Empire of Horses

by John Man

  For Zhizhi to deal with the Wusun proved a relatively easy task, aided by the death of the emperor, a severe famine in 48–47 BC, power struggles in the new court, and cost-cutting by the frugal new emperor, 27-year-old Liu Shi, ruling as Emperor Yuan. Zhizhi also bought himself time by exchanging envoys with Han, asking for the return of his son, who had been kept as a hostage for the last decade. For the sake of a quiet life, the new emperor agreed. A general named Gu Ji volunteered to deliver the prince all the way to Zhizhi’s court as a way to guarantee his gratitude. But, says Sima Guang, ‘for some obscure reason’ – perhaps in revenge for Han’s generosity towards his brother – ‘Zhizhi went berserk and ordered his men to execute Gu Ji and his retainers.’

  Now Han had cause to join Zhizhi’s brother Huhanye in an attack. Realizing his mistake, Zhizhi decided to lead his people further west, out of harm’s way. Kangju – a small kingdom beyond Wusun in the steppes west of Lake Balkhash, on the Syrdarya River, in present-day Kazakhstan – saw a chance to assert itself by allying with Zhizhi, attacking Wusun and making Zhizhi King of the Wusun. ‘Zhizhi loathed the Wusun with vile bitterness, so when he heard the proposal it was beyond his wildest aspiration.’ There followed a further migration into Kangju – an unrecorded epic across 2,400 kilometres, taking three months – an exchange of daughters as wives between the chanyu and the Kangju king, and finally a successful joint attack on the Wusun.

  Back in the Xiongnu heartland, Zhizhi’s migration had left the centre ground vacant. Since Huhanye’s people had pretty much exhausted the reserves of their new estates, he led his people north again, back to the more fertile pastures from which he had fled not long before. He was a vassal now, but safe, and with a guaranteed income from the emperor to whom he owed his power.

  If you ever drive along the new motorway between Zhangye and Wuwei, turn off to the left about 60 kilometres short of Wuwei, leave the snowy peaks of the Qilian Mountains behind you, and follow the gentle slope down to the town of Yongchang. The approach road ends at a roundabout, in the middle of which is a strange sight: three lumpish concrete statues, the central one being obviously Chinese, flanked by two that are equally obviously Roman, a woman and a soldier, his hand on his armoured chest in a Roman salute.

  This is the story of how these two Romans came to be here. Actually, it is three stories. The first concerns real Romans, and an astonishing suggestion that once upon a time, when the Han were building their Great Wall and reaching out westward, some Romans settled in China, not far from Yongchang. The second is the story of the story, of how a historical will-o’-the-wisp arose from nothing, and strengthened, and became an obsession – locally, nationally, internationally – and is now fixed in stone by these statues, and therefore (to many) true. The third is the continuing story of the Xiongnu collapse, which in this case involves the death of the appalling Zhizhi.

  The starting point is Rome in 59 BC, when Julius Caesar, Pompey and Crassus formed a triumvirate, the first of several that governed the empire. Crassus was a property tycoon who craved military glory. In 54 BC he led an army of seven legions, some 42,000 men, against Parthia (roughly present-day Iraq and Iran). The Parthians were horse-archers, the Romans specialists in hand-to-hand combat. In May, 53 BC, they met at Carrhae (now Harran in Turkey). The Parthians kept their distance and riddled the Romans with arrows, often faking a retreat and firing over their shoulders – the famous ‘Parthian shot’, which English corrupted into the ‘parting shot’ in a war of words. The Romans gambled that the Parthians would run out of arrows. They didn’t because the Parthian commander had organized relays of camels with new supplies. The Romans’ only response was to form a testudo, a ‘tortoise’ of interlocking shields. Still the arrows came, lobbed in high trajectories over the shields, flashing in underneath them into Roman legs, until their feet were nailed to the ground (as Plutarch put it in his life of Crassus). Nearly 20,000 were killed and another 10,000 captured, many of whom were sent over 2,000 kilometres across Parthia to guard its eastern frontier against the Xiongnu. At this point, they drop out of western history.

