid governor of Baghdad to support his nearby complex that included a madrasa, moṣallā (prayer hall), and tomb (Janabi, pp. 111-46). The building itself comprises a huge (45 x 27 m), two-story hall spanned by eight transverse arches supporting stepped vaults crowned with domes on squinches. The sophisticated roofing system, which allows light to reach the interior, shows that the governor considered the commercial construction on a par with the other parts of the complex that have not survived.

Secular structures tend to deteriorate through constant use, and much of the construction that survives from the Il-khanid period post-1295 concerns the new state religion of Islam. Ḡāzān ordered a mosque and bath built in every town, with the revenue from the latter to be used to support the former (Rašid-al-Din, p. 1087; tr., p. 743). Many standing congregational mosques were rebuilt or restored, as at Ardabil (early 14th century) and Isfahan (710/1310; Wilber, nos. 40, 48; van Berchem). In other cases, new mosques were constructed. The new ones take a variety of forms. Smaller examples, such as those built in the 1330s in the Isfahan oasis along the Zāyandarud at Dašti, Kāj, and Ezirān and perhaps as far downstream as Barsiān, comprise a single dome chamber (Wilber, nos. 69-71). Larger examples usually contain ayvāns, ranging from one or two (Aštarjān, 715/1315-16; Wilber, no. 49) to the standard four (Hafšuya, early 14th century; Ker-mān, 750/1349; Wilber, no. 97).

The most stunning is the mammoth single-ayvān mosque constructed by the vizier Tāj-al-Din ʿAlišāh (q.v.) at Tabriz around 1315 (Wilber, no. 51). Set on the qebla side of a large courtyard, the ayvān, which was flanked by a madrasa and ḵānaqāh (hospice for Sufis), sprang from walls 10 meters thick and 25 meters deep (Plate IV). Although the vault spanning some 30 meters has since collapsed, the walls still bear pockmarks from cannon shots lodged when it served as a citadel (arg) in the 19th and early 20th centuries. It was lauded in its own time by Ḥamd-Allāh Mosawfi (Nozhat al-qolub, ed. Le Strange, pp. 76-77; tr. Le Strange, p. 80) as a congregational mosque with a ruined hall (ṣoffa) larger than the Sasanian palace at Ctesiphon, and was considered one of the wonders of the world (Kārang, pp. 240 ff.).

PLATE IV. Tabriz, Mosque of ʿAli Shah, qebla wall from rear. Courtesy of the author.View full image in a new tab

The best-preserved example of the standard, four-ayvān plan is the mosque constructed by a local family of viziers serving the ruler Abu Saʿid Bahādor Khan at Varāmin south of present-day Tehran (722-26/1322-26; Wilber, no. 64). The plan of four ayvāns set around a courtyard with a dome chamber behind the qebla ayvān had been standard since the renovations of the Friday Mosque at Isfahan under the Saljuqs in the late 11th and early 12th centuries. The dome chamber at Varāmin also shows the classic tripartite elevation developed in the Saljuq period of square base, zone of transition with four squinches alternating with four blind arches, and dome. The Il-khanid mosque at Varāmin is distinguished from its Saljuq antecedents, however, by its attenuated proportions, small courtyard, and extensive but routine use of tile mosaic.

Il-khanid builders also elaborated the standard, four-ayvān plan, and mosques in Yazd and its environs show a distinct variant with rectangular halls covered with transverse vaults added on the qebla side flanking the south ayvān and dome chamber. This plan appears in the congregational mosque at Yazd, rebuilt in 725/1324-25 by a local notable Šams-al-Din Neẓāmi and distinguished by a monumental tiled portal with two tall minarets (Wilber, no.66; Afšār, no. 32). As the patron had spent much of this time in Tabriz, where he married Rašid-al-Din’s daughter, such other new features as the galleries on the upper floor and the easy flow of space between sanctuary and side halls were probably copied from now-lost buildings in the Il-khanid capitals of northwest Persia. In the same way, texts (e.g., Jaʿfari, pp. 88-89) tell us that plans for other buildings in Yazd, such as Šams-al-Din’s own tomb funerary complex in Yazd, known as the Madrasa-ye Šamsiya (begun 727/1327; Wilber, no. 107; Afšār, no. 137), were sent from Tabriz (Jaʿfari, apud Afšār, II, pp. 591-92). The buildings in central Persia have survived better than those in the northwest, and their superb tile revetment also allows us to trace the development of tile mosaic from three or four colors (dark and light blue, white, and black) to the full seven-color palette (with yellow, green, and unglazed brick) and the gradual shift from angular to floral patterns.

The greatest architectural projects from the period are tomb complexes. The first Il-Khanids, like their Mongol forebears, concealed their grave sites; but Muslim rulers beginning with Ḡāzān followed local Islamic practices in building monumental domed tombs, which were typically incorporated into sizeable pious foundations with public services such as mosques, madrasas, hospices for Sufis, hospitals, residences, and the like. The complexes for Ḡāzān and Rašid-al-Din at Tabriz are substantially destroyed (for a reconstruction of the later, see Blair, 1984, pp. 67-90). The best surviving example is the one built by Sultan Oljāytu/Öljeytü at his new capital of Solṭāniya, 120 kilometers northwest of Qazvin on the road to Tabriz (Wilber, no. 47; for complete bibliography, see Akhavan Tavakoli). The tomb itself is an enormous octagon, measuring some 38 meters in diameter and oriented almost cardinally, with a rectangular hall (15 x 20 m) appended on the south. Crowned by eight minarets and surmounted by a dome almost 50 meters high, the octagon contains a vast hall surrounded by eight arched openings with balconies overlooking the interior (Plate V). On the exterior, blocked off from the interior, is another ring of galleries, which display a bewildering array of carved plaster motifs painted in red, yellow, green, and white. Many of the designs there resemble those found in contemporary book illumination, suggesting that Il-khanid designers were the first to use paper patterns to execute similar designs on different scales (Plate VI). The lofty interior of the tomb, one of the largest uninterrupted spaces of medieval times, and the subtle interpenetration of volumes make this building a landmark of world architecture in which the sultan’s desire for immortality was deftly realized, and the tomb provided the model for the Gur-e Mir, Timur’s tomb in Samarkand, and ultimately the Taj Mahal in Agra.

