Yuan Empire

The second capital

Drafted from (eds), The Cambridge History of China, Volume 6: Alien Regimes and Border States, 907-1368 (1994)formation

The Yuan begins with a disputed election and ends up, for that reason, as a Chinese dynasty rather than a Mongol one.

1260

When the Great Khan Möngke died on campaign in 1259, the succession should have been settled by a quriltai — an assembly of the Mongol nobility, held in Mongolia, with the princes of the whole empire present.

Kublai did not wait for one. He held his own assembly at Kaiping in Inner Mongolia, among the commanders of the armies he had been leading in China, and had himself proclaimed. His younger brother Ariq Böke held the proper kind of assembly at Karakorum, with the proper people, and was proclaimed too.

The war between them lasted four years. Kublai won it because he held north China and could feed his army from it while Karakorum, which had never been able to feed itself, depended on grain convoys from the south that he cut.

That is the whole argument of the dynasty in one campaign. The brother with the farmland beat the brother with the steppe, and the empire's centre of gravity moved with the result.

What winning that way cost

It cost the empire. Kublai's election was procedurally indefensible and the western khanates said so; the Golden Horde and the Chagatai ulus effectively stopped treating the Great Khan as anything but a distant relative, and the Ilkhans in Iran, who did recognise him, were four thousand kilometres away.

After 1260 there was no single Mongol empire in any operative sense. There were four states with a common ancestry, of which Kublai's was the largest, richest and most administratively developed, and which spent the next century minding its own business.

Becoming a dynasty

He then did something no previous Great Khan had done: he took a Chinese dynastic name.

In 1271 the state was proclaimed the Great Yuan — the name taken from the Book of Changes, meaning the primal origin — and with it came the whole apparatus of Chinese legitimacy: a dynastic title rather than a people's name, ancestral temples with posthumous imperial titles conferred on Chinggis and his successors, a court calendar, and the compilation of dynastic histories for the Liao, the Jin and eventually the Song.

He built a capital to match at Dadu, on the site of the Jin capital and the future Beijing: a walled rectangle on a grid, laid out on classical principles by a Chinese official, with the palace at the centre and the ancestral temple and altars of soil and grain in their canonical positions.

And he kept Shangdu — Kaiping, where he had been elected — as a summer capital, and went there every year with the court.

Two capitals, one for each half of what he was. The word travelled to Europe through Marco Polo, and six hundred years later Coleridge wrote a poem about it under the name Xanadu.

What he never did

He never became Chinese, and the state never pretended he had.

The Mongols did not adopt the examinations for another forty years and never relied on them. The population was formally divided into four ranks — Mongols, central and western Asians, northern Chinese, southern Chinese — with the last two excluded from the senior posts. Mongolian was the language of the court; Persians, Turks, Uyghurs and at least one Venetian held offices that Chinese scholars expected to hold.

A dynasty with a Chinese name, a Chinese capital, a Chinese calendar and a foreign ruling caste that would not let the Chinese govern is a contradiction that lasted ninety-seven years, which is roughly how long such things last.

Xiangyang

Drafted from (eds), The Cambridge History of China, Volume 6: Alien Regimes and Border States, 907-1368 (1994)expansion

The Southern Song took forty-five years to conquer, which is longer than the Mongols spent on anything else, and the reason is water.

Why it was hard

Everything that had made the Mongol armies unstoppable across Eurasia worked less well south of the Huai. The country is wet, cut by rivers and canals, terraced, and in summer it is hot in a way that killed both the horses and the men. The Song had the largest navy in the world, walled cities in enormous numbers, gunpowder weapons, and a state that — whatever its failures against the Jin — could still put very large armies into the field.

Möngke had died in 1259 on that campaign, of dysentery or wounds depending on the account, which is the fact that produced the succession war and the division of the empire.

The siege

Xiangyang and Fancheng faced each other across the Han river, and between them they were the lock on the middle Yangtze. Everything the Mongols needed to do depended on getting past them.

The siege lasted from 1268 to 1273 — five years. The Mongols built a ring of forts, blockaded the river, and were still unable to take the walls. Song relief convoys got through by river more than once.

What broke it was artillery, and the artillery was imported. Kublai sent to his nephew the Ilkhan in Iran for engineers, and two Muslim specialists — the sources name them Ismail and Ala al-Din — arrived and built counterweight trebuchets of the Western type, larger and far more powerful than anything in Chinese use. The Chinese sources call them Huihui pao, Muslim catapults.

Fancheng fell first. Xiangyang surrendered shortly after, its commander given terms and taken into Yuan service.

