Why Humans Stopped Evolving Smarter 2,000 Years Ago - David Reich
Why It Matters
The stall in genetic intelligence evolution means economic and societal progress now hinges on cultural, educational, and technological investments rather than biological change.
Key Takeaways
- •Genetic selection for intelligence peaked during Bronze Age, 4–2k years ago.
- •No detectable natural selection on cognitive genes in past two millennia.
- •Selection strength then equated to two standard deviations, unusually high.
- •Modern industrial era hasn't altered genetic architecture of intelligence.
- •Cultural advances now drive cognitive gains, not biological evolution.
Summary
David Reich’s lecture examines why human brains have not become genetically smarter over the last two thousand years. By analyzing ancient DNA from white British populations, he tracks variants linked to modern IQ test performance, a proxy for cognitive ability, and maps their frequency changes across millennia. The data reveal a pronounced surge in selection for intelligence‑related alleles between 4,000 and 2,000 years ago, roughly the Bronze Age, with a selection coefficient equivalent to two standard deviations—far stronger than typical polygenic traits. In contrast, the subsequent two millennia show virtually no detectable natural selection on these loci, despite the rise of industrialization and complex societies. Reich emphasizes the paradox: “the impact in the last 2,000 years is almost nothing…there’s no evidence of natural selection at all.” This stark drop suggests that cultural and technological innovations, rather than genetic changes, now drive the rapid expansion of cognitive capabilities. The findings imply that future gains in human problem‑solving will stem from education, institutions, and technology, not from further biological evolution of intelligence. Policymakers and investors should therefore focus on nurturing environments that amplify human capital rather than expecting genetic breakthroughs.
Comments
Want to join the conversation?
Loading comments...