Shaybanid Khanate

The qualification Timur lacked

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)formation

Timur's whole career turns on a technicality. He was not descended from Chinggis Khan, could therefore never legitimately be khan, and spent thirty-five years ruling as amir beside a puppet Chinggisid he kept at court. That constraint explains a good deal about how he ruled: a man who cannot command by right has to be followed for some other reason, and the reasons available get used up.

The Shaybanids are the people who had the right.

Shiban was the fifth son of Jochi, Chinggis Khan's eldest, and his descendants held territory in the eastern half of the Jochid ulus — the steppe north and east of the Aral, in the part of the Golden Horde's world that fragmented earliest. In the fifteenth century one of them, Abu'l-Khayr Khan, assembled a tribal confederation out of the Turkic and Turco-Mongol groups of that steppe. They were called Uzbeks, after a fourteenth-century khan of the Horde, and the name has since attached to a country, a language and a nation none of which existed then.

Abu'l-Khayr's confederation broke up after a defeat by the Oirats, and part of it split away and became the Kazakhs, which is where that distinction begins. His grandson Muhammad Shaybani spent his youth as a mercenary commander for whoever was hiring in Transoxiana — including, at times, for Timurid princes — and reassembled the following.

Then he used it. In 1500 he took Bukhara and Samarkand. By 1507 he had Herat. The Timurid century in Central Asia ended, and the man who ended it required no puppet, no marriage into a khan's family, and no title that meant something other than khan. He simply was one.

What it did to Babur

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)expansion

Zahir al-Din Muhammad Babur was a Timurid prince of Farghana, a great-great-great-grandson of Timur, and he spent his youth trying to hold Samarkand. He took it three times and lost it three times, twice to Muhammad Shaybani. He wrote about all of it afterwards in Chagatai, in a memoir that is one of the great autobiographies in any language and is unusually frank about being beaten.

The Shaybanid conquest is therefore visible from inside, from the losing side, which is a thing that almost never survives. Babur's account of Shaybani is not the account of a barbarian invasion — he records the man as a competent commander with a legitimate claim, and records his own difficulty in raising troops against someone the Timurid amirs regarded as a plausible sovereign.

In 1504 Babur gave up on Central Asia and took Kabul. He came back once: in 1511, after Shah Isma'il had killed Shaybani Khan at Merv, he retook Samarkand with Safavid help — and lost it within a year, at Ghijduvan in 1512, partly because accepting Shi'i backing and the Safavid red cap had cost him the support of a Sunni population that had been prepared to have him. It is the last time a Timurid ruled Timur's capital.

He went back to Kabul and, in 1526, took northern India, where the dynasty he founded ruled for three centuries and called itself Gurkani after Timur's own title — the one claim to the Timurid inheritance that anybody ever made good.

The chain is worth stating plainly. The Shaybanids took Transoxiana because they had Chinggisid descent and the Timurids did not; a dispossessed Timurid went south and founded the Mughal empire; and the reason the Mughals were in India at all is that a Jochid house had a better claim to a steppe title than they did.

Khanzada Begim

Drafted from Babur, The Babur-nama in English (Memoirs of Babur) (1922)expansion

The Shaybanid conquest of Transoxiana cost the Timurids their cities, and Babur records what it cost him personally in four sentences, in the middle of a list of his sisters.

The transaction

He had taken Samarkand in 1500 despite being beaten at Sar-i Pul, and held it through a five-month siege. No beg or ruler anywhere near sent any help or reinforcement at all. In the end he left it in despair and got away.

In that throneless time Khan-zada Begim fell to Muhammad Shaibani Khan.

That is all he says about the mechanism, and the phrase — that throneless time, fatrat — is the one he uses throughout for the years when he had no country. She was his full sister, five years older than he was, the eldest daughter of Umar Shaikh Mirza. She was handed to the conqueror as part of the terms on which he was allowed to leave his own capital.

Babur was eighteen.

The ten years

She had a child by Shaybani — a pleasant boy, Babur calls him, named Khurram Shah, who was given the Balkh country and died a few years after his father.

