After coming of age, in 135 BC Han Wudi had signed another ‘peace-through-kinship’ treaty with the then shanyu. The matter had sparked a heated debate in Chang’an, and this had flared again in 134 BC when, with Xiongnu suspicions disarmed by the latest tribute bonanza, prospects for a surprise counter-strike, not to say a perfidious one, seemed particularly favourable. Old arguments were rehearsed, revenge of Gaozu’s defeat at Pingcheng was reinvoked, and Wudi now sided with the hawks; an elaborate plan was approved for luring the shanyu into an ambush in the town of Mayi (in northern Shanxi).
This time there was no disaster, just dismal failure. The Xiongnu got wind of the trick, wheeled about, vanished into the steppe, and repudiated the treaty. Five years of ‘phoney war’ ensued. Xiongnu raids continued but so did the frontier markets, at which cross-border trade flourished as never before. It was all part of the plan. In autumn 129 BC, when the markets were at their busiest, Han armies swooped on four of them. Despite the element of surprise, only one attack was moderately successful. Xiongnu losses were put in the hundreds, Han’s in the thousands, and Pingcheng remained unavenged. But two years later Wei Qing, brother of the emperor’s favourite consort and one of half a dozen charismatic generals to emerge at this time, redeemed the Qin First Emperor’s conquests by retaking the Ordos. It was ‘the first major setback for the Xiongnu since the days of Maodun’.6 Qin’s ‘Great Wall’ defences were reoccupied and settlements re-established on either side of the Yellow River’s northern bend.
This brought retaliation from the Xiongnu both east and west of the new salient and was followed by devastating countermoves from the Han. Throughout the 120s BC scarcely a year passed without ever larger Han expeditions probing ever deeper into Xiongnu territory. By 119 BC they were pushing north right across the Gobi desert into Outer Mongolia and north-west through Gansu to Ningxia. Han armies were now matching the Xiongnu for mobility and could support themselves in the field for several months. New commanderies, crammed with labour camps and soldier-settlers, ensured the security of the ‘Great Wall’ frontier, while Xiongnu losses, especially of livestock and pasturage, induced dissent within the nomadic confederation. The consequent defections may partly account for the improved performance of the Han forces as these new allies were deployed against their erstwhile comrades.
Success was real, but the price high. Sima Qian follows official practice by ‘scoring’ each engagement as if it were a rubber of bridge, or an exam paper. From his totting up of the hundreds of thousands of troops involved, the tens of thousands slain, the numerous generals and chieftains captured, and the vast herds of sheep, cattle and horses corralled, it appears that, for the Han, acceptable losses ran as high as 30 per cent and sometimes reached 90 per cent. If anything, the Xiongnu fared better in this respect – and they needed to; for while Han resources of manpower and provisions were practically inexhaustible and constrained only by the logistics of deployment, nomadic numbers were finite, their livelihood in terms of flocks and herds was vulnerable, and their only asset lay in the limitless terrain.
Having reclaimed the northern frontier and scattered the enemy, the Han might have scaled down their operations after 119 BC. That they would do no such thing looks to have been due to the intelligence-gathering of explorer Zhang. For although the exact chronology is uncertain, it seems to have been at about this time that his information on central Asian affairs was reviewed, he himself re-examined, and a new direction given to Han’s expansionist momentum. Instead of pushing ever farther north into the unrewarding wastes of Mongolia, Chang’an’s troops would veer west and set their sights on the flourishing states of central Asia, as reported by Zhang.
Initially Zhang’s discoveries had served as a distraction. In the course of his western odyssey, the explorer had crossed the deserts of Xinjiang, scaled the bleak Pamirs and descended to both the Syr Darya (Jaxartes River) and Amu Darya (Oxus River). The region along the former was called ‘Dayuan’, otherwise Ferghana and now eastern Uzbekistan, that along the latter ‘Taxia’, otherwise Bactria or northern Afghanistan. Since the Yuezhi had just overrun Bactria, they showed no interest in returning east of the Pamirs to oblige Chang’an in its feud with the Xiongnu. Their future lay south of the Himalayas, where they would be known as the Kusana (Kushan), would found one of northern India’s greatest empires, and in the first century AD would repay Zhang’s visit by sending to China the first Buddhist missionaries. But as yet ignorant of ‘the Enlightened One’, the Yuezhi in the early 120s BC opened explorer Zhang’s eyes only to the importance of the strange world into which he had blundered.
