#polygenic selection

By: Razib Khan

Published: May 6, 2026

For decades, many scientists insisted that human evolution essentially stopped 50,000 years ago. A groundbreaking new paper proves them wrong.

In 1859, Charles Darwin published On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life. It revolutionized our understanding of biology. Darwin also inadvertently centered evolutionary theory in the intellectual debates of Victorian-era Britain. Is man indeed a beast? Does he have an immortal soul? Is he a special creation? Darwin’s work triggered a change in the understanding of humans in the universe.

Still, though Darwin created the modern understanding of human origins, his theory and the work that followed him for decades left open the question of how humanity has evolved in more recent millennia. Yet for decades it was common for evolutionary biologists to insist that human evolution had essentially stopped tens of thousands of years ago.

The question of recent human evolution has existed to a large degree in the shadow of famed paleontologist and naturalist Stephen Jay Gould. Gould’s theory of punctuated equilibrium held that genetic evolution proceeded in fits and starts. And for Gould, humanity was in a period of genetic stasis, because humans were now adapting to new conditions by changing culture and lifestyle, not genes. “There’s been no biological change in humans in 40,000 or 50,000 years,” Gould argued in an interview published in 2000. “Everything we call culture and civilization we’ve built with the same body and brain.”

With no way to look into the ancient genome itself, Gould’s view that human genetic change belonged to the past became predominant. The idea that genes could both shape and be shaped by lifestyle was discarded. And biologists like E.O. Wilson, who deviated from this line, were shunned and dismissed as racists, under no more that the weak supposition that digging into the last few millennia of human origins might reveal something unpalatable.

Now the conventional wisdom has imploded. A tour de force of big-data science published in Nature last month shows how a mountain of genetic data backs up the ideas about human evolution that date back to Darwin. By using cutting-edge genomic methods, the researchers showed that natural selection has been ubiquitous across the last 10,000 years, reshaping our appearance, immune system, and even predispositions to mental illness.

In fact, it now appears that culture did not replace genetics as the vehicle for human adaptation, nor did it put an end to humans’ biological fluidity over time. On the contrary, the paper in Nature, from a team at David Reich’s pioneering genetics lab led by Harvard’s Ali Akbari, shows how the movement of people and the rise of agriculture unleashed a surge of genetic changes.

Don’t be put off by the highly technical title of the paper, “Ancient DNA reveals pervasive directional selection across West Eurasia.” Akbari used cutting-edge 21st-century genomics to show something profound: that we are an adapting animal not just in theory, but in reality.

After analyzing ancient DNA from exactly 15,836 West Eurasians—that is, Europeans and West Asians who lived over the last 18,000 years—the Harvard team reported 479 gene variants where there was strong directional selection over the millennia. Skin color, eye color, celiac disease, schizophrenia, bipolar risk, and intelligence are just some of the traits that have changed since the end of the last ice age 12,700 years ago. We are not simply cavemen wearing suits; we are different beasts altogether.

This fine-grained understanding of evolution became achievable due to technology. Until about 25 years ago, biologists could examine individual genes, but did not have the ability to delve into how distinct genes could work in tandem to influence traits, including personality and intelligence. But in recent years, scientists have developed ever better tools to analyze the full set of 19,000 or so human genes, creating the new field of genomics. That, in turn, has provided population geneticists with a fire hose of data. Akbari and his colleagues arguably gathered more information from a single ancient genome than human evolutionary geneticists could amass in the entire 20th century.

That brings us to the details of what Akbari and his colleagues discovered. One key finding: West Eurasians—the peoples of Europe and the Near East—developed a host of adaptations in and around the class of genes that control immune function. This provides a clearer demonstration of how the pressures of natural selection worked in humans. The rise of endemic diseases and pandemics, along with the acceleration of trade and the growth of cities, seem like excellent candidates for selection pressures since the end of the last ice age. Constant selection pressures buffet a world of plagues, twisting and turning genomes generation to generation.

The research’s most startling and potentially controversial results, though, come from its analysis of polygenic traits. These are characteristics controlled not by single genes but by whole sets of variants, an analysis made possible by advances in computational horsepower that let Akbari sift through a mountain of paleogenetic data.

