Timurid Empire

The man who could not be khan

Drafted from Manz, The Rise and Rule of Tamerlane (1989)formation

Timur was born around 1336 into the Barlas, a minor Turkicised Mongol clan in the western Chagatai lands south of Samarkand. He was not a descendant of Chinggis Khan, and under the political rules of post-Mongol Central Asia that meant he could never legitimately hold the title of khan. He never claimed it.

He ruled for thirty-five years as amir. He married a Chinggisid princess in order to style himself gurkan, son-in-law of the house. He kept a puppet khan of the correct descent at his court for the entire period, issued documents in that khan's name, and when one died he installed another.

It looks like a technicality and it is the key to the whole career. Manz's argument is that the constraint explains the man. A ruler who cannot command by right has to be followed for some other reason, and the reasons available were plunder, success and personal obligation — all three of which are consumed by use and have to be renewed. Timur's following was not an army of a state; it was an army that was the state, assembled out of Chaghatay tribal contingents whose leaders had to be kept satisfied and out of personal nökers who owed him directly and could be used against the tribal amirs.

That is why the campaigns never stopped. There was no year of Timur's rule in which he was not preparing, conducting or returning from one, and the standard explanation — restless ambition, the will to conquer — describes the behaviour without accounting for it. A confederation held together by success cannot afford a quiet decade, and Timur appears to have understood this better than anyone writing about him since.

What the campaigns were for

Drafted from Manz, The Rise and Rule of Tamerlane (1989)expansion

The list is long and the pattern in it is more interesting than the extent. Iran and the Caucasus repeatedly from the 1380s. The Golden Horde, whose capital he sacked twice, in 1391 and 1395. India in 1398, where Delhi was taken and its population massacred. Syria in 1400, where Aleppo and Damascus were sacked and the craftsmen of Damascus were deported to Samarkand. The Ottomans at Ankara in 1402, where he destroyed the army and captured the sultan. He died in 1405 on the road to invade Ming China.

Almost none of it was annexed. Timur took submissions, plunder and craftsmen, installed or reinstalled local rulers, and went home; when the submission lapsed he came back and did it again, which is why Iran appears in the list four times. Manz's account treats this not as a failure to consolidate but as the operating method of a polity whose income was campaign proceeds and whose cohesion was the expectation of the next one. The campaign proceeds were not what an army happened to pick up along the way. They were structural income, budgeted for, and the empire could not stop earning them without coming apart.

Two of the campaigns changed other people's histories more than his own. Sacking Sarai and New Sarai wrecked the northern caravan route across the steppe and broke the Golden Horde's grip on its Russian tributaries, which is a substantial part of why Moscow rose in the following century, and entirely incidental to anything Timur intended. And Ankara removed the Ottoman state for a decade, restored the Anatolian emirates the Ottomans had absorbed, and postponed the fall of Constantinople by fifty years.

He was also, throughout, building. The plunder went into Samarkand: the Bibi Khanum mosque, the Gur-i Amir, the Shah-i Zinda, and gardens ringing the city named after the places he had taken. Deported craftsmen built them. Few buildings anywhere have so direct a relationship to a conquest: it is a straightforward transfer, and Delhi and Damascus paid for Samarkand.

Delhi, as the conqueror's own books tell it

Drafted from Various, The History of India, as Told by its Own Historians: The Muhammadan Period, Volume III (1871)expansion

The Indian campaign of 1398 is unusual among Timur's wars in that two accounts of it survive from inside his own house, and they can be read against each other. The Malfuzat-i Timuri is written in the first person and presents itself as his memoirs; Sharaf al-Din Yazdi's Zafarnama narrates the same weeks in the third. Whether the memoirs are genuinely his, or a Persian construction worked up to agree with the Zafarnama, has been argued over since the eighteenth century and is not settled. What can be said is that both books were made inside the dynasty for the dynasty, and that the passages below are therefore not accusations. They are the version the Timurids wanted kept.

Three days before the battle for Delhi, the army was camped east of Loni. Amir Jahan Shah and Amir Sulayman Shah and the other senior commanders came and raised a problem: since crossing the Indus the army had taken something over a hundred thousand Hindu prisoners, and they were all in the camp. The day before, when the Delhi forces had come out to probe, the prisoners had made signs of rejoicing and cursed their captors, and it was thought that on the day of the real battle they would break their bonds, plunder the tents and go over. The commanders' advice was that on such a day a hundred thousand men could not be left with the baggage, and that setting idolaters free was contrary to the rules of war. The first-person account records the conclusion without any distance at all: When I heard these words I found them in accordance with the rules of war, and the order went through the camp that every man holding prisoners was to kill them, and that anyone who failed to do so would be executed himself and his property given to whoever informed on him.