Had it not been for his decade-long detention by the Xiongnu, Zhang would have found Greek-speaking kings with names like Euthydemus and Menander still ruling in Bactria. Relics of Alexander the Great’s expedition, these Bactrian kings had been ejected by the Yuezhi in 130 BC, just months before Zhang’s arrival. Their magnificent gold and silver coinage was still in circulation and surprised Zhang by its novel use of portraiture. Each coin, he reports, ‘bore the face of the king [and] when the king died, the currency was immediately changed and new coins issued with the face of his successor’. Such a practice had never been known in China; since it could be construed as ennobling commerce and demeaning the sovereign by association with it, nor would it be.
West of Bactria stretched the great kingdom of ‘Anxi’, otherwise Parthia or Persia (Iran), where the Seleucids, also legatees of Alexander’s empire, had earlier been overthrown by the Parthian Arsacids; this Anxi extended to what Zhang calls ‘the western sea’, which is thought to be the Gulf rather than the Mediterranean. East of Bactria, the kingdom of ‘Shendu’ was of more interest. While exploring the Bactrian bazaars, Zhang had noticed ‘cloth from Shu [Sichuan] and bamboo canes from Qiong [also in Sichuan]’, both of which were said to have been imported via ‘Shendu’. Although some of explorer Zhang’s place-names are hard to identify, there is no question that ‘Shendu’ was India; its inhabitants ‘rode elephants into battle’ and even the Romans knew the country as ‘Sindu’ (after the Sind, or Indus, River). Since it was said to be several thousand kilometres east of Bactria, Zhang reasoned that it must ‘not be very far away from Shu’. Silk cloth and bamboo canes must therefore be reaching India direct from China’s extreme south-west.
Of all Zhang’s revelations, this was the one that had at first excited the most interest in Chang’an. As his ten years of Xiongnu captivity had demonstrated, access to central Asia was as yet fraught. Across the deserts of Xinjiang the Xiongnu controlled the route north of the ‘Great Swamp’ (nowadays the salt desert of Lop Nor), while Xiongnu allies, the proto-Tibetan Qiang, controlled that south of it. Only the cane-and-silk route from Sichuan to India and Bactria looked to offer a way of circumventing both. When Zhang had proposed that he lead a secret expedition to explore it, ‘the emperor was delighted’.
This must have been in c. 125 BC, the year after Zhang’s return from ‘the western regions’; for by 123 BC the explorer was back again, the south-western route having proved a cul-de-sac. There was indeed some unofficial trade between Sichuan and India; and Zhang stressed the importance of an intervening kingdom called ‘Dianyue’, whose people also rode elephants; it was probably Burma. But the hillsmen of Yunnan, between Sichuan and Burma, had been unimpressed by the affable Zhang, opposed his progress and murdered his colleagues. Further efforts would require military support, and this would not be forthcoming until twelve years later. Coinciding with the storming of Panyu (Canton) and the subjugation of Nanyue, Han troops would then force their way into Yunnan. Sima Qian, the ‘Grand Historian’, accompanied them, certainly as far as the now provincial capital of Kunming. But farther to the south-west their progress was halted in the vicinity of the Lancang (upper Mekong) River. The perpendicular terrain, as much as the population, would in fact keep the silk-and-cane trail shrouded in mystery for centuries. Only with the construction of the Burma Road in the run-up to the Second World War would a serviceable trade route finally link China and India over the dripping passes at this extremity of the eastern Himalaya.