  And so, perhaps, into Chinese history, eighteen years later, thanks to the Xiongnu. That at least is what was claimed in a 1957 monograph by the Professor of Chinese at Oxford, the wonderfully named Homer Hasenpflug Dubs.4 Dubs was as eccentric as his name suggests, ‘well liked, but a bit odd’, as a senior member of his college, University College, put it. Dubs was much struck by the possibility that a village near Yongchang had once borne the name by which (he claims) the Chinese once referred to Rome: Li Qian. Dubs gives an elaborate and ingenious explanation as to how this could have happened.

  ‘Between 110 and 100 BC, there arrived at the Chinese capital an embassy from the King of Parthia. Among the presents to the Chinese Emperor are stated to have been fine jugglers from “Li Qian”.’ They were, he claimed, from Alexandria. The Chinese, who disliked polysyllables and initial vowels, dropped the A, transliterating the second and third syllables, lexan, as lí qián. They then used their version of Alexandria as their name for the empire of which it was a part. Thus Li Qian equalled Rome. If this was so, it didn’t last long. By the sixth century, the Chinese were referring to ancient Rome and its empire as Da Qin, Great Qin, as in the dynasty and empire created by the First Emperor, and later still Luo Ma, an approximation of Roma.

  Now we return to Zhizhi, in Sima Guang’s account. He had set himself up on the Talas River in southern Kazakhstan, just beyond the far borders of the Western Regions, threatening to gain control of the Silk Road. He was not a popular leader, for military successes had made him ‘insufferably insolent’. He argued with his royal Kangju ally, killed his wife (the king’s daughter) in a fit of temper and then turned on the Kangju nobility, ordering hundreds to be killed, dismembered and tossed into a river. Then, in full dictatorial control, he had his subjects build a fortress. Three times, Han sent envoys to demand Gu Ji’s insignia as a symbol of surrender. Zhizhi bought time by promising to submit, but never did. A Han minister warned that he should be taken out.

  In 36 BC a Chinese commander of the frontier, intent on glory, led 40,000 men westward, sweeping up fifteen minor kingdoms as allies and capturing a Kangju noble who was happy to brief the commander on the layout of Zhizhi’s castle, with its double palisade of huge logs, moat and rammed-earth ramparts. Hundreds of armed men crowded the battlements, yelling, ‘Come and fight!’ Horsemen galloped about, challenging the attackers. A band of foot soldiers drilled in close formation. In the siege that followed – engineers undermining the walls, infantry shielding crossbowmen, fires set against the wooden palisades – Zhizhi was wounded by an arrow in the face and fell down the battlement steps. Night fell, the fire spread and at dawn, to the boom of war-drums, the ‘ear-splitting shrieks’ of war-horns, troops piled earth against the inner ramparts, stormed up and over, and the fortress fell. Zhizhi died of his wounds, 1,518 of his force fell – a remarkably exact figure, so perhaps true – 1,000 more surrendered, and 145 were captured. With Gu Ji’s imperial tallies and his silken instructions found and with Zhizhi’s head sent off to Chang’an to be hoisted above the main street, victory was complete. The Han generals sent a message to the emperor suggesting he issue an imperial edict ‘disseminating a message to all our vassal states reading, “Kings! Whosoever dares to flout Han, no matter how near or far-flung, we shall have you executed!” ’

  It was the group of foot soldiers that seized Dubs’s attention. In his translation of Ban Gu’s text (a source for Sima Guang’s), the attackers noticed that ‘more than 100 foot soldiers, lined up on either side of the gate in a fish-scale formation, were practising military drill’. This, Dubs suggested, was a Roman testudo (tortoise) of interlocking shields. In addition, he argued, the idea of a double palisade was a standard Roman design, unknown elsewhere. The 145 prisoners were – Dubs surmised – the ‘more than 100 foot soldiers’, spared because they were not Xiongnu. Dubs’s conclusion was that these ‘Romans’ had escaped their Parthian
masters and fled to the Xiongnu, who had taken them on as mercenaries with some interesting military insights, and now found themselves in Chinese hands.

  These men would have become good frontiersmen, helping to guard the Wall, the narrow Gansu Corridor and the way west. It would have been natural for them to be settled in their own frontier town, named after themselves: Rome, or Li Qian. That’s as far as the story goes.