PLATE V. Solṭāniya, mausoleum of Sultan Öljeytü, exterior. After Survey of Persian Art VIII, p. 381.View full image in a new tab

PLATE VI. Solṭāniya, mausoleum of Sultan Öjeytü, interior. After Survey of Persian Art VIII, p. 382.View full image in a new tab

Il-khanid royal tombs are marked by enormous size (and prodigious endowments), but other dignitaries and notables built smaller versions, both for themselves and for local saints and Sufis, some living and some dead (Golombek, pp. 419-30). At Naṭanz, in central Persia, for example, in the first decade of the 14th century, the vizier Zayn-al-Din Mastari refurbished the local mosque and added a three-story minaret and ḵānaqāh around the tomb he built for the Sufi shaikh ʿAbd-al-Ṣamad (d. 699/1299?; Blair, 1986). The glory of the tomb is the moqarnas vault, the finest to survive from the period. Marks in the plaster show that the tomb’s lower walls were once revetted with luster tiles made at nearby Kāšān, and many tiles, including a concave hood that once surmounted the prayer niche (meḥrāb) and a large frieze with birds whose heads were later defaced by a zealous iconoclast, can be traced to museums around the world. The upper walls of tomb and mosque were decorated with superbly carved stucco, part of it signed by the master craftsman Ḥaydar, who also designed the finest sculptural achievement of the age, the stunning meḥrāb added to the winter prayer hall in the mosque at Isfahan in Ṣafar 710/June-July 1310, following Sultan Üljeytü’s conversion to Shi’ism (van Berchem, pp. 367-78; Blair and Bloom, 1994, fig. 12).

Smaller, freestanding tomb towers were also built. The octagonal one at Solṭāniya known as the tomb of Čelebi Oḡlu marks the grave of the crypto-shamanic dervish Barāq Bābā (d. 708/1308; q.v.; Wilber, no. 80). To judge from its form and details, it served as the model for the Emāmzāda Jaʿfar at Isfahan, built to commemorate the grave of an ʿAlid Shaikh and descendant of the Fifth Imam, who died in 725/1325 (Wilber, no. 68). A tower’s form therefore did not mark sectarian affiliations. Such similar buildings also show that specialized workers moved from site to site. These included not only builders but also tile workers. The ones employed at Naṭanz in the first decade of the 14th century, for example, then went to Solṭāniya in the second. The interiors of the fanciest tombs and shrines, like the palace at Taḵt-e Solaymān, were revetted in luster tiles imported from Kāšān, which were used to cover the walls, meḥrāb, and cenotaph. The breakup of the Il-khanid empire following the death of the last major ruler Abu Saʿid in 736/1335 heralded the end of luster production. On the later pieces the quality of the painting declines, and the inscriptions on one tile dated 739/1338 ask God to protect the city from the accidents of time. Such a plea, perhaps in face of declining commissions, went unheeded, for a single tile dated 740/1339 marks the end of 140 years of production.

IL-KHANIDS iii. Book Illustration

The Il-khanid period is no doubt the historical moment during which the art of painting, in particular in illustrated manuscripts, witnessed a dramatic increase in number, subject matter, artistic output, and patronage as compared to the late Saljuq epoch in Persia, to the late ʿAbbasid period in Baghdad, and to the rule of the Atābaks (q.v.) of Mosul in the Jazira region in northern Mesopotamia that preceded the advent of the Mongols in the area. It is fair to surmise that increasingly fewer works on paper are likely to survive as one goes back in time, so it is to be expected that, comparatively, more manuscripts from the Il-khanid period have been preserved than earlier ones, but this is not sufficient to challenge the notion that the late 13th century and especially the first quarter of the 14th can be regarded as perhaps the most important formative period in the history of Persian painting, an epoch of great changes, and a moment of fervor and stimulation on the part of both patrons and artists. As a matter of fact the number of complete, illustrated manuscripts that survive from this period is not so large, just a dozen or a few more; but dispersed folios, the miscellaneous material included in albums in Istanbul and Berlin, as well as documentary evidence, help to suggest that book illustration became one of the most popular art forms in Il-khanid Persia, trickling down from the patronage of the courtly elite to that of powerful viziers, governors, and less wealthy patrons in just a few decades.

From the artistic point of view, the single major change that took place in Persian painting was due to the strong influence and incorporation of eastern Asian, mostly Chinese, elements that freely circulated in Persia, an area that represented just one section of the pan-Asian highway that was open to all kinds of traffic during the Mongol period. It is not clear whether Chinese painters were actually transferred to Persia and trained local artists or if it was mostly a matter of wide circulation of Chinese scrolls and other works on paper in Persia that made a special impression on Persian artists. There was, however, a clear evolution in the selection, integration, elaboration, and ultimately reinterpretation of Chinese elements in a context (the manuscript or codex), and in pictorial compositions (images illustrating an Arabic or Persian text in prose or poetry) that were familiar to local patrons, calligraphers, and painters. It is this integration and elaboration of exotic constituents into the traditional fabric of book illustration that marked a new chapter in the history of Persian painting, one that would carry its effects all the way to the new heights reached during the Timurid period in the 15th century.

It is not surprising that the Il-khanid city first subjected to these changes, at least according to the extant works, was the former ʿAbbasid capital Baghdad, where a so-called Arab school of painting had flourished around the second quarter of the 13th century (Ettinghausen, 1962, pp. 59-124). There is no question that artistic activities almost came to a halt for a few decades following the Mongol sack of the city in 1258, but it is interesting that the first three dated, illustrated manuscripts that survive from the Il-khanid period were produced in Baghdad or nearby in Iraq, whose central cultural role obviously had not been entirely wiped out, having actually been revived under the control of local Arab or Persian governors working for the Il-khanids. One of these was ʿAṭā Malek Jovayni (1226-83); an early copy of his historical work Tāriḵ-e jahāngošā (Bibliothèque Nationale, Paris, MS suppl. persan 205), dated 689/1290, almost certainly produced in Baghdad, is the earliest Il-khanid work on paper that includes a frontispiece incorporating Chinese elements such as peony-like flowers and clouds with a “flaming pearl” (Richard, 1997, p. 41). These elements are, however, isolated or outright awkwardly positioned in the painting (for example, large peonies on a pomegranate tree; Plate I) and therefore suggest that this process of integration had just started and was taking its first steps in Baghdad. Two other earlier works from this period with illustrations were also produced in Iraq. One is the earliest known copy of Zakariyāʾ b. Moḥammad Qazvini’s ʿAjāʾeb al-maḵluqāt wa ḡarāʾeb al-mawjudāt (q.v., finished in nearby Wāseṭ and dated 1283, just three years after its author’s death; Bayerische Staatsbibliothek, Munich, MS Cod. Ar. 464; Bothmer); the other is a copy of the Rāsʾel Eḵwān al-ṣafāʾ, completed in Baghdad in 1287 (Süleymaniye Library, Istanbul, MS Esad Efendi 3638; Ettinghausen, 1962, pp. 98-102). Unlike the Tāriḵ-e jahān-gošā, the compositions and artistic styles of these two manuscripts are closer to the pre-Mongol “Arab” tradition and do not lead one to foresee the great changes that began to appear only a few years later in the Jovayni manuscript.

