Seldom can more mismatched adversaries have squared up to one another. The Song, with a massive army but a wretched military record, had been retreating, on and off, for nearly three hundred years; the Mongols, though fewer and occasionally repelled, had been advancing for fifty years, had never lost a war, and had conquered most of the known world. Their onslaught, more sudden than that of the Arabs, more widespread than that of Alexander, and more traumatic than either, was like nothing that history had ever known – or would again. According to both Christian and Muslim sources, bleached bones and charred timbers marked their trail; the slaughter of those who offered resistance was generally as comprehensive as steel and muscle could manage. Urbanisation in much of Asia was halted in its tracks, fragile agricultural systems never recovered, and population figures are thought to have nosedived. Like the plague, which they may unintentionally have spread, the Mongols came from nowhere and were suddenly everywhere.
It had all begun in north-east Mongolia in 1206 when a resourceful and audacious twenty-year-old called Temujin, scion of a junior branch of the neither numerous nor powerful Meng people of that region, having united the Meng and overrun more formidable neighbours, was acclaimed their ‘Chinggis Khan’ (‘Oceanic Qaghan’ or ‘Universal Ruler’). A great white banner was unfurled at the windswept gathering near the source of the Onon River where this ceremony took place, a basic military-cum-administrative organisation was set up, and plans laid for further triumphs. The unity of the peoples of the Mongolian steppe, no less than the authority of their new ruler, depended on maintaining the momentum of success. Only by more raids, more proof of invincibility and more redistributed plunder – livestock, gold, furs, textiles, foodstuffs, craftsmen, presentable women and enslaveable children – could loyalty be rewarded and the allegiance of others attracted.
Though ‘Meng’ or ‘Meng-wu’ was just the Chinese rendering of ‘Mongol’, ‘Mongolian’ better describes the composition of Chinggis’s following. Turkic-speaking neighbours, some of whom had adopted Nestorian Christianity, were already more numerous than native Mongols. Included, too, were the Tatars, another Mongolian people whose name both Chinese and European writers would often prefer to that of ‘Mongol’ – much as the Mongols would prefer ‘Jin’ (or ‘Chin’) for Song China. Other early adherents included numerous Uighurs from the northern city-states of Xinjiang. They would play a major part in Mongol administration. Their script was adapted for Mongol use, and their confessional odyssey – through Buddhism to Manichaeism and Islam – lent an ecumenical diversity to the profile of the Mongol court. While much of contemporary Eurasia counted professing the wrong faith a greater incitement to war than the possession of conspicuous wealth or irresistible women, the Mongols begged to differ. Their demands, though exhorbitant, rarely extended beyond the portable and the serviceable; freedom of worship, for those who lived to enjoy it, was officially guaranteed.
First among their prosperous southern neighbours to feel the impact of Mongol power was Tangut Xia, or the ‘Great State of White and High’. Mongol armies rode in from the Gobi in 1209 and laid siege to the Tangut capital beside the upper Yellow River in Ningxia. It was the first Mongol assault on a sedentary people with a fortified city and it fared indifferently. A Mongol attempt to flush out the defendants by destroying their irrigation system flooded the Xia defences but also swept through their own positions. The Mongols withdrew, wet, undefeated and yet not exactly victorious. They had secured Xia’s submission, plus a Tangut bride for Chinggis, but left the state intact. This would turn out to be a mistake.
The Xia campaign served as a trial run and strategic prelude to the more rewarding task of raiding the northern territories of the Jurchen Jin. Begun in 1211, the raids had climaxed with the capture of the Jin capital (the later Beijing) in 1215 and the attachment of much of Manchuria, whose remaining Khitans transferred their allegiance to the Mongols. Chinggis then turned west. The submission of the Kara Khitai (or Western Liao) of Turkestan in 1218 was easily won, though not so that of their more powerful neighbour, the Muslim kingdom of Khwarazm (roughly Uzbekistan). While its shah took refuge on an island in the Caspian Sea, where he died in 1221, his son went on the run and led the Mongols a none-too-merry dance through what are now northern Iran, Afghanistan, Pakistan and then back again. This bloodiest of campaigns ended in 1223 with Khwarazm’s collapse and the dispersal of its forces. Many entered India, where they swelled the ranks of the Muslim conquistadors under the Delhi sultanate; others gravitated towards Syria and there carved for themselves ‘a small niche in history as the Muslim force that (more or less incidentally) finally evicted the Crusaders from Jerusalem in 1244’.14
Further Mongol probes in the 1220s into Azerbaijan, Georgia and Russia presaged the great advance into Europe of 1241. Meanwhile Chinggis Khan ended his foreign adventures as he had begun them, with an attack on Tangut Xia. From end to end the ‘Great State of White and High’ was blackened and brought low before its capital was again besieged. It was still holding out when in 1227 Chinggis died, apparently from complications resulting from a fall from his horse. (Mongols were forever falling off horses, according to Brother William of Rubruck, a missionising Franciscan who, with encouragement from Louis IX of France, reached the Mongol capital in 1253; for men who lived and often slept in the saddle, it was an occupational hazard.) Days later, Xia surrendered and the khan’s corpse could be taken home to be interred on a mountain in eastern Mongolia. As for the ‘Great State of White and High’, within a couple of decades it too had disappeared under the sands of history.
