Prehistoric Tibet

Several rumblings around genetics Twitter have come along this week concerning Hongru Wang's protohistory of Tibet. Tonight we'll consider Bernard Sécher. Not all of it at once; it's a massive blog which even the Turtle cannot digest as of yet.

First up: all the ancient Tibetans come from a genetic mixture between an ancient population of northeast Asia (90%) and an old population not yet identified (10%). Is that missing tenth the indigenous Stone-Age human population? Are these the humans who mixed with the Denisovans, with the newcomers subsequently mostly-swamping sometimes-mixing with these? That "Sino-Tibetan" language-family: likewise from the northeast? - I consider that last as probable.

Starting "5260" years ago so, I guess, 3240 BC the source population of Tibet was in place, complete with that one Denisovan gene for the high life. This was still prior to wheat and barley agriculture (maybe they had millet?). Then the Tibetans started mutually differentiating. Sécher transmits a fine map to explain Where and Who. Nagpu is the centre. Lhasa-plus-Shannan is south by southwest (as it were). Chamdo is southeast. Rounding it out: Ngari far west, whence the Yarlung Tsangpo flows down to Lhasa; and Yushu far northeast. So five clusters.

Implicitly Lajia around 2200 BC introduced the cereals but without introducing notable genetics. Yay trade!

Up to 500 BC, in Nagpu, the locals there were like those of Chamdo. But by the early AD 400s Lhasa had extended its (different) population into this centre of the plateau. That move encroaches upon the Gansu trade. To Sécher, this was intentional and implies the Tibetan Empire.

So Yushu would be, what exactly: Tanggut?