Meanwhile Han operations against the Xiongnu had continued. In 121 BC they had been rewarded with the surrender of one of the shanyu’s subordinate kings, who brought with him 40,000 men and control of the Gansu corridor. Then known as Hexi (‘west of the [Yellow] river’), this vital neck of cultivation was secured, settled (700,000 were compelled or induced to remove there) and fortified over the next few years to as far as the Jade Gate (Yumen), the terminal of the extended ‘Great Wall’ near Dunhuang. The Han now had the equivalent of ‘a covered way’ leading into Xinjiang. It was this that led to explorer Zhang’s recall from obscurity in c. 119 BC (in the interim he had been demoted and nearly beheaded as a scapegoat for a recent defeat by the Xiongnu), and to his again being quizzed about ‘the western regions’.
According to Zhang, all the kingdoms of the far west so valued the produce and political endorsement on offer from Han China that they could be induced to accept some kind of feudatory status. The more martial peoples of Ferghana and other northern states would be keen to join the Han against their common Xiongnu enemy; and the more commercial peoples of Bactria, Parthia and India would comply with tributary conventions if they could be assured of Chinese trade. In this way, argued Zhang (or perhaps the Grand Historian on his behalf), all could be brought within the Han scheme of things. The emperor would be gratified by a constant stream of exotic products and visitors, ‘his might would become known throughout all the lands within the four seas’, and in time their rulers would ‘acknowledge themselves our foreign vassals’.7
It is noteworthy that neither trade nor alliance was seen as an end in itself; arguably they never would be, at least for as long as the Chinese empire lasted. Both were perceived as inducements whereby the lands discovered by explorer Zhang could be satisfactorily fitted into a traditional scheme of sino-centric geography. Modelled on the cosmos, this conceived of ‘All-under-Heaven’ being disposed in rings of concentric dependency that radiated outwards from the universal sovereign (that is, the Han emperor) and which shaded from the directly administered commanderies of the empire itself to its less directly ruled kingdoms, various indirectly ruled tributary states on the frontier, and finally feudatory dependants beyond it.
Resourceful as ever, in c. 119 BC Zhang responded to the emperor’s renewed interest with another master plan. Bactria this time took a back seat in it. The priority was to secure Xinjiang and its transit routes. Deprived of Chinese grain and manufactures, the Xiongnu had been resupplying themselves from Xinjiang’s oases. Each a small city-state, the oases had made Xinjiang what Zhang called ‘the right arm of the Xiongnu’. It must be ‘cut off’, and he knew just the people to do it. They were the Wusun, a tribe that had fallen out with the Xiongnu and, like the Yuezhi, been displaced by them. Now established somewhere in the north of Xinjiang (perhaps in the vicinity of modern Urumqi), they would be flattered by Han overtures and easily bribed into an offensive alliance. ‘The emperor approved of this suggestion,’ reports Sima Qian.
Reinstated in high office, in 115 BC explorer Zhang readied himself for a last odyssey into the land of ruddy cheeks and bushy beards, of fat-tailed sheep and shaggy camels with wobbly humps. To the emperor it must have sounded like a realm of make-believe, albeit an expensive one. Gifts comprising ‘gold and silks worth a hundred billion cash’ and ‘tens of thousands of cattle and sheep’ were entrusted to Zhang, along with an escort of 300 and a staff sufficient to supply ambassadors to all the countries beyond.
As an arena for imperial expansion, inner Asia has somewhat improbably been called ‘the rough equivalent of the Mediterranean Sea’ in Graeco-Roman history.8 Though as dry as the Mediterranean is wet and as harsh as it is balmy, the sands of what was later known as Turkestan glowed at sunset like ‘the wine-red sea’ and did indeed produce grapes; Zhang had earlier noted that they were fermented into a fine liquor that improved with age; and it was from Xinjiang that the vine would speedily be introduced into China. More to the point, the trading oases lay strung around the desert like an archipelago of islands. Inexperienced voyagers could safely navigate from one to another; and autonomous but vulnerable, they could then serve as ports of call and supply or as permanently garrisoned strongholds. The key to the wider world beyond lay in securing the ‘sea-ways’ of Xinjiang.