Both books then record the same detail, and it is the one that survives reading. Maulana Nasir al-Din Umar was a scholar and a counsellor who in his whole life had never killed a sparrow. He had fifteen Hindus in his train, and in execution of the order he killed all fifteen with his own sword.

Then one man in ten was told off to guard the property and the cattle and the captives' wives and children, and at the midday prayer the army marched for the Jumna.

The sack of the city three days later is narrated in a completely different register, and the difference is the point. The Delhi massacre had a plain military rationale and is reported plainly. What happened inside the walls had none, so the first-person account opens with a list of reasons why armed men happened to be in the city at all — a party of soldiers idling at a gate who began helping themselves and the amirs sent in to restrain them; ladies of the harem who had asked to see the thousand-column palace, and their escort; the officials gone in to collect the levy; a thousand troopers gone in for grain, oil, sugar and flour; and regiments sent in to seize the Hindus who had taken refuge there with their families and goods, with orders to ignore any protest. Six separate innocent errands, and each of them put soldiers inside the walls.

What follows is not softened. The refugees drew their swords rather than be taken; the fires spread from Jahanpanah and Siri to Old Delhi; fifteen thousand Turks spent Thursday and all Thursday night killing and plundering, and on Friday morning the whole army went in and was no longer under anybody's control. Saturday was the same. Every man got fifty to a hundred captives and none got fewer than twenty. Only the quarter of the sayyids and the ulama was spared. On Sunday a body of Hindus who had fortified themselves in the congregational mosque of Old Delhi were cleared out of it and killed, and Old Delhi was plundered too.

And then the sentence the whole preamble was built for: By the will of God, and by no wish or direction of mine, all the three cities of Delhi had been plundered. The sermon proclaiming his sovereignty had been read in the city, which is an assurance of protection; it was therefore his earnest wish that no evil should come to the people; but God had ordained that the city be ruined and had inspired its inhabitants with a spirit of resistance, so that they brought upon themselves a fate that was inevitable.

One order, before the plunder was distributed, is worth more than the whole apology. The artisans were to be picked out of the prisoners and set aside; some thousands were, and were parcelled out among the princes and amirs. But the builders and the stonemasons were reserved to the sovereign personally, because he had determined to build a congregational mosque at Samarkand which should have no rival in any country.

That mosque is Bibi Khanum, and the men who built it were taken out of Delhi in the same week as the elephants.

What the Castilians saw

Drafted from Clavijo, Narrative of the Embassy to the Court of Timour at Samarcand (1859)peak

In the autumn of 1404 an embassy from Henry III of Castile reached Samarkand, and one of the ambassadors kept a journal. Clavijo is the only Western eyewitness of Timur's court, he arrived at the exact top of it, and he had no literary ambitions whatever — which is why the account is useful. He writes down what he was shown, what it cost and how long he was kept waiting.

He arrived into a festival. Timur had ordered every trade in the city — merchants, cooks, butchers, bakers, shoemakers, all of them — to come out of Samarkand, pitch their tents on the plain where the horde was camped, sell their goods there instead of in the city, and each trade to perform a game and walk it through the camp for the amusement of the army. So they did: the skinners and butchers came dressed as lions and foxes, there were masquerades on a carpeted ground, two of Timur's grandsons were married, and there was wine and fruit for everybody.

On the same ground he had a great number of gallows set up, and announced that in this festival he knew how to be merciful and kind to some and severe to others.

The first was the chief magistrate of Samarkand, the greatest officer in the whole country, whom Timur had left in charge of the city when he departed six years and eleven months earlier. He had neglected his duties. He was hanged and his goods confiscated, and Clavijo notes that the punishment of so great a man caused terror among the people. A councillor who interceded for him was offered his life for four hundred thousand bezants of silver; he handed over everything he had, was tortured for more, had no more, and was hung up by the feet until he died. Another man had been left with three thousand horses and could not produce all of them; he was hanged, although he offered to produce six thousand if given time. Traders who had sold meat above its worth were executed and others fined. Clavijo records the convention without comment: a great man is hanged, the meaner sort are beheaded.

That is a working description of what the empire ran on. There is no institution in it. There is a man who went away for seven years, came back, and audited his officials personally, and whose audit had one penalty.