PLATE I. Left-hand side frontispiece from the Tāriḵ-e jahāngošā, Iraq, probably Baghdad, 689/1290. Courtesy of Bibliothèque Nationale de France, Paris, MSS Or., Suppl. persan 205, fol. 2r.View full image in a new tab

Baghdad would remain the most important center for the production of illustrated manuscripts until the Il-khanid court officially converted to Islam and at the same time became more and more persianized, adopting manners, customs, and ultimately the legendary pre-Islamic history of Iran, the nation they had conquered. Thereafter, Baghdad and the region of Iraq would become just a province, and its great manuscript-producing tradition declined for the rest of the Il-khanid period. The last example of important artistic activities in Baghdad is the earliest known copy of Saʿd-al-Din Varāvini’s Marzbān-nāma (Archaeological Museum Library, Istanbul, MS 216), a Persian collection of moralizing stories, which was finished in 1299 and contains three paintings that still show an awkward integration of Chinese floral elements in otherwise traditional compositions, similar to the Jovayni codex of 1290 (Simpson, 1982, pp. 94-106).

After the Il-khanids chose Marāḡa in Azerbaijan as their capital and they subsequently moved to nearby Tabriz, the court developed a strong interest in patronizing manuscripts. Scientific subjects evidently made a special impression on the Il-khanids, since the earliest known example is a translation from Arabic into Persian of a zoological text on the usefulness of organs and other body parts of animals (Manāfeʿ-e ḥayawān) ordered by Ḡāzān Khan after he ascended to the throne in 1295 (Pierpont Morgan Library, New York, MS M. 500). Although it is heavily restored and partially repainted, this manuscript, dateable to ca. 1297-1300 and copied in Marāḡa, is busily illustrated following the tradition of Arabic scientific texts that ultimately drew from Greek sources. As expected, at the close of the 13th century, its pictorial style is eclectic, including a less awkward insertion of East Asian elements than in previous codices; but it also shows the traditional approach to drawing, color, and composition that was present in the area before the arrival of the Mongols (Schmitz, pp. 9-24; Plate II).

PLATE II. "The Simorḡ,” Marāḡa, 697 or 699/ca. 1297-1300, illustration from the Manāfeʿ-e ḥayawān. Courtesy of The Pierpont Morgan Library, New York, Ms. M. 500, fol. 55r.View full image in a new tab

In addition to another copy of the Manāfeʿ-e ḥayawān, now dispersed and surviving in just a few folios, two other manuscripts created during Ḡāzān’s rule (1295-1304) or in the very first years of Öljeitü/Oljāytu’s (1304-16) show a similar eclectic approach. One is a partially surviving copy of the abovementioned ʿAjāʾeb al-maḵluqāt wa ḡarāʾeb al-mawjudāt, a text that had been conceived by Zakariyāʾ b. Moḥammad Qazvini toward the beginning of Il-khanid rule in Iraq in the 1260s-70s and compiled in honor of his governor and mentor ʿAṭā Malek Jovayni (British Library, London, MS Or. 14140). Here, the traditions upon which the three, possibly more, painters that contributed to the codex drew are even wider than in previous works, since they include those from the Jazira around the capital Mosul as well as those from southeastern Anatolia. They, however, represent the beginning of the blended, eclectic new style that became typically Il-khanid (Carboni, 1992; Plate III). This process was virtually completed in a copy of Abu Rayḥān Biruni’s Ketāb al-āṯār al-bāqia ʿan al-qorun al-ḵālia (known as “The Chronology of Ancient Nations”) dated 707/1307-8 and most likely copied and illustrated in the area of Tabriz or nearby northern Iraq (Edinburgh, University Library, MS Arab 161). This unique, illustrated codex, a treatise of calendar systems, is clearly a step forward in both the choice of subject matter and complexity of painted compositions. It demonstrates a specific Il-khanid interest in patronizing works that deal with different religions of the past and present, emphasizing the prominence of Islam above all the others and in particular of Shiʿite Islam, which became the official religion under the Il-khanids during the rule of Öljeitü in those very years. The last two illustrations in the manuscript, both of obvious Shiʿite content and the largest in the codex, make an everlasting impression. They can be regarded as a turning point in Persian painting in that their compositions demonstrate that the artists who conceived and executed them had come to terms with the new artistic language developed in the previous years and had finally blended all elements, the East Asian and the multiple indigenous ones, into successful works, thus creating a truly original style (Soucek; Hillenbrand, 2000; Plate IV).

PLATE III. "The tale of the river Nile,” Iraq, possibly Mosul, ca. 1295-1330, illustration from the ʿAjāʾeb al-maḵluqāt wa ḡarāʾeb al-mawjudāt. Courtesy of The British Library, London, MS Or. 14140, fol. 62v.View full image in a new tab

PLATE IV. "The investiture of ʿAli at Ghadir Khumm,” northern Iraq or northwestern Iran, 707/1307-8, illustration from the Ketāb al-āṯār al-bāqia ʿan al-qorun al-ḵālia. Courtesy of Edinburgh University Library, Ms Arab 161, fol. 162r.View full image in a new tab