The Tanguts of Xia had brought all this upon themselves by failing to support ongoing Mongol operations against the Jin and then signing a unilateral peace treaty with them. From 1216 to 1223, while the main Mongol forces were subduing Khwarazm in the west, one of Chinggis’s generals had rampaged up and down the Yellow River from Shandong to Shaanxi at the head of an army composed principally of Khitan and other ex-Jin subjects, including many Han Chinese. Experience in siegecraft was gained and sophisticated weaponry amassed; but the Tangut absence, and then defection, had limited its territorial gains.
Once the succession to Chinggis had been resolved with the elevation of his third son, Ögödei, as Great Khan (and with the installation of other sons as subordinate khans elsewhere in Eurasia), offensive operations in China were resumed. In 1230 Ögödei himself campaigned in Sichuan, ever the strategic key to an all-China dominion. Then in 1233–34, with some gleeful assistance from the Southern Song, he completed the conquest of the last Jin redoubts in Henan. All northern China down to the Huai River was now under the Mongols’ control and at the mercy of their military commanders. A sinicised Khitan official favoured by Ögödei remonstrated against indiscriminate pillage and supposedly convinced his master that depriving China of its people in order to turn it into pasturage would be less rewarding than sparing the population in order to tax their labours and enjoy their produce. He set up a basic administration, reinstated the examination system (though practically no successful candidates were entrusted with office) and briefly curbed the excesses of military rule. But with Ögödei’s death in late 1241, this first Mongol experiment in Chinese-style government ended. The examinations lapsed and tax collection was farmed out among a coterie of central Asian merchants and moneylenders who had attached themselves to the Mongol leadership. Their extortionate demands and brutal enforcement methods laid waste the countryside. The northern Chinese, though spared annihilation to make way for grass, thus got an early taste of a most pernicious form of colonial oppression. ‘It is difficult to imagine a more ruinous or exploitative economic system,’ says a contributor to The Cambridge History of China.15
Succession crises occurring on and off through the 1240s brought a lull in Mongol operations in China. They ended with the elevation of Möngke, a grandson of Chinggis, as Great Khan in 1251. Like Ögödei, Möngke apportioned responsibilities for the wider Mongol empire among his kin. A cousin headed the Kipchak khanate in Russia, otherwise the Golden Horde (‘horde’ being derived from the Mongol ordos or ordu, meaning an ‘encampment’, like ‘Urdu’, the ‘language of the camp’, as spoken by those Mongols, or ‘Mughals’, who would one day set their sights on India). A brother of Möngke, Hülegu, was dispatched to Iran, where he would found the Persian Il-khanate, capture Baghdad and overthrow the Abbasid caliphate. And another brother, Khubilai, was given responsibility for operations in China.