In 1510 Shah Isma'il of Persia destroyed the Uzbek army near Merv and killed Shaybani Khan. Khanzada Begim was in Merv when it happened.

For my sake he treated her well, giving her a sufficient escort to Qunduz where she rejoined me.

The shah who had made a drinking cup of her husband's skull sent her back to her brother, with an escort, as a courtesy.

The recognition

Then comes the passage, and it should be read slowly.

We had been apart for some ten years; when Muhammadi kükildash and I went to see her, neither she nor those about her knew us, although I spoke. They recognized us after a time.

Two people who had been children together, meeting in a Central Asian fortress after ten years — one of them a queen of the house that had destroyed her family, the other a landless prince who had lost three kingdoms — and neither the woman nor her household could tell who he was until he had been standing there for a while.

Although I spoke. He expected his own voice to be enough, and it was not.

Why it is worth having

Because it is the only thing in any of these sources that gives the conquest a price in the currency people actually pay in, and because of who wrote it.

Babur is not making a case. The sentence sits in a genealogical list between his half-sisters' names and their mothers' names, and he goes on immediately to Mihr-banu Begim, two years younger than himself. He is a man who lost everything to the Shaybanids, wrote a long book about it, and put the single worst thing that happened to him in a family register without raising his voice.

She outlived all of it, and Babur — who records in the same list that another sister was eight years younger than he was, and that two more were born after their father's death — never says another word about what those ten years had been like. The memoir is famous for its frankness. This is the one place where it simply stops.

A head, two hands and two legs

Drafted from Babur, The Babur-nama in English (Memoirs of Babur) (1922)expansion

The Uzbek conquest of Transoxiana is recorded in detail by the man it happened to. Babur held Samarkand against Shaybani Khan for five months in 1501, lost it, and wrote the siege up thirty years later from the inside of the walls.

The five months

The fighting he reports as a professional. Four of his men ran up to a stretch of rampart the Uzbeks were climbing and drove them all off by energetic drubbing; one beg at the Needle-makers' Gate held out well to the end; another led a sortie out of that gate and chased the Uzbeks as far as a village, unhorsing some and coming back with a few heads. He names them all, and notes which of them did best.

Then the supply.

It was now the time of ripening rain but no-one brought new corn into the town. The long siege caused great privation to the towns-people; it went so far that the poor and destitute began to eat the flesh of dogs and asses and, as there was little grain for the horses, people fed them on leaves.

He records which leaves worked — mulberry and elm — and that some people scraped dry wood and gave the damp shavings to their horses. That is a besieged commander's memory, and nobody who had not been inside would keep it.

Shaybani Khan did not assault. For three or four months he kept the fort invested at a distance and moved round it from post to post. Then, at midnight, when the defenders were off their guard, his men came up to the Turquoise Gate, beat their drums and flung out their war-cry. Babur was in the college, undressed. There was great trepidation and anxiety. After that they came night after night, drumming and shouting.

Why nobody came

The passage that follows is the most honest thing in the book, and it is about Babur rather than about the Uzbeks.

Although envoys and messengers had been sent repeatedly to all sides and quarters, no help and reinforcement arrived from any-one. No-one had helped or reinforced me when I was in strength and power and had suffered no sort of defeat or loss; on what score would any-one help me now?

Then the maxim he had been taught and was now testing:

The old saying was that to hold a fort there must be a head, two hands and two legs, that is to say, the Commandant is the head; help and reinforcement coming from two quarters are the two arms and the food and water in the fort are the two legs.

He had the head. He had neither arm and both legs had gone. He left in despair and got away.

The sister

The price of the withdrawal is recorded elsewhere in the memoir, in the section where Babur lists his father's daughters, and he does not dwell on it.

His elder sister Khanzada Begim — who, he says, had been with him from his birth — fell to Shaybani Khan in the throneless time that followed. She bore him a son, who was given Balkh and died a few years after his father. Nine years later Shah Ismail destroyed the Uzbeks near Merv, where she was, and sent her back to Babur with an escort, for my sake.