Zhang led his men with the confidence born of long years of hard travel. The Wusun were located, their strength ascertained, and his gifts well received. Interpreter-guides were provided for his envoys to Bactria, India, Parthia, Ferghana and Sogdiana (the later Samarkand). But the Wusun faced a leadership crisis and were ill informed about the might of Han China. Before committing themselves to an alliance, they preferred to send their own envoys to the Han court. Zhang accompanied them back to Chang’an. He had the satisfaction of seeing ‘their appreciation of Han considerably enhanced’ and of receiving the various delegations that his envoys to the other central Asian states would escort back to Chang’an.9 When Zhang died, probably in 113 BC, he was hailed as the pioneer ‘who had opened the way’ to the Western Regions. ‘All the envoys who journeyed to these lands in later times relied upon his reputation to obtain a hearing. As a result of his efforts, the foreign states trusted the Han envoys.’10
But Han comings and goings across Xinjiang were not trusted by the Xiongnu, who during the last decade of the century mounted major assaults on both the Gansu corridor and the northern frontier. These were counterproductive with regard to the Wusun, who, stampeded by Xiongnu threats, soon embraced the Han alliance envisioned by explorer Zhang. Signed in 105 BC, the treaty was not so much one of ‘peace-through-kinship’ as of ‘war-through-kinship’. Han Wudi got 1,000 horses, a willing feudatory and a doughty defender of Han interests in Xinjiang and beyond, while the Wusun welcomed lavish gifts, a powerful patron and a tearful imperial bride.
This Princess Xijun and her sizeable entourage proved the most influential element in the package. Married to a grey-bearded Wusun leader, then to his grandson – with neither of whom she was able to converse – the princess yet exercised considerable influence among her new kinsmen and would on occasion act as the Han emperor’s representative. But she never became reconciled to her fate and famously composed a song that would be remembered long after the Wusun themselves had been forgotten. The song tells of Princess Xijun’s exile in ‘a strange land on the other side of heaven’, where her house was a tent of felts, and her food just ‘meat with fermented milk as a sauce’. ‘I live with constant thoughts of my home,’ sang the princess, ‘my heart is full of sorrow. I wish I were a yellow [-beaked?] swan winging back to my home country.’11 The song is included in the Hanshu, the second of the ‘Standard Histories’ (so sequel to Sima Qian’s Shiji), which was compiled by members of the Ban family. One of them was Ban Zhao, the sister of the main author; and it was surely she, a noted scholar in her own right and in this case a peculiarly sympathetic one, who was responsible for incorporating so poignant a reflection of the homesick plight of a Chinese bride.
Painted T-shaped silk banner from Mawangdui, Changsha (Hunan). Excavated in the 1970s, Mawangdui’s three tombs yielded a dazzling array of artefacts. The banner was found draped over the innermost coffin of the Lady Dai (d. c.150 bc), a contemporary of the Former Han empress Lü.
Inscribed turtle plastron from Anyang (Henan). The authenticity of the Shang dynasty (c.1600 bc–c.1045 bc) was substantiated through study of the written characters inscribed on such shells. Early evidence of a written tradition, they record oracular responses obtained from the cracking of the shells when subjected to heat.
Elongated bronze figure from a sacrificial pit at Sanxingdui in Sichuan. Dated to c.1200 bc, this nearly two-metre-tall figure remains a stylistic anomaly. Such finds suggest that technically sophisticated cultures flourished beyond the parameters of the Yellow River’s classic ‘Three Dynasties’ (Xia, Shang and Zhou).
Bronze vessel (a zun with pan) from the tomb of the Marquis of Zeng (d. 433 bc) in Hubei. The casting of often colossal bronze vessels was one of the supreme achievements of ancient China. None were more encrusted with decoration than those found in the ‘warring state’ of Chu, to which Zeng was tributary.
The 1974 discovery of the Qin First Emperor’s ‘terracotta army’ was probably the greatest archaeological find of the twentieth century. Thousands of warriors and hundreds of horses are now on display at the site near Xi’an. More exotic artefacts, like four-horse chariots in bronze chased with silver and gold, have since been excavated, and the main chamber of the emperor’s tomb has still to be opened.
