In a one-shot prisoner’s dilemma, the dominant strategy is defection: whatever the opponent does, you do better by defecting. But this produces mutual defection and a worse outcome for both players than mutual cooperation — a Nash equilibrium that is Pareto-inferior. In iterated games, the dynamic reverses. Future interactions make cooperation rational because defection invites retaliation and destroys the cooperative surplus.
Robert Axelrod’s computer tournaments (1980s) invited strategists to submit programmes for iterated prisoner’s dilemmas. The winning strategy — tit-for-tat (Anatol Rapoport) — succeeded through four properties: niceness (starts cooperative), retaliation (punishes defection immediately), forgiveness (returns to cooperation as soon as opponent does), and clarity (simple and predictable). Were tit-for-tat complex or opaque, it would lose its strategic advantage.
W.D. Hamilton (1964) showed that genes promoting altruism toward relatives can spread if the benefit to the recipient, weighted by relatedness, exceeds the fitness cost to the actor (Hamilton’s rule: rB > C). Robert Trivers (1971) extended this to non-kin: reciprocal altruism evolves when the probability of future interaction is sufficiently high. Had evolution favoured only self-interest, neither human cooperation nor animal societies of the complexity we observe would have emerged.
Direct reciprocity operates between the same two agents (“I help you because you helped me”). Indirect reciprocity operates through reputation: “I help you because others observe and will help me in return.” Martin Nowak and Karl Sigmund showed that indirect reciprocity can sustain large-scale cooperation in societies where most interactions are non-repeated, provided reputational information flows efficiently.
Inverted: “Were self-interest the only evolutionary pressure, altruism would not exist.”
“Were tit-for-tat not forgiving, it would lose its cooperative advantage rapidly.”
“Were defection the evolutionarily stable strategy, cooperative societies could not persist.”
“Were reputation to disappear, indirect reciprocity would collapse.”
Topic connection: the whole game-theory argument rests on counterfactuals — what would happen if the conditions changed.
Inverted: “Should a player encounter the same opponent again, cooperation becomes rational.”
“Should defectors become too numerous in the population, cooperative networks begin to exclude them.”
“Should you find yourself in a repeated interaction, kindness is not naïve — it is optimal.”
Register note: Should… inversion is characteristic of formal academic and legal writing.
Inverted: “Had Axelrod run only one round, tit-for-tat would not have won.”
“Had evolution favoured only self-interest, neither kin selection nor reciprocal altruism would have emerged.”
“Had the players met only once, defection would have dominated.”
Topic connection: evolutionary history is inherently counterfactual — had the selective pressures been different…
× “Were it not for cooperation, life had been very different.” → use would have been (not had been) in the result clause.
× “Should the strategy would fail…” → no modal after Should: “Should the strategy fail…”
× “Had they cooperated, they would cooperate more.” → keep past hypothetical in result clause: would have cooperated.
Were… “Were we to define rationality as…” / “Were tit-for-tat unforgiving…” / “Were we to take that seriously…”
Should… “Should two agents never meet again…” / “Should your cooperative behaviour become known…”
Had… “Had tit-for-tat been more complex…” / “Had early human groups lacked…”
Answer Key
Why Kindness is Rational
Vocabulary
Vocabularyaltruism · reciprocity · kin selection · iterated game · fitness cost
Grammar
GrammarInverted conditionals omit if and move the auxiliary verb before the subject. This produces a more formal, academic register — and is characteristic of the counterfactual reasoning that runs throughout evolutionary biology and game theory.
Rule: Remove if → move auxiliary before subject → keep result clause unchanged.
Inverted: “Were self-interest the only evolutionary pressure, altruism would not exist.”
“Were defection the evolutionarily stable strategy, cooperative societies could not persist.”
“Were reputation to disappear, indirect reciprocity would collapse.”
“Were tit-for-tat not forgiving, it would lose its advantage rapidly.”
