Scientists Compared the DNA of Ancient Chinese to Modern Chinese People — The Results Are Shocking

Added: 2026-05-24

[00:00:00]In 2023, scientists extracted DNA from 55 ancient remains in China's Yellow River basin some 9,500 years old. They compared the genomes to modern Chinese DNA. The result shocked everyone. The ancient people buried there were not the ancestors of modern Chinese. The founders of Chinese civilization, the first Yellow River farmers, their DNA had been overwritten, >> [music] >> replaced, gone. The civilization didn't move. The people did. This was so politically explosive the scientists published with extreme caution. China's entire national identity rests on one claim, one people unbroken since the beginning. The genome said otherwise.

[00:00:37]Harvard's Reich Lab confirmed it. 166 ancient individuals, 10,000 years of data, at least three completely separate populations contributed to the modern Han Chinese gene pool. This wasn't evolution in place, it was replacement.

[00:00:51]But before we get into who replaced whom you need to understand how scientists read 9,500-year-old DNA. Because without that none of what follows makes sense. Now before we get into the full picture, you need to understand how scientists extract DNA from 9,500-year-old bones and why the results from that DNA are far more reliable than you might expect. Ancient DNA, which researchers abbreviate as aDNA, degrades over time.

[00:01:16]Every decade, every century, the long strands of genetic code inside a bone fragment break apart a little more, get contaminated by soil bacteria, get overwritten by environmental DNA. For most of the 20th century, scientists assumed that anything older than a few thousand years was simply unreadable.

[00:01:33]Then the sequencing revolution changed everything. Starting in the mid-2010s, advances in high-throughput DNA sequencing, which is the process of reading millions of tiny genetic fragments simultaneously, the way you'd reassemble a shredded document by reading every individual scrap and using overlap patterns to reconstruct the whole, made it possible to pull meaningful genetic data from bones once considered useless. The technique involves extracting petrous bone, the densest part of the human skull, from ancient remains. That particular bone preserves DNA at concentrations up to 100 times higher than any other part of the skeleton. Think of it like this.

[00:02:09]Imagine the rest of the skeleton as an old cassette tape left in the sun, warped, degraded, partially unplayable.

[00:02:16]The petrous bone is the same tape stored in a climate-controlled vault. Same recording, far better preservation. From that bone, scientists can extract mitochondrial DNA, passed only through the maternal line, and increasingly, nuclear DNA, which carries the full diploid genome and can reveal ancestral origin with extraordinary precision. By comparing the ancient genomes against a reference panel of modern and ancient populations from across Asia, Europe, and Africa, researchers can calculate mathematically, not speculatively, what percentage of a modern person's ancestry came from which ancient source population, and when that mixing occurred. That technique is called admixture analysis. Admixture, meaning the genetic blending of previously separate populations.

[00:03:00]>> [music] >> And when scientists applied it to ancient China, what they found was not a single river of descent running [music] clean and unbroken from prehistory to the present. What they found was a delta, a vast, braided, constantly shifting delta of intermingling populations. Each one leaving its mark on the people who came after. Here's where things start to get interesting, because the picture that emerges from the data is not one story. It is at least four distinct stories happening in sequence. Start at the beginning. 9,000 years ago, the Yellow River basin in what is now northern China was home to a population of early millet farmers.

[00:03:33]People scientists designate by the technical label Yellow River Neolithic.

[00:03:37]Their genetic profile was unlike anything that survives today in pure form. They carried a distinct ancestral component that researchers have labeled Ancient North Chinese, a genetic signature that connects them not to modern Han Chinese populations, but to a broader ancient East Asian cluster, whose closest surviving relatives are found, surprisingly, among certain populations in Siberia, among the Tibetan Plateau's earliest inhabitants, and among some indigenous communities in the Americas. Think about what that means. The people who planted the first seeds in the Yellow River Valley, who are credited in Chinese historical tradition as the ancestors of Chinese civilization itself, carried DNA that has more in common with indigenous Siberians and the distant ancestors of Native Americans than with the vast majority of modern Chinese people. The DNA doesn't lie. The Yellow River Neolithic farmers were genetically distinct from the Han Chinese, who would eventually claim their cultural legacy.

[00:04:30]Meanwhile, roughly 1,000 km to the south, a completely separate population was developing its own independent agricultural tradition along the Yangtze River. These people, designated Yangtze River Neolithic in the genetic literature, cultivated rice rather than millet, built different architectural traditions, developed different pottery styles, and carried a completely different genetic profile. Their ancestral component, labeled ancient South Chinese, connects them to populations that would eventually spread across Southeast Asia, Taiwan, and the Pacific Islands. These are two separate founding populations, two separate agricultural revolutions, two separate gene pools developing independently in the same general geographic region at roughly the same time. And here's where the science gets even more fascinating.