In the same years as the making of Biruni’s Chronology, which may or may not represent a courtly commission, during the rules of Ḡāzān (1295-1304) and Öljeitü (1304-16) and in particular under the influence of their most powerful vizier, Rašid-al-Din Fażl-Allāh (d. 1318), the role of the élite in the creation of this new style and in bestowing royal status on the production of manuscripts becomes evident. This is the time when Ḡāzān commissioned the vizier to compose a history of the Mongols beginning from the life of Čengiz Khan. The initial single volume developed under Rašid-al-Din’s supervision into a four-volume work that represents the earliest world history ever written, a text in both Arabic and Persian that became known as the Jāmeʿ al-tawāriḵ or “Compendium of Chronicles.” The universal scope of the work was very much in step with the developments that were happening in the art of illustration, and it is not surprising that also its artistic sources were drawn from around the known world in order to illustrate new subjects: from China to Byzantium, from the Europe of the Franks and the Crusaders to Nepal and Tibet, from Buddhist, Christian, and Jewish models. In general, Chinese sources, mostly scroll painting, predominate (Plate V). It must be kept in mind, however, that only about two hundred pages with just over one hundred illustrations survive from the original Arabic work of 1315, corresponding to the second half of the second volume and therefore to about one-eighth of the entire work (Edinburgh, University Library, Ms. Arab 20; London, Khalili Collection, MS 727). If one considers that a second set of volumes in Persian was created in the same year and that Rašid-al-Din’s endowment stipulated that two illustrated copies were to be produced every year, one can surmise that we are allowed today only a glimpse of the complexity of the sources utilized to complete these colossal achievements. What, however, unifies these manuscripts is the Mongols’ sense of their role in history, whereby the rulers, warriors, and heroes of the past throughout Eurasia are represented in Mongol guise, an anachronism that served only one purpose, that is, the celebration of Mongol power (Rice; Blair, 1995).

PLATE V. "Mountains between Tibet and India,” illustration from the Jāmeʿ al-tawāriḵ, Tabriz, 714/1314-15. Courtesy of The Nasser D. Khalili Collection of Islamic Art, London, MS 727, fol. 261r.View full image in a new tab

Fragments from the Jāmeʿ al-tawāriḵ and other examples of paintings and drawings from the Il-khanid period are also preserved in three albums in the library of the Topkapı Saray in Istanbul (Hazine 2152, 2153, 2160) and in three out of five codices, known as the Diez Albums, in the Staatsbibliothek in Berlin (Diez A, foll. 70-72; Plate VI). The works that are now in Germany, collected by the Prussian Heinrich Friedrich von Diez at the end of the 18th century, were also compiled from material originally assembled at the royal Ottoman palace. This practice of arranging in albums miscellaneous examples of calligraphy, illustrations, and drawings detached from their original context actually started during the Timurid period, when works from the Il-khanid era were regarded as important models from the past. We are fortunate that these albums preserve Il-khanid works and help fill in the rather limited view of Persian painting from the 14th century (Ipşiroğlu; Roxburgh, 1995).

Figure 1. "The conquest of Baghdad” by Hülegü in 1258; illustration from the Diez Albums, Iran, 14th century. Courtesy of Staatsbibliothek zu Berlin–Preussischer Kulturbesitz, Orientabteilung, Diez A fol. 70, S. 7.View full image in a new tab

Illustrations that are clearly derivative from the format and the Chinese-inspired color-washed chromatic choices that are present in the Jāmeʿ al-tawāriḵ are also found in a poetic anthology copied in 1314-15 by a scribe from Kāšān (British Library, London, MS 132). Hardly of great artistic importance and rather repetitive, this manuscript testifies nonetheless to the popularity of that style of painting at the beginning of the 14th century (Robinson, pp. 3-10).

Like many other conquerors, the Mongols made an effort to find a place in history by adopting and reinterpreting events of the past. The Il-khanids found themselves in an ideal position to do so, because Persia’s literary past had preserved both the factual and the legendary history of the nation. The Il-khanids soon understood that, if they blended in a single work the power of the written word and the highest expressions of Persian poetry with the intensity and immediacy of illustrations, they would create an everlasting vehicle to transmit to posterity their legitimacy to rule over their new homeland. This vehicle was the Šāh-nāma, by then a classic of Persian literature, although there are no extant illustrated copies before the advent of the Il-khanids. It is highly unlikely that it was only under the Mongols that the first illustrated copies of the Šāh-nāma were made, but there is no doubt that the Il-khanids enthusiastically sponsored its production, particularly because they could reinterpret it through its images in order to make a powerful statement of legitimacy. The undisputed masterpiece of Il-khanid painting is a fragmentary copy of the book, known as the “Demotte Šāh-nāma” (q.v.) or, more recently, as the Great Mongol Šāh-nāma. What remains of the manuscript after several centuries and, especially, after its dismemberment at the hands of the dealer George Demotte in the early 20th century, is about one-third to one-fourth of the illustrations (according to the most recent reconstruction, 57 out of the estimated 190 survive, at present widely dispersed in public and private institutions worldwide) and much less of the original text, parts of which were also rewritten in recent times. Study of this codex is made even more difficult by the fact that some illustrations have been pasted over unrelated text and that a number of extant paintings have been restored. There is little doubt, however, that the surviving illustrations suggest a careful, deliberate, and sophisticated choice of subjects to be portrayed, almost as if every Il-khanid ruler could be identified with a sovereign or a hero of the Iranian past and thus carry the epic significance of the Šāh-nāma into contemporary times and serve a specific agenda. From an art historical point of view, many of its illustrations are true masterpieces as regards to composition, fully and successfully integrated elements from different pictorial traditions, interpretation of the subject, craftsmanship, and use of color. Since the scenes are transported into Mongol times, the attention to architectural details and the illustration of portable objects are extremely useful to contribute to the scant information we have on the material culture of the royal Il-khanid milieu. Court and funeral scenes help us to understand Mongol customs and manners as adapted to the Iranian environment (Plate VII). Some of the scenes are memorable because they express a kind of pathos, unbridled emotions, or simply a sense of action and dynamism that are unique to this phase of Persian painting (Plate VIII). The fragmentary condition of the manuscript is especially unfortunate, because its beginning and end are lost and therefore its patron or dedicatee and calligrapher and/or painters are unknown. Scholars largely agree, however, that it was produced toward the end of Il-khanid rule under the reign of Abu Saʿid Bahādor Khan (1316-35, q.v.) or perhaps shortly after his death, that it was a royal manuscript, that several artists worked on its illustrations in an established atelier, and that it can be regarded as one of the greatest expressions of the arts of the book in Persia (Grabar and Blair; Blair, 1989; Soudavar, 1996; Blair and Bloom).