Khubilai resumed Ögödei’s offensive in Sichuan and continued south through its steep ravines into still-little-known Yunnan. There in 1253 he defeated the proud kingdom of Dali, which had succeeded that of Nanzhao in the tenth century. ‘Thus it was the Mongols who first made Yunnan a directly administered province of China.’16 In the course of these operations, contact was established with Tibet, or rather with its dominant lama-hood. Mongol–Tibetan relations were thereafter close, though their nature is obscure and controversial. Chinese sources imply a conventional subordination with military and administrative arrangements that provoked much Tibetan ‘banditry’ and necessitated several punitive invasions. Tibetan sources take little account of political relations and stress the influence of Tibetan Buddhism within the Mongol court. In neither case is it clear which of Tibet’s several peoples was involved or what part of the Tibetan region was affected. After careful research, Herbert Franke has concluded that no Mongol incursion reached central Tibet, that ‘most of Tibet proper remained outside the direct control of the Sino-Mongol bureaucracy and that even the border regions were throughout the [Mongol] Yuan dynasty an unruly and troubled region’.17
Since he had hitherto been supposed the least martial of Mongol princes, Khubilai’s success in Yunnan caused some surprise, while his willingness to spare lives and reinstate the king of Dali looked like a repeat of Chinggis’s mistake in dealing with Xia. From Yunnan an expeditionary force descended the Red River into Vietnam, whose Tran dynasty took fright and formally recognised Mongol suzerainty. The landward encirclement of the Southern Song was now complete. It remained only to cut off any seaborne assistance from the Korean kingdom, an objective that was attained, after numerous bloody incursions, when the Koreans submitted in 1259.
Meanwhile Möngke had come south in person in order to deliver the coup de grâce to the Southern Song. In time-honoured fashion a many-pronged attack was planned with the main thrust coming from Sichuan in the west. Möngke advanced there in 1258, Khubilai pushed down to the Yangzi through Hubei in 1259, and in the same year another Mongol prong prodded into Anhui. Southern Song resistance proved unexpectedly resolute; yet it was already crumbling when another Mongol succession crisis brought an unexpected reprieve. For in 1259, either from dysentery or a direct hit from a Song missile, Möngke died. Operations were immediately suspended. Khubilai, who was poised to cross the Yangzi to Wuchang, did so, but then reluctantly headed north to contest the succession. Before he got there, one of his brothers mobilised against him, whereupon Khubilai declared himself Great Khan. There then ensued a many-sided war fought largely in Mongolia between the contending Mongol princes (1260–64).
The war effectively ended Mongol unity. Khubilai emerged as the Great Khan with direct control of the Mongolian homeland, northern China, Manchuria and Korea, plus nominal authority over the entire empire as successor to Chinggis; but his khanate was just one of four, all vast Eurasian powers that now acted more like fraternal states than constituent parts of a single empire. The so-called pax Mongolica was deceptive. Travellers with information to share, merchandise to sell or expertise for hire passed freely through the Mongol lands; in what would prove to be the swansong of the Silk Road, east and west engaged in a fruitful exchange of technologies and ideas as well as luxury goods. But familial ties among the khanates were no longer sustained by a common purpose, and the overarching claims of the Great Khan provoked more dissension than collaboration. Thus Khubilai and his successors in China, though the one celestial dynasty to rule something approximating ‘All-under-Heaven’, would do so only in theory.
Nearly fifty when in 1264 he emerged triumphant from the Mongol war of succession, Khubilai Khan is traditionally portrayed by Chinese historians as an example of a ‘barbarian’ ruler who responded well to Chinese acculturation. He had already located his capital south of the steppe within what had been Jin territory at a place called Shangdu, about 300 kilometres (185 miles) north-west of Beijing near the Hebei/Inner Mongolia border. Marco Polo would wax lyrical about Shangdu’s palace and hunting park; and Samuel Taylor Coleridge would make it sound even more exotic, not least by spelling Shangdu as ‘Xanadu’. In Yunnan and then in Mongolia, Khubilai’s success had owed much to the manpower, revenue and advice available from his Chinese subjects. His fellow Mongols despised such indulgence of a conquered people, decried his adoption of a settled lifestyle and ridiculed him for preferring Chinese forms of sovereignty over the cut-and-thrust charisma of traditional steppe dominion. Indeed, the war of succession between the Mongol princes had been seen as a conflict between opposing styles of rulership as much as contending personalities. But whether willing convert or wily statesman, Khubilai seemed enthralled by his Chinese dominion at the expense of his Mongol heritage.
