In 1957, American sinologist Homer H. Dubs published A Roman City in China, a book detailing the academic’s theory that a group of Roman soldiers worked as border guards for the Western Han Dynasty at the empire’s western edge. These ancient expatriates, Dubs suggests, were survivors of Rome’s catastrophic loss to Parthia at the Battle of Carrhae in 53 BCE, afterwards moving to the eastern front of the Parthian Empire before eventually finding their way into battle with Chinese troops. In defeat, Dubs claims, the out-of-place legionaries were moved by the Western Han Dynasty to “a specially created frontier city, to which the Chinese of course gave their name for Rome, which was Lijien (now Liqian).”
Today, Liqian is a small village of earth-rammed homes located in Gansu province and, in the decades since the publication of A Roman City in China, Dubs’ theory has led researchers, archeologists and even geneticists to visit the town, all looking to answer the same tantalizing questions: did a Roman legion settle in ancient China? And, if so, are those living in Liqian today the descendants of these lost troops?
On an arid expanse of earth at the edge of the mighty Gobi Desert lies the county of Yongchang. Located in Gansu province’s north-central region, the area is a four-hour drive away from the provincial capital of Lanzhou via the National Highway G30, and three and a half hours by rail.
The journey by slow train, while scenic, puts Yongchang’s remoteness into perspective. China’s iconic green and yellowstriped trains travel the barren landscape like aged serpents, slowly passing forgotten town after forgotten town, crumbling, abandoned industrial infrastructure, derelict, earthen homes and the occasional golden domed mosque. It’s a three-and-a-half hour ride from Lanzhou to Jinchang, the nearest train station to Yongchang, and when I make the journey on a cool afternoon in late September, the train is virtually empty.
From the prefecture-level city of Jinchang, Yongchang town is a 40-minute drive over long stretches of rocky, unpaved road. While more remote places certainly exist within the boundaries of the Middle Kingdom, the area feels about as far away from the comforts and convenience of first-tier cities as one could imagine.
Yongchang, which had a population of around 235,000 in 2010 according to census records, has a long history. The area is set within the Hexi Corridor, formerly a key section of the ancient Northern Silk Road, and human habitation here dates to before the Common Era. It is within the boundaries of Yongchang county that one finds the town of Liqian, the fabled city founded by Roman survivors of the slaughter at Carrhae.
Liqian, according to Dubs, is the “most ancient Chinese name for Rome,” and the name first found its way into the history books when it appeared on the record of Chinese countries and cities for the year 5 CE. The settlement existed until 746 BCE, when, Dubs writes, “the Tibetans overran that part of China.” Inspired by Dubs’ research, Australian scholar and author David Harris suggested the town of Zhelaizhai could be the ancient city of Liqian. At the conclusion of the 20th century, Zhelaizhai officially adopted the ancient moniker of Liqian.
It is late in the afternoon when I arrive in Yongchang proper. While wandering the small city’s central area, the rumored Roman heritage of the region is on full display. The lampposts running along the city’s roads sport plate-sized metal plaques featuring Roman architecture and a helmet-clad legionary, and along the main thoroughfare stands a towering, concrete statue of two Romans flanking a Western Han Dynasty official."
“The statue was built in 1983 or ’84,” Mr. Zeng, a security guard at the Yongchang Museum, tells me. When I arrive at the small museum, which is located a short stroll from the aforementioned station, Zeng is the sole person at the site.
At the mention of my research into the county’s alleged Roman heritage, Zeng – who is eclectically dressed in a suit two sizes too big and a pair of weathered red Nikes – becomes visibly excited.
“The locals believe this story about the Romans,” he tells me, adding that descendants of ancient Romans still live in the area. Zeng leads me into one of the museum’s three exhibition halls, walking me through a small display on the area’s Roman roots, the centerpiece of which is a glass-topped box housing skeletal remains. The bones, he claims, are those of an ancient Roman immigrant to the region.