PLATE VII. "The Funeral of Isfandiyar,” illustration from the Great Mongol Šāh-nāma, Iran, probably Tabriz, ca. 1330s. Courtesy of The Metropolitan Museum of Art, New York, Joseph Pulitzer Bequest, 1933, inv. 33.70.View full image in a new tab

PLATE VIII. "Rostam slaying Šagad,” illustration from the Great Mongol Šāh-nāma, Iran, probably Tabriz, ca. 1330s. Courtesy of The British Museum, London, inv. 1948.12-11.025.View full image in a new tab

The popularity of the Šāh-nāma as an illustrated text in the Il-khanid period is evident from a number of small-format copies, two of them dispersed and known as “The First Small” and “The Second Small Shahnameh” and one known as “The Gutman Shahnameh” (New York, Metropolitan Museum of Art, inv. nos. 1974.290, 2003.330), which include lively compositions and consistent use of gold, but which were clearly made for patrons less affluent than the royal family and perhaps were not even specially commissioned. Their lack of a surviving colophon and the relatively modest attention to details make their place of manufacture and date rather uncertain, with suggestions ranging from Baghdad at the beginning of the 14th century (Simpson, 1979) to northwestern Persia around 1330, except for the Gutman manuscript, which has been convincingly attributed to an atelier in Isfahan in around 1335, when the city was indirectly under Il-khanid control (Swietochowski and Carboni, pp. 67-145).

That Isfahan was indeed a center of production of illustrated codices is proven by internal evidence in an anthology of poetic artifices compiled by a self-appointed poet, Moḥammad b. Badr-al-Din Jājarmi, who lived most of his life in Isfahan and finished his collected works in 1341 without a specific dedication (Dār-al-Āṯār al-Eslāmiya, Kuwait, MS LNS 9). The codex, however, includes a double frontispiece that seems to suggest that it was indeed made for a princely Il-khanid couple. One of its chapters, now dispersed, is also heavily illustrated in simple, didactic compositions with red backgrounds dictated by the nature of the poetic text (Swietochowski and Carboni, pp. 8-66). Red or yellow backgrounds and other details in the frontispiece relate this provincial Il-khanid production to a series of manuscripts, the earliest of which is perhaps a copy of the Kalila wa demna dated 707/1307-8 (British Library, London, MS Or. 13506; Waley and Titley), which were produced in Shiraz under the Injuids and are therefore outside the scope of the present discussion. It is not surprising, however, to notice a mutual influence in two cities that were relatively close to each other but far from the centers of Il-khanid power in northwestern Persia.

The last dated, illustrated manuscript in a clear Il-khanid style, completed well after the demise of the dynasty, is a copy of the epic poem Garšāsp-nāma (q.v.) of Asadi Ṭusi finished in 1354 (Topkapı Saray Library, Istanbul, MS Hazine 774). In its five paintings, the compositions are as sophisticated as some of the best earlier Il-khanid illustrations, whereas their vivid, bold chromatism shows that the painter was moving forward to the developments that would take place in Baghdad and Tabriz under the new rulers, the Jalayerids (r. 1336-1410; Ettinghausen, 1959, pp. 60-65, Carboni, 2002, p. 216). It was indeed at the beginning of Jalayerid rule that a new style started to develop parallel to the Il-khanid, while at the same time drawing inspiration from it. The earliest surviving illustrations are in the double-page frontispiece of an alchemical treatise dated 739/1339 (Topkapı Saray Library, Istanbul, MS Ahmet III 2075; Farès, pp. 156-60; Grube, 1978, pp. 18-19; Berlekamp). The last flourishing of Il-khanid painting and the first steps of the new Jalayerid style correspond to the years during which, according to Dust Moḥammad two centuries later, the master painter Aḥmad Musā changed the style of Persian painting while working for both Il-khanid and Jalayerid rulers, thus lifting “the veil from the face of depiction” (Chagtai; Thackston, pp. 3-17). The very fragmentary picture of Il-khanid book illustration that survives today seems sufficient to confirm this assessment of Persian painting in the 16th century.

IL-KHANIDS iv. Ceramics

This entry deals with glazed wares and tiles of the so-called “Sultanabad” (Solṭānābād) group, lajvardina (< Pers. lājvard “lapis lazuli”) wares, and luster wares produced in the Il-khanid period. The period extends from the fall of Baghdad in 1258 to the last dated luster tiles made in 1339 (dated pieces are listed by Watson, 1985 and Ettinghausen, “Dated Faience”). Changes in the style, technology, and iconography of tile revetments can be traced at Taḵt-e Solaymān (q.v.), Tabriz, Solṭāniya (modern Arāk), and several tomb complexes. Typical of changes affecting vessels are decorative, iconographic, and stylistic innovations from the Far East copied from textiles (Lane, 1971, pp. 3-6) and imported Chinese ceramics (Morgan, 1991). Usually dismissed as “chinoiserie,” dragons, phoenixes, lotus blossoms, and cloud collars motifs were, in varying degrees, closely controlled symbols of Mongol imperial power. Other, largely technical, changes in ceramics were brought about by the exchange of skills between artisans drawn from different craft traditions, who were working together on Mongol architectural projects. A decline in the use of nesbas in the Il-khanid period suggests small-scale craft production diminished as major tile work commissions increased. Advances in mosaic tile work at the end of the period were probably undertaken by artisans trained in Anatolia.

Production. Despite the lack of excavated domestic sequences, earthenware and glazed earthenware vessels remained largely unchanged, whereas stonepaste vessels and tile work were more sensitive to Far Eastern influence. The body of stonepaste wares (soft-paste porcelains, faience or frit wares in earlier works) is an artificial mixture of frit (glass), white clay, and silica. The recipe given in Abu’l-Qāsem Kāšāni’s text of 1301 (Allan; Ritter et al.) is confirmed by scientific analysis (Dayton and Bowles; Mason). The treatise deals only with stonepaste wares, particularly vessels of “two firings,” that is, overglaze painted and luster wares, and with underglaze painted wares. Kāšāni cautions against smoke in the kiln, recommending the use of saggars to protect overglaze decoration; to counter the smoke problem, artisans of Kāšān burned soft firewood (hizom-e narm), while barkless willow wood was used in Tabriz, Baghdad, and other cities (Kāšāni, p. 346). Although production in Tabriz is to be expected, it is not the case in Baghdad, where tiled monuments are scarce, as are suitable mineral resources. His remark that haft rang overglaze painted mināʾi wares had disappeared by his time is largely confirmed by the absence of dated pieces after the last one was produced in Ṣafar 616/April 1219 (see Watson, 1994, for a list of dated mināʾi wares). A restricted mināʾi technology continues in lajvardina wares—cobalt blue glazed vessels and tiles decorated in a limited range of overglaze colors (red, black, and white) in a glass-based matrix, and gold leaf fixed in a muffle kiln. Kāšāni refers briefly to underglaze painting but does not mention underglaze red used on Syrian Euphrates wares.