We had been apart for some ten years; when Muhammadi kukildash and I went to see her, neither she nor those about her knew us, although I spoke. They recognized us after…

What the Uzbeks had done

Babur's account is the best evidence for the character of the conquest, and what it shows is not a raid.

Shaybani did not storm Samarkand. He sat down outside it for five months, starved it, ensured that no grain came in from the countryside at harvest, and broke the defenders' sleep for the last stretch. That is the method of a man who intends to keep the city rather than sack it, and he kept it: the Uzbek dynasties held Transoxiana for the next four centuries.

And Babur's complaint — that nobody came because he had never helped anybody either — is a description of what the Timurid system had become. A country full of princes of the same house, each holding a town, none of whom would combine, was taken one town at a time by a confederation that would.

The river becomes a border

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)expansion

In 1510 Shah Isma'il met Muhammad Shaybani near Merv, destroyed the Uzbek army, and killed him. He then had the skull mounted in gold as a drinking cup, and the skin stuffed with straw and sent to the Ottoman sultan. Isma'il was twenty-two and had been shah for nine years.

The Uzbeks recovered — Ubaydullah Khan held Transoxiana and beat Babur and the Safavids at Ghijduvan in 1512 — but Khurasan and Herat were gone, and something larger had changed. Two years earlier this had been a dynastic quarrel over which house held which cities. After Merv it was a confessional war: a Sunni khanate whose legitimacy rested on Chinggisid descent against a Shi'i empire whose legitimacy rested on the imams, fighting along the Amu Darya for two centuries.

The consequences are still legible on a map and they run in both directions.

Transoxiana was cut off from Iran, with which it had been a single cultural region since the Samanids, whose world had held Bukhara and Nishapur in the same conversation. The Persian literary culture continued on both sides of the river and stopped circulating across it. Central Asian scholars went to India instead, which is part of why the Mughal court became what it was.

And the frontier acquired a slave trade. Uzbek raiding into Khurasan carried off Iranians who, being Shi'i, could be argued by Central Asian jurists to be lawfully enslaved in a way Sunni Muslims could not; the markets at Bukhara and Khiva were supplied that way into the nineteenth century. It is the ugliest single consequence of the settlement of 1501, and it belongs here rather than in an account of the Safavids because these were the people who profited from it.

Abdullah Khan

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)peak

Abdullah Khan II ruled in his father's name from the 1550s and in his own from 1583, and he is the one Shaybanid who tried to make the khanate into something other than a family holding land in common. He spent thirty years reducing the appanages of his own relatives by force, centralised what he could, reformed the coinage, and built — caravanserais, bridges, reservoirs, the covered markets of Bukhara that are still standing.

Then he used the instrument. In 1588 he took Herat and much of Khurasan from a Safavid state that was in the middle of its own succession crisis, with a sixteen-year-old shah who had just come to the throne over a country the Ottomans were also invading. It is the widest extent Uzbek power ever reached.

It did not stick. The sixteen-year-old was Abbas I, who made peace on bad terms to buy time, broke the Qizilbash, built a standing corps of ghulams, and then came back. Khurasan and Herat were Safavid again by 1603.

The reach is therefore a measure of something narrower than it looks. The khanate could take a great deal when the neighbour was distracted and could not hold it when the neighbour was not, because what it had was tribal cavalry and what it did not have was a paid standing force and a revenue system to support one. That is the difference between the sixteenth century and every one before it, and it was settled on this frontier within twenty years.

Abdullah Khan died in 1598. His son was murdered within months by his own kinsmen, and the Shaybanid male line ran out.

White felt and Zamzam

Drafted from Howorth, History of the Mongols from the 9th to the 19th Century, Part II: The So-Called Tartars of Russia and Central Asia, Division II (1880)peak

When Abdullah Khan II was raised to the khanship in 1583, the ceremony was performed on a piece of white felt that had been dipped at Mecca in the water of the well of Zamzam.

That object is the whole Shaybanid settlement in one sentence.

The ceremony

His father Iskandar died while he was away at Khujand, and the head of the clergy and the other notables buried the khan and then rode out to find him, meeting him on the road. He summoned them to a council — Howorth calls it by its Mongol name, the quriltai — to consider the succession, and the chief cleric did not take long to proclaim him the fittest man for the post of khaqan.