Jade burial suit of Princess Tou Wan. In the Han period (c.200 bc–ad 220) royalty were often buried in tailored jade. Besides advertising the rank of the deceased, jade was supposed to have preservative qualities. Each suit consists of hundreds of platelets, like the armour of the ‘terracotta warriors’, that were knotted together with silk, silver or gold thread.
The so-called ‘Tarim Mummies’ from the deserts of Xinjiang caused an archaeological furore when identified as not just un-Chinese but decidedly European. Though the origins, language and technologies of this ‘Charchan Man’, or ‘Ur-David’, are still uncertain, it seems that in the first two millennia bc much of what is China today was populated by non-Mongoloid peoples.
Though typical of the structures associated with the long northern frontier, these fortified remains are not those of the Great Wall but of a military granary. Located near Dunhuang in the Gansu corridor, they date from the Later/Eastern Han dynasty (ad 25–220).
The Three Kingdoms period (ad 220–265) is best known as the setting for the later Romance of the Three Kingdoms. A scene from this enduringly popular novel is here depicted on a lacquer-coated box of the Ming period.
Under the Tang dynasty (618–907) imperial China stretched out along the Silk Road to achieve its greatest extent. Even the peoples of Mongolia, Kazakhstan, Kyrghystan and Uzbekistan tendered tribute, typically in horses (seen here being ridden by their escorts in a silk painting of the Song dynasty). Central Asia’s bloodstock was prized, its sports (like polo) were adopted and its fashions aped. Ceramic figurines of the period depict elaborate female head-dresses and booted cameleers with noses of most un-Chinese proportions.
Seventeenth-century Tibetan fresco of the Buddha’s First Sermon. Buddhism reached China in the second century ad, some five hundred years after the Buddha lived and preached in India. It spread rapidly during the long ‘Period of Disunion’ (220–581) and thereafter enjoyed imperial patronage under the Sui and Tang dynasties.
The ‘Ancestor Worshipping’ cave at Lungmen (Longmen) outside Luoyang in Henan was commissioned in the 670s by Empress Wu, wife of Tang Gaozong and China’s only female ‘emperor’. The tradition of rock-cut images of the Buddha had come from India by way of Afghanistan (Bamian) and Xinjiang.
Traffic along the mountain-and-desert trails known as the Silk Road was two-way. Buddhist missionaries heading east into China were soon complemented by Chinese pilgrims heading west for India. Most passed through Dunhuang in Gansu, from where came this ninth-century silk painting of a travelling monk.
While the princess was accustoming herself to life in a yurt, operations to secure the Xinjiang route had continued. Though by no means an unqualified success, in 104 BC they were sufficiently advanced for the dispatch of a military expedition to the capital of Ferghana (it lay between Khojend and Samarkand). This was much the most ambitious foray made by the Han into central Asia, both in terms of distance (about 4,000 kilometres – 2,500 miles – from Chang’an) and of troops (6,000 cavalry and over 100,000 infantry). Its purpose was twofold: to punish Ferghana for its reluctance to be drawn into a subordinate relationship with Chang’an, and to obtain some of its famously ‘blood-sweating horses’. Equine bloodstock, always an important item in nomad–Han trade, was a speciality of the region and would long remain so. Li Guangli, the general who commanded the expedition, was anticipating William Moorcroft, superintendent of the English East India Company’s stud farm, who in the 1820s would spend six years tracking the finest stallions in Asia to exactly the same region and so launch European participation in both the exploration of central Asia and the ‘Great Game’.
Moorcroft would die on his quest, and at first Li Guangli fared only slightly better, being defeated by the terrain as well as the enemy. Returning to Dunhuang in disgrace, he was refused re-entry into Han territory and sent back to Ferghana in 102 BC to try again with augmented forces. This time he succeeded. He won a string of victories, obtained the king of Ferghana’s head (his subjects had decapitated him), commandeered 3,000 assorted horses, and left the neighbouring states in no doubt as to the seriousness of Han intentions. After such a demonstration ‘all the states of the Western Regions were shocked and frightened’, says the Hanshu.12 Most of the Xinjiang oases now sent missions to Chang’an, and Bactria and Parthia would follow suit.