  But the threads wind on down the centuries to the present. Folklore took over. These Romans, now men in their forties with unrivalled experience, would have married local women and raised families, creating a community with looks very different from the locals. So different, perhaps, that it is possible to see their origins in some of today’s ‘foreign-looking’ inhabitants. That was the rumour. That was what drew me to Yongchang, and onwards, to the town supposedly built by Romans 1,300 years before China established any other links with the West.

  I was with Michael, seeking evidence to back the rumours. He had given me the background, so he was as intrigued as me. Those statues lent weight to his words. There they stood, the two Romans, with the heavy-duty stone plaque underneath: ‘In Memory of Li Qian’, challenging anyone to doubt the story.

  He questioned a passer-by. Yes, everyone knew that there were people around who ‘looked foreign’. We should ask at the local government office, which was just around the corner, up a side alley. There we were taken in hand by Song Guorong, a man with slicked-back hair and a silver tooth who was, by chance, an expert. He had edited a little book of papers on the subject. It had not yet been published, but he presented me with a proof copy, warning me that the subject was rather sensitive.

  ‘If you want to see a man who looks like a foreigner,’ he said, ‘you have to register with the police.’

  ‘But why?’

  ‘Because other foreigners come here and make trouble. So many other foreigners! Italians come here and see the local foreigners and call them brothers. People are tired of them, so it becomes difficult.’

  But not too difficult, because he immediately telephoned a friend.

  ‘How many have come?’ I asked, through Michael, as he put the phone down.

  ‘More than hundreds. They stayed for weeks. There was an Italian film producer, talking about making a film next year. Then a scholar came from Oxford. He looked foreign, like you. Maybe he was your brother?5 He wanted to research, but there was the SARS scare, you remember, and he couldn’t travel anywhere. But you can see. Mr Luo is coming. He looks foreign. He will show you. If you want to discover the truth,’ he added, ‘you could pay Mr Luo, maybe fifty yuan for half a day.’

  Mr Luo, who by an astonishing coincidence was waiting downstairs, was certainly anything but mainstream Chinese. He had hazel eyes, a thick head of hair, heavy black eyebrows, saturnine looks and a supercilious manner that would have served him well if he ever went for a bit part in a Mafia movie. He was a sharp dresser, too. Black polo-neck, lightweight grey jacket, black trousers, black slip-on shoes.

  Luo took us in hand. We would go to his village, Zhelai, where the foreign-looking people were. There used to be many more, he said. They had mostly gone away looking for work in Lanzhou, Beijing, Shanghai. But there were still seventeen or eighteen left.

  We set off along a dirt road under the motorway back towards the Qilian Mountains.

  ‘Mr Luo, how do people feel about looking different?’

  ‘Before 1978, they felt bad. Some had yellow hair, which they thought was bad, because people came to see them; so they dyed their hair black. But since China opened up to the outside world, they have become proud.’

  As we climbed, terraced fields gave way to big, open country, with glorious views towards the snows and glaciers of the Qilian Mountains. A beacon tower stood guard over stubbly wheat, which meant that the Han Great Wall was, or had been, somewhere around.

  ‘There used to be lots of towers round here,’ said Luo. ‘But villagers made them into fields.’

  Past a stand of yellow poplars, we came to a score of mud-brick houses and courtyards and thatched barns, all of medieval simplicity. This was Zhelai, once Li Qian – Dubs’s ‘Rome’ – where people supposedly looked foreign. It was hard to tell, because there was no one around. Bar a distant cockerel, the place was as silent as the quite large tomb which we were now approaching. It turned out not to be a large tomb, but a small section of earth wall with a fence around it. ‘Li Qian Historic Ruin’, said the sign, giving the impression that this was the very wall created by the Romans when they came to live here. It looked suspiciously like any other piece of the Han Great Wall, which you can see running all along the Gansu Corridor paralleling the new motorway. It must once have run right through the village. All the rest having long since vanished under ploughs, this 27-metre slab of rammed earth had been saved by being declared Roman.