Lajvardina wares. There are few dated Il-khanid lajvardina objects. Richard Ettinghausen listed a large star tile dated 1315 (Ettinghausen, “Dated Faience,” p. 1691) and a smaller example dated 1304 (ibid., pp. 1666-96). A lotus bowl in the Preussisches Kunstbesitz, Berlin, said to have been found at Nišāpur, is dated Rajab 776/December 1374-January 1375 (Klein et al., eds., no. 222). The decoration of lajvardina wares is almost exclusively non-figural, and the remains of disassembled lajvardina tile work come largely from funerary contexts (Plate I). There are numerous frieze tiles with Koranic inscriptions. A large inscriptional tomb tile in the form of a prayer niche (meḥrāb) is inscribed with the name of an unidentified vizier, Jalāl-al-Din Esmāʿil (British Museum OA G, 1985, 500; see Porter, fig. 26). A comparable tile published by Friedrich Sarre may have come from Solṭā-niya (Sarre, fig. 87), where fragments of lajvardina star tiles have been found in the funerary area east of the Gonbad-e Oljāytu/Öljeitü and elsewhere on the site. Fragments of large figural plaques with hunting scenes were excavated at Taḵt-e Solaymān (Masuya, fig. 98, cat. no. 94). A large (51 x 45 cm), unprovenanced inscriptional lajvardina frieze tile with a dragon and phoenix border, almost certainly from Taḵt-e Solaymān (British Museum OA 1896.3-13, 32; Porter, fig. 29), adds weight to the suggestion that the lajvardina style was created for Abaqa Khan’s palace at Taḵt-e Solaymān. Its choice may mirror restrictions on the distribution of blue clothing in China to members of the Yüan court (Morgan, 1995). The extensive use of gold leaf can be compared with the Il-khan’s unrestricted use of gold in drinking ceremonies, in elixirs, religious contexts, jewelry, and brocades. Cobalt blue colorants were amongst the most costly materials used by potters.

PLATE I. Lajvardina tile ca. 1310. Reżā ʿAbbāsi Museum, Tehran.View full image in a new tab

The last significant use of lajvardina is in the tile work of the tomb of Qosām b. ʿAbbās in the Šāh-e Zenda in Samarkand (Tomb A), dateable to about 1335. The aesthetic continued in tombs using both fired gold leaf or gold stencils and transfers: in the Timurid tombs in Samarkand, in the 15th century Masjed-e Moẓaffariya in Tabriz, in the mosaic tile portal of the tomb of Shah Neʿmat-Allāh Wali in Māhān, Kermān, and in the tomb chamber of Shah Esmāʿil I Ṣafawi in Ardabil.

Lajvardina vessels have been excavated at Taḵt-e Solaymān, Solṭāniya, and Saray Berke (the capital of the Golden Horn on the Volga); the sources of most museum pieces are unknown. Taken en masse, forms are limited, and closed vessels are copied from pre-Mongol inlaid metalwork forms, as are the decorative panels of “Y” and “T” frets. Open forms are copied from L’ung chuan celadons from southern China, particularly flat-rimmed dish and hemispherical (lotus) bowls that were produced in three sizes. The range of lajvardina bowls and dishes is similar and the usually radial decoration expanded to fit. The reduction in formal variety may be evidence of new modes of production that in China saw a decline in quality as craft production was overtaken by export-driven factory production. A piriform jug is the only lajvardina vessel that can be traced to a nomadic prototype, a strip-sewn leather bottle for holding liquids (Plate II).

PLATE II. Lajvardina jug. Museum für Islamische Kunst, Berlin.View full image in a new tab

Blue and white tiles. The white relief decoration of the blue and white star, frieze, and tomb tiles used to decorate tomb walls and cenotaphs is often covered with gold leaf. The earliest surviving blue and white cenotaphs are those of the Saljuq rulers of Rum in Konya made in the early 1220s. Cobalt (q.v.) is a flux and tends to run into the glaze and diminish decorative clarity. To counteract this failing, the cobalt colorant was often applied in a compartmentalized fashion around raised inscriptions and decorations. A Kāšān luster bowl with a blue and white inscription decorated by Abu Zayd (Bahrami, pp. 37, 39) is the earliest dated use of the technique (Šawwāl 609/March 1213). At Taḵt-e Solaymān the blue and white of tiles with molded decoration is frequently masked by overglaze decoration and gold leaf. The inscription around Ḡāzān’s tomb in Tabriz, said by the historian Šehāb-al-Din ʿAbd-Allāh Waṣṣāf to have been blue, was probably made of blue and white tiles (Wilber, 1955, p. 126, for field evidence from Šenb, collected in 1937 and 1939). Fragmentary blue and white star tiles and cavetto moldings from tomb contexts have been found at the Rabʿ-e Rašidi in Tabriz (survey collections, British Institute of Persian Studies, Tehran) and in and around the Gonbad-e Öljeitü in Solṭāniya, and similar star tiles without inscriptional borders were used in the decoration of an Il-khanid building at Bisotun, possibly the site of Solṭānābād Čam-čemāl (the second capital built by Öljeitü at the foot of Bisotun mountain; see Luschey-Schmeisser, pp. 221-40).