But the cost of the exercise had been all too commensurate with its achievement. Of the over 100,000 men who had marched out of Dunhuang with Li Guangli, only 10,000 straggled back. Even Han China could not long sustain human losses on this scale, nor the financial expenditure involved in winning allies like the Wusun. The new state monopoly on liquor probably produced as much resentment as revenue, and voices were increasingly raised in protest over expansionist policies that, while impoverishing the whole empire, gratified only the imperial court.
Among these voices, though somewhat tentatively, was that of Sima Qian. In 99 BC, when the Grand Historian had been working on his Shiji for about seven years, a general called Li Ling served under Li Guangli’s command in an expedition against the Xiongnu during which Li Ling suffered defeat and was forced to surrender. Li Ling’s courage was not in question; with just 5,000 infantry, a ridiculously inadequate force, he had fought on for weeks against impossible odds until not an arrow was left. Sima Qian knew Li Ling and respected him, although he says he was not a particular friend. In interceding with the emperor on his behalf, the Grand Historian meant only ‘to broaden His Majesty’s views’ and counteract the calumnies of others. But somehow ‘our Enlightened Ruler did not wholly perceive my meaning’. Sima Qian was accused of trying to exonerate Li Ling in order to disparage the great Li Guangli. This in turn was construed as an attempt to deceive the emperor, a capital offence.13
Faced with such a charge an official was expected to commit suicide. His honour would thereby be partially redeemed, the emperor would be unimplicated in his fate, and the laws and punishments would remain in utopian abeyance. For in the Confucian ideal, officials attracted by the magnetic effect of the emperor’s moral example were supposed to be sufficiently righteous and high-minded to recognise their guilt and penalise themselves. Not to do so would be to acknowledge their inadequacy for office in the first place and so cast a slur on the judgement and moral calibre of the emperor, a surefire way to a death yet more painful.
Except as a delaying tactic, it was pointless to plead one’s innocence. Imprisonment pending trial was a euphemism for torture pending confession; the trial itself was intended merely to hear the confession and award punishment; and for an accused to be exonerated was almost unheard of. But of late there had arisen a custom whereby an accused might, at great expense, purchase commutation of a death sentence into something less draconian, such as demotion or exile. This was what explorer Zhang had done when disgraced in battle with the Xiongnu. The practice seems to have been favoured by Han Wudi as a means of raising revenue and was much abused for this very purpose.
Grand Historian Sima Qian declined both these options. He could neither raise the necessary funds for commutation nor resign himself to a suicide that would mean leaving his great work unfinished. His father had started the Shiji and had entrusted him with the task of completing it. ‘How can I, his son, dare to neglect his will?’ he asked. Filial piety as well as personal attachment dictated that he persevere. And though horrible to contemplate, there was in fact a third option. Worse than suicide, more disgraceful than execution, and too humiliating to be named in the long letter he wrote explaining his decision, it would at least permit him to finish his life’s work.
This extreme penalty, ‘the punishment of rottenness’ as it was called, involved a short detention in ‘the silk-worm chamber’, a cell reserved exclusively for judicial castrations. After repeated beatings and the inevitable confession, Sima Qian saw his final emasculation as an abject surrender, just like Li Ling’s. ‘Together we became a sight for all the world to laugh at in scorn. Alas, alas! Matters such as these, it is not easy to explain in detail to ordinary people.’ The physical shame, the ignominy brought on his ancestry and family, and the contempt in which he was popularly held for the rest of his life drew from the Grand Historian a bitter outpouring. Such was the price of scholarship, such the debt that posterity owes for the first and finest of China’s Standard Histories.
When I have truly completed this work, I shall deposit it in the Famous Mountain. If it may be handed down to men who will appreciate it, and penetrate to the villages and the great cities, then though I should suffer a thousand mutilations, what regret should I have?14
By the time of his death he had brought his great work up to the reign of Han Wudi. He had also included enough innuendoes to alert posterity to both the emperor’s failings and his own likely bias. Whether he actually outlived his tormentor is not certain.