Every game so far has assumed a calculating player — Nash's optimizer, Schelling's strategist, Selten's backward-reasoning chooser. John Maynard Smith, a biologist by trade, asked a startling question: what if the players cannot reason at all? In the living world, strategies are not chosen but inherited, and they compete not for satisfaction but for survival, whose currency is offspring. Maynard Smith showed that game-theoretic equilibrium survives this stripping-away of mind entirely — that an equilibrium emerges from blind natural selection just as surely as it does from reason. The discovery vastly widened the reach of the theory, and it hands the volume a new way to think about how a cooperative strategy might take hold: not because anyone decides to cooperate, but because cooperation, in the right conditions, simply out-survives its rivals.
Maynard Smith's move was to reinterpret every element of a game in the language of evolution. The strategies are no longer options weighed by a mind but heritable behavioral programs — to fight or to flee, to share or to grab, to cooperate or to cheat — written into an organism by its inheritance. The players are the organisms carrying those programs. And the payoff is neither money nor pleasure but Darwinian fitness, the number of surviving offspring a strategy yields. With these three substitutions, the machinery of selection replaces the machinery of choice. A strategy that earns higher fitness leaves more copies of itself in the next generation and so spreads; a strategy that earns lower fitness dwindles toward extinction. The population's mixture of strategies shifts across the generations until it settles — and the resting point at which it settles is an equilibrium, reached without a single calculation, by creatures that merely live and reproduce and die.1
The equilibrium concept Maynard Smith defined is the evolutionarily stable strategy, and it captures the idea of a strategy that, once established, cannot be displaced. A strategy is evolutionarily stable if, when almost the whole population already plays it, no rare mutant playing anything else can invade — that is, the established strategy earns at least as much fitness in this population as any mutant would, so the mutant cannot multiply and gain a foothold. Such a strategy is uninvadable: small deviations are selected against, and the population returns to it. The concept is a close cousin of the Nash equilibrium — indeed every evolutionarily stable strategy is a Nash equilibrium, for if some mutant could earn more it would not be held in check — but it adds a condition of stability against the ceaseless drift of mutation, asking not merely that no one could gain by deviating but that the equilibrium actively repel the deviations that biology will forever throw at it. Equilibrium becomes not just a resting point but a resilient one.2
The illustration that made the idea vivid is the contest of hawk and dove. Imagine animals of one species meeting again and again to contest a resource — food, territory, a mate — and suppose each carries one of two programs. The hawk always escalates, fighting until it wins or is injured; the dove makes a display but retreats the moment its opponent escalates. Pit them against each other and the fitness payoffs emerge. A hawk meeting a dove wins the resource outright, the dove fleeing unhurt. Two doves share or settle the resource cheaply between them. But two hawks fight in earnest, and the risk of serious injury makes that encounter very costly to both. Now ask which population is stable. A population of all doves is not: a single hawk, winning every contest against the gentle doves, would multiply and spread through it. A population of all hawks is not stable either: when hawks are everywhere, the constant injurious fights grow so costly that a rare dove, by simply declining them, does better and gains ground. Stability lies between — at a mixture of hawks and doves, or a population in which each animal escalates only part of the time, where the two strategies earn exactly equal fitness and neither can invade the other.3
The hawk-dove result explained something real that had long puzzled naturalists: why animal conflict is so often ritualized and restrained, settled by display and posture rather than by fights to the death, even when a fight to the death might secure the prize. The naive expectation, that selection should make every creature as ferocious as it can be, is simply wrong; for in a world already full of fighters, ferocity is self-defeating, its injuries outweighing its winnings, and restraint earns the higher fitness. The evolutionarily stable mixture builds a measure of peace into the population, not from any benevolence but from the cold arithmetic of survival — too much aggression, and the aggressors destroy one another. The same logic reaches far beyond animals. It illuminates stable sex ratios, the limits of cheating in a symbiosis, the persistence of both bold and cautious temperaments within a species; and it lends a precise meaning to the old intuition that markets, like ecosystems, can reach equilibrium through the differential survival of strategies — successful firms and practices spreading, failures dying away — as if the participants were rational even where no foresight guides them at all.4
Maynard Smith's lesson is that equilibrium has two roots, not one. The classical theory reached it through reason — players calculating best responses to one another. Evolutionary game theory reaches the same kind of resting point through survival — strategies multiplying or dwindling by their fitness until only the uninvadable remain. The second root is in a sense the deeper, for it requires no mind at all, and it suggests that the strategic logic the volume has traced is woven into the fabric of life itself, not merely into the deliberations of the clever. And it hands the volume's central question a new and hopeful angle. The prisoner's dilemma warned that cooperation cannot survive among the self-interested; but if strategies compete by survival across many encounters rather than one, perhaps a cooperative strategy could arise that, once common, no defector could invade — cooperation made evolutionarily stable. Whether that is so, and under what condition it holds, is the question the flagship answers, by setting the prisoner's dilemma to be played not once but again and again. The next plate is the reciprocity.5
Maynard Smith freed game theory from the need for rational players. Reinterpret a game in the terms of evolution — strategies as inherited programs, players as organisms, payoffs as fitness — and equilibrium still appears, reached not by calculation but by selection. The evolutionarily stable strategy is the resting point: a strategy that, once common, no rare mutant can invade, a Nash equilibrium hardened against the drift of mutation. The hawk-dove game shows such an equilibrium to be a stable mixture of aggression and restraint, and explains why real conflict is so often ritualized rather than lethal — for ferocity, when everyone is ferocious, destroys itself. Equilibrium, then, has two roots, reason and survival, and the deeper of them is survival, which needs no mind. This reframing offers the volume's question a fresh hope: that a cooperative strategy, tested across many encounters, might prove uninvadable after all. To find out, one must play the dilemma again and again. The next plate is the reciprocity.