In 1981, Robert Axelrod ran a tournament.
He invited game theorists, economists, mathematicians, and computer scientists to submit strategies for the iterated Prisoner's Dilemma — a game where two players repeatedly choose to cooperate or defect, and the payoffs depend on what both players do. Cooperate together and both benefit moderately. Defect while the other cooperates and you win big. Both defect and both lose.
Fourteen strategies were submitted. Some were elaborate. Some were devious. Some used complex probabilistic models to predict opponent behavior and exploit it.
The winner was the simplest strategy in the tournament. Anatol Rapoport submitted "Tit for Tat": cooperate on the first move, then do whatever the other player did last.
That's it. No exploitation. No prediction. No cleverness. Just reciprocity.
Axelrod ran the tournament again with sixty-two entries. Tit for Tat won again. He ran computer simulations with thousands of generations, letting strategies compete, reproduce, and die based on their success rates. Cooperative strategies dominated every time. The exploiters won early rounds and then went extinct — because once they'd driven out the cooperators, they had no one left to exploit but each other.
The takeaway was striking: in any system where agents interact repeatedly, cooperation isn't just moral. It's the mathematically optimal survival strategy.
The Genetics of Cooperation
For decades, the evolutionary puzzle of cooperation was: how does a trait that benefits others at cost to yourself survive natural selection? Selfish individuals should outcompete altruists every time.
The answers have come from multiple directions.
William Hamilton's kin selection theory, published in 1964, showed that altruistic behavior toward relatives makes genetic sense — you share genes with your kin, so helping them reproduce helps propagate your own DNA. This explains why worker bees sacrifice themselves for the hive and why parents sacrifice for children.
But kin selection doesn't explain cooperation between unrelated individuals. Robert Trivers's reciprocal altruism theory (1971) does: I help you now, you help me later. The cost of helping is recovered through future reciprocation. This works in any population where individuals interact repeatedly and can remember who cooperated and who cheated.
The genetic basis for these behaviors is increasingly well-documented. The OXTR gene, which codes for oxytocin receptors, influences trust, empathy, and social bonding. Variants in the AVPR1A gene affect pair bonding and prosocial behavior. The COMT gene influences emotional regulation and stress response — traits that affect how people behave under social pressure.
These aren't "cooperation genes" in a simplistic sense. They're genetic variants that influence the neural and hormonal systems that make cooperation possible — the capacity for trust, empathy, delayed gratification, and social memory.
Cooperation isn't just a choice. It's a capacity. And that capacity has a heritable component.
What Happens When You Remove the Cooperators
This is where the research gets uncomfortable.
In the 1970s, the Soviet geneticist Dmitri Belyaev ran a famous experiment on silver foxes. He selectively bred them for a single trait: tameness. Within just a few generations, the tame foxes began exhibiting other changes — floppy ears, curled tails, spotted coats, changes in brain chemistry. Selecting for one behavioral trait produced a cascade of related changes.
The same principle works in reverse. If you selectively remove cooperative individuals from a population, you don't just get a less cooperative population. You get a cascade of related changes — increased aggression, reduced social trust, diminished capacity for collective action.
This isn't hypothetical. History has run this experiment repeatedly.
Stalin's purges systematically removed the most competent administrators, officers, and scientists from Soviet society — precisely the people whose combination of intelligence and institutional cooperation made complex systems function. The result wasn't just a political loss. It was a genetic one. Millions of people with high-cooperation, high-competence trait profiles were eliminated from the Soviet gene pool.
The Khmer Rouge emptied Cambodia's cities and murdered anyone with education, professional skills, or the cooperative social networks that came with them. Teachers, doctors, engineers, even people who wore glasses. The country's recovery took decades — not just because the infrastructure was destroyed, but because the human capital, including its genetic substrate, was devastated.
Mao's Cultural Revolution targeted intellectuals and "class enemies" — disproportionately the educated cooperators who formed the connective tissue of Chinese society. The resulting damage to China's institutional capacity lasted a generation.
Every authoritarian purge follows the same pattern: remove the cooperators, then discover that the society can't function without them.
The Defector's Paradox
Axelrod's tournaments revealed something else. In a population of cooperators, a single defector can thrive — exploiting trust, extracting resources, rising to dominance. This is why cheaters, con artists, and authoritarian leaders succeed in the short term. They're exploiting a cooperative equilibrium that they didn't build.
But the success is self-terminating.
As defectors multiply and cooperators withdraw or die, the environment shifts. There's less trust to exploit, fewer cooperative networks to parasitize, less surplus to extract. The defectors end up in a population of other defectors — and in Axelrod's simulations, populations of defectors collapse. Every time.
This is the trajectory of every extractive regime in history. The early phase is profitable. The late phase is catastrophic. The transition happens because the regime consumed the cooperative infrastructure that produced the wealth it was extracting.
The Roman Republic's transition to Empire followed this pattern. The early Empire was productive because it inherited the Republic's cooperative institutions. As those institutions were hollowed out by autocratic extraction, the Empire's capacity to manage its own complexity declined. The late Empire couldn't maintain its roads, its armies, or its tax base — not because it lacked resources, but because it had destroyed the cooperative networks that turned resources into institutional capacity.
The Fatal Flaw
This is the argument that I kept circling back to while writing The Genesis Protocol.
Morrison's THRESHOLD program is designed to reduce the global population by four billion through targeted genetic deployment. The targeting is based on genetic markers — the "Pattern Eye" system categorizes millions as "preserve" or "eliminate" based on genetic profiles associated with resilience, cognitive flexibility, and adaptability.
But here's what Morrison's model doesn't account for: the traits that make populations resilient are inseparable from the traits that make populations cooperative. Cognitive flexibility, stress tolerance, pattern recognition — these are the same traits that underpin trust, reciprocity, and collective action.
THRESHOLD doesn't just reduce the population. It eliminates four billion carriers of cooperative traits from the human gene pool. The surviving population would be optimized for individual resilience but devastated in its capacity for collective behavior.
In Axelrod's terms, THRESHOLD creates a world of defectors. A population selected for individual survival in the absence of cooperative infrastructure. The strongest individuals on a planet that no longer has the social tissue to support a civilization.
The math Morrison presents to Sarah Chen is rigorous. His population projections are defensible. His ecological models are peer-reviewed. What's missing is the variable he never included: the cooperative capacity of the population he'd be destroying.
Sarah doesn't articulate it in these terms. She's a geneticist, not a game theorist. But she knows something Morrison doesn't: the traits he's selecting for and the traits he's selecting against are entangled. You can't optimize for survival while eliminating the genetic basis for the cooperation that makes survival meaningful.
Morrison's model predicts a stable post-THRESHOLD population. Axelrod's math predicts extinction.
The Cooperative Imperative
The research points in one direction: cooperation is not a luxury that civilizations can afford once they've solved harder problems. Cooperation is the hard problem. It's the genetic and cultural infrastructure that makes everything else possible — agriculture, trade, governance, science, art.
Remove it, and you don't get a leaner civilization. You get no civilization at all.
Every authoritarian who has tried to build a society by purging the wrong people has discovered this. The society they build is brittle, paranoid, and unable to solve complex problems. It collapses — not from external pressure, but from the internal absence of the cooperative capacity it destroyed.
Cooperation is genetic. It's heritable. It's the most valuable trait in the human genome.
And it's the first thing that every system of concentrated power tries to eliminate.