Then they raised him on the felt.

Raising a new khan on a felt carpet is a steppe rite older than Chinggis; it is how the Türk khagans were enthroned and it is what the word for enthronement meant on that grass. Zamzam water is the water of the Hajj, brought back by pilgrims from the well beside the Ka'ba.

A Muslim clergy, in a Mongol assembly, raising a descendant of Jochi on a Turkic felt sanctified in Arabia. Nobody in the room thought any of it incongruous, and that is the point about this dynasty: it was not a Mongol regime that became Muslim, nor a Muslim regime with Mongol trappings, but a working synthesis in which both sets of forms were operative at once.

What he had already done

Howorth's assessment is unequivocal — Abdullah was the greatest of the Abu'l-Khayrid line — and his reason for saying so is structural rather than military.

Under cover of his father's shield he had crushed or displaced every rival from the Kazakh deserts to the frontiers of Kabul, so that the Uzbegs were for the first time subject to a single strong grasp, instead of being broken in pieces among a number of appanages.

That sentence describes the correction of the defect that had wrecked every Chinggisid state in this history. Muhammad Shaybani had conquered Transoxiana and then divided it among his kinsmen as appanages, exactly as Chinggis, Dayan Khan and every other steppe founder had done; and the khanate had spent eighty years being a federation of cousins who fought each other. Abdullah spent his father's lifetime removing them one at a time before taking the title himself.

Thus the apple which had been so long in ripening fell on his knees.

Mashhad

With a unified khanate he could do what no Uzbek ruler had managed since 1512: attack Khurasan seriously. Persia was between strong shahs; the Uzbeks raided across the frontier and withdrew when the Persian army appeared. In 1585, with Shah Abbas occupied against the Ottomans, Herat fell after a nine-month siege, its governor and several chiefs were executed, the city was plundered, a large number of prisoners were carried off to Bukhara, and north-eastern Khurasan was laid waste.

And then something happened that is more revealing than the campaign.

The guardians of the shrine of the Imam Reza at Mashhad — who administered the enormous endowment of fields, gardens and benefices attached to it — wrote to Abdullah Khan, and asked him a question. How did he reconcile with his religious feelings the destruction of the Imam's property, and the waste of the income by which so many thousand pious pilgrims were supported, including many Sunnis?

The reply

The mullahs in Abdullah's camp answered with a treatise.

By their principles and their faith, they wrote, the followers of the Shia were worse than unbelievers, whose destruction God himself had ordained. If it was every Muslim's duty to make war on unbelievers, how much more on those who had wandered from the right path and had fallen into grievous sin in spite of living around the bones of the saint.

As for the shrine's fields and gardens, they knew perfectly well that these were devoted to pious uses. But it was an open question, they said, who had the better right to the enjoyment of them: the warriors fighting for the cause of God and right and deprived of all means of subsistence, or those who had sinned against God and treated the guardians of the faith with contempt.

The Shia scholars, Howorth notes, had a rejoinder ready, and proposed a council of equal numbers from both sides to determine whether the Shia were in fact traitors to the faith.

What the correspondence settles

It is easy to read the Uzbek-Safavid wars as dynastic quarrels dressed in religion. This exchange shows the doctrine doing real work, in writing, at the moment of application.

A formal legal opinion that the property of Shia Muslims is lawful plunder for Sunni soldiers is the same opinion that makes Shia Muslims lawfully enslavable, and it is the reason the slave markets of Bukhara and Khiva were stocked with Iranians for the next three hundred years. The argument was not invented for the occasion by cynics. It was argued out with the other side, by name, in letters, and both parties were prepared to submit it to a joint commission.

That is what an ideological frontier looks like from the inside while it is being built.

Bukhara, and the decline that has to be argued for

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)aside?What aside means

The standard story about Central Asia after 1500 is decline: the Portuguese rounded the Cape, the overland routes died, the caravan cities withered, and the region became a backwater until Russia arrived. It is a tidy story, it was told for a century, and it does not survive contact with the evidence.