In the year 53 BCE, Marcus Licinius Crassus launched an unprovoked invasion of the Parthian Empire, marching an army of 35,000 heavy infantry, 4,000 light infantry and 4,000 cavalry into Upper Mesopotamia.
Crassus was an extraordinarily wealthy politician and, along with Julius Caesar and Gnaeus Pompeius Magnus (otherwise known as Pompey the Great), he was a member of First Triumvirate, a political alliance established to counter the Roman Senate. Crassus launched war with the Parthians in an effort to further his political aspirations and to earn himself a military reputation in line with that of his partners, Caesar and Pompey.
His grand military ambitions came to a head near the town of Carrhae, close to the present-day settlement of Altınbasak, southeastern Turkey. On his advance across the barren landscape, Crassus spotted an army of Parthian mounted troops. The commander ordered his troops to execute a large, hollow square formation, with heavy infantry forming the four boundaries of the square and his 4,000 cavalry inside. The battle formation was intended to prevent his army from being flanked by the enemy.
He then commanded his light infantry forward to engage the Parthians, although they were swiftly driven back towards the relative safety of the square by a flurry of arrows fired by 9,000 horse archers.
Parthian forces, which were dreadfully outnumbered by about four to one, then moved on Crassus’ square of soldiers, launching arrows at the Roman heavy infantry while maintaining a distance that kept them out of range of Roman projectiles.
Crassus then dispatched his son, Publius, to lead a charge of 1,500 mounted soldiers to push back the Parthian horse archers. While this tactic initially appeared successful, Publius and his men were baited into an ambush by 1,000 heavily armored mounted troops called cataphracts. The majority of Publius’ force was slaughtered in the maneuver and Publius is believed to have killed himself.
What happened next was a bloodbath, with both Parthian mounted archers and cataphracts descending on Crassus and his square formation of soldiers. Arrows rained down on the helpless men, while cataphracts directly engaged the Roman heavy infantry. The slaughter continued until nightfall, when the surviving Romans were able to flee.
The following day, Crassus was summoned to meet with the commander of the Parthian forces, Surena, to negotiate a truce. Crassus was killed in a scuffle that broke out during the assembly, and legend has it that molten gold was poured into his mouth by the Parthians to signify the Roman leader’s ‘thirst for wealth.’
When the dust of war settled, 20,000 Roman men were killed, and 10,000 captured. While estimates on Parthian losses vary greatly, it is widely accepted that the number of Parthian dead was relatively low. A meager one-fourth of Crassus’ army is believed to have escaped to Syria.
The fate of the 10,000 captured Roman troops is a key element of Dubs’ lost legion theory. He notes in his book that little is known about what happened to these prisoners, before adding that Pliny (presumably a reference to Pliny the Elder, a Roman author, philosopher, naval and army commander) claimed the captives were moved to Margiana “to guard the eastern frontier of Parthia.”
Margiana is a historical region that is now part of Afghanistan, Turkmenistan and Uzbekistan; the area was home to the Silk Road oasis of Merv.
There are no records indicating how many of the Parthian’s Roman prisoners made it to Margiana. Dubs notes that the distance between the battlefield at Carrhae and Margiana is vast and that captive invaders “would hardly have been treated kind on such a march.”
Dubs suggests that some of the Roman survivors of Carrhae did manage the journey to the eastern frontier of Parthia, and that the next time they appear in the historical record is on a battlefield facing down soldiers of the Western Han Dynasty.
The Battle of Zhizhi took place in 36 BCE in what is present day Kazakhstan. The belligerents were the Western Han Dynasty and the chieftain Zhizhi Chanyu of the nomadic Xiongnu, with Chinese troops laying siege to a fortified settlement occupied by Zhizhi and his troops. Dubs notes that 100-145 Roman troops who fled Parthia were possibly present at the battle, fighting as mercenaries on behalf of Chief Zhizhi. To offer weight to his claim, Dubs cites an account of the battle by Chinese historian Ban Gu: “Outside the wall horsemen gallop about, and more than a hundred foot-soldiers are lined up on either side of the gate in a fish-scale formation.”