The other thing Clavijo watched was the building. Timur had put up a mosque in honour of his wife's mother, and it was the most respected in the city; when it was finished he decided the entrance was too low, and had the whole thing pulled down. To get it rebuilt faster he took charge of one part of the work himself and set his mirzas over the other. By then he could not walk or ride and went about in a litter, and he came out to the site every day and stayed a while. The work stopped when the snow began to fall.

The ambassadors, meanwhile, could not get an audience. On the first of November they went as ordered and waited from morning until noon, and were told he could not speak to them that day because he was occupied with his grandson, the king of India, who was going home. On the Saturday he did not come out of his tent, being ill; they waited until noon and were sent away. On the Sunday they waited a long time and were told to go back to their lodgings, because the lord would not see them.

He was seventy, he was building a mosque he had already built once, and he was four months from setting out to invade China. The Castilians went home without their answer.

A government without a monopoly of force

Drafted from Manz, Cambridge Studies in Islamic Civilization (2007)aside?What aside means

Manz's later book asks a question about Shahrukh's Iran that applies to most pre-modern states and that the Timurid case states more clearly than most: how did a government that did not monopolise force control a society it could not administer directly, and why was the arrangement stable for decades at a time?

The Timurid answer was that the government did not try to be the only power. The dynasty and its military servitors were one set of actors. City elites — the notable families of Herat, Shiraz, Yazd, who supplied the officials and held the local standing — were another, with their own resources and their own continuity across changes of ruler. The religious classes, both the ulama and the Sufi shaykhs whose followings were large enough to matter politically, were a third. Provincial rulers who acknowledged the Timurids and governed as they liked were a fourth.

Rule consisted of holding those constituencies in a relationship, mostly by distributing things: appointments, revenue assignments, endowments, marriages, tax remissions. It broke down when there was nothing left to distribute or when a succession forced everyone to choose.

Every ordinary description of the machinery is true and misleading. The army was a tribal levy and it was also men holding land against service. The succession was by appanage, in the strong sense: Timur assigned his sons and grandsons territories as personal patrimony and they fought over them at every death, as the Seljuks and the Mongols and both Turkmen confederations did. What none of that conveys is that the state's stability came from a negotiated relationship with people who were not part of it, and that conducting the negotiation was the government's principal activity.

It also answers a question that keeps arising about states which look chaotic in the political narrative and produce extraordinary things anyway. Ulugh Beg's observatory and Bihzad's workshop were funded by a state that was losing control of its provinces, and the two facts are not in tension: the patronage was part of how the relationship was held.

Shahrukh

Drafted from Manz, Cambridge Studies in Islamic Civilization (2007)peak

Timur died in 1405 having designated a grandson, and the empire went into four years of war among sons and grandsons. Shahrukh, Timur's fourth son, won it, and then ruled for thirty-eight years — longer than his father, and over a state that looked nothing like his father's.

He governed from Herat rather than Samarkand, and that is the substantive change. The centre of the empire moved off the steppe frontier and into Khurasan, and the polity stopped being a Central Asian conquest machine with Iranian provinces and became an Iranian state with a Central Asian dynasty. Samarkand went to his son Ulugh Beg as a viceroyalty.

The rest followed. Shahrukh stopped campaigning annually; he fought the Kara Koyunlu in the west three times and otherwise defended what he had. He patronised the ulama and the Sufi orders heavily, presented himself as a Muslim sovereign ruling by the sharia rather than by the Mongol yasa, and dropped the puppet khan his father had needed. That is a real change in what the dynasty claimed to be, made in one reign, and it is easy to miss because the family kept Timur's name for everything else.

His wife Gawhar Shad is one of the substantial figures of the period in her own right: she endowed the mosque at Mashhad that still carries her name and a mosque and madrasa complex at Herat, and she was politically active in the succession disputes for decades after his death, until she was executed at about eighty.

What Shahrukh did not fix was the succession, and the reason is on the previous page. The appanage system that had produced the war of 1405 was still in place in 1447, because it was not a policy but the structure of the family's claim to its territories.

Ulugh Beg

Drafted from Manz, Cambridge Studies in Islamic Civilization (2007)peak

Ulugh Beg governed Samarkand for his father for nearly forty years and was an astronomer. This is not a hobby recorded politely by later admirers: he built an observatory on a hill outside the city with a meridian arc of about forty metres' radius, sunk into a trench so that the instrument could be large enough to be accurate, and he assembled a working staff of mathematicians around it — al-Kashi, who computed pi to sixteen decimal places and wrote on decimal fractions, and Qadi Zada, and after them Ali Qushji.