The trade did not die. It reorganised. The volume of goods moving between India, Iran, Russia and China through Transoxiana appears to have held up or grown in the sixteenth and seventeenth centuries; Bukhara and Balkh were substantial commercial cities; Indian merchant communities were resident in Central Asian towns in numbers; and the Russian trade through Astrakhan expanded. What changed is that the region stopped being on the main road between Europe and Asia and became the hub of a set of regional systems instead, which is a different thing from collapse.

What the Shaybanids built with the money is still there. Ubaydullah Khan's Mir-i Arab madrasa at Bukhara, finished in the 1530s, faces the Kalyan mosque across a square and is the visual centre of the city; Abdullah Khan's markets and caravanserais are the fabric of its bazaar quarter. The waqf endowments that funded the madrasas made the ulama an independent economic power in a way that shaped Central Asian politics for three centuries — the same mechanism the Mamluks ran in Cairo, for the same reason, with a different ruling class.

Two things follow. The first is that nothing about this khanate's size or its military record would tell a reader what the sixteenth century was actually like in Bukhara, which was busy, literate and building.

The second is a caution about the word itself. "Decline" is a judgement, and it is almost always a judgement made from somewhere else about where the important traffic went. The same verdict has been passed on the Ottoman seventeenth century and on the Safavid one, by people counting the things their own century happened to count, and it has had to be argued back in every case.

The caravan that stopped

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)contraction

The Shaybanid khanate was built on a crossroads that was closing, and the closing had nothing to do with anything anybody in Bukhara did.

Three walls going up at once

To the south, Iran became Safavid and Shi'i after 1501, and the Uzbeks were Sunni. What had been a frontier between two Muslim states became a confessional one, fought over continuously — Merv, Herat and Khurasan changed hands repeatedly — and the road south stopped being a road and became a front.

To the west, Russia took Kazan in 1552 and Astrakhan in 1556, which put a Christian power across the lower Volga and ended the old northern route to the Black Sea as a Muslim thoroughfare.

To the east, the Ming had shut the maritime expeditions down in the 1430s and was managing its western frontier through a tightly restricted tribute trade that let a fixed number of missions in and treated everything else as a security problem.

A state whose whole position was that it sat between Iran, Russia, China and India found, within about half a century, that three of the four approaches had become either a war or a permit.

And the sea

The fourth thing was happening a long way away and was not visible from Transoxiana at all.

Portuguese ships reached India in 1498 and the Indian Ocean trade began moving by sea. That did not empty the overland routes overnight — the older view that the caravan trade collapsed the moment the Cape route opened is too neat, and Central Asian caravans went on running well into the seventeenth century. But the direction was set. Bulk goods between India, Iran and Europe increasingly went round Africa, and the share of the world's long-distance trade that had to cross Central Asia declined from something enormous to something regional.

The khanate's revenue was transit and its cities were built on it. There was nothing to be done about any of this from Bukhara, and nobody in Bukhara appears to have understood that it was happening until it had.

What was done instead

The one thing available: taking Khurasan.

Abdullah Khan II spent his reign fighting the Safavids for it and took Herat in 1588, and it did not solve the problem, because the problem was not that the Uzbeks lacked territory. Khurasan was a rich province at the end of a road that was carrying less every decade, and holding it required a permanent war with a larger neighbour.

The other thing available was building, and the Shaybanids did a great deal of it. The madrasas, mosques and covered bazaars that give Bukhara and Samarkand their sixteenth-century character were put up by this dynasty out of the last good decades of the caravan revenue.

That is what the contraction of this state looks like: not ruin, but an increasingly magnificent city on an increasingly quiet road, paid for by a frontier war that could not be won and could not be given up.

The house ends and the khanate does not

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)end

Abdullah Khan II died in 1598 and his son Abd al-Mu'min was murdered by his own amirs after six months. There was no Shaybanid of sufficient standing left; the male line of the house had been thinned by a century of appanage quarrels and by Abdullah Khan's own campaigns against his relatives.

