Ilkhanate

A standing army sent west

Drafted from Jackson, The Mongols and the Islamic World: From Conquest to Conversion (2017)formation

The Ilkhanate is the only one of the four Mongol successor states that was created on purpose. The other three are inheritances that hardened into kingdoms. This one was a commission.

The assignment

In 1251 the assembly that made Möngke Great Khan resolved on two great campaigns: one against the Song in the south, under Kublai, and one against the west, under their brother Hülegü.

The western brief was specific. Reduce the Ismaili strongholds in the Elburz. Bring the caliph in Baghdad to obedience. Then go on to Syria and Egypt. Hülegü was given, by the accounts, two men in ten from the whole empire's forces, and a train that included Chinese siege engineers — the empire's most valuable technical asset, moved four thousand miles because a campaign required it.

He left in 1253 and travelled for two years.

Alamut

The Nizari Ismailis had held a chain of mountain fortresses in northern Iran for a hundred and sixty years, and no army had ever taken them. Their method was to be impossible to besiege and dangerous to leave alone: castles on crags with their own water and granaries, and a practice of political assassination that had made every Seljuk and Khwarazmian ruler negotiate rather than campaign.

Hülegü campaigned. The fortresses fell one at a time through 1256, and in November the Imam Rukn al-Din surrendered Alamut itself and was sent east to the Great Khan — and killed on the way.

The library at Alamut was catalogued before it was burned by Juvaini, who was Hülegü's own secretary and who preserved in his history what he chose to preserve of a doctrinal tradition he had just helped destroy. Almost everything known about the Nizaris of Alamut comes from the men who took it.

What kind of state this was going to be

Two things about the founding shaped everything after.

The first is that Hülegü was not conquering for himself. He was an imperial army commander with a mandate, and the territory he took was the Great Khan's; the title il-khan means subordinate khan, and the coinage struck in Iran carried the Great Khan's name for decades. It is the only one of the four successor states that formally acknowledged a superior, and it went on doing so after the superior was three thousand miles away and out of touch.

The second is that he arrived with the whole apparatus of empire behind him rather than as a prince taking a share. That is why the Ilkhanate looks different from the Golden Horde: it inherited a functioning Persian administrative tradition, kept it, staffed it with Persian bureaucrats, and produced within two generations the most sophisticated state in the Mongol world.

What it did first, though, was destroy the institution the Islamic world had been organised around for five hundred years.

The eagle's nest

Drafted from Howorth, History of the Mongols from the 9th to the 19th Century, Part III: The Mongols of Persia (1888)expansion

The first thing Hülegü was sent to do was destroy a state that had made itself invulnerable by making itself unnecessary to attack — and that everybody had nevertheless been afraid of for a hundred and sixty-six years.

What Alamut was

The Nizari Isma'ilis held a chain of mountain fortresses across northern Iran and Syria, of which Alamut in the Elburz was the chief. They had no field army, no cities and no revenue worth the name. What they had was the ability to put a man next to anybody.

Two Abbasid caliphs, a Seljuk vizier, princes, generals and the ruler of Jerusalem had died at their hands; the Crusaders, who met the Syrian branch, brought home the word assassin and an etymology — from hashish — which Howorth repeats and which has never been substantiated. The Khwarazmshah Jalal al-Din had campaigned against them and got nowhere. Every power in Iran had at some point paid them or negotiated with them, and none had taken their castles.

They were also, and this is the part usually left out, a place of learning. The library at Alamut was one of the great collections of the Islamic world, and the most learned man of the age, Nasir al-Din al-Tusi, was living there — having left Baghdad in a rage after the caliph's vizier sneered at one of his poems and the caliph had it thrown into the Tigris.

Why they were first

Because the great khan had been asked.

Envoys from the Isma'ilis had appeared at the assembly in Mongolia, and the western campaign that Möngke gave to his brother had two objects listed in order: destroy the Isma'ilis, then deal with the caliph. Qadis and lawyers from Qazvin had been petitioning the Mongol court for years about neighbours they could not defend themselves against.

A sect that survived by being impossible to besiege had spent a century and a half making itself the one enemy everybody agreed on.

How it ended

Not in a great siege. Hülegü brought up Chinese engineers and mangonels, took the outlying castles one at a time, and negotiated.

The Isma'ili ruler, Rukn al-Din Khurshah, surrendered in 1256 on terms and then ordered his own garrisons to give up their fortresses, which most of them did. He was kept alive and used, sent from castle to castle to talk the commanders out, and when there were no more castles left he was sent east to the great khan, who is said to have asked why he had been brought so far for nothing, and had him killed on the road home.

At Alamut the library was burned. Tusi was taken into Hülegü's service, where he advised on the campaign against Baghdad and afterwards built the observatory at Maragha.