Many unprovenanced star tiles and frieze tiles are probably from cenotaphs. Three large dated tomb tiles are known: Jomādā II 711/October-November 1311 (Klein et al., no. 185; 719/1319-20, Wiet, pl. II, no. 179; 722/1322, Metropolitan Museum, no. 37; see Plate III). The second example made by ʿAli b. Abi Ṭāleb b. Abi Naṣr and a foundation plaque in the Masjed-e Kala in Qohrud, a village between Kāšān and Isfahan, made by Yūsof b. ʿAli, the potter of Kāšān in 710/1310 (Watson, 1975, pl. VIIIb; Qučāni, 1992b, figs. 38-40) suggest that they are all products of Kāšān or its satellite ateliers. Large blue and white plaques with cloud-collars and lotus blossoms which once decorated the mausoleum of Bāyazid Besṭāmi (q.v.) in Besṭām (see Survey of Persian Art, pl. 350, for a distant photograph) and the blue and white chinoiserie tiles of the cenotaph of Qosām b. ʿAbbās in Samarkand, around 1335 (Bulatova et al., pl. 121), suggest that overglaze decoration was abandoned in religious structures in the reign of Ḡāzān Khan (q.v., 1295-1304). Blue and white tile cenotaphs, typified by conspicuous chinoiserie decoration, were used to commemorate religious figures in later 14th-century tombs in Ḵiva and Organj.

PLATE III. Part of a blue and white inscriptional tile dated 722/1322. Metropolitan Museum of Art, New York. Available at www.metmuseum.org.View full image in a new tab

For technical reasons, a tradition such as that of Yüan China, using multiple shades of blue, did not fully develop in Persia until the Safavid period. Persia’s contribution to the development of Chinese blue and white (after ca. 1325) lay in the export of cobalt blue colorants.

“Syrian” wares. The underglaze decorative techniques described by Abu’l-Qāsem Kāšāni must refer to the so-called “Syrian” wares (Hobson, fig. 73 and p. 65). “Syrian” wares were decorated with black (siāh qalam, mozarrad, the latter so-called after the name of a deep black stone found in Khorasan; see Kāšāni, pp. 340, 345), probably chromite, cobalt blue, and manganese purple. In the decoration of related laqabi wares produced in Syria before 1220 (?), sgraffito outlines had been used to control underglaze colors. The same colors, used to decorate tiled cenotaphs in the Sayfiya Madrasa in Sivās, Anatolia (4 Šawwāl 617/1 December 1220; Meinecke, II, pp. 435-36), continued in use until about 1240. This style’s Syro-Anatolian antecedents informed Robert Hobson’s choice of terminology, but the weight of evidence suggests that they were made in Persia, perhaps by or under the influence of Anatolian or Syrian craftsmen. Despite some suggestion of later tampering, most provenanced examples were found at Taḵt-e Solaymān; and the earliest group was made during the period of the site’s restoration: a lotus bowl (667/1268-69) in the Chicago Art Institute (Pope, “The Ceramic Art,” p. 1637), a pilgrim flask (Rajab 671/January-February 1273) in the Reżā ʿAbbāsi Museum in Tehran (Plate IV), and a pedestal bowl (Ramażān 672/March 1274) in the Victoria and Albert Museum in London (Ettinghausen, 1935, pl. 13). Other pieces include a large dish (706/1306-7, Plate V) in a private collection in Geneva (Soustiel, no. 263), a unique plaque (722/1312-13) in the David Collection in Copenhagen (von Folsach, no. 151), and a bowl (729/1329) in the Royal Ontario Museum (Lane, 1947, pl. 95A). The typical decoration includes reserved inscriptions, hares, jackals, and interlaced motifs, but there are no human figures. Forms are usually related to Islamic metalwork, Chinese celadons, and Mongol metalwork and are unusually small. Many inscribed pieces have Arabic blessings in relief white on a blue ground.

PLATE IV. "Syrian” ware, a pilgrim flask dated Rajab 671/January-February 1273. Reżā ʿAbbāsi Museum, Tehran.View full image in a new tab

PLATE V. Large “Syrian” ware dish (706/1306-07?). Private collection, Geneva.View full image in a new tab

Sultanabad wares. When Pope undertook an archeological survey of villages near Solṭānābād (modern Arāk) in the 1930s, he concluded that they were the nucleus of a major Il-khanid ceramic industry and that commercial excavations had uncovered complete vessels in a buried shop. These so-called “Sultanabad” wares include Colored Ground wares, ʿErāq wares, Black underglaze painted turquoise glazed wares, and blue and black Panel wares.

The underglaze blue, black, and turquoise colors used in Panel wares divide the surface into radial or vertical panels filled alternately with reserved inscriptions, interlaced palmettes, and dotted or hatched fields, a style in which Arthur Lane saw the influence of Mongol brocades such as the Heinrichsgewänder preserved in the treasury of the Alt Kapelle in Regensburg. The style, also popular in Syria and Egypt, is found on only one group of tiles, on spolia in the doorway of the tomb of Shaikh Aḥmad in Jām (Khorasan). Black underglaze painted turquoise glazed ware tiles were used to decorate a cenotaph in the Masjed-e Kermāni next to the tomb of Shaikh Aḥmad and also are on the cavetto moldings of the tomb of Qosām b. ʿAbbās. Although Black underglaze painted turquoise vessels were developed before the Mongol period, notably on a range of fenestrated ewers produced in Kāšān around 1200-1220, this ware became more popular under the Il-khans; and complete pieces and shreds have been found at Taḵt-e Solaymān. A dated bowl (Rabiʿ I 676/August 1277) in the Victoria and Albert Museum (C. 350-1929) was probably made in Kāšān; and another one (760/1358-59) was made by ʿAli Zāhed, a potter from Ḵodāšāh, a small town about 50 km northwest of Sabzavār (Karimi and Kiāni, no. 72). Firm evidence of production at Solṭā-niya is provided by wasters (whereabouts unknown) recovered in the 1930s (Talbot Rice, pl. III), and the technology seems to have been widely dispersed after about 1325. The decoration of most pieces consists of stylized vine leaves and bunches of grapes, which, coupled with the relatively large number of storage jars in the same style, suggests that they belonged to wine sets in imitation of Chinese wares.