Traditionally his imperial reign is dated 1279–94, so prolonging to the maximum the Song’s exclusive claim on the Mandate and foreshortening the duration of Mongol rule. In reality Khubilai had been active in China from the early 1250s and supreme throughout all but the south from 1260. In that year he adopted a Chinese reign-name and calendar; it was evidence of imperial intent. In 1264, victorious over his Mongol rivals and acknowledged as Great Khan, he moved his capital still farther south into core China. Shangdu was retained as a summer capital, but for most of the year court and government would now operate from what had been the Khitan Liao and Jurchen Jin capital in the heart of Hebei. Renamed Dadu (‘Great Capital’), and later incorporated into the Ming city known as Beijing, its construction began in 1266/67 and took eight years. The layout was painstakingly Chinese. Based on an idealised city as outlined in that repository of ultra-traditionalism, the Zhouli (‘Rites of Zhou’), it advertised Khubilai’s assumption of the Mandate. Within high rectangular walls aligned with the compass points and pierced with towered gateways, the grid of Dadu’s wide thoroughfares radiated from an inner walled city wherein stood the south-facing palace. Though commissioned by a Mongol emperor and then realised by a Muslim architect, its remains constitute the earliest surviving example of a quintessentially Chinese imperial capital. As emperor, Khubilai would be the first to rule all China from the future Beijing; and as Great Khan he was the first to preside over Mongol Eurasia from China.
In 1272, while the Southern Song yet reigned in the south, Khubilai clearly signified acceptance of the Mandate. He did so in a formal announcement declaring that his dynasty was to be known by the unusual title of (Da) Yuan, ‘(Great) Originator’. Historians ought to be eternally grateful. The tiresomely confusing habit of recycling old dynastic names had finally been broken. As he proudly explained, ‘Yuan’ derived neither from his state of origin, like Qin and Han, nor from some feudal dukedom, like Sui and Tang. ‘In all these cases, they [the dynasties] fell prey to the ingrained habits of common people . . . [and] adopted momentary measures of expediency for the sake of control,’ he declared.18 Clearly they were doomed. Khubilai, successor to the ‘sage-like’ Chinggis Khan and now ruler of the largest empire Heaven had ever seen, was above such parochialism. On the best possible advice, he had sourced his dynastic title in the ‘Book of Changes’ (Yi-jing, I-Ching), perhaps the most venerable of all the ancient classics. This transcended previous practice; and such was the exalted provenance of the new title that there could be no question of sharing the Mandate with someone else or conceding its duplication. The Southern Song were, by implication, delegitimised and must bow in submission.
To that end, military, and more especially naval, operations were again under way. Abandoning the idea of an advance from Sichuan, Khubilai had resolved on a frontal attack across the Yangzi. The prerequisite for this was gaining control of the great river’s Han tributary, on which the twin cities of Xiangyang and Fancheng (now amalgamated as Xiangfan) constituted Song’s formidable northern bastion. Facing one another across the Han River, both were heavily fortified and well supplied with the shipping with which to intercept any Mongol armada or bring in reinforcements and provisions if besieged.
They were indeed besieged. Invested in 1267/68, the twin cities held out until 1273. In one of the most protracted and celebrated actions in China’s history, the Mongols soon discovered that cavalry were useless against ships and unhappy under firebomb attack. Summoning shipwrights and sailors from Korea and Shandong, experts in siege engines and ballistics from the Muslim west, engineers experienced in the use of gunpowder from the north, and vast contingents of Han infantry, the Mongols pressed the blockade in most un-Mongol fashion.
Heroic endeavours are recorded on both sides, none being more desperate than that of two Song colonels, both called Chang. When in 1272 Fancheng’s outer defences were breached, the Changs assembled a supply convoy of a hundred giant paddle-boats and, churning their way to the relief, tried to run the Mongol blockade. They succeeded, though at a heavy cost in men, ships and supplies. One of the Colonels Chang was lost on the approach, the other while trying to escape downriver under cover of darkness. Later in the same year, Song engineers built a pontoon bridge of timbers across the river to connect the two cities. The Mongol commander reportedly met this challenge with craft equipped to cut through the bridge with specially designed ‘mechanical saws’. Needham interprets this to mean that the boats were again paddle-powered and that the, presumably, circular saws were so mounted as to be run by the river’s current turning the paddle-wheels, and so the saws, when the boats were stationary against the bridge.19 More prosaically the pedalling crew could have been linked directly to the saws in an arrangement like that of a treadle grindstone.
