There's an article in The New York Times on the recent paper which extracted genetic material from remains in Xinjiang dated to 4,000 years ago. Remember that these remains exhibited male lineages which were west Eurasian, specifically R1a1, while the female lineages (mtDNA) were more heterogeneous, both eastern and western. This particular twist in history is of very strong interest. I think there are three reasons for this. First, it is counter-intuitive, as we don't have a good grasp of how mobile ancient nomadic populations were. Most of their history is unaccessible because they were often not literate. We know of them by and large by the shadows that they cast upon the literate civilizations. Our map of the past is strongly skewed toward civilized sedentary groups, where lack of mobility was the norm. Second, there has been a strong attempt by the Chinese government to control and massage these findings for a generation, in large part because of concerns about separatist movements in western China. The Uyghurs of Xinjiang are almost certainly in part descended from these ancient Europoid populations. Finally, there is sublimated race pride on the part of white people.* Since white people created the modern world as we know it, it may seem rather trivial to exhibit interest in lost white civilizations. But the Chinese are apparently proud of a partly fictitious 5,000 year history based on descent from the Yellow Emperor,** while black nationalists have long ludicrously grappled onto continuity with Carthaginians and Egyptians. The fame of peoples long gone seems to have a stronger proportional weight than more recent contributions.***
One of the major points in regards to the Indo-European peoples of ancient Xinjiang are their strange affinities to populations on the western rim of Eurasia, the textile patterns which seem almost Celtic for example. But you have to be careful how you frame data. Consider this from The New York Times article:
The language spoken by the people of the Small River Cemetery is unknown, but Dr. Mair believes it could have been Tokharian, an ancient member of the Indo-European family of languages. Manuscripts written in Tokharian have been discovered in the Tarim Basin, where the language was spoken from about A.D. 500 to 900. Despite its presence in the east, Tokharian seems more closely related to the "centum" languages of Europe than to the "satem" languages of India and Iran. The division is based on the words for hundred in Latin (centum) and in Sanskrit (satam).
There is actually some dispute as to significance of the centum-satum division as phylogenetically informative, but here's a map which shows the
modern 3,000 year old distributions of the two groups (pretty close to modern distributions as some groups, like Celtic, were replaced by languages in the same category, Romance):
As you can see, the Baltic and Slavic languages are satem. In other words, satem is not purely extra-European, and the Tocharian languages are an outlier of centum on the "other side" of the satem group. Assuming this division is informative there is a easy way to explain this: the sateum-centum split was already in existence when the Indo-European tribes covered a much smaller geographical area. It would not be particularly surprising if one centum tribe went east while the others went west.
The Tocharians were who they were, and too often contemporary ideologues try and foist modern categories onto the past. As I've noted before, they show up in the early Chinese records, and if you trust those records their physical appearance should be no surprise. Additionally, the presence of some South Asian lineages among the Tocharians should be no surprise either; the famed Chinese Buddhist translator Kumarajiva> was from the Tocharian city of Kucha. His mother was a native while his father was an Indian immigrant. This highlights the significant trade connections between Xinjiang and northwest India.
Ancient DNA extraction seems to be producing a lot of surprises. This will probably make things uncomfortable for ideologues of all stripes, because reality has a funny way of often not suiting our preferences.
* Sublimated because public espousals of white pride are obviously taboo in the Western world, in contrast to more particular ethnic pride (e.g., Jewish or Italian pride). In contrast, in the United States racialized pride by non-whites is not nearly as constrained, while in East Asia it has never been particularly closeted.
** Chinese identity has deeper roots than the Shang dynasty, but history can go no further back than this period.
*** Just as ancient esoteric wisdom is still something that modern people seek after, even though we live in an ancient of real magic, science.
Xinjiang has never been culturally or racially Chinese, at least at not until this century. During eras of Chinese expansion(Han, Tang, Ming, And Qing) Xinjiang was politically and militarily controlled by China some or most of the time.
Can't remember where specifically I found this, but I think it's consensus that the Tocharians and Hittites were the first to branch off from the IE trunk and thus have diverged furthest from the others and each other, and that that satum languages developed and split off later. So Tocharian, Hittite, and the Western languages were all centum languages because when Tch. and Htt. split off all IE languages were centum, but Tch. and Htt. are not otherwise close to the western IE languages.
Craig Benjamin's The Yuezhi traces the history of this tribe (neighbors of China defeated by the Xiung-nu ca. 200 BC) and from his book you can construct this history: the Tocharians migrated east before the nomadic lifestyle had developed and brought many new things to Chinese culture and technology. They were a major power for many centuries but by the 200 BC, under pressure from nomads, they had been weakened and were defeated. They fled through Central Asia and then to Afghanistan and Pakistan where they founded the Kushan Empire ca. 100 BC. A few hundred years later they Kushan Empire was destroyed, and the only Tocharians left were outliers in Xinjiang, who survived until ~900 BC. These outliers might have been left behind when the Kushans-to-be (Yuezhi) went south and West, or they could have been trade colonies.
Pretty tentative, but the Tocharians have been a mystery of sorts for over a century, and 20 years ago they still were, but a coherent story is emerging.
During eras of Chinese expansion(Han, Tang, Ming, And Qing)
not ming. also, the qing absorbed xinjiang as the manchu first. that is, the union of xinjiang with china proper was analogous to that being hanover and the united kingdom in the 18th century. in the 19th century the UK and hanover split as hanover followed salic law i believe, so victoria could not become monarch. in the 19th century the qing decided to integrate xinjiang fully with china proper.
your overall point is valid of course.
So Tocharian, Hittite, and the Western languages were all centum languages because when Tch. and Htt. split off all IE languages were centum, but Tch. and Htt. are not otherwise close to the western IE languages.
i think the assumptions about tocharian are going to be tentative because fewer records. but the idea that hittite separated early seems modestly supported from what i can tell. ergo, the language family "indo-hittie," of which indo-european and hittite are the two branches.
finally, though valid to observe that the tocharian domains were never culturally or racially chinese, i think it is probably true that they were within the cultural orbit of the chinese sphere of influence from around 0 AD to about 1000. after 1000 there was a shift toward the world of islam, though the islamicization of the uyghurs seems to have occurred in the period after genghis khan.
The IE speaking Assamese are furthest east. Happy St. Patty's Day, when everybody's green.
Razib, have you written anything about the muslim Rohingyas in Bangladesh/Burma? Are they historically Cham/Malay speakers, or a politically marginized peripheral Sino-Muslim group like the Uighurs?
The direction of flow was the other way though, Buddhism from India (the Kushans again) to C. Asia and Xinjiang to China. The Tang dynasty had wide involvements, but it was also less Chinese and more xenophile, with the heaviest Buddhist influence and also with strong Turkish, Muslim, Sogdian, etc. influence.
The Shape of Ancient Thought: Comparative Studies in Greek and Indian Philosophies
By Thomas McEvilley
rohingyas speak a dialect similar to that of chittagong (southwest bangladesh), and generally look bengali. perhaps a bit more burmese, but many bengalis already look a bit that way. like many south asian muslims groups they've made up somewhat of a fake history; though in their case it is arab instead of turco-persian as is the north in north india.
The direction of flow was the other way though
this is fair. though i would say that the chinese influence on xinjiang and central asia has been obscured by turkic and muslim overlay.
also, minor note, there's a fair amount of circumstantial evidence that buddhism came into china via sea at the same time as through the silk road. specifically, one of the earliest temples is in shandong on the coast.
The division is based on the words for hundred in Latin (centum) and in Sanskrit (satam).
Not so much on Sanskrit ÅatÃ¡m as on Avestan (Old East Iranian) satÉm. But of course it's all the same thing.
but here's a map which shows the modern distributions of the two groups:
Ouch. That's a reconstruction of, like, 1000 BC! It's by no means modern. Look at it. :-)
I think it's consensus that the Tocharians and Hittites were the first to branch off from the IE trunk and thus have diverged furthest from the others and each other, and that that satum languages developed and split off later.
Correct. Don Ringe's guest posts on Language Log a year ago treated this in some detail.
So Tocharian, Hittite, and the Western languages were all centum languages
No, "centum" doesn't just mean "not satÉm", it refers to another sound shift. Hittite (and the rest of the Anatolian languages, an entirely extinct branch) and Tocharian (itself at least 2 distinct languages) were neither centum nor satÉm languages.
(And neither is Albanian, it now appears; Armenian looks satÉm, but probably got there independently.)
R1A seems to correspond to Satem languages and R1B to Centum languages
Any possibility of a link to the legend of Prester John?
prester john was the kara-khitai
"Prester John, the Mongols and the Ten Lost Tribes", by CF Beckingham, collects a lot of the legends of Prester John.
His supposed location wandered from India to Central Asia to Ethiopia, and he was looked for for centuries. His legend was part of a vast, confused wad of lore also involving Alexander's Gate, Gog and Magog, the Three Kings (who brought gifts to baby Jesus), the lost tribes of Israel, and other things I've forgotten. A lot of fun if you're interested in textual and literary studies.
During the Crusades an Armenian Prince mediating between the Mongols and the Crusaders helped spread the legend of Prester John as part of his diplomatic attempt to unite the Mongols and the Catholics against Islam. Some but not much success.
Most legends of Prester John were more or less loosely based on actual Christian peoples in India, Ethiopia, and Central Asia. there were many Christians among the Mongols including Genghis Khan's daughter.
However, the Mongol Empire had a religious foundation which trumped Christianity. Christianity among the Mongols was on the one a private, individual enterprise, and on the other, a tool of Mongol diplomacy. But not really an authority the way it was in Europe.
Robert Silverberg's THE REALM OF PRESTER JOHN is a great source for the history of the legend of Prester John.
there were many Christians among the Mongols including Genghis Khan's daughter.
follow up to my comment above, kulchug, the last khara-khitai leader, was from a christian background, as were many of his naiman tribal compatriots. though kulchug seems to have expressed pro-buddhist policies against the muslim subjects of the khara-khitai, as his the old ruling class and his wife were buddhist.