[00:05:14]The Han Chinese of today carry genetic contributions from both of these founding populations. They are, in a very real sense, the product of a merger of 5,000-year-long process of contact, conflict, assimilation, and mixing between the millet farmers of the north and the rice farmers of the south. But the story doesn't stop there, because then the steppe arrived. Approximately 4,000 to 5,000 years ago, something dramatic happened across the entire Eurasian continent. Populations living on the great grassland belt, stretching from the Pontic Steppe, what what now Ukraine and southern Russia, began moving in every direction, westward into Europe, where their descendants would become the ancestors of most modern Europeans, southward into Iran and India, where their genetic signature would fuse with existing farming populations to produce the ancestral South Asian genome, and eastward across Central Asia into Xinjiang, Mongolia, and eventually the northern margins of what we now call China. The steppe populations carried a distinctive genetic signature that scientists can identify with precision. A mixture of Eastern European hunter-gatherer DNA and ancient North Eurasian ancestry. And when researchers look at the genomes of Bronze Age individuals buried in northern China between 3,000 and 4,500 years ago, they find it. A layer of steppe ancestry that simply wasn't present in the Yellow River Neolithic samples from a few thousand years earlier. This genetic arrival roughly coincides, intriguingly, with the period that Chinese historical tradition associates with the founding mythological figures of Chinese civilization, the Yellow Emperor, the mythic ancestors. The transition from a world of scattered agricultural communities to something that begins to look like organized statehood. The genetic record cannot tell us whether specific individuals from the steppe became mythologized as founders, but it can tell us that new people arrived, mixed with existing populations, and left a permanent genetic mark on the population of northern China. Then, approximately 2,000 to 3,000 years ago, during and after the Qin and Han dynasties, something equally dramatic happens in the genetic record from a completely different direction. The Han Chinese expansion southward, as the centralized Chinese state pushed into regions previously inhabited by non-Han populations speaking Austroasiatic, Austronesian, and Tai-Kadai languages, resulted in one of the most consequential population replacements in human history. The indigenous populations of what is now southern China, Vietnam, Laos, and parts of mainland Southeast Asia were not eliminated. They were absorbed, genetically, linguistically, culturally assimilated into an expanding Han Chinese identity. And simultaneously, the expanding Han population itself picked up genetic contributions from those absorbed populations, meaning that the modern Han Chinese of southern provinces like Guangdong, Fujian, and Guangxi carry measurably more ancient South Chinese and Southeast Asian ancestry than their counterparts in Beijing or Shanxi. Here's what that means in plain language. Modern Chinese people from the same country, identifying with the same ethnic label, speaking dialects of the same language, carry measurably different genetic profiles depending on which part of the country their families are from. A person from Guangzhou and a person from Harbin are both Han Chinese, but their genomes tell different stories about which ancient populations their ancestors came from. The Han ethnic identity is one of the most powerful cultural constructs in human history. It unified hundreds of millions of people across an enormous geographic space under a single civilizational banner.

[00:08:37]But the genetic data reveals that this identity was built not on genetic uniformity, but on genetic synthesis, on the extraordinary capacity of a cultural framework to absorb, assimilate, and ultimately transform people of wildly different ancestral origins into a shared civilization. And now we reach the single finding that made the scientists run their models three times before publishing. When researchers from the Shanghai Institute for Nutrition and Health analyzed ancient genomes from the core region of Chinese civilization, the Central Plains, the Yellow River heartland, the region that Chinese historical tradition calls Zhongyuan, the Central Land, >> [music] >> and compared them against modern Han Chinese from the same region, they found something that rewrote the fundamental assumptions of Chinese genetic history.