  The wall was the centrepiece of the village’s claim to fame, a claim commemorated by one of the most incongruous pieces of architecture you can imagine. Raised on a low platform of pink granite stood four fluted pseudo-Roman columns, set in a square, supporting an architrave and a flat roof. It might have been copied from a tomb or shrine, but all it shielded was a stele recording that ‘Roman troops settled here’, and its date, 1994. It stood to the side of a courtyard or parade ground. Today, it seemed to be used for winnowing, for chaff blew around the waist-high walls.

  Alongside the village, but across a dry ravine in a much more barren area of hard sand and tussocks, were more walls, in better shape, the remains of a substantial fortress or campsite. It seemed a strange place to build a camp.

  ‘In ancient times,’ explained Luo, glancing round at the wasteland and raising his voice over a dusty breeze, ‘this was grassland. The ravine was a big river flowing from the Qilian Shan. People called the place “Treasure Basin” and saw the shape of a sleeping Buddha in the mountains. But the population grew, and so did the numbers of sheep and cattle, and the grasslands died, and the river dried up, so the people left.’

  Perhaps that explained why the place was empty?

  No, it was not empty. There were still 300 people here, in twenty-eight families. We would go and see. Buffeted by the hot wind, we walked back, through clumps of brittle grass that crackled in the breeze like wildfire. Luo promised to find a girl who was the most foreign-looking of them all. Xiao Dan was her name: Little Dan. She would be in the school.

  She wasn’t, because the school was shut. Luo led the way to her house. Little Dan’s aunt, who looked thoroughly Chinese, ushered us into a tiled compound and then into an airy marble-floored room, with a kang, a sleeping-platform, heated from underneath in winter, big enough for fourteen people. Little Dan was off on a picnic with the other children. She wouldn’t be back until much later. We couldn’t wait.

  ‘Could I see a picture of her at least?’

  The aunt hurried out, and back with a photograph. Little Dan, the most foreign-looking of all, had a fine, typical Chinese face. The only thing different about her, to my eyes, was that she had a hint of red in her dark hair. It is a little unfair to make a judgement on the basis of a one-day visit and a photograph, but the only evidence of foreignness in Zhelai seemed to be a girl whose hair was not completely jet black.

  Dubs’s paper might simply have become an academic oddity but for the passionate interest of an Australian writer, David Harris. As a post-graduate in Armidale, New South Wales, he heard of Dubs from a linguistics professor who had been at Oxford in Dubs’s time. The story seized him with such force that he set out to find the ‘lost city’. No one could have been worse prepared. He knew little of China, had no funds, and was trying to prove that westerners formed an important element in Chinese history, not a claim likely to win official support. His naivety produced some excellent material for his book, Black Horse Odyssey, but he managed no more than two quick visits to ex-Li Qian. He saw the bits of wall, and the ruins of the fortress, and came away determined to return, with good financial and archeological suppor
t. Bureaucracy stymied him. He never did go back.

  But his conviction and his search made a great story, which broke in December 1989 when the People’s Daily reported the Dubs-Harris hypothesis. Then local officials, among them Song Guorong, the editor of the book, took up the theme. If this could become better known, it would do wonders for local tourism, for there was really no other reason for Great Wall and Silk Road travellers to stop off at Yongchang. Up went the statues, and the fence around the old wall in Zhelai, and the Roman pavilion.

  More publicity followed. Google ‘Romans China’ and you will see what I mean. The story almost made it to the silver screen. North of Yinchuan, on the great loop of the Yellow River, a Ming Great Wall fortress has been turned into a film lot, the China West Film Studio, which has provided settings for some eighty movies. One of them was a Chinese-British made-for-TV co-production called Homeward Bound, being the story of what happened to the Romans when they tried to get back to Italy (not the point of the legend, but a nice twist). Much work was done, much film shot. Then what happened? I have been unable to find out. The money ran out, or some new executive pulled the plug. In any event, nothing more has been heard of the film.

  Now let’s see what all this amounts to. In brief: it’s a fine example of how a hypothesis becomes folklore, and then drifts towards being accepted as a historical fact.

  Glen Dudbridge, Dubs’s successor at Oxford, was scathing about the matter when I asked for his opinion. ‘Over the years, I have received a long string of enquiries about Dubs and his Roman city in China,’ he e-mailed. ‘The story refuses to go away, even though Dubs’s claims were methodically refuted by generation after generation of specialists during the past 50 years.’