Playbook
Tasteful Skills
“Tasteful Skills” argues that the best agent skills are not documentation or best-practice lists.
Back to Podcast Digest
Which Human Traits Are Evolving Fastest? – David Reich
TL;DR
Immune genes are the clearest hotspot of recent natural selection — David Reich says the strongest signals in their dataset were enriched roughly 4- to 5-fold around immune-related traits, with metabolic traits also standing out.
Weak signals for behavior do not mean behavior wasn’t evolving — Reich explicitly says behavioral, psychiatric, and cognitive traits were also under selection, but they’re spread across many genes of tiny effect, making them much harder to detect than immune traits.
The genome is ‘vibrating with natural selection’ rather than sitting still — even if most detected frequency shifts are small, Reich argues selection is acting broadly across the genome, nudging DNA positions in one direction or another instead of leaving them quiescent.
The apparent paradox is that fast recent selection didn’t produce huge population differences — Reich says variants can rise at 1%+ selection rates and double over dozens of generations, yet Europeans and East Asians still don’t show lots of 100% frequency differences because this selection seems historically recent and intensified.
Selection appears to have accelerated in the last 5,000 years — comparing the Bronze Age and later to the prior 5,000 years in Europe and the Middle East, Reich says enrichment around immune and metabolic traits gets stronger in the later period.
Agriculture and dense living likely created an evolutionary ‘shock’ — his framing is that genomes shaped in hunter-gatherers were suddenly dropped into farming, pastoralism, animal contact, and urban density, producing an evolutionary mismatch that selection then started correcting.
The Breakdown
The Genome Isn’t Quiet — It’s ‘Vibrating’
Reich opens with the big claim: the genome is not mostly sitting there untouched by evolution. His team’s analyses suggest natural selection is everywhere, with many weak effects and hundreds of high-confidence positions already identified, so even small frequency changes add up to a picture of DNA being constantly tugged in one direction or another.
What Happens When You Map Selection Against Traits
They took those strongly selected positions and asked a simple question: are they randomly scattered, or clustered around known traits from genome-wide association studies? Looking across about 100 traits, they found a huge enrichment — around four- to fivefold — for immune traits, plus strong enrichment for metabolic traits tied to obesity, fat biology, and type 2 diabetes.
Why Behavioral Traits Look Quiet Even When They’re Not
At first glance, behavioral and psychiatric traits showed almost no detectable enrichment. Reich immediately pushes back on the obvious misread: that does not mean behavior hasn’t been under selection over the last 18,000 years; it means those traits are genetically diffuse, built from many genes of weak effect, unlike immune traits, which often involve fewer genes with larger effects.
The Detection Problem, Not the Selection Problem
That distinction matters because their analysis is strongest at finding big, sharp signals. Reich says the method is effectively “querying” immune traits but not very effectively querying behavioral traits, so the weak signal is mostly about statistical power, not absence of evolutionary change. He even says they can prove the conclusion “behavioral traits are not selected” is wrong.
The Puzzle: Fast Selection Without Total Divergence
Then he frames the deeper paradox. If some variants are rising at selection rates of 1% or more — enough to double over dozens of generations — then across 1,500 to 2,000 generations separating Europeans and East Asians, why don’t we see many variants fixed in one population and absent in the other?
Reich’s First Answer: We’re Looking at an Accelerated Period
One answer is historical timing: Europe and the Middle East seem to be in a period of intensified natural selection, not a constant background rate stretching forever into the past. Comparing the last 5,000 years — the Bronze Age and after — to the previous 5,000 years, Reich says the concentration of selection around immune and metabolic traits clearly intensifies.
Farming, Animals, Cities — and Evolutionary Mismatch
His explanation is deeply human: this population lived through a massive lifestyle shock. Farming begins in the Middle East around 11,000 to 12,000 years ago, expands into Europe after about 8,500 years ago, and later the Bronze Age packs people and animals together at much higher densities, creating new disease pressures and biological demands.
Hunter-Gatherer DNA in a Farmer World
Reich’s closing image is essentially evolutionary whiplash. A genome adapted for hunter-gatherer life gets dropped into agriculture, pastoralism, urban density, and constant pathogen exchange, producing what he calls an evolutionary mismatch — and the selection they’re measuring may be the DNA of that population reacting in real time to the shock.
Share
Keep Reading
Two Ways to Follow
The Weekly Echo. The inbox-shaped summary of what mattered.
New editorials announced here.