Howorth's historian Von Hammer notes the coincidence that in the same year two of the most famous libraries in the East were destroyed — the one at Alamut, and the one at Medina, which burned in a fire and was then plundered by the Arabs.

What it proved

That the thing which had protected the Isma'ilis was not their castles but the unwillingness of any Muslim power to spend a decade on them.

The Mongols were willing. They had siege engineers from China, no local politics to manage, a general with one instruction, and no reason to make terms. It took about a year to end an institution that had outlasted the Seljuks, and it demonstrated to everyone west of the Elburz — including the caliph, who had refused to send troops to help — what was now possible.

Two years later they found out what it meant.

The master of this house

Drafted from Howorth, History of the Mongols from the 9th to the 19th Century, Part III: The Mongols of Persia (1888)expansion

How the Abbasid caliphate actually ended is better documented than almost any other event of the century, and the documentation is mostly about a quarrel over a poem, an argument between two astrologers, and a joke about a beard.

The riot at Karkh

The village of Karkh, outside Baghdad, was almost entirely Shia. In 1256 a dissension broke out between its people and the Sunnis, and the caliph's troops, under his dawatdar, sacked it: the houses of the descendants of Ali robbed, the people of the Prophet's own clan taken prisoner, and the women carried through the public streets on horses' cruppers with their faces and feet bare.

The caliph's vizier, Ibn al-Alqami, was himself a Shia. He wrote to the chief of Hilla, the great centre of Shia scholarship, complaining that the dishonour done to Husayn's household had been done again. The reply he received is the hinge of the whole episode:

The heretics must be put to death and destroyed, and their race be uprooted. If you will not side with us you will be lost. You will be despised in Baghdad, as henna, which delights women, is despised by rough men, and as a ring is despised by him who has had his hand cut off.

The vizier then wrote to Hülegü, who had just taken Alamut, pointing out the weakness of Baghdad and inviting him to come.

He had already prepared the ground. He had persuaded the caliph that with so many powerful vassals he had no need of such a large force, that the money saved could buy off the invaders, and that the army should be cut from a hundred thousand men to twenty thousand.

The astrologers

Hülegü was not confident. The caliph's troops had beaten Mongol forces twice already, and he consulted the astrologer Husam al-Din, whom the great khan had sent with him.

Husam al-Din, a Muslim, predicted six consequences of attacking the House of Abbas: all the horses would die and the soldiers be struck with pestilence; the sun would not rise; rain would not fall; there would be hurricanes and earthquakes; plants would cease to grow; and the emperor would die within the year.

Hülegü made him write all of it down.

Against him stood the Mongol shamans, the amirs, and Nasir al-Din al-Tusi — the astronomer taken out of Alamut, a Shia, and a man with a grievance of his own. He had sent the caliph a poem; the vizier had written a sneering note on the back of it about the composer's knack for putting other people's thoughts into his own writing; and the poem had ended in the Tigris. He advised that the expedition would succeed.

Hülegü sent his summons from Hamadan. He denounced the caliph for not helping against the Isma'ilis, reminded him what had happened to the Khwarazmshahs, the Seljuks, the rulers of Daylam and the atabegs — all of whom had at some point been masters of Baghdad — and asked why its gates should be shut to him. Do not strike with your fist against an iron spike, nor mistake the sun for a taper.

Let us see what you can give us

The city fell in February 1258. The Georgians in Hülegü's army, who had their own long account to settle, breached the walls; the Georgian chronicle claims they opened the gates for the Mongols.

On the tenth of February the caliph came out with his three sons and three thousand others — sayyids, imams, judges, grandees. On the fifteenth, Hülegü went into the city to look at the palace, had the caliph brought to him, and said:

You are the master of this house and I am your guest. Let us see what you can give us.

The caliph broke some locks and offered two thousand sets of robes, ten thousand gold dinars and a quantity of jewels. Hülegü would not take them. It is unnecessary to point out what is patent; disclose your hidden treasures. The caliph told them where to dig, and they found a cistern full of gold pieces of a hundred miskals each.

Wassaf, the Persian historian, records what the soldiers got: gold and silver vessels from the caliph's kitchens handled as though they were lead, stuffs and goods of Greece and Egypt and China, Arab horses and mules, Greek and Alan and Kipchak boys, Turkish and Chinese and Berber slave girls, past counting. Men broke their swords off at the hilt and filled the scabbards with gold. Enough of it reached Shiraz that the wretchedly poor there became rich.

The caliph's harem was counted: seven hundred wives and concubines and twelve hundred servants. He asked for a hundred of the women back, on whom the sun and moon had never shone, and was allowed to choose; he chose his relatives.

The two basins

Wassaf tells one more story about the treasure, and it is the best thing in the account.