The output was the Zij-i Sultani, a star catalogue of more than a thousand stars observed rather than copied from Ptolemy, and tables of planetary motion. The positions were the most accurate produced anywhere until Tycho Brahe a century and a half later, and the value it gives for the length of the sidereal year is within about a minute of the modern figure. The catalogue reached Europe in the seventeenth century and was printed at Oxford.

He succeeded Shahrukh in 1447 and was a bad ruler, which the sources agree on and which the succession made worse: he could not hold the Timurid territories together, lost to his nephews, and in 1449 was deposed by his own son Abd al-Latif and beheaded on the road while ostensibly leaving for the pilgrimage. The son was killed six months later.

The observatory was abandoned and eventually demolished; the trench containing the arc was rediscovered in 1908. Ali Qushji left for the Ak Koyunlu court and then for Istanbul, where he entered Mehmed II's service — one of the several threads by which Samarkand's learning reached the Ottoman and Safavid and Mughal courts.

Ulugh Beg's forty years and his father's are the summit of the dynasty, and they have to be read together. The decades in which the Timurid state produced the most were the same decades in which it stopped being able to defend its own succession.

The princely vision

Drafted from Lowry, Timur and the Princely Vision: Persian Art and Culture in the Fifteenth Century (1989)aside?What aside means

The Timurid court style is one of the most widely inherited things in the history of the Islamic world that is neither a religion nor an alphabet, and it was made largely at Herat and Samarkand between about 1410 and 1500 by workshops the dynasty paid for directly.

Lentz and Lowry's argument is that this was a programme rather than an accident of wealth. The kitabkhana — the book house — was a court institution that employed calligraphers, illuminators, painters, binders, and also designers whose drawings turn up executed in tile, in metal, in textile and in architectural ornament. It produced a coherent visual language across media because the same people were designing for all of them, on salary, for a patron who cared. The point of it was princely: to display a kind of cultivated sovereignty that could be recognised by anyone in the eastern Islamic world.

The names are worth having. Bihzad, working at Herat under Sultan Husayn Bayqara, is the painter against whom Persian miniature was measured for two centuries afterwards, and his figures are the first in the tradition that behave like individuals doing something rather than like types arranged in a composition. Mir Ali Shir Nava'i, Husayn Bayqara's companion and vizier, ran much of the patronage — and did something else with consequences beyond art: he wrote major poetry in Chagatai Turkish and argued in prose, at length and polemically, that Turkish was capable of everything Persian could do. Courts had written Persian as a matter of course. After Nava'i there was a literary Turkic tradition with a canon to point at.

Then the workshops dispersed, and the dispersal is the reason three later empires look the way they do. Bihzad went to Tabriz and worked for the Safavids, who took the Herat school as their starting point. Timurid architectural forms — the double dome, the four-iwan court, the garden divided in four — are the vocabulary of Safavid Isfahan and of Mughal India, where Babur laid out Timurid gardens from memory and his descendants built the Taj Mahal out of the same grammar. The Ottoman court took the book arts and the ceramic ornament.

Political weight and cultural output come apart everywhere. The Timurids are the largest case of it and also the strangest, because here the output was the thing that lasted: the empire was gone by 1507, and three of the four largest Muslim states of the following two centuries were still working in a visual language its workshops had assembled.

After Ulugh Beg

Drafted from Browne, A Literary History of Persia, Volume III: The Tartar Dominion (1265-1502) (1920)contraction

Ulugh Beg was murdered in 1449 on the orders of his own son, and the eastern half of the empire never reassembled. The fifty-eight years between that and the Uzbek conquest are the part of Timurid history nobody tells, and they contain the best court the dynasty ever had.

The scramble

The pattern is monotonous and it is worth seeing in one place, because it is what appanage succession does when there is nothing left to conquer.

Abd al-Latif murdered his father and was killed himself within months. Abu Sa'id, a grandson of Timur's son Miranshah, had been attached to Ulugh Beg's court, was thrown into prison by the parricide, escaped through the negligence of the sentries to Bukhara, and came out of the wreckage holding much of Iran for two decades. Mirkhwand calls him supreme among the princes of the House of Timur in high enterprise, lofty rank and perfect discernment, and says the lands from Turkistan to Mazandaran reached the height of their prosperity under him. He was captured and executed by Uzun Hasan of the Ak Koyunlu in 1469.

