Points of Contact: A Chinese Exile in Ancient Armenia

Was a medieval Armenian dynasty really founded by an exiled Chinese prince?

Image: Tiridates III, from Illustrated Armenia and the Armenians by Ohan Gaidzakian (1898), via Wikipedia.

I'm launching a new occasional series on this blog, Points of Contact. The name is inspired by A Point of Contact (1922), a (chronologically absurd) short story by Arthur Conan Doyle, in which Odysseus on his travels runs into the young David son of Jesse, future King of Israel: the series will examine points at which the legends of different cultures appear to intersect.

In C.E. 287, the exiled Armenian prince Trdat - more usually called by his Greek name, Tiridates - returned to his homeland with Roman support, to declare himself king and drive out the occupying Persians. Edward Gibbon, in The History of the Decline and Fall of the Roman Empire (1782), tells us that:

"Among the Armenian nobles appeared an ally, whose fortunes are too remarkable to pass unnoticed. His name was Mamgo, his origin was Scythian, and the horde which acknowledge his authority had encamped a very few years before on the skirts of the Chinese empire, which at that time extended as far as the neighborhood of Sogdiana. Having incurred the displeasure of his master, Mamgo, with his followers, retired to the banks of the Oxus, and implored the protection of Sapor. The emperor of China claimed the fugitive, and alleged the rights of sovereignty. The Persian monarch pleaded the laws of hospitality, and with some difficulty avoided a war, by the promise that he would banish Mamgo to the uttermost parts of the West, a punishment, as he described it, not less dreadful than death itself. Armenia was chosen for the place of exile, and a large district was assigned to the Scythian horde, on which they might feed their flocks and herds, and remove their encampment from one place to another, according to the different seasons of the year. They were employed to repel the invasion of Tiridates; but their leader, after weighing the obligations and injuries which he had received from the Persian monarch, resolved to abandon his party. The Armenian prince, who was well acquainted with this merit as well as power of Mamgo, treated him with distinguished respect; and, by admitting him into his confidence, acquired a brave and faithful servant, who contributed very effectually to his restoration."

In 1845, Gibbon's editor H. H. Milman annotated this passage with the information that

"Mamgo (according to M. St. Martin...) belonged to the imperial race of Hon, who had filled the throne of China for four hundred years. Dethroned by the usurping race of Wei, Mamgo found a hospitable reception in Persia in the reign of Ardeschir. The emperor of China having demanded the surrender of the fugitive and his partisans, Sapor, then king, threatened with war both by Rome and China, counselled Mamgo to retire into Armenia. 'I have expelled him from my dominions, (he answered the Chinese ambassador;) I have banished him to the extremity of the earth, where the sun sets; I have dismissed him to certain death.' ... Vou-ti, the first emperor of the seventh dynasty, who then reigned in China, had political transactions with Fergana, a province of Sogdiana, and is said to have received a Roman embassy..."

When I first read this passage as an undergraduate, my imagination was instantly seized by the idea of this exiled prince who had led his followers across the entire breadth of Asia. That he lived in the third century made it still more exciting: for that was the era of the Three Kingdoms, whose battles and intrigues reverberate through Chinese legend, and whose heroes became gods. The Persian names too are ones much garlanded with later tales of romance and adventure.

However, even before we begin to investigate, we can see that Milman's notes contradict Gibbon. One Mamgo is "Scythian", the other a prince of the Han Dynasty; one is received in Persia by Ardashir I, the other by Shapur I. A glance at the chronology of the relevant Persian and Chinese monarchs makes the whole story seem even shakier. The Han Dynasty fell in 220, 67 years before Mamgo is supposed to have joined Tiridates, and four years before Ardashir overthrew the Parthian Empire and founded Sasanian Persia.

Milman appears to imply that Mamgo himself had reigned as a Han Emperor, or was due to succeed to the throne. He might, of course, be referring not to the original Han Empire but to the state of Shu, whose rulers claimed relationship to the fallen Emperors, and which was itself conquered by Wei in 263 - but that is too late for any connection with Ardashir, who died between 240 and 242. Furthermore, "Vou-ti" was not a ruler of Wei: it is a French rendition of Wu Di, one of the names of the warlord who was born Sima Yan and became the Emperor known as Wu of Jin, reuniting China at the end of the Three Kingdoms period. Sima Yan was an enemy of Wei (though his forebears had served the kingdom), and conquered it in 266.

To make any sense of this, we have to ignore the mention of Ardashir and interpret Milman's "Hon" as referring to Shu. Then we have a hypothetical Shu prince, fleeing in 263, still pursued by Sima Yan after 266, sent to Armenia by Shapur I, and falling out with the Sasanians some time after Shapur's death in 270. Such a person could easily still have been leading armies in 287. (Whether any such prince could actually have existed may be another matter. The last King of Shu, Liu Shan, was made prisoner by Wei in 263, as were all his surviving children.) This still doesn't have anything much in common with Gibbon's "Scythian". The original Scythians were an Indo-European people who occupied roughly what is now Ukraine and southwestern Russia in the early to middle Iron Age, and spoke a language closely related to Persian: but the name was conventionally applied to steppe peoples of various ethnicities, over a much wider chronological and geographical range. Sogdiana, a region covering parts of what are now Tajikistan, Uzbekistan, and Kyrgyzstan, had indeed been on the fringes of the Han Empire at one time - but not by the Three Kingdoms era. It is quite impossible that anyone alive in 287, or even the 260s, could have remembered a Chinese presence there. In fact, after a long stretch of being largely ignored by outside powers, it actually came under Persian rule in the reign of Shapur. He therefore acquired Sogdian servants without any need for them to flee to the west, or travel at all. Does this explain Mamgo? We'll consider further below.

Where did Gibbon and Milman get this story? Milman attributes it to Antoine-Jean Saint-Martin, a celebrated enthusiast for Armenian history and culture; Gibbon, writing before Saint-Martin was born, cites no source. However, the ultimate sources are two fifth century chronicles, both entitled History of Armenia - one by (to use their Western names) Moses of Chorene, the other by Faustus of Byzantium. Moses' chronicle, at least, had been published in Latin in Western Europe by Gibbon's time and was available to him; certainly Saint-Martin used both. So what do they have to say about the mysterious Mamgo?

The name, in that form, comes from the Latin edition of Moses' history published in London by William and George Whiston in 1736. (Moses in the original called him "Mamik", supposedly giving rise to the Mamikonian or Mamikonean family who claimed descent from him; Faustus never names the family's reputed East Asian ancestor.) Moses reports that Mamik came from "Chem" or "Chenk", rendered by the Whistons as "the region of the Zenensi", situated "between the rising sun and the north-east"; and that the region's king, Chenbakir (Zenbacurius in Latin, also referred to as Arbocus, which appears to be a title), had exiled him towards the end of Ardashir's reign in Persia.

Here matters become confused: there is no English translation of Moses' history, and I read no Armenian. A modern English summary of Moses' account, repeated on Wikipedia, has it that Mamik and his brother Konak were Chenbakir's half-brothers, had revolted against him, and fled together in defeat; but the Latin text says that his brother's name was Beldochus, that the king was not their half-brother but their uncle, and that Mamik fled after being accused by his brother of plotting Chenbakir's death. The English summary is several steps removed from the original text and contains errors (for instance, Persia is confused with Parthia), so I prefer to trust the Latin. Chenbakir then reportedly threatened war; Ardashir having died before he could reply, Shapur sent Mamik to Armenia, and placated the eastern ruler with the news, as quoted by Milman, that he had dispatched him to the ends of the earth and near certain death.

This account dates Mamik's arrival in Armenia quite precisely to the early 240s, provides several names which do not sound at all Chinese, and indicates that Chenbakir was a ruler whose enmity Shapur feared - suggesting one both closer to Persia than any Chinese kingdom at the time, and more powerful than any known polity in between. Although various Central Asian origins for "Chenk" have been suggested, the lyrical passage which follows, detailing that land's wealth, fertility, culture of study, and love of peace, makes it quite clear that Moses was thinking of China, even if that was not intended by his source. Faustus uses the same term (or "Chenac"), referring to it as the land where the Mamikonians "were kings" before "a quarrel among brothers" compelled their departure, which in his account they later claim puts them above the Arsacid rulers of Armenia. (He gives no chronological information, nor does he mention any Persian dimension - his Mamikonians seem to have travelled directly from "Chenk" to Armenia, suggesting that Faustus did not think of it as somewhere quite so distant as China.)

The only Chinese upheaval of this period was the rebellion in Wei of Gongsun Yuan, crushed in 238 by the royal general Sima Yi, grandfather of the Sima Yan who would later become Emperor. Gongsun Yuan's power base was east of Sima Yi's, making any rebel's flight to the west unlikely; and there is no indication that any close relative or high ranking supporter of his survived the rebellion. Furthermore, Moses has dropped this story into the midst of that of Tiridates III, and returns in the following chapter to events of the 280s and 290s.

There might be some confusion here with Tiridates II (217-52), who was King of Armenia when Ardashir died. Confusion of some sort there certainly is, for a couple of chapters later Tiridates is fighting Shapur II in what appears to be the Romano-Persian war of the mid 330s, and the same Mamik of Chenk is reported as being sent "with his people" against pro-Persian rebels in the highlands along the border with Caucasian Albania. This is a clear mistake. Tiridates was dead by the time of this campaign; Shapur II had not warred with Rome or Armenia before; and the actions attributed here to Mamik appear to belong to Vache Mamikonian, the first recorded head of the by then established family. This episode, at least, can be safely discounted: it tells us nothing about Mamik.

Is there a Chinese name that could have given rise to "Mamik"? Ma was a well known family name in the third century. As well as the famous mechanical engineer Ma Jun in Wei, there was a prominent Ma family in Shu Han. Ma Liang of the White Brows and his younger brother Ma Su were military officers there in the early part of the century, beginning their service under King Liu Shan's famous father Liu Bei, who after his death would become a patron god of shoemakers and is still worshipped in parts of China today. The cunning Ma Su was reportedly responsible for spreading rumours in Wei designed to get Sima Yi removed from office, so that Shu Han should have an easier fight. However, though both brothers were reportedly brilliant tacticians, both were eventually defeated, Ma Liang falling in battle in 222 and Ma Su dying in captivity in 228. Ma Su was captured in southwestern Wei: although his wife and children were reportedly also taken prisoner, and treated well, it is just about possible to imagine a member of his family fleeing in the general direction of Ardashir's Persia - though they would have had a lot of other territories to cross first.

This is, however, extremely tenuous. It seems much more likely that, if there is any truth in the story of Mamik, he did not come from anywhere quite so distant as China. Perhaps he fled from the near-collapse of the Kushan Empire (centred in northern India, bordering on Sogdiana at its northernmost extent) around 240, or was a Sogdian who entered Persian service after the conquest of his native land. Certainly the names cited by Moses sound more Turkic than anything else, suggesting a Central Asian origin. It might be objected that no Sasanian Great King would have cause to fear a Kushan emperor whose realm was imploding, or a Turkic warlord whose forces would be dwarfed by Persia's: but he might still prefer them as friends rather than foes. Besides, it is hard to see that distant and war-torn China could be any greater threat.

It's a pity to lose the connection with the Three Kingdoms: but the most that can be salvaged is the faint possibility that Mamik and his people, if they were the nomadic mercenaries imagined by Gibbon, had fought there at some point before arriving in Shapur's empire. This might perhaps account for Moses' apparent identification of China as their homeland - but it is equally possible that he, or the Mamikonian family, simply conjured this connection out of the tradition that their ancestor had come from the East. Sadly, we'll never know for sure.