▶ Use for present or future hypotheticals: if…were → Were…
Inverted: “Should a player encounter the same opponent again, cooperation becomes rational.”
“Should defectors become too numerous, cooperative networks begin to exclude them.”
“Should your cooperative behaviour become known, third parties are more likely to help you.”
▶ Use for formal future conditions: if…should → Should…
Inverted: “Had Axelrod run only one round, tit-for-tat would not have won.”
“Had evolution favoured only self-interest, altruism would not have persisted.”
“Had early human groups lacked reputation transmission, large-scale cooperation would not have been possible.”
▶ Use for past hypotheticals: if…had → Had…
× “Were it not for cooperation, life had been very different.” → would have been in result clause.
× “Should the strategy would fail…” → no modal after Should: “Should the strategy fail…”
× “Had they cooperated, they would cooperate more.” → past hypothetical result: would have cooperated.
Reading
ReadingThe Rationality of Kindness
In a single interaction between two strangers who will never meet again, the game-theoretic case for cooperation is weak. Whatever the other player does, you do marginally better by defecting — a conclusion that mathematicians formalised as the prisoner’s dilemma. Had rational self-interest been the only variable, defection would be universal and cooperation would be an evolutionary anomaly.
It is not. Cooperation is ubiquitous in the natural world — from the cellular machinery of every living organism to the complex reciprocal arrangements of human societies. The question is not whether cooperation exists, but why rational agents would choose it.
Robert Axelrod’s computer tournaments in the 1980s provided a compelling answer. When the prisoner’s dilemma is played repeatedly by the same agents, the dynamics invert. Were a player to defect in an iterated game, their opponent could retaliate in all subsequent rounds, eliminating the cooperative surplus both had accumulated. The shadow of the future — the prospect of future interactions — makes cooperation rational in the present.
The winning strategy in Axelrod’s tournaments, tit-for-tat, succeeded through four properties: niceness (it cooperates first), retaliation (it punishes defection immediately), forgiveness (it returns to cooperation as soon as the opponent does), and clarity (its behaviour is predictable). Should tit-for-tat fail to retaliate, it would be exploited. Should it fail to forgive, it would lock both parties into permanent mutual defection.
Kin selection (Hamilton, 1964) explains altruism toward relatives: genes promoting cooperation with others who share those genes can spread, because they effectively help copies of themselves. But this cannot explain the scale of human cooperation, which routinely extends to non-relatives and even strangers. Reciprocal altruism (Trivers, 1971) extends the model: had the probability of future interaction been negligible throughout our evolutionary history, reciprocal altruism could not have taken hold. The small, stable social groups in which humans evolved provided exactly the conditions required.
Indirect reciprocity — helping strangers because observers will cooperate with you in turn — extends this further. Should your reputation as a cooperator become established, you attract cooperative partners without ever needing to repay individuals directly. The entire apparatus of social norms, moral emotions, and reputational concern that characterises human societies can be understood, in part, as a system for making kindness individually rational at large scale.
Listening
ListeningThe Game Behind the Gift — Key Points
Speaking
SpeakingDiscuss in groups of three or four. Use evidence from the reading and listening. Use at least two inverted conditionals in your contributions.
Use: “Were kindness truly strategic and nothing more…” / “Had the science shown otherwise…”
Writing
WritingOption A — Personal / Analytical B2–C1: Describe a situation where acting kindly also turned out to be strategically rational for you. Does the rational basis for your kindness change how you think about it?
Option B — Argumentative C1: “The science of cooperation does not explain away kindness — it vindicates it.” Evaluate this claim, drawing on the evolutionary and game-theory evidence in the lesson.
Option C — Opinion / Evaluation C1: What specific changes to educational institutions would follow from taking the science of cooperation seriously? Address likely objections.
Word count: 250–300 words
Were… at least once · Had… or Should… at least once · at least three vocabulary terms from Handout 1 · a clear thesis · a concession