Meanwhile the cousins fought each other. One of them occupied Herat, defeated a cousin, was preparing to crush him at Astarabad, and was suddenly attacked by Jahanshah of the Kara Koyunlu; he abandoned Herat, fled, came back when the invader withdrew, and was then beaten by Abu Sa'id. He died on the road from Damghan to Mashhad in 1458, and his cousin and rival was killed the same year.

Every one of those men was a great-grandson of Timur.

Herat

What was left after 1469 was Khurasan, and Sultan Husayn Bayqara ruled it from Herat for thirty-seven years.

By any political measure it is a rump: a province, hemmed in by Turkmen confederations to the west and Uzbeks to the north, spending its revenues on a capital instead of an army. By any other measure it is the summit of the whole dynasty. Husayn was a serious patron and a writer himself, in Turki as well as Persian. His friend and minister Ali Shir Nava'i made Chaghatay Turkish into a literary language capable of everything Persian could do, and argued the case in print. Jami was at that court. So was Bihzad.

What the fifty-eight years were

It is tempting to call this a decline with a cultural consolation prize, and that gets it backwards.

The Timurid state after 1449 could not defend Khurasan, could not recover Transoxiana, and could not keep its princes from killing each other. It also paid for the workshops out of which the visual language of the next three empires came, and for a court where two literary languages were being worked on at once by men who knew each other.

Those are not two separate stories. A prince who cannot expand and cannot be displaced has revenue and no use for it, and time, and rivals to outshine who are doing the same thing in the next city. The Timurid golden age is what happens when a dynasty stops being able to conquer and has not yet stopped being rich.

Babur's inventory of Herat

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

Sultan Husayn Bayqara died on the march against Shaibani Khan on 11 Dhu'l-Hijja 911 — 5 May 1506 — which is worth stating precisely, because a date of January 1501 for his death is in wide circulation and is five years out against the contemporary who wrote the following. When he died, Babur — who was twenty-three, held Kabul, and would spend the rest of his life elsewhere — stopped his memoirs and wrote out the whole of the Herat court from memory and report: the ruler, his sons, his wives, his begs, his scholars, his poets, his painters, his musicians, each with a sentence or two of assessment. It runs for dozens of pages and it is the best single description of a Persianate court that anybody wrote, partly because Babur admired it and partly because he was incapable of flattering anyone.

He begins with the body. Husayn was slant-eyed and lion-bodied, slender from the waist down; old and white-bearded he still wore silks of fine red and green, and a black lambskin cap or a felt qalpaq, except on a feast day, when he would wind a little three-fold turban broad and badly, stick a heron's plume in it and go to prayers in that.

Then the record, which Babur gives without arranging it to flatter. On taking Herat he thought of having the Twelve Imams named in the Friday sermon, and Ali Sher Beg and others talked him out of it; after that everything important was done by orthodox law. He could not perform the prostrations because of trouble in his joints and he kept no fasts. He was lively and pleasant, rather immoderate in temper, with a tongue to match it. He once surrendered a son of his own to the avengers of blood and had him taken to the court of judgment. He was abstinent for six or seven years after taking the throne and then, in Babur's phrase, degraded himself to drink: in nearly forty years of rule there may not have been a single day on which he did not drink after the midday prayer, and not one on which he drank before it. He was personally brave — Babur, who knew what he was talking about, says no man of Timur's line has been known to match him at slashing with a sword. He put a divan together in Turki under the pen-name Husayni; many of the couplets are not bad, though the whole thing is in one metre. And, says Babur, great ruler though he was by the length of his reign and the breadth of his lands, he yet kept fighting-rams, flew pigeons and fought cocks, like little people.

The court is catalogued the same way. Sultan Ali of Mashhad, the finest of the calligraphers, copied thirty couplets a day for the Mirza and twenty for Ali Sher. Of the painters, Bihzad's work was very dainty, but he did not draw beardless faces well and used to lengthen the double chin too far — bearded faces he drew admirably. Shah Muzaffar painted portraits and rendered hair beautifully, and died young, on his way up to fame. Khwaja Abdullah played the dulcimer better than anyone. Qul-i Muhammad the lutenist added three strings to the guitar. And there is Shaykhi the flautist, who had played since he was twelve, who produced an air at one of the prince's assemblies that Qul-i Muhammad could not reproduce on the guitar — so that Qul-i Muhammad pronounced the guitar a worthless instrument, whereupon Shaykhi took it out of his hands and played the air on it, well and in tune. He could hear a tune once and say whose flute it came from.