That is the Mongol empire working exactly as designed: a siege in central China broken by Persian engineers summoned from Tabriz, because one family held both ends of the road.

The last six years

After Xiangyang it went quickly. Bayan took the Yangtze in 1275, Hangzhou surrendered in 1276 and the child emperor and the dowager empress were sent north, and the court that would not surrender kept moving south by sea with a fleet and two more child emperors.

It ended at Yamen, off the Guangdong coast, in 1279: a naval battle in which the Song fleet was chained together and destroyed, and the chief minister Lu Xiufu put the last emperor — seven years old — on his back and jumped into the sea.

The accounts say tens of thousands went into the water after them. Whatever the number, it is the only Chinese dynasty that ends in a mass suicide at sea rather than an abdication ceremony, and Chinese memory has never treated it as anything else.

What was acquired

For the first time since 907, one government held everything from the Amur to Hainan. It was also, in that moment, the richest and most populous territory on earth, with the most advanced economy — printing, paper money, iron production, canal transport, ocean shipping — and it had all been captured intact by a regime that had spent seventy years learning not to destroy such things.

Kublai's two failed invasions of Japan, in 1274 and 1281, both broken by storms; the failed expeditions to Vietnam, Champa and Java — those are what happened when the same machine kept going after the useful conquest was finished. The sources for the Java expedition of 1293 record it being outmanoeuvred by a Javanese prince who used it to destroy his own rival and then turned on it.

An empire that had just absorbed the wealthiest state in the world spent the next fifteen years losing fleets in the tropics, which is what a conquest machine does when there is nothing left worth conquering.

A steppe dynasty that took all of China

Drafted from (eds), The Cambridge History of China, Volume 6: Alien Regimes and Border States, 907-1368 (1994)peak

The Liao took part of northern China. The Jin took more. Kublai took all of it, finishing the Southern Song in 1279 and becoming the first ruler from the steppe to govern the entire country.

He did it by becoming, in part, a Chinese emperor: proclaiming a dynastic name in the Chinese manner in 1271, building a capital at Dadu on the site of modern Beijing, performing the imperial rites. He also did not: Mongols and Central Asians were kept as a separate and privileged legal category above northern and then southern Chinese, the civil service examinations were suspended for decades because they would have staffed the administration with the wrong people, and much of the revenue apparatus was run by Muslim and Uyghur financial administrators brought from the west.

That deliberate refusal to assimilate is the same problem the Liao solved with two administrations and the Chagatais failed to solve at all, and the Yuan answer — rule China with Chinese methods but not with Chinese people — held for under a century.

The Phags-pa script is the sharpest illustration of the ambition. Kublai commissioned a Tibetan monk to design a single alphabet capable of writing Mongolian, Chinese, Tibetan and Uyghur in one system, for an empire that spanned all four. It was promulgated in 1269, used on seals and coinage and official documents, and abandoned when the dynasty fell.

The two monks

Drafted from Sauma, The Monks of Kublai Khan, Emperor of China (1928)aside?What aside means

In the early 1270s two Nestorian Christian monks left Khanbaliq — Kublai's own capital — to walk to Jerusalem, and the account one of them left behind is the only description of the Yuan's western frontier written by someone who actually crossed it on foot while the empire was at war with itself.

Where they were from

Rabban Sauma was born in Khanbaliq itself, the son of a Nestorian nobleman of the city, ordained a priest there and later a hermit in the hills nearby. His companion Markos was a younger monk from Koshang, a town of Önggüd Mongols further west who had been Christian for generations. Neither man was Chinese, and neither was a stranger to the court: the community they belonged to had held office and standing in the Mongol capital since Chinggis's day, and Sauma's own account takes that standing for granted rather than explaining it.

They decided to go to Jerusalem to pray at the tombs of the martyrs, gave away their belongings, and went to the capital to find a caravan.

Kublai's sons-in-law

Word of two respected monks leaving the country reached Könchi and Ay-buqa, sons- in-law of Kublai Khan himself, who had the pair brought before them and tried, with real affection, to talk them out of it.

Why are ye leaving this country of ours and going to the West? For we have taken very great trouble to draw hither monks and fathers from the West. How can we allow you to go away?

Sauma's answer is the whole of a monk's case in one line: we have cast away the world; as long as we live in the society of men there will be no peace to us.