Colored Ground wares and ʿErāq (i.e., ʿErāq-e ʿAjam) wares probably belong to the same production cycle; and, although forms are similar, each decorative technique is dominated by a distinctive iconography. Colored Ground wares are coated with a slip usually ranging from grey to purple, to which the decoration is applied in thick white relief paint with additional fine details added in black. Forms include monumental jars, open celadon-like forms, bowls with vertical rims, jugs, and wine ewers. The borders of bowls and dishes are often decorated with pheasants or other birds (Plate VI) and occasionally with water birds pursued by hawks, a theme repeated on ceramics and metalwork found at Karakorum and one of the few specifically “Mongol” elements in the decoration of Il-khanid wares. The tondo is normally decorated with hares, deer, and Mongol figures, often wearing feathered hats and seated on a folding stool often depicted in contemporary illustrations of Rašid-al-Din Fażl-Allāh’s Jāmeʿ al-tawāriḵ. A unique bowl in the British Museum is decorated with several seated figures alternately wearing turbans or feathered hats, with a standing camel reserved on a blue ground at the center (British Museum OA, 1928, 1-21, 1). A short-tailed bird is a common subject and was widely copied on luster tiles painted in the “Sultanabad” manner (Plate VII). Archeologically attested Colored Ground wares are rare; a few sherds were found at Taḵt-e Solaymān and in surveys from Sistān and the Persian Gulf. Production probably began quite late, perhaps during the reign of Öljeitü, a supposition supported by two dated pieces. The first, dated Jomādā I 713/August-September 1313, is in the Bāzargān Collection of the National Museum in Tehran (Qučāni, 1992b); and the other (716/1316-17) is in the Ṭāreq Rajab collection in Kuwait (Fehérvári, no. 47A-B). No tile work is known, despite attempts to identify molded luster tiles in situ on the soffit of the arch over the tomb chamber in the Pir-e Bakrān in Lenjān as Colored Ground. Although many vessels carry pseudo-inscriptions around the rim, only one legible poetical inscription has been located (Oxford, Ashmolean Museum, 1978.1619), a fact that may help to identify potential purchasers as non-Persian (Morgan, 1995, pls. 1 a-c). The recurring decorative themes certainly suggest that the clientele was Turco-Mongol.

PLATE VI. Colored Ground ware bowl. Reżā ʿAbbāsi Museum, Tehran.View full image in a new tab

PLATE VII. Panel of molded luster star tiles with flowers and birds in flight and Koranic inscriptions. © Victoria and Albert Museum, London.View full image in a new tab

ʿErāq wares, a dealer’s term, appeared in Europe following their discovery during house building “near Tehran” and entered museum collections in the 1880s. The dominant colors are cobalt blue, turquoise, and black, which are applied beneath a transparent colorless glaze. Some objects are of high quality, and other evidence indicates investment in technical innovation in which cobalt blue is, for the first time, successfully modified to produce a pale blue underglaze colorant (Plate VIII). The decorative subject matter is wide ranging, but deer, hares, and pheasants are well represented, as in Colored Ground ware, although again there are no legible inscriptions. Similarly, there are no ʿErāq ware tiles, and only two installations where blue inscriptions are applied on a turquoise ground are known. The first is an unprovenanced sequence of frieze tiles with quotations from the Šāh-nāma (Melikian-Chirvani, figs. 51-62), and the second is a dated external panel (725/1325) at the base of the minaret of the tomb and ḵānaqāh of Shaikh ʿAbd-al-Ṣamad in Naṭanz.

PLATE VIII. ʿErāq ware lotus bowl with radial designs. Fitzwilliam Museum, Cambridge.View full image in a new tab

Turquoise glazed vessels were probably widely produced; but, with the exception of sherds from Taḵt-e Solaymān, information is scant. Monochrome turquoise tiles were made at Taḵt-e Solaymān, and tomb tiles were manufactured elsewhere for both emāmzādas (q.v.) and private citizens. Iraj Afšār’s work on the monuments of Yazd has shown that molded tomb tiles were personalized by the addition of the deceased person’s name and that they may well have been made in Yazd. The earliest Yazd tomb tile (1257) is located in the Friday Mosque of Haftādor (Afšār, I, pp. 52, 472). Qom is equally rich in such pieces; the earliest (Moḥarram 667 /September-October 1268) is from the tomb of Šāh Jaʿfar, a grandson of Imam Musā al-Kāẓem (Modarresi Ṭabāṭabāʾi, II, pl. 15, p. 38). Monochrome turquoise tomb tiles are found throughout Persia; turquoise glazed meḥrābs are less common.

Luster ware. The output of dated luster tiles declined after the death of Jalāl al-Din Ḵᵛārazmšāh in 1227. A meḥrāb in the Āstān-e Qods-e Rażawi (640/1242) was made by ʿAli b. Moḥammad b. Abi Ṭāher at a time when Khorasan was governed by a Uighur Körgüz, a Muslim convert from Buddhism, and it indicates that Kāšān luster potters could, under favorable circumstances, produce tiles for religious institutions. After Hülegü’s arrival, luster production increased visibly, and two monumental luster jars (the Hermitage and Hirsch jars) may be dated to the 1260s. The same decade saw the manufacture of large, non-figural star and cross tiles with Koranic inscriptions (October 1261 and January 1263) for the Emāmzāda Yaḥyā in Varāmin (London, 1976, no. 379) and figural star and cross tiles with poetic inscriptions for the portal of the Emāmzāda Jaʿfar in Dāmḡān (1267). The imagery and inscriptions of these are repeated on numerous star and cross tiles from Taḵt-e Solaymān made between 1269 and 1274 (Qučāni, 1992b, pp. 37-40). Thereafter, cross tiles are plain or molded turquoise and cobalt blue (but see an undated series in the Reżā ʿAbbāsi Museum, ca. 1340). Although non-figural tiles generally have Koranic inscriptions and figural tiles poetical inscriptions, a set of unprovenanced tiles with identical birds in flight (Plate VII), which have Koranic inscriptions, are probably from the tomb of a shaikh like those from the Pir-e Bakrān and the tomb of ʿAbd-al-Ṣamad in Naṭanz. Birds, the most common motifs in such contexts, are linked metaphorically to the spiritual quest, as outlined in Farid-al-Din ʿAṭṭār’s Manṭeq al-ṭayr. Luster tiles with hawks attacking birds may be compared with the hawk and dove simile used in the context of the challenge to Islam posed by Buddhist monks (baḵši, q.v.) and shamans, rather than viewed as simple hunting scenes (Digby). The animal subject of star tiles from Dāmḡān and Taḵt-e Solaymān may be given a quasi-religious interpretation by reference to Rumi’s reinterpretation of animal fables and popular stories. Although dragons and phoenixes are commonly associated with the new iconography, dragons usually do not appear on Persian vessels, with the exception of one tile other than those from Taḵt-e Solaymān.

IL-KHANIDS v. Carpets

IL-KHANIDS viii. CARPETS. See CARPETS viii

IL-KHANIDS vi. Clothing

See CLOTHING ix