In Marco Polo’s ‘Description of the World’, it is claimed that Polo himself was present towards the end of the Xiangfan siege. In fact, he supposedly introduced the Mongol command to the two catapult experts – ‘a German and a Nestorian Christian’ – who devised the trebuchets, each capable of launching rocks weighing over 100 kilograms (220 pounds) that eventually battered the defenders into surrender. This is almost certainly a fabrication. Other sources say the experts were Muslims, and modern scholars doubt whether Polo had yet reached China. But whether Polo himself was responsible for making the claim is uncertain. It could just as well have been inserted by others, either by the far from reliable writer who allegedly wrote down Polo’s word-of-mouth narration or by one of the many who further elaborated on his story over the following centuries.
So problematic is the provenance of the various versions of this celebrated text, and so apparently minimal Polo’s authorial control, that revisionists have had a field day. Some question whether Polo ever got to China at all. The fact that he never notices the Great Wall is offered as evidence that he didn’t, though since ‘long walls’ are little mentioned after the Sui, and since the Great Wall, as we now know it, was a creation of the Ming, this could be taken as evidence that he did. Other omissions – tea-drinking, foot-binding and the Chinese system of writing – deserve consideration. Yet, while in travel-writing the commonplace is notoriously easy to forget, incidental description is hard to invent. It has been suggested that he could have plagiarised some now lost, possibly Persian, source. The process of transcription must then have been one of elaborate collusion, a feat that its slapdash composition argues strongly against. In the end, trying to disprove his narrative proves as inconclusive as trying to substantiate his itinerary. And whatever Polo did, or wherever he went, the text itself retains an immediacy that it would be perverse to ignore.
When at last the twin cities fell in 1273, there was a year’s delay before the Mongol advance to the Yangzi, and on down that river towards Hangzhou, could begin in earnest. Again it met stiff resistance. A seasoned official called Bayan was given command of the Mongol armies; Polo calls him ‘Hundred Eyes’. Khubilai himself took no part in the campaign other than to urge clemency wherever possible. Bayan did just that, enticing defections, conciliating opponents, massacring the population of only one or two major cities, and winning a series of great battles. In Hangzhou, the weak young Southern Song emperor (Song Duzong, r. 1265–74) chose this moment to die. It was no great loss. His place was taken by a five-year-old and his powers by a dowager empress. Meanwhile Jia Sidao, the last in a long line of disposable Song ministers, took the blame. For the widespread disaffection and the military defeats, he also paid the price and was quietly murdered. Though savaged by later commentators, Jia Sidao was probably no more culpable than Cai Jing or Han Tuozhou. Not even a correct understanding of the Supreme Ultimate could by now have saved the Song. In late 1275 ‘Hundred Eyes’ Bayan was approaching Hangzhou.
Happily the surrender and occupation of the great city early in the following year belied the Mongols’ fearsome reputation. The city was spared, the imperial tombs protected (though later plundered), officials pardoned, and the dowager empress treated with respect. Taken north to Dadu, she would be befriended by Khubilai’s empress, while the last Song emperor would opt for a monastic life and was eventually exiled to Tibet. Even the discovery that courtiers had smuggled two of his brothers out of Hangzhou and down the coast by sea brought no reprisals.
It did, though, prolong resistance. Leapfrogging from port to port ahead of the Mongol advance, the young Song princes and their court-in-exile remained on the run for three years. Each prince in turn was acknowledged as emperor, one succeeding when the other died. From the Leizhou peninsula in the extreme south (near Hainan island), an escape was contemplated to Champa (in southern Vietnam), then abandoned in favour of a return to the offshore islands at the mouth of the Pearl River. There, somewhere in the archipelago that includes Hong Kong island, the Song flotilla was finally surrounded and the young emperor’s ship sunk. It is said that a loyal follower took the seven-year-old emperor in his arms and leapt overboard as the vessel went down. The year was 1279.
With the incorporation of the south into Yuan China, four centuries of division came to an end. Not since the disintegration of the Tang empire had all China been united under one ruler. The conquest of the Southern Song was Khubilai’s supreme achievement, ‘and it was indeed a mighty conquest’, says Polo, ‘for in all the world there was no kingdom worth half as much’.20 The population of southern China, no less than its wealth, was probably double the north’s and may have exceeded that of all the rest of the Mongols’ domains put together. With such assets the Yuan dynasty might well have asserted its authority over the other Mongol khanates and so added a reunited Mongol empire to that of China. World dominion beckoned, and Khubilai seems to have been tempted. Yet within less than a century the Mongols would be bundled out of China back to the windswept steppe from which they had come. For the Yuan, as for the Qin and the Sui, assembling an empire was one thing, consolidating it quite another.