Failing to change their minds, the princes did what a Mongol grandee did for anyone travelling far on his goodwill: they loaded them with riding animals, gold, silver and clothing, over the monks' own objections that they needed none of it. Accept these gifts as a loan, the princes said. If necessity does not arise, and ye arrive safe and sound, distribute them among the monasteries that are there. This was, in effect, a Mongol travel warrant made of goods instead of paper — the same function a paiza or yarlikh performed for an official envoy, extended informally to two men of no rank at all, because they were going to pray for the khan's own western Christian subjects.

What the road actually looked like

The Yuan's western reach was not a line on a map. It was a country being fought over, and the monks walked straight through the middle of that fight.

They came to Tangut — the old Xi Xia country, by then a Yuan province — and were welcomed with gifts by a population Sauma describes as ardent believers. Then two months across a waterless desert to Khotan, with drinkable water found on only eight days of the whole crossing. And there they arrived in the middle of an actual war: Kublai's forces against a rebel commander who had fled into the country and, in Sauma's plain words, destroyed thousands of men therein. The roads were cut. Grain could not be found. People were dying of hunger around them.

They waited six months for it to pass, then went on to Kashgar and found the city empty of its inhabitants, because it had been already plundered by the enemy.

Kaidu

Then they reached Talas, and the camp of the man Sauma's editor correctly identifies as Kublai's own cousin: Qaidu, the Ögedeid prince who had refused to recognise Kublai's succession in 1260 and had been at open war with him, on and off, for the whole of the monks' lifetime.

They went into his presence, blessed his kingdom, and asked for a safe-conduct through his territory. He gave it to them.

Two monks travelling on gifts and goodwill from Kublai's own sons-in-law walked into the camp of the man Kublai's empire had been fighting for a decade, and were received, blessed, and sent on their way with the same courtesy the Yuan court had shown them. Whatever Qaidu's war with Kublai was actually about, it was not a war on travellers, and it was certainly not a war that recognised borders the way a modern reader expects a war to.

Why this belongs beside the ortogh and the paper money

Because it is evidence of the same thing from a completely different direction.

The Yuan financed a merchant aristocracy that ran silver the length of the empire regardless of which khanate it was crossing. It printed a currency an Ilkhanid observer's cousin marvelled at. And it produced two monks who could walk from Beijing to Khorasan through a live civil war, unarmed, protected at every stage not by soldiers but by a network of obligation — Mongol courtesy to travellers, Christian hospitality from Tangut to Talas to Maragha, and a rebel prince's willingness to honour a blessing from men who served a court he did not recognise.

The empire that could not agree on who its Great Khan was could still move two harmless men four thousand miles through the middle of the argument without either side laying a hand on them. That is not a small thing to be true of a civil war, and it says something about what the word empire still meant, even when it had stopped meaning one government.

They reached Khorasan with most of what they had been given already spent or lost on the road, and went on to Maragha, where the story of one of them ends and the story of the Ilkhanate's church resumes it.

The ortogh

Drafted from (eds), The Cambridge History of China, Volume 6: Alien Regimes and Border States, 907-1368 (1994)aside?What aside means

The Yuan was funded by a Chinese land tax and the salt monopoly, which supplied the largest single share of the money. Both of those would be true of any Chinese dynasty of any period.

What they leave out is the part that belonged to nobody else.

Merchant partnerships with imperial capital

The ortogh were associations of merchants, largely Central Asian and mostly Muslim, financed directly by Mongol princes and by the imperial household. A prince put up silver; the partnership traded with it across the length of the empire and beyond it; the profits were divided.

They also farmed taxes. A partnership would advance the state a sum against the revenue of a district and then collect it, at whatever it could extract. And they lent — to peasants, at rates the Yuan's own edicts periodically tried to cap, against the security of land and people.

So the imperial family was simultaneously the government, the largest investor in the empire's long-distance trade, the creditor of its taxpayers, and the beneficiary of the tax-farming contracts. Venice and Genoa were states run by merchants. The Yuan is the inverse: merchants run as an arm of the ruling family's private capital.

Why it disappears from the summaries

Ask where a pre-modern state's money came from and the answers available are land, heads, tolls, tribute, plunder, mines, monopolies. They are all categories of extraction — of taking a share of something somebody else produced.

The ortogh are not extraction. They are finance: the ruling house putting silver into a trading partnership and taking a return on it, with the state's own courier network, tax farms and coercive power available to the partnership as working capital. There is no ordinary way to describe a government whose rulers were also its principal investors, which is why the arrangement tends to vanish from any short account of how the Yuan was funded.

A land tax and a salt monopoly are not a wrong description of Yuan finance. They are simply the description that fits every other dynasty too, applied to the one government in Chinese history whose treasury was entangled with a commercial partnership running to Tabriz.

The river, and the paper

Drafted from Polo, The Book of Ser Marco Polo (1903)contraction

The Yuan was killed by a currency and a river, in that order, and the second would not have been fatal without the first.

The money

The Yuan ran the first economy in history in which paper was the only legal currency. Not paper alongside coin — paper instead of it. Silver and copper were called in; the notes were backed by a silver reserve, exchangeable, and legally required for all transactions.

For about forty years it worked, and it is one of the genuinely remarkable administrative achievements anywhere. Marco Polo gives it a chapter, and cannot quite get over it:

The Emperor's Mint then is in this same City of Cambaluc, and the way it is wrought is such that you might say he hath the Secret of Alchemy in perfection, and you would be right!

He then describes the process exactly: the fine white bast between the wood and the outer bark of the mulberry, made into sheets, cut into denominations from half a tornesel up to ten bezants of gold, each signed by a succession of officials and finally stamped in vermilion by the Khan's own deputy — the Money is then authentic. Any one forging it would be punished with death.

Two of his observations are the whole fiscal system. First: the Khan has this made every year in a quantity which costs him nothing, that it must equal in amount all the treasure in the world. Second: foreign merchants arriving with gold, silver, gems or pearls are forbidden to sell to anyone but the emperor, who has twelve experts appraise the goods and pays for them in paper — and the merchants accept readily, because nobody else pays as well and nobody else pays as promptly.

A Venetian looking at that saw alchemy. What he was actually describing is a state buying the world's luxury imports with money it printed, and a bullion monopoly enforcing the currency at the point where bullion entered the country.

Then it did what unbacked paper does. Military campaigns, the court's spending, the subsidies paid to Mongol princes, and the loss of revenue from a countryside in trouble were all covered by issuing more, and the reserve stopped matching the issue. There were currency reforms in 1287 and again in 1309, each of which devalued the existing notes against a new issue. The last, in 1350, was catastrophic: prices rose beyond the ability of the state to pay its own bills in its own money, and people went back to barter and to hoarded coin.

A government whose entire fiscal system is a promise loses everything the moment the promise is not believed, and by the 1350s it was not.

The river

The Yellow River carries more silt than any large river in the world, builds its own bed above the surrounding plain, and periodically breaks out and changes course by hundreds of kilometres. It had done so before and would again.

In the 1340s it did it repeatedly, flooding the north China plain, wrecking the Grand Canal and cutting the grain supply to a capital that could not feed itself. Famine followed, and plague — the same pandemic that reached Europe in 1347 was moving through Asia in the same decades, and the Chinese records describe severe epidemics in the 1330s and 1340s.

In 1351 the government did the responsible thing and mobilised, by some accounts, a hundred and fifty thousand conscripted labourers to return the river to its old channel.

The work succeeded. The mobilisation destroyed the dynasty. A hundred and fifty thousand hungry men from a ruined province, working under corvée for a government whose money had stopped being worth anything, was the largest concentration of grievance in China, and the White Lotus preachers were already among them. The Red Turban rising began that year.

What the court was doing

Fighting itself. Between 1307 and 1333 there were nine emperors; two were murdered, one reigned for fifty-three days, and the succession was decided by whichever faction of Mongol princes and commanders had the palace guard.

That instability has a structural cause and it is the one the dynasty was founded on. The Mongols never adopted a fixed succession rule, because their own tradition was election by assembly and the assembly had been abandoned in 1260. What replaced it was neither: not a quriltai, not primogeniture, just the strongest faction at court. Every reign began with a purge and ended with one.

The last twenty years

By 1360 the Yuan controlled the north, the canal and not much else, while its generals in the south fought the rebels and each other, and the court in Dadu argued.

In 1368 a Ming army came north. The emperor Toghon Temür left the capital without fighting for it, and rode back to Mongolia with his court.

That is the least discussed fact about the fall. The dynasty did not end at Dadu; it went home, kept the dynastic name, kept the seals, and went on claiming to be the legitimate government of China from the steppe for another two centuries. The Ming called them the remnant Yuan and campaigned against them for two hundred years without settling it, and in 1449 one of those campaigns lost a Ming emperor.

Why it lasted less than a century

Drafted from (eds), The Cambridge History of China, Volume 6: Alien Regimes and Border States, 907-1368 (1994)end

The Yuan held all of China for eighty-nine years, which is short. The Liao managed two hundred and nine on a fraction of the territory, the Jin a hundred and nineteen, the Qing later two hundred and sixty-eight.

Several reasons compound. The Mongol elite was small and its legal privileges made it visibly foreign in a way the Khitan and Jurchen arrangements had softened. The suspension of the examinations for most of the dynasty cut off the route by which Chinese literati normally attached themselves to a regime, leaving a large educated class with no stake in it. Revenue was farmed to Central Asian financiers whose methods were efficient and detested. Paper currency was issued against no reserve and inflated.

Then the Yellow River changed course in the 1340s, the conscripted repair works gathered hundreds of thousands of labourers in one place, plague arrived, and the rebellions began in the exact populations the dynasty had least reason to trust.

The ending is an unusual one. The Yuan court was not destroyed; it withdrew to the steppe in 1368 and continued as the Northern Yuan for two more centuries, still claiming the mandate. Almost every other conquest dynasty here ended where it stood. The Mongols had somewhere to go back to, and that is a consequence of never having fully come inside.

The assembly and the throne

Drafted from (eds), The Cambridge History of China, Volume 6: Alien Regimes and Border States, 907-1368 (1994)end

Every other dynasty that held the whole of China lasted two or three centuries. The Yuan lasted less than one, and the largest single reason is that it never changed how it chose an emperor.

The Mongol rule was an assembly of the eligible, raising a successor from among all the eligible.

What that is for

On the steppe it is a good rule. A confederation whose survival depends on the competence of the man leading it cannot afford to be handed a child, and an assembly of the leading families choosing among adult claimants will usually produce somebody who can command. It is a large part of why the Mongol empire reached the size it did.

In a settled empire it is a war. The pool of eligible Chinggisids is enormous, every one of them has a following, the assembly can be packed or convened twice, and the loser of a disputed succession has an army rather than a monastery.

The Yuan never replaced it. Kublai's own accession in 1260 was decided by two rival assemblies and four years of fighting his brother. Between 1307 and 1333 the throne changed hands eight times, and in 1323 an emperor was murdered by his own guard commanders at the summer camp.

Why they did not fix it

This is the part worth dwelling on, because the alternative was neither obscure nor untested.

The Chinese answer — a designated heir apparent, named in advance, with the rule written down — was available, thoroughly documented, and being used by every dynasty whose records the Yuan's own historians were at that moment compiling. The Ming would codify it within a generation. The Han had run on it for four centuries, at the cost of a great many child emperors.

Adopting it would have meant conceding that the emperor of China was something other than the Great Khan: that the throne descended by a Chinese rule rather than being conferred by Mongol nobles. And the Yuan ruler was answerable to three constituencies at once — the Chinggisid princes who elected him, the Chinese bureaucracy that administered for him, and the population being ruled. Each required a different account of why he held the throne. Fixing the succession meant choosing among them.

They did not choose. They got eight emperors in twenty-six years, and then a rising among conscripted labourers on the Yellow River, and then the Ming.

The one that took a whole empire

Drafted from (eds), The Cambridge History of China, Volume 6: Alien Regimes and Border States, 907-1368 (1994)afterlife

Four states came out of the Mongol division: the Ilkhanate in Iran, the Chagatai khanate in Central Asia, the Golden Horde on the Russian steppe, and this one.

Three of the four took a share of somewhere. The Yuan took a settled empire entire — the whole of China, north and south, for the first time in three and a half centuries — and that is the difference everything else follows from.

It shows up in what the dynasty said entitled it to rule, which was three things at once: descent from Chinggis, the mandate of Heaven, and the fact of conquest. The Ilkhanate's rulers eventually became Muslim sovereigns of Iran and stopped needing to say they had conquered it. The Yuan never got there. Its own legal system sorted the population into four ranked categories with Mongols at the top and the southern Chinese — the population of the Song, the majority of the empire's people — at the bottom, and it kept the examinations suspended for the first thirty years and quota-restricted thereafter.

A dynasty that claims the Mandate of Heaven while running a legal hierarchy in which the people it claims it over are ranked last is making two arguments that cannot both be load-bearing.

What withdrew

The dynasty had exactly one successor, and it was itself.

In 1368 the court did not fall. It left — north, with the seals, the archives and the reign titles — and its successors went on issuing them from the steppe for two more centuries. That is why the Ming lost an emperor to a Mongol force at Tumu in 1449, and why it spent so much of its revenue on a wall.

The two states are the same fact from opposite sides: a dynasty that says it is still the dynasty, and a dynasty that has to keep an army in the field about it for two hundred years. Calling 1368 a conquest is true about the territory and wrong about the claim.