What followed is unusual. The khanate did not fall, fragment or get conquered. Its throne passed by marriage to the Toqay-Timurids — also called the Janids or Astrakhanids — who were descendants of Toqay Timur, a different son of Jochi, and who had arrived in Bukhara as refugees when Russia took Astrakhan in 1556. They ruled the same state from the same capital with the same institutions until 1785.

It is a change of family and nothing else: nobody conquered anything, no territory changed hands, and the officials, the coinage and the law went on without interruption.

The reason the transfer was possible without a war is the reason this whole page exists. What made a man eligible to be khan in Transoxiana was Jochid descent, and the Toqay-Timurids had it. Any of Chinggis Khan's descendants would do; a Timurid, however capable, would not. That rule had governed the politics of this country since the thirteenth century, it had forced the greatest conqueror of the fourteenth to spend his life pretending to serve a nonentity, and it was still deciding who sat in Bukhara in 1599.

That is what a rule about legitimacy is worth when it has been obeyed long enough. In 1599, in a city three thousand miles from Mongolia, the throne of Transoxiana passed between two branches of a family nobody in it had seen for three hundred and seventy years, and the transfer was so obviously correct that it required no fighting.

The Uzbek name

Drafted from (eds), The Cambridge History of Inner Asia: The Chinggisid Age (2009)afterlife

The Shaybanids gave their confederation's name to a country, and the country is named after a khan who was not one of them and who ruled two thousand kilometres away.

Where "Uzbek" comes from

The name attaches to Uzbeg Khan of the Golden Horde, who ruled at Sarai from 1313 to 1341 and made that khanate definitively Muslim. The nomads of the eastern Jochid lands — the ulus of Shiban — came to be called Uzbeks after him, and it was under that name that Muhammad Shaybani led them south across the Syr Darya at the end of the fifteenth century and took Transoxiana from the Timurids.

The people who moved were nomads. The country they took was one of the oldest settled regions in the world. Within a few generations the name had transferred from the confederation to the population of the country it had conquered, which is what usually happens, and the modern Republic of Uzbekistan carries it.

The Kazakhs are the other half of the same split: the groups who declined to follow the Shaybanids south in the 1450s and stayed on the steppe took a different name and kept the nomadic life for another four hundred years. Two modern nations, one confederation, divided by a decision about whether to go and take the cities.

Three khanates and a protectorate

What the Shaybanids left institutionally lasted until the twentieth century.

The state passed to the Toqay-Timurids without a break, and then, in the eighteenth century, fragmented into three: the emirate of Bukhara, the khanate of Khiva in Khwarazm, and the khanate of Kokand in the Ferghana valley. All three were governed on the pattern the Shaybanids had established — a Sunni Muslim court, an ulama with real authority, a revenue drawn from irrigated agriculture and what was left of the transit trade.

Russia conquered them in the 1860s and 1870s and left two of them standing as protectorates. The emir of Bukhara and the khan of Khiva were still on their thrones in 1917, which means the political order the Uzbeks established in 1500 outlived the Romanovs by a few months.

The buildings

The other inheritance is visible from any street in Bukhara.

The Shaybanid century is when the city acquired the form it still has: the great madrasas facing each other across a square, the covered crossroads bazaars, the Kalyan complex, the domed markets for the moneychangers, the hat-sellers and the jewellers. Samarkand's Registan was completed under their successors on the plan they worked in.

Those buildings are the last great expenditure of the overland caravan trade before it stopped mattering, and they were put up by a dynasty of nomads who had been in the country for two generations, in the styles of the Timurids they had driven out.

What the descent was worth to the end

The last thing worth saying about this dynasty is the thing the whole Mongol world turns on.

The Shaybanids ruled because they were descended from Jochi. When their male line failed in 1599 the throne went to another line descended from Jochi, without a war, because the qualification was the descent rather than the family. The men who actually governed — the Uzbek tribal amirs — could no more take the title in Bukhara than Timur could take it in Samarkand two centuries before.

Chinggis Khan died in 1227. In 1785, in Bukhara, the answer to who was entitled to be khan was still: one of his descendants. It is difficult to name any other rule about legitimate authority anywhere that ran for five and a half centuries and was obeyed by people who had every practical reason not to.