The caliph al-Nasir had left behind two cisterns filled with gold. His grandson al-Mustansir was sitting one day with a trusted companion and said he hoped not to live long enough to have to spend it, and the companion laughed. Asked why, he said that he had once come into the grandfather's presence in that same room when one of the basins was not yet full, and the old man had said he wished to live only until he had filled them both.

Al-Mustansir spent all of it on good works, including the college that carries his name.

The point of this story, Wassaf adds, is that when Musta'sim came to the throne he once more filled up these basins by his avarice, and finally emptied them as well.

The beard

Hülegü might well have spared him. The Muslims around him warned that if the caliph's blood were shed on the ground there would be an earthquake, and Husam al-Din said so again.

Tusi answered that no such portents had occurred when John the Baptist was killed, nor the Prophet Muhammad, nor the Imam Husayn, and that they were not likely to happen now. Others argued that a living caliph would raise every Muslim in the army and outside it. Asked for his advice, the vizier — who had brought the whole thing about — replied only, the vizier has a long beard, repeating a jibe that had been used against him, from the Arab proverb: long in beard, short in wit.

The caliph asked leave to make his ablutions and objected to the five Mongols sent to escort him. He recited a verse:

In the morning we dwelt in a house like paradise or heaven, In the evening we had no longer a dwelling as if we had been homeless.

On the twentieth of February he was put to death at the village of Waqf with his eldest son and five eunuchs who had stayed with him. Rashid al-Din does not say how, and his editor suggests the method was deliberately kept secret.

The Georgian chronicle says that when he was brought in and ordered to kneel he refused, and stood, and said: I am an independent sovereign, who am dependent on no one. If you choose to set me free I will submit to you; if not I will die before becoming any man's slave.

Husam al-Din the astrologer was executed four years later, because his prophecies had proved false.

What the office was holding

Drafted from Jackson, The Mongols and the Islamic World: From Conquest to Conversion (2017)expansion

In February 1258 the Mongols took Baghdad and killed the Abbasid caliph. In September 1260 a Mamluk army destroyed a Mongol force at Ain Jalut in Galilee. Thirty-one months separate the two, and between them they fixed the shape of the Middle East for the next two and a half centuries.

The office

Baghdad had been the caliphal capital since 762, and by the thirteenth century the caliph's actual authority ran over Iraq and not much else. What made the city worth taking was not its army and not even its treasury. It was that the office was the constitutional foundation of Sunni Islam: sultans ruled by his investiture, and the Friday prayer from Morocco to Bengal named him.

Hülegü removed it in a fortnight.

What the office had been holding

Every Sunni state in the world had derived its legal standing from an investiture by a man who no longer existed. The Mamluks in Cairo dealt with the problem within three years by producing an Abbasid survivor and installing him as a caliph of their own, which nobody outside Egypt took very seriously and which lasted until the Ottomans took Cairo in 1517.

The Hafsids in Tunis were briefly acknowledged as caliphs by the sharif of Mecca. Delhi went on naming the dead caliph in the khutba for years because there was nobody else to name.

An institution that had been decorative for three hundred years turned out, when it was removed, to have been holding up more than anyone realised.

Ain Jalut

Then the expansion stopped, and the reason is partly the Mamluks and partly the death of a Great Khan.

Möngke died in 1259 and Hülegü withdrew east with the bulk of his army to be near the succession, leaving a much smaller force under Kitbuqa in Syria. Qutuz and Baybars came out of Egypt and destroyed it at Ain Jalut.

It is often called the battle that stopped the Mongols, and it should be qualified in both directions. The force beaten was a detachment, not the army that had taken Baghdad, and the Ilkhans went on attacking Syria repeatedly for sixty years — they took Damascus more than once. But they never held it, and the Mamluk state that beat them was the one power in the region that had been built specifically to produce heavy cavalry out of the same steppe the Mongols came from, bought as boys through the Black Sea trade the Golden Horde controlled.

And the Golden Horde

Which is where the second consequence comes in.

Berke of the Golden Horde had become a Muslim, and his reaction to the killing of the caliph by his own cousin was to go to war with him. From 1262 the two Mongol states fought each other in the Caucasus, and the Golden Horde allied with the Mamluks — supplying, among other things, the slave recruits the Mamluk army was made of.

So the Ilkhanate's northern frontier was a Mongol enemy, its western frontier was an Egyptian enemy the northern one was arming, and its eastern frontier was the Chagatai ulus, which was also hostile. Within four years of Baghdad, a state created to conquer the west was fighting on three sides and expanding on none.

What it did instead was govern, and it turned out to be unexpectedly good at it.

The church of the Third Bazar

Drafted from Abu'l-Faraj), The Chronography of Gregory Abu'l Faraj, the Son of Aaron, the Hebrew Physician, Commonly Known as Bar Hebraeus (1932)expansion

The Ilkhanate's foundation is usually narrated from the outside — a fortnight's siege, a caliph executed, an office abolished. There is an account of the same fortnight by a man who lived under the regime it founded, in a community it did not kill, and it is a different kind of document.

Bar Hebraeus was a Syriac bishop, later primate of his church in the east, who spent his career in Mongol-ruled Mesopotamia and knew Ilkhanid officials. He wrote the chronicle in the 1280s. His account of 1258 is circumstantial, cold, and in places quietly devastating.

The first day

Baghdad won it.

The Mongol advance guard under Baiju met the Baghdad troops west of the city and the company of Baju Nawin was broken, and the Baghdad troops were triumphantly victorious. Then Bar Hebraeus gives a conversation.

In the evening an old commander called Bar Kurar said to a young one called Dawithdar that God had given them the victory this time, that it was right to go into their houses and rest, and then to sally out and attack again. The young man, however, was puffed up, and did not want to go in, and both of them passed the night outside.

The Baghdad camp was on low ground. At midnight the Mongols cut the banks of the great canal from the Tigris, and the water came down on them; they fled from it with their bows, their arrows and the sheaths of their swords soaked. At daybreak the Mongols came back and by the ninth hour the army was destroyed. The old man was killed. The young man got into the city.

That is the entire military history of the fall of Baghdad, and it turns on a decision not to go indoors.

The embassy

With Hulegu's camp east of the city and the bombardment falling on the palace front, the caliph sent out his last offer, and Bar Hebraeus lists the party: the vizier, a senior official, and Mar Makika, the Catholicus of the Church of the East. They were to take gold in abundance, royal possessions and Arabian horses; to bring the Mongol ambassadors out of prison, dress them well and load them with gifts; and to ask for the lives of the caliph and his children, explaining that what had happened had been the work of evil counsellors. If the lives were granted they would become his slaves and pay tribute.

Hulegu kept the envoys and did not let them go back, and continued the war with still greater ferocity.

The Mongols breached a tower and entered, and the citizens drove them out again. The next day they took the whole circuit of the walls. The people of Baghdad hid in their houses and in the underground chambers.

What the Catholicus did

Then the passage the chronicle exists for.

And the Mongols drew their swords, and slew all the people of Baghdad — tens of thousands of men … And the Catholicus collected all the Christians into the church of the Third Bazar, and there he protected them, and none of the Christians was injured.

Bar Hebraeus does not pause on that, and immediately writes the next sentence:

And the wealthy Arabs brought out vast amounts of their riches to the Catholicus, that peradventure if they were able to save themselves they might retain their property, but all of them were killed.

Two sentences. In the first, a Christian bishop gets his own people through the sack of a city by putting them in one building. In the second, the Muslims of that city, understanding exactly what protection was available and on what terms, bring their money to him — and are killed anyway.

He also notes which contingent did the worst of the killing, and it is not the Mongols: the Iberians especially effected a great slaughter — the Georgian troops in Hulegu's army, Christians fighting for a pagan khan against a Muslim capital.

The seven days

The caliph and his two sons came out on the Saturday. Hulegu had iron fetters put on him and kept him under guard in a tent for seven days, until he had gone in person to the palace, inspected the treasures old and new, which were hidden away and laid up in stores, and had them all brought out. Then he passed judgement and condemned him, and had him sewn into a piece of sackcloth.

A week is how long it took to count it.

Why this witness matters

Because of what the new empire looked like from the position Bar Hebraeus occupied.

The Ilkhanate would spend forty years as a pagan state ruling a Muslim country, favouring Christians and Buddhists, corresponding with popes and kings about a joint attack on the Mamluks, before Ghazan converted and it became a Persian Muslim monarchy like its neighbours. For the Syriac and Armenian churches those were the best decades they had had in six centuries, and the price was being visibly on the side of the people who had done this.

Bar Hebraeus writes about it without triumph and without much comment. He records that the Christians were saved and that the Arabs who paid for the same were killed, in consecutive sentences, and moves on to the next year.

The chao

Drafted from Howorth, History of the Mongols from the 9th to the 19th Century, Part III: The Mongols of Persia (1888)aside?What aside means

In September 1294 the Ilkhan Gaykhatu introduced paper money to Iran. It was the first fiat currency issued outside China, and it lasted about two months.

The idea

The treasury was empty and the vizier was desperate. A man whom Wassaf refers to only with indignation suggested a plan that would restore the public prosperity and be without reproach: put into circulation, in place of metal, a paper currency like the chao in China.

The vizier liked it. Gaykhatu liked it, and took counsel with Bolad Chingsang, the great khan's own representative at the Ilkhanid court, who knew how the Chinese system worked. The most intelligent of the Mongol grandees warned against it and was told he had mercenary motives.