In genetic terms, it looks like Europeans have been selected for higher IQ, higher educational attainment, and less schizophrenia and bipolar disorder over the last 10,000 years. This is a startling result: The selection for educational attainment seems to predate even the most ancient education by thousands of years. Surprising as that is, it seems to be the case, with later samples reflecting selection for the combinations of genes associated with years of schooling in modern studies. Similarly, Neolithic farmers in the dataset carry more variants associated with higher intelligence in modern samples than Mesolithic foragers. It’s a strong sign that the emergence of farming and village culture changed human brains, too.

These claims about polygenic selection have faced scrutiny from researchers wary of both social implications and of the methodology. Geneticists like Iain Mathieson at the University of Pennsylvania, an alumnus of the Reich lab himself, are skeptical of polygenic selection results. Extraordinary claims do demand extraordinary evidence, and Mathieson and like-minded scientists claim that Akbari and his colleagues simply have failed to prove that the changes they see are not due to other evolutionary forces like genetic drift—random variation in gene frequency—rather than selection.

But moving beyond the nuts and bolts of population genetic interpretation, is it plausible that there was selection for greater intelligence in the last 10,000 years? It is. Consider the Neolithic town of Çatalhöyük, in modern-day Turkey. It was home to around 1,000 people 8,000 years ago, just 30 to 40 generations after these ancient Anatolians switched from foraging, most likely in bands of a dozen or so, to farming. Today, humans vary in personality, intelligence, and likelihood of mental illness—and all these traits affect success or failure in multiple ways, as they almost certainly did in ancient Çatalhöyük.

It was certainly the same in the Bronze Age villages of Europe, with their myriad cultural innovations, not the least of them new modes of warfare. The ancient world was a crucible that tested human ingenuity and indefatigability. In fact, these results imply that selection may have accelerated with the shift to farming, with our Neolithic ancestors adapting furiously as they left behind the hunting and gathering that had been synonymous with our species for millions of years.

Though the specific results apply only to the northwest corner of the Afro-Asian landmass over a very specific time, the lessons here are general. Researchers will repeat this work for most human populations with enough samples. A new preprint from the Reich lab deploys the same research methods in East Asia, and replicates many of the same findings.

The bottom line is that Akbari’s paper does not just tell us more about the past. Its effect is to remove the taboo on talking about human evolution in the historic era. This work is relevant not only to the deep past, but to our present, and our future. We now live in a world where populations are on the verge of crashing due to declining fertility. Educated urbanites now have birth rates far below replacement, but the crash has swept over nearly the entire world in a generation. The modern world could turn out to be an evolutionary proving ground just as intense as the Bronze Age, forcing genetic changes that will keep shaping humanity in ways that we ignore at our peril.

––

Abstract

Ancient DNA has transformed our understanding of population history1, but its potential to reveal as much about human evolutionary biology has not been realized because of limited sample sizes and the difficulty of distinguishing sustained rises in allele frequency increasing fitness—directional selection—from shifts due to migrations, population structure, or non-adaptive purifying or stabilizing selection. Here we present a method for detecting directional selection in ancient DNA time-series data that tests for consistent trends in allele frequency change over time, and apply it to 15,836 West Eurasians (10,016 with new data). Previous work has shown that classic hard sweeps driving advantageous mutations to fixation have been rare over the broad span of human evolution. By contrast, in the past ten millennia, we find that many hundreds of alleles have been affected by strong directional selection. We also document one-standard-deviation changes on the scale of modern variation in combinations of alleles that today predict complex traits. This includes decreases in predicted body fat and schizophrenia, and increases in measures of cognitive performance. These effects were measured in industrialized societies, and it remains unclear how these relate to phenotypes that were adaptive in the past. We estimate selection coefficients at 9.7 million variants, enabling study of how Darwinian forces couple to allelic effects and shape the genetic architecture of complex traits.

==

I got all nostalgic the other day when I encountered my first proper god-believing evolution denialist in… I can't even remember how long.

These days it's been overwhelmingly the evolution denialists who think sex differences are a "social construction" or "socialization."