This is the summit of the dynasty and the state underneath it was a province. It could not recover Transoxiana, could not hold Khurasan against the Uzbeks and did not survive its ruler by two years. Babur's own explanation for writing the inventory at such length is not stated, but the shape of it says what he thought: he was a Timurid recording what a Timurid court had been, at the moment it stopped existing, for the benefit of a reader who would never see one.

Taken by cousins

Drafted from Manz, Cambridge Studies in Islamic Civilization (2007)end

After 1449 the eastern half never reassembled. Abu Sa'id held much of Iran for two decades and was captured and executed by Uzun Hasan of the Ak Koyunlu in 1469, which is where Timurid power west of Khurasan ends. What was left was Herat under Sultan Husayn Bayqara, and that is the subject of the previous chapter.

The end came from the north and from relatives. Muhammad Shaybani Khan led the Uzbeks — Turco-Mongol tribes out of the Qipchaq steppe under a Chinggisid line, which is to say people with exactly the descent Timur had lacked. He took Samarkand in 1500 and Herat in 1507, and the Timurid line in Central Asia finished.

One Timurid got away. Babur, a great-great-great-grandson of Timur, lost Samarkand to Shaybani twice, gave up on Central Asia, took Kabul, and in 1526 took northern India. The dynasty he founded is the one Europe calls the Mughals, and it did not call itself that. It called itself Gurkani, after Timur's own title — the son-in-law, the word Timur had used because he could not call himself khan.

A man who could not hold Ferghana carried his great-great-great-grandfather's apology for not being a Chinggisid two thousand kilometres south and made it the name of an empire.

Shaybani Khan was killed by Shah Isma'il of the Safavids in 1510 near Merv, and his skull was made into a drinking cup. Babur, briefly allied with Isma'il, held Samarkand for eight months in 1511 as a Safavid client and lost it again, which is the last time a Timurid ruled Timur's capital.

What the empire left was not territory. It was a court style — the architecture, the manuscript, the garden, the Chagatai verse — carried forward by three empires that had each defeated the family that made it.

A cow and a road

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

The empire ended in 1507. One Timurid got away, and he wrote down what it was like — which is why the Timurid afterlife is the only one in this period that can be read in the first person.

The boy with a province

In the month of Ramadan of the year 899, and in the twelfth year of my age, I became ruler in the country of Farghana.

That is how the Baburnama opens. Babur was a great-great-great-grandson of Timur on his father's side and a descendant of Chinggis Khan on his mother's, which by the arithmetic of that world made him one of the best-connected landless men in Asia. He took Samarkand twice and lost it twice, lost Ferghana, and spent his early twenties moving between mountain valleys with a few hundred men.

What that was actually like

The book does not dramatise it, which is why it lands.

At Dikh-kat, in the hills, he records that he constantly made excursions among the mountains round about, and generally went barefoot, and from doing this so much his feet became so that rock and stone made no difference to them. On one of these walks, between the afternoon and evening prayers, somebody saw a cow going down a narrow ill-defined road. I wonder which way that road will be going, Babur said. Keep your eye on that cow; don't lose the cow till you know where the road comes out. One of his companions made a joke: if the cow loses her way, what becomes of us?

In the winter several of his soldiers asked leave to go home to Andijan, since there were no raids to be made. His beg pressed him to send something of his own wear to his brother by them. He sent his ermine cap.

And then the sentence the whole memoir turns on: It passed through my mind that to wander from mountain to mountain, homeless and houseless, without country or abiding-place, had nothing to recommend it.

That is a man in his early twenties who has been a king since he was eleven, deciding that this is no way to live.

What he did about it

He took Kabul, held it for twenty years, and in 1526 went down into India and won at Panipat. The dynasty he founded ruled northern India, in one condition or another, until the British deposed the last of them after the Mutiny of 1857.

It never called itself Mughal; Europe did. It called itself Gurkani — from gurkan, the son-in-law, the title Timur had adopted because he could not lawfully call himself khan. The apology of a man who was not a Chinggisid became the name of an empire in a country he never saw.

And the book

Browne notes what happened to it afterwards, and the trail is itself a small history of the dynasty. Babur wrote in Chaghatay Turkish. His grandson Akbar had it translated into Persian by one of his generals in 1589. That Persian version is what Leyden and Erskine worked from for the first English translation in 1826, and the Turkish original was not printed until 1857 — the year the empire it describes the founding of came to an end.

A prince who could not hold a valley in Ferghana produced the only autobiography of its kind written by any ruler in the Islamic world before the modern period, and three centuries of his descendants had it read to them in a language he did not write it in.