The order to create the chao was issued in May 1294.

What the note looked like

It is worth describing, because the object is the whole thirteenth century in one rectangle of paper.

An oblong sheet with several words in Chinese characters along the sides. At the top, on both faces, the Muslim profession of faith: there is no God but God, and Muhammad is his prophet. Below it, the name the Buddhist lamas had given Gaykhatu at his accession — Irinjin Dorji, in Tibetan. In a circle in the middle, the denomination, from half a dirham to ten dinars. And underneath:

The sovereign of the world has issued, in the year 693, this propitious chao. Whoever defaces it will be punished with death, with his wives and children, and his goods will be confiscated.

A Chinese financial instrument, carrying the Islamic creed and a Tibetan Buddhist name, denominated in Persian units, issued by a Mongol in Tabriz.

The rollout

A mint in every province, with its governor, scribes and cashiers. Metal currency prohibited throughout the kingdom. Gold and silver forbidden for bowls and for cloth of gold, except for the Ilkhan's own wardrobe and his chief officers'. The goldsmiths thereby put out of work were given orders on the chao banks. Old notes exchangeable for new at ten per cent discount. Only the merchants of the Persian Gulf trading abroad might exchange notes for gold at the treasury — and they were to be watched closely.

The first issue was at Tabriz on 12 September 1294, with an edict read by criers in the streets: death for refusing it, death for buying or selling in any other money, death for not bringing your coin to the mint to be exchanged.

Gaykhatu had been told that once gold was displaced there would be no more poor in the land and provisions would be cheap. Poets wrote in praise of the paper money.

Eight days

Fear of the penalty held for eight days. Then the shops and markets emptied.

Nothing could be bought in Tabriz. People began to leave the city. The hungry went out to the gardens outside the walls to take fruit. Prices, where anything was traded at all, went arbitrary: Wassaf records the sellers of a horse not worth more than seventy gold pieces asking seven hundred and fifty in paper.

Gaykhatu rode through the bazaar one day and noticed the shops were shut, and asked why. The vizier told him that a great magistrate had died and that it was customary for the citizens to close the bazaars on such occasions.

On the Friday the congregation broke out into lamentation in the mosque. Open imprecations were shouted against the author of the scheme; attempts were made on the vizier's life and his people's; the troops had difficulty holding the crowd. The vizier's brother had to authorise buying food with coin, and several men were executed for taking part in the disturbances.

Then coin came back for provisions, and then for everything, and then the chao was suppressed altogether — amidst universal joy. Commercial dealing had effectively stopped for two months. The mints had cost a fortune to build; the one at Shiraz alone cost five gold tumans. Wits and poets competed at lampooning the paper money and the men who made it.

Why it failed and what it shows

The Chinese chao worked because the Yuan state could enforce it over a closed economy with a long habit of paper instruments and a tax system that took the notes back. The Ilkhanate had none of those things. It had an open frontier with the Mamluks and with India, a merchant class whose business was long-distance trade in metal, and a government that had been farming its own revenue to the highest bidder for thirty years and was trusted by nobody.

A currency is a promise. Gaykhatu's government could impose the death penalty and could not make a promise anyone believed.

But notice what the experiment demonstrates about the thing this state belonged to. A Mongol prince in Azerbaijan could adopt a Chinese fiscal technology within a generation of its invention, on the advice of an official sent from Beijing who happened to be at his court, and implement it simultaneously in every province from Anatolia to Khurasan.

Nobody in that part of the world had been able to do anything of the kind before, and nobody would again for six hundred years.

The conquerors convert

Drafted from Jackson, The Mongols and the Islamic World: From Conquest to Conversion (2017)peak

Forty years of the wrong religion

The Ilkhanate's first four decades were run by and for a Mongol military elite that was shamanist and Buddhist, with a marked preference for Christians and Jews in the administration over Muslims, and with the Persian bureaucratic class doing what it had done for every conqueror since the Arabs. Nomad pasture requirements displaced agriculture across large parts of the plateau, and the tax system collapsed repeatedly under short-term farming of revenue.

Ghazan

In 1295 Ghazan took the throne and took the faith, and the dynasty that had killed the caliph became a Muslim dynasty within a year.

It was not a private change of heart, whatever it was privately. His accession depended on the support of the amir Nawruz, who was a Muslim; the conversion was public and it was followed by policy. Buddhist temples across Iran — built by his own father and grandfather — were pulled down or converted. The Christian and Jewish communities lost the privileged position they had held under a shamanist regime and went back to the status they had had under Muslim rulers, with the jizya and the restrictions. The coinage carried the profession of faith. He took the name Mahmud and the title padishah of Islam, and had the Friday prayer said in his name.

