In this blog post, we’ll examine how human altruism developed, focusing on the repetition-reciprocity hypothesis, and explore the nature of altruistic behavior together.
We tend to view altruistic people more favorably than selfish ones. We tend to view someone who betrays a colleague for personal gain as a bad person, while we see someone who donates to the Salvation Army kettle in winter as a good person. However, upon closer reflection, the existence of altruistic humans is quite a mystery. This is because altruistic individuals, who sacrifice themselves for the benefit of others, seem like they should have gone extinct after being exploited by selfish individuals who free-ride on their kindness. Therefore, to resolve this contradiction between reality and common sense, scientists have long studied human altruistic nature and proposed various theories. One of these is the Repeated Reciprocity Hypothesis, which argues that altruism evolved through the process of returning an action exactly as it was received—much like the saying, “an eye for an eye, a tooth for a tooth.” It posits that the essence of altruism lies in responding to a kindness with a kindness, and to betrayal with betrayal.
Before examining the repeated reciprocity hypothesis in detail, it is necessary to understand the “Prisoner’s Dilemma,” which systematically explains altruism and selfishness. This is a type of game that posits a scenario in which two suspects of a crime are arrested and interrogated separately in different rooms. They know that the other is being asked similar questions in a separate interrogation room, but they cannot communicate with each other. According to the law, if both confess, the charges are proven, and they each receive a five-year prison sentence; if both deny the charges, they each receive only a one-year sentence due to lack of evidence. On the other hand, if only one confesses and the other denies the charges, the confessor is released, while the one who denied the charges receives a harsher punishment of seven years in prison. At first glance, it seems like the best option for everyone would be for the two suspects to agree to remain silent until the end and each receive only a one-year sentence. However, upon closer examination, this outcome is unlikely to occur. From one person’s perspective, if the other confesses, it is better to confess as well and receive a five-year sentence than to remain silent until the end and receive a seven-year sentence. Conversely, if the other person remains silent, it is more advantageous to confess and be released immediately than to remain silent together and receive a one-year sentence. In other words, regardless of the other person’s choice, confessing is the rational choice. This is the most basic starting point for explaining the enigma of altruistic humans.
If, like in the Prisoner’s Dilemma, betrayal were always the best option, there would be no way to explain altruism. However, the repeated reciprocity hypothesis explains that if this game is repeated, altruistic behavior becomes the best strategy through reciprocity. This simple explanation gained significant attention through a famous computer simulation tournament organized by American political scientist Robert Axelrod. Axelrod recruited computer programs to participate repeatedly in a league-style format in games such as the Prisoner’s Dilemma. The winner was the program that achieved the highest payoff—that is, the lowest prison sentence. A total of 15 programs participated, ranging from random strategies to highly complex ones, but surprisingly, the winning program was the simplest one. This program initially chose to cooperate and then simply mimicked the action its opponent had chosen immediately before. It was a strategy where, if the opponent confessed, it would confess as well; conversely, if the opponent cooperated, it would cooperate as well. This strategy is called “Tit for Tat.”
Let’s think a little more about why Tit for Tat becomes the dominant strategy when the game is repeated. From an intuitive and empirical perspective, while betraying someone you see every day to leave a bad impression might yield a large immediate benefit, it is highly likely to result in a much greater loss in the long run.
Since the Prisoner’s Dilemma is a one-time scenario, defection is the best strategy; however, in real human society, interactions and cooperation with the same people often occur repeatedly. For example, if a butcher deceives a customer by selling poor-quality meat, he might make more profit in the short term, but over time, he will lose all his regular customers.
The process of mathematically analyzing the Repeated Reciprocity Hypothesis with strict rigor is somewhat complex. However, simply examining what happens between a simple conditional cooperation strategy and a strategy of always defecting is enough to understand why this hypothesis is mathematically valid. Let’s consider a strategy where one cooperates if the other cooperates but defