Geographic locations of the studied sites in Gansu and nearby regions (color-filled circles) and published ancient sites in East Asia co-analyzed in this study (white-filled circles).
Archaeologists have long warned that “pots are not people”. The distribution of a particular style of pottery does not necessarily trace the movement of the people who first made it, because pots can be traded, copied or carried through a succession of neighbouring communities. Much the same principle applies to crops, livestock and agricultural techniques: wheat can travel without wheat farmers, sheep without shepherds, and an effective method can spread simply because other people adopt it.
Distinguishing between the movement of people and the movement of their culture is therefore essential when reconstructing prehistory. Agriculture sometimes spread by demic diffusion, in which expanding farming populations moved into new territory and brought their domesticates with them. In other cases, it spread principally by cultural diffusion, as existing populations acquired crops, animals or knowledge through trade, imitation and intermediate exchange networks. Archaeological remains alone do not always reveal which process occurred, but ancient DNA can now test the alternatives directly.
Gansu, in north-western China, provides an especially important case. Situated between the Yellow River basin, the Tibetan Plateau, the Mongolian Steppe and the routes leading towards Central Asia, the region became a major corridor of prehistoric trans-Eurasian exchange. Crops and livestock domesticated much farther west eventually appeared there, raising the obvious question of whether they arrived with migrating western farmers or passed between the established populations of Inner and East Asia.
A multidisciplinary study published in Cell has now examined genomic data from 149 people who lived at eleven Gansu sites between approximately 4,700 and 3,000 years ago. Western-derived crops or livestock were present at nine of those sites, yet the researchers found no corresponding influx of western Eurasian or Central Asian ancestry. The principal genetic components instead derived from populations associated with the Yellow River basin, the south-eastern Mongolian Plateau and the Tibetan Plateau. A few genetic outliers testify to regional mobility and contact, but not to wholesale population replacement.
The evidence therefore indicates that the western domesticates reached Gansu largely through cultural transmission—possibly along the Inner Asian Mountain Corridor or through chains of intermediary communities across the Mongolian Steppe. Crops, animals and ideas crossed Eurasia farther and more readily than the people among whom they had originated. The finding also shows why neither pottery, farming nor any other cultural innovation can safely be treated as a proxy for mass migration.
This carefully reconstructed picture bears little resemblance to the simplistic human history imagined by biblical literalists. It contains no recently dispersed descendants of a single family spreading from Babel with fixed cultures and divinely assigned identities. Instead, it reveals long-established, genetically differentiated populations connected by shifting networks of exchange, occasional mobility and selective cultural borrowing. People did not need to migrate en masse for useful innovations to spread across a continent.