‘Fish-scale formation’ is a historical clue for Dubs, who notes that the term is wholly unique in Chinese literature. He likens the concept of a fish-scale battle formation to that of the famous Roman testudo formation, a defensive tactic in which a legion’s members press their shields together in a manner that, visually speaking, is not unlike that of fish scales.
A Roman City in China also states that Zhizhi’s troops made use of a wooden palisade during the battle, something Roman forces frequently utilized “to strengthen their ditches, especially before gates.”
This evidence, coupled with the fact Roman troops were (in theory) within walking distance of the site of the conflict, leads the Roman-curious sinologist to conclude that survivors of the Battle of Carrhae were on the field of battle when troops from the Western Han Dynasty laid siege to Zhizhi’s fort.
At the conclusion of the confrontation, Dubs surmises that the Roman mercenaries surrendered to the Western Han Dynasty. The ancient foreigners were then settled on the fringes of the empire in a settlement their conquerors gave their name for Rome: Liqian.
Back in Gansu, I arrive at the village of Liqian. The site is now split into two distinct – and vastly different – sections: the town itself and the recently built Jinshan Temple. Both sites lie within a couple kilometers of each other, roughly a 25-minute drive from Yongchang county’s urban center.
Jinshan Temple, which according to the locals was constructed some time in the past decade, is where I’m dropped off. As I meander passed the toll collector stationed at the gate to the gravel parking lot outside the temple, the man in the booth eyes me curiously, his vivid green eyes and light brown hair immediately set him apart from other parking lot attendants I’ve seen in China.
Passing the decidedly European-looking toll man, I wander into the deserted entrance of the Buddhist temple complex, an area occupied by a bizarre, Coliseum-inspired visitor’s center that, among other Roman-esque elements, features 11 large, life-like statues of Roman legionaries.
There’s a gift shop, of course, and inside Rome-themed trinkets and faux suits of armor are available for purchase.
Beyond the entrance and through the gate of a large stonewall, Jinshang Temple is anything but traditional by Chinese standards. The grand, column- and dome-clad structure draws serious inspiration from neoclassical architecture, and a sign outside the building states it was designed in the style of a Roman temple – “skillfully combining Buddhist and Roman culture.”
It’s an interesting spot, but it’s hard to ignore the obvious: Jinshang Temple feels rather gimmicky, an empty Disneyland for Chinese tourists looking to snap a selfie at the alleged convergence point of two ancient powers.
“It’s rarely busy here,” a barbecue stand attendant tells me on my way out of the temple. “Except on holidays.” The middle-aged woman of Hui ethnicity says that she’s been living in the old town of Liqian for over 20 years. I head there next.
While all roads led to Rome, only one road leads into modern day Liqian, and visitors traveling through must walk past a massive gate – part of a wall erected last year that surrounds most of the village.
Inside the barrier, Liqian is a ghost town, aside from the occasional camel grazing on the yellow-tinged grass that protrudes in patches from the rocky earth.
Earthen walls surround old homes that aren’t aging very gracefully. A short walk from the new town gate are the remains of the old city wall; nearby stands a sign declaring the ruins a cultural relic of the county. The sign also states that in the late 1970s, villagers digging near the old wall recovered shards of pottery and Wuzhu coins from the Western Han Dynasty.
“No Roman artifacts have been found here,” Yan tells me, echoing what I’d heard from others on my way to the town. The 43-year-old man lives just around the corner from the ancient Liqian wall with his retired parents and works as a cattle rancher.
“The people here don’t know if they are descended from Romans,” says Yan. “It’s uncertain.” It’s a vague but fair answer, being situated on the Silk Road, many have passed through this area over the past 2,000 years and the prevalence of European features among locals is by no means proof of Dubs’ lost legion theory.
