David Reich – Why the Bronze Age was an inflection point in human evolution

What It Covers

Harvard geneticist David Reich presents findings from a large-scale ancient DNA study covering 18,000 years of human history across Europe and the Middle East. Using roughly 10 million genomic positions, Reich and colleague Ali Akbari demonstrate that natural selection has been pervasive rather than quiescent, with the Bronze Age emerging as a critical inflection point for biological adaptation across immune, metabolic, and cognitive traits.

Key Questions Answered

• •Natural Selection Scale: Ancient DNA analysis of ~10 million genomic positions reveals approximately 3,600 confirmed sites under selection in the last 18,000 years, with the genome showing selection signals at virtually every position. Despite representing only 2% of total frequency change — the remaining 98% driven by migration and genetic drift — this 2% produces measurable, population-wide biological shifts of roughly one standard deviation across multiple complex traits.
• •Bronze Age as Biological Inflection Point: Selection pressure intensifies markedly between 5,000 and 2,000 years ago — the Bronze Age and Iron Age — not at the earlier Neolithic farming transition as previously assumed. Immune and metabolic traits show the strongest acceleration during this window, suggesting that high population density, proximity to domesticated animals, and urban living created greater biological stress than the initial shift from hunter-gatherer to agricultural subsistence roughly 10,000 years ago.
• •Immune Traits Under Strongest Selection: Genome-wide association data reveals a four-to-fivefold enrichment of selection signals in immune-related traits compared to other categories. Metabolic traits — including obesity risk, fat mass, waist-to-hip ratio, and type 2 diabetes susceptibility — show the next strongest enrichment. The TiK2 variant, a major tuberculosis risk factor, rocketed upward in frequency before 3,000 years ago, then reversed sharply, suggesting a pathogen-driven environmental shift around that period.
• •Cognitive Trait Selection Peaks in Bronze Age: Polygenic scores predicting performance on modern IQ tests and years of schooling show selection peaking between 2,000 and 4,000 years ago, with a two-standard-deviation effect size during that window versus one standard deviation across the full 18,000-year period. Crucially, no detectable selection on these traits appears in the last 2,000 years. A cross-validation using Chinese GWAS data produces a five-to-six standard deviation correlation with European trajectories, ruling out statistical artifact.
• •Migration Dwarfs Selection but Both Matter: European hunter-gatherers score approximately three standard deviations below the modern mean on polygenic predictors of cognitive performance, while early farmers score near zero — a gap driven entirely by population replacement, not selection. The steppe migration around 4,500 years ago shifted allele frequencies by 40–80% across the genome. Separating migration effects from genuine directional selection requires treating the dataset as an archipelago of isolated sub-populations and testing whether allele changes consistently point in the same direction across all of them.
• •Thrifty Gene Hypothesis Supported by Data: Genetic variants associated with obesity, high BMI, and type 2 diabetes decrease by roughly one standard deviation over 10,000 years in Europe and the Middle East, consistent with the thrifty genes hypothesis. Populations with longer agricultural exposure — including Europeans — show greater genetic protection against type 2 diabetes than populations with shorter exposure, such as Native Americans. This implies that stable food access from farming, despite periodic famines, reduced selective pressure for fat storage compared to hunter-gatherer boom-bust cycles.
• •No Fixed Genetic Differences Explain the Cognitive Revolution: A 2016 analysis found no selective sweeps shared across all modern humans dating to within the last 400,000–500,000 years, meaning no single genetic mutation explains the behavioral modernity seen in the archaeological record around 50,000–100,000 years ago. The implication is that the cognitive toolkit enabling farming, art, and state-building was already present in the common ancestral population of roughly 70,000 years ago — and possibly as far back as 300,000 years — making the 12,000-year delay before agriculture a cultural and climatic puzzle, not a genetic one.