He also kept the Mongol law, the Mongol army, the Mongol titles and the Chinggisid claim, and went on fighting the Mamluks — who now had to explain why a Muslim ruler was attacking them.

Why it came too late

The state fell apart after Abu Sa'id died without an heir in 1335, and nobody conquered it. A state whose ruling caste held power by descent and whose administrators held none of their own has nothing left to obey once the descent runs out.

And the conversion did not save it. Ghazan died in 1304 aged thirty-two; his brother Öljeitü converted to Shi'ism and back; Abu Sa'id came to the throne at twelve and died at thirty without a son. The dynasty ran out of men faster than it ran out of money, and the reforms had been the personal project of one ruler and one vizier, both of whom were dead by 1318 — the vizier executed on a charge of having poisoned his own master.

The dervish and the assignation

Drafted from Howorth, History of the Mongols from the 9th to the 19th Century, Part III: The Mongols of Persia (1888)peak

The most detailed account anywhere of how a Mongol state actually extracted money from a conquered country was written by the man who reformed it, and it is a description of a machine designed to steal from everybody including itself.

How it worked before Ghazan

Each province had a hakim, a governor, responsible for collecting the taxes. The amount he had to account for was fixed, and so were the charges on it. Everything else about the arrangement was an opportunity.

He began by setting aside the sum he owed the treasury. Then, every time a commissary arrived from the court demanding money, he levied a fresh qubchur — a supplementary collection. He made as many as ten of these in a year, and in some places twenty.

The more commissaries came, Rashid al-Din records, the better pleased the governor was: he levied a contribution for their lodging, their entertainment, their food and their presents, kept a large part of it himself, and gave the rest to the garrison commander and the treasury clerks to make sure none of it appeared in the official accounts.

The circle

What makes the account remarkable is that it explains how the system fed itself.

The provincial revenue was mostly consumed by current expenses and by assignations — treasury orders drawn on a province's taxes, issued to soldiers, officials, pensioners and creditors. In Khurasan four-fifths of them went unpaid. The holders would come back to the treasury waving their orders. They would be told the province still owed part of its contribution and that it must be paid at once, and a severe ordinance would be issued to that effect.

So they went back to the province with it — and the governor, on their report, immediately made another levy, explaining to the taxpayers that many commissaries were waiting and their needs must be supplied. And nobody dared point out that he should be paying them out of the collections he had already made, which greatly exceeded the proper quota.

Two-thirds of each new contribution went to the governor and his people. The holders of the unpaid orders got a third, which barely covered the cost of travelling back and forth.

Meanwhile the officials at court did not enquire what any province actually received, and granted assignations indiscriminately — orders that were never paid but that temporarily satisfied whoever was making a nuisance of himself. The vizier and the provincial governors had an understanding: unless certain secret marks were put on a treasury order, no money was produced when it was presented.

The public treasury, in consequence, received nothing from the provinces at all, and the provinces were charged with a fifth of the normal payments.

What it did to people

Salaries and pensions simply stopped.

In spring, claimants were told the treasury must be satisfied first and promised payment at harvest; at harvest they were told to stand aside until the crowd of commissaries had been paid. Men waited for their salaries, hungry and naked, from year's end to year's end. The adroit ones got half of what was owed from the governor's officers by signing a receipt for the whole, and then had to take payment in goods valued at twice their worth. Anyone who got a quarter of what was due thought himself lucky, because so many got nothing.

If one of them made the journey to the court and laid his case before the council, and the governor was asked why the claim had not been met out of the first receipts as it should have been, the governor replied that he had not managed to collect the whole levy — and issued an order of payment against the arrears, which were arrears of illegal exactions he had already collected several times over.

Villagers who could not meet the repeated demands abandoned their villages and their families. If the last levy was then remitted, the clerks entered it in the books as still owing.

Rashid al-Din's summary is worth having exactly:

All the viziers who have held authority to this time have shared in this guilt.

Then his illustration. A dervish or a sheikh or some other deserving poor man would be granted an assignation of five hundred dinars. Overjoyed, he would spend a hundred of it on a mount and the expenses of the journey to go and collect the rest.

Why it matters

Two reasons, and the second is the larger.

The first is that this — not the massacres, which were over in a generation — is what ruined Iran under the Ilkhans. Land goes out of cultivation when the man working it cannot predict what will be taken; a village empties when the collectors come twenty times a year; and no amount of cavalry can make a depopulated district pay.

The second is who wrote it down. Rashid al-Din was Ghazan's vizier, the head of the system he is describing, and he set out its corruption in detail as the justification for the reform he was about to carry out — fixed rates published on stone tablets in the provinces, restrictions on the use of the post, standard weights and coinage, waste land granted out to get it back under the plough.

A government explaining, in its own official history, exactly how thoroughly it had been robbing the country, in order to argue that it had stopped. Very few states have ever produced such a document, and the reason this one did has everything to do with the kind of man Rashid al-Din was.

The first world history

Drafted from Jackson, The Mongols and the Islamic World: From Conquest to Conversion (2017)aside?What aside means

Rashid al-Din Hamadani was a physician, a Jewish convert to Islam, and Ghazan's vizier. At his master's instruction he compiled the Jami al-Tawarikh, the Compendium of Chronicles, and the result is unlike anything written before it.

It contains a history of the Mongols; then histories of the Arabs, the Persians, the Turks, the Chinese, the Jews, the Franks and the Indians — each attempted from informants and texts belonging to those traditions, with Chinese and Kashmiri and European material gathered through people the empire had at its disposal. The Frankish section works from a Latin chronicle. The Chinese section used Chinese scholars resident in Iran.

It is the first serious attempt at a history of the world that does not place the author's own community at the centre of it, and the reason it was possible is purely political: for about fifty years one administration had access to informants from China to the Mediterranean, and a ruler with a motive for commissioning a universal account of an empire that claimed universal authority.

Rashid al-Din was executed in 1318 on a charge of poisoning a previous ruler, and his endowed suburb at Tabriz was looted. The Ilkhanate outlived him by seventeen years.

Nothing about the Ilkhanate's size, reach or duration would lead anyone to expect it. A conquest state of seventy-nine years, on three hostile frontiers, produced the first attempt anywhere to write the history of the whole world as one subject — because for about forty years it was the only place on earth where informants from all of it were in the same room.

What was actually in it

The Jami al-Tawarikh was commissioned as a history of the Mongols and grew into something nobody had attempted.

The first part is the history of the Turks and Mongols, drawn from an official chronicle in Mongolian that Rashid al-Din was given access to and which has not otherwise survived — which makes him, for the early Mongols, a primary source standing in for a lost one. The second part was to cover everybody else: the pre-Islamic kings of Iran, the Prophet and the caliphs, the dynasties of Iran and Anatolia, the Turks, the Chinese, the Jews, the Franks, and India.

He wrote each section with informants to hand. There was a Chinese scholar at the court, and a Kashmiri Buddhist monk, and Mongol commanders who remembered the steppe, and European friars and merchants passing through Tabriz. The section on the Franks is built on a Latin chronicle. The section on China was checked against Chinese texts. The section on India came from the monk.

That is the thing worth pausing on. A world history with chapters on China, India and Europe, written in Persian in Azerbaijan around 1310, each chapter informed by somebody from the place it describes — because they were all in the same building.

The suburb

He built the profits of office into a quarter of Tabriz named after himself, the Rab'-i Rashidi: a foundation with a hospital, a library said to hold sixty thousand volumes, paper-makers, a scriptorium, lodgings for scholars brought in from Egypt, Syria, China and India, and an endowment to pay for all of it.

Part of the endowment deed survives and specifies the arrangement for the books. Two complete copies of his own works, one in Arabic and one in Persian, were to be made every year, on good paper, checked against the master copy for errors, and sent out to the cities of the empire.

An author with the state's revenues behind him instituting an annual print run, two centuries before printing.

The end

He was a Jewish physician's son from Hamadan who converted to Islam around thirty, served three Ilkhans, ran the finances of Iran for two decades and wrote the first history of the world that tried to be one.

In 1318 his enemies accused him of having poisoned the Ilkhan Öljeitü. He was seventy-three. He was executed, along with his son; his property was confiscated; the Rab'-i Rashidi was plundered by the mob.

His works survived because he had spent twenty years having copies made and sending them to other cities.

No heir

Drafted from Jackson, The Mongols and the Islamic World: From Conquest to Conversion (2017)end

The Ilkhanate ended in 1335 because a khan died at thirty without a son, and nothing in the state survived the discovery that there was nobody to obey.

Abu Sa'id

He came to the throne as a boy in 1316 and reigned nineteen years. The reign was not a disaster: peace was finally made with the Mamluks in 1323, ending sixty years of war on the Syrian frontier, and the administration went on working.

He died in 1335, on campaign against the Golden Horde, aged about thirty. He had no son.

That is the whole of it. Within a year there were rival puppet khans in different provinces, each a Chinggisid of some line, each maintained by a non-Chinggisid commander who could not take the title himself. Within twenty years there were no khans at all, and the men running the provinces stopped bothering to produce one.

Why a succession problem was fatal here

Every Chinggisid state had disputed successions and most survived them. The Ilkhanate did not, and the reason is the thing that had made it work.

It was a conquest elite ruling a Persian population through a Persian bureaucracy. The bureaucracy was competent, indispensable, and had no independent claim to authority: viziers ran the finances and were periodically executed for it — Rashid al-Din, the greatest of them, was put to death in 1318 on a charge of poisoning a previous khan, and his quarter of Tabriz was plundered.