I press Yan about the skeleton in the Yongchang Museum and he says he’s aware human remains of a “large size” have been found in the area, before adding that he has no idea who they belong to.
Yan is skeptical of the area’s alleged Roman heritage, believing that the theory is played up to attract tourists to the sleepy town. Despite his suspicion, he admits “there are people living in the region that look different than your average Chinese person, people with yellow hair and blue eyes.”
Before we part ways, Yan encourages me to head to the edge of town. There, he says, is a monument to the town’s alleged Roman founders. After a 10-minute stroll, I come to an empty, windswept patch of earth at the edge of a small cliff. At what feels like the end of the world stands a white, Romanesque pavilion, complete with Doric-style columns.
A sign near the monument states that the pavilion was originally built in 1994 – 2,046 years after the Battle of Carrhae.
The idea that a rag-tag band of survivors from one of the worst military defeats in the history of Rome managed to cross thousands of kilometers of potentially hostile land and find a new home in China is undoubtedly intriguing.
For one, if Roman and Chinese soldiers did cross swords at the Battle of Zhizhi, it would be the only known time in history that citizens of the two powers engaged militarily.
Also, if the theory is true, it could mean that citizens of ancient Rome were responsible for founding a settlement that still exists in China to this day.
Unfortunately for the believers, not everyone buys into Dubs’ theory.
“In the academic world, that theory is conclusively false,” Zhang Defang, a retired researcher with the Gansu Archaeology Institute, tells That’s Senior Staff Writer Tristin Zhang. “The story is made up by local authorities.”
Zhang is not alone in his assessment. In a China Daily report published in 2010, Professor Yang Gongle of Beijing Normal University addressed Liqian residents’ European features, pointing out that interracial marriages were not uncommon on the old Silk Road, and that light-colored hair, blue eyes and hooked noses are not sufficient proof that the villagers are descended from survivors of The Battle of Carrhae.
“The ‘foreign’ origin of the Yongchang villagers, as proven by the DNA tests, does not necessarily mean they are of ancient Roman origin,” said Yang.
The genetic tests Yang is referring to occurred in 2005, when blood samples were taken from more than 90 Liqian residents. The results of the test revealed that some of the villagers’ DNA was up to 56 percent of Caucasian ancestry.
In 2007, a follow up DNA test yielded results that were far less exciting to believers in the legend of Roman settlers in Liqian: 77 percent of the locals’ Y chromosomes were limited to East Asia, according to the Journal of Human Genetics.
“The Liqians are closely related to Chinese populations, especially Han Chinese populations, whereas they greatly deviate from Central Asian and Western Eurasian populations,” states a Journal of Human Genetics article published on the subject back in June of 2007.
According to Song Guorong, though, the results of the 2007 genetic test do not invalidate Dubs’ theory. “The Roman Empire covered vast territories, and many soldiers in the legion were mercenaries,” he told China Daily.
The lack of Roman artifacts found in the region is another major factor that detractors of the theory often point to, although it’s prudent to point out that the likelihood of Roman captives being allowed to carry any of their personal belongings with them after surrendering at Carrhae seems extremely low.
The only real, hard evidence that Romans settled in the area is the skeleton housed in the Yongchang Museum. The remains were discovered back in 2003 in a 2,000-year-old grave, according to the Economist, and are of a 1.8-meter-tall male, much larger than the “average Chinese” back then. The height of the ancient man alone seems to be what’s led to the conclusion he was a Roman immigrant, with China Daily writing, the “human skeleton is believed to belong to a Roman soldier because it is 1.8 meters tall.”
While disagreement is widespread on Dub’s theory, scientific consensus seems to have labeled A Roman City in China a myth, with the physical traits of Liqian’s inhabitants easily attributed to Silk Road commerce and travel, and the lack of hard evidence being a major sticking point.
One thing everyone does seem to agree on though is that the theory has introduced tourism to the remote region.
“The name change,” Zeng, an ardent believer in Dubs’ theory, tells me, “that was to attract tourists.”