[00:09:19]The ancient Yellow River farmers, the genetic founding population of what became Chinese civilization, contributed far less to the modern Han Chinese gene pool than anyone had predicted. Read that again. The people who built the earliest settlements, who fired the first oracle bones, who established the agricultural foundations that made Chinese civilization possible. Those people are not the primary genetic ancestors of modern Han Chinese. The primary genetic ancestry of modern Han Chinese comes instead from a southern expansion, from populations moving northward out of the Yangtze basin and surrounding regions. Over the last 3,000 years, the genetic dominance in modern Han Chinese is southern in origin, not Yellow River Neolithic, not ancient North Chinese, but a mixture in which the Yangtze agricultural tradition and its descendants carry the heaviest weight. Think about the cultural irony embedded in that finding. The Yellow River is the river. It is the mother river of Chinese civilization. As every schoolchild in China learns, the mythology of Chinese origin is centered on the Yellow River. The imperial capitals were built along the Yellow River. The foundational texts of Chinese philosophy emerged from the Yellow River basin. And yet the genetics say that the genetic weight of modern Han Chinese ancestry skews south toward the Yangtze, toward regions that were historically considered peripheral, frontier, less central to the Chinese civilizational core. The civilization moved north to south culturally. The genes moved south to north demographically. And the modern Han Chinese are the product of both movements running simultaneously in opposite directions. And this brings us to a bigger question, one that cuts to the heart of what we mean when we use the word Chinese at all. The Chinese state's official historical narrative, embedded in school curricula, national museums, and the foundational rhetoric of modern Chinese national identity, rests on a concept called continuous civilization, the idea that the Chinese people are one people, stretching back unbroken to the Yellow Emperor, to Yao and Shun, to the earliest millet farmers of the Yellow River basin. The idea of civilizational continuity is not merely a historical claim, it is a political one. It underpins arguments about territorial sovereignty, about the legitimacy of governance over ethnically diverse regions like Tibet and Xinjiang, about the relationship between modern Chinese citizens and the ancient peoples whose artifacts fill Chinese museums.

[00:11:29]The genetic record does not destroy that story, but it complicates it profoundly.

[00:11:33]What the aDNA data reveals is that Chineseness as a genetic category does not mean what the continuity narrative implies. The modern Han Chinese are not a genetically homogeneous population descended from a single founding group in the Yellow River Valley. They are instead a dynamic synthesis, the outcome of at least four major population events across 10,000 years involving farmers from two separate river systems, migrants from the Eurasian steppe, and the absorbed remnants of dozens of smaller non-Han ethnic groups who were folded into the Han identity during the imperial expansions. Lower genetic continuity between ancient and modern northern Chinese doesn't mean modern northern Chinese are less authentic. It just means that the biological reality of human populations is messier, richer, and more interesting than any national narrative can contain. And here is where the finding lands with the most force.

[00:12:20]The ethnic minority populations of modern China, the Tibetans, the Uyghurs, the Mongolians, the various southern minority groups, have often been framed in official discourse as populations that need to be integrated into or measured against the Han Chinese civilizational standard. But the genetic data reveals something quietly extraordinary. The Tibetan people carry genetic ancestry from the Yellow River Neolithic population, the original founding population of Chinese civilization, at higher concentrations than most modern Han Chinese do. Think about what that means. The people who are most closely related genetically to the ancient founders of Chinese civilization are not the Han Chinese. In key respects, they are the Tibetans. The genome has its own politics, and they are not the politics of the 20th century. The China story is not unique.

[00:13:03]It is not an exception. It is one of the clearest examples of a pattern that ancient DNA research has now documented on every continent. The populations living in a place today are almost never the direct unbroken descendants of the populations that lived there 5,000 or 10,000 years ago. In Europe, ancient DNA research has shown that modern Europeans are themselves a three-way mixture of Paleolithic hunter-gatherers, Anatolian farmers who arrived around 9,000 years ago, and Pontic Steppe herders, who swept across the continent 5,000 years ago, genetically replacing up to 90% of the previous population in some regions.

[00:13:36]The people buried in Stonehenge are not the ancestors of modern Britons. Modern Britons replaced them. In South Asia, ancient DNA has revealed that the populations now living in India carry genetic ancestry from three separate source populations: ancient South Indians, ancient Iranian farmers, and Steppe migrants in proportions that vary dramatically between northern and southern India and between upper and lower castes. In the Americas, ancient DNA has shown that the first people to arrive on the continent, roughly 15,000 to 20,000 years ago, were themselves partially replaced or absorbed by later migration waves, meaning that the people European colonizers encountered when they arrived in the 15th century were not the same genetic populations as the people who first crossed the Bering Land Bridge. Everywhere scientists look with the tools of ancient genomics, the story is the same. Populations are not rivers running purer and clean from one point to another. They are braided deltas, constantly shifting, mixing, absorbing, diverging, and reconverging. The concept of ethnic or national genetic purity is not just politically dangerous. It is genomically impossible. And China, the civilization that has invested more intellectual and political capital in the idea of genetic and cultural continuity than perhaps any other on Earth, turns out to provide one of the clearest scientific demonstrations of why that concept has always been a human story projected onto biological reality, rather than a biological reality confirmed by science. And this brings us to perhaps the most viscerally shocking dimension of the entire research program. Because at a certain point, the geneticists stopped simply reading the DNA. They started using it to reconstruct faces. In 2019, a team of forensic geneticists and physical anthropologists, working with ancient remains from Neolithic China, used a technique called DNA phenotyping, which is the process of predicting physical appearance directly from genetic code, the same technology now used by law enforcement to generate suspect composites from crime scene DNA to model what the Yellow River Neolithic farmers actually look like, not artist's impressions based on skull morphology, not guesswork. Probabilistic facial reconstructions driven directly by the genetic data buried inside 7,000-year-old Petrus bone. The results were published with careful scientific hedging, but the images themselves were not subtle. The ancient Yellow River farmers did not look like modern Han Chinese. Their predicted facial features, broader nasal structures, different epicanthal fold patterns, distinct cranial [music] proportions reflected a population that had not yet undergone the specific selection pressures and admixture events that would eventually produce the physical characteristics most commonly associated with modern East Asian populations today. They looked, in the assessment of several researchers who reviewed the reconstructions, more similar to certain modern indigenous Siberian populations than to the people now living in the same river valley where those ancient farmers were buried. Think about that for a second. Stand on the banks of the Yellow River in Henan province today.

[00:16:15]Look at the modern Chinese citizens living their lives in the cities built on top of those 7,000-year-old settlements. Now, imagine the people who originally farmed that same soil buried beneath your feet, whose faces the genome has partially reconstructed, and understand that those two groups of people, separated by 7,000 years, would not immediately recognize each other as kin. That is not a statement about superiority or inferiority. It is a statement about time, about what 7,000 years of migration and admixture and population replacement actually does to a people. It changes them completely at the level of the face. But the facial reconstruction story is only the beginning of this act. Because when researchers began asking not just what did the ancient people look like, but where did they go? The answers started arriving from directions nobody anticipated. The Yellow River Neolithic population didn't simply disappear.

[00:17:01]Populations don't vanish. They get absorbed, diluted, pushed to margins, or in some cases, they migrate. And when scientists began tracing the genetic descendants of the ancient Yellow River farmers using the full toolkit of modern population genomics, they They them in places that should, by any conventional historical logic be completely unrelated to ancient China. A significant genetic signal from the Yellow River Neolithic founding population, the genome of the people who planted the first millet, who built the first rammed earth walls, who established the cultural DNA of what would become Chinese civilization, shows up in measurable concentrations among the Tibetan people. We touched on this earlier, but the depth of that signal deserves its own full examination. The Tibetan population, which diverged from a common ancestral group with early East Asians somewhere between 9,000 and 15,000 years ago, preserved genetic material from the founding Yellow River population at levels that have been reduced in modern Han Chinese through subsequent admixture events. The Tibetans went up, literally. Their ancestors moved onto the Tibetan Plateau, one of the most extreme environments on Earth averaging over 4,500 m above sea level, and in doing so, they removed themselves from the subsequent mixing events that transformed the genetic profile of lowland China. The plateau became a vault, and inside that vault, they preserved genetic material from the ancient founding population of Chinese civilization more faithfully than the people living in what is now considered the Chinese heartland. The Tibetan people are, in a very specific and scientifically measurable sense, more closely related to the original founders of Chinese civilization than the modern Han Chinese who administer governance over Tibet. That is not a political argument. That is a genomic observation.

[00:18:35]But it is an observation that lands with extraordinary weight in the context of contemporary geopolitics. In 2023, a researcher at Fudan University in Shanghai, one of China's most prestigious institutions, gave a public lecture about the ancient DNA findings from the Yellow River basin. She was careful. She was precise. She framed everything in the language of scientific nuance. But at one point, speaking to an audience of graduate students, she said something that transcended the data entirely. She said, "The bones don't know what dynasty they belong to." She was right. The bones don't know. The DNA doesn't know. The genome is not a citizen. It carries no passport. It swears no allegiance to any dynasty, any empire, any modern nation-state. It simply records what happened, every migration, every mixing event, every moment when two groups of people who had been separated for thousands of years came into contact and created something new. What the ancient DNA of China reveals is not a story of one people in one place across 10,000 years. It is a story of human beings doing what human beings have always done, moving, adapting, merging, building civilizations out of the accumulated experience of multiple ancestral traditions. Then telling a simpler story afterward, a story of unity and continuity, because human beings need that story, because civilization requires it, because the complexity of the genomic truth is harder to put on a flag. But here's the thing about that complexity, it doesn't diminish what was built, the Great Wall and the oracle bones and the poetry of the Tang Dynasty and the bronze vessels of the Shang.

[00:20:00]None of that becomes less extraordinary because the people who built it turned out to be a synthesis of millet farmers, rice farmers, and steppe migrants and absorbed southern minorities. If anything, the synthesis makes it more extraordinary. What we are looking at in the genetic record of ancient China is evidence that civilization itself is a form of alchemy, the transformation of diverse human raw materials into something unified, something enduring, something that outlasts the genetic particulars of any single founding group.