So the state had a ruling caste whose only qualification was descent, and an administrative class with no qualification at all beyond usefulness. Remove the descent and there is nothing left that anybody is obliged to obey. The Mongol commanders — Jalayirid, Chobanid, Injuid — divided the country between them, and each of them needed a khan for exactly as long as it took his neighbours to stop pretending.

What went with it

Iran did not return to anything. The period from 1335 to Timur's arrival in the 1380s is a patchwork of regional dynasties, and it is the least legible half century in Iranian history precisely because there was no centre producing records.

The Ilkhanate had also spent seventy years doing something the Mongols did nowhere else: rebuilding. Ghazan's fiscal reforms after 1295 had fixed tax rates, curbed the tax farmers, restored the postal system to something the peasantry could survive, and re-established the coinage on a uniform standard. Rashid al-Din's charitable foundation outside Tabriz was a small town of libraries, a hospital and a scriptorium.

Almost none of that outlasted the dynasty by a generation. The Rab'-i Rashidi was sacked when its founder fell. Tabriz changed hands repeatedly. The observatory at Maragha, built for Nasir al-Din al-Tusi at Hülegü's expense and the best-equipped in the world when it was made, was derelict within a century.

The one thing that did last

The books.

Rashid al-Din's Compendium of Chronicles, commissioned by a Mongol khan and written by a Jewish convert to Islam serving as vizier, is a history of the world — the Mongols, the Chinese, the Indians, the Franks, the Jews — assembled from informants of each nation who were at the court because the court was the centre of an empire that reached all of them.

Nothing like it had been attempted anywhere, and nothing like it was attempted again for centuries, because the conditions that made it possible lasted about forty years and then stopped.

What Tabriz kept

Drafted from Jackson, The Mongols and the Islamic World: From Conquest to Conversion (2017)afterlife

The Ilkhanate lasted seventy-nine years and left more behind it than either of its longer-lived Mongol cousins, most of it visual and most of it produced by the collision of two artistic traditions that would not otherwise have met.

Chinese painting in Persian books

The Mongols moved craftsmen the way they moved troops. Chinese painters, silk-weavers, potters and their pattern books came west with the administration, and Persian workshops absorbed them.

What came out of that is the Persian miniature. The dragons and phoenixes, the cloud-bands, the twisted rocks and the treatment of landscape that define Persian book painting for the next four centuries are Chinese in origin and arrived in Iran under the Ilkhans. So did the blue-and-white ceramic tradition, in both directions: Persian cobalt went east to Jingdezhen, and Chinese porcelain forms came west.

The great illustrated Shahnama made at Tabriz in the 1330s — the epic of Persian kingship, in a manuscript whose battle scenes are painted with Chinese brushwork — is the whole transaction in one object. A Mongol court paid Persian painters using Chinese methods to illustrate a poem about the Iranian kings the Mongols had replaced.

The observatory

Maragha did not survive as a building and did survive as a method.

Nasir al-Din al-Tusi persuaded Hülegü to fund an observatory on a scale nobody had attempted, with instruments, a library said to run to hundreds of thousands of volumes, and a staff drawn from across the Islamic world and from China. The planetary models developed there — the Tusi couple and the work of Ibn al-Shatir after him — are the mathematical devices that turn up, two centuries later and without acknowledgement, in Copernicus.

How they got to Europe is argued and unresolved. That the same constructions appear in both places, in the same order, is not.

The idea of a world history

Rashid al-Din's Compendium is the Ilkhanate's most characteristic product because it could only have been made there.

A history of the world requires informants from the world, and the Ilkhanid court had them: Chinese scholars, Buddhist monks from Kashmir, a Frankish envoy, Mongol nobles who remembered the traditions, and a vizier with imperial funds and imperial authority to ask. The result treats the Franks and the Chinese as subjects of history in their own right rather than as barbarians at the edge of a Muslim narrative, which is not how history had been written in Arabic or Persian before.

It is also, for the Mongols themselves, a primary source of the first importance. Much of what is known about the early Mongol confederation comes from a Persian vizier writing in Tabriz from Mongol oral tradition, because the Mongols had not written it down and the people who could remember it were at his court.

The frontier that stayed

The last inheritance is a line on a map.

Sixty years of Ilkhanid war with the Mamluks fixed a frontier in northern Syria that neither side could cross, and the peace of 1323 confirmed it. That division — an Iranian power east of the Euphrates, an Egyptian-Syrian power west of it — is the same division the Ottomans and the Safavids fought over two centuries later, in the same places.

The state that drew it had ceased to exist within twelve years of the treaty, and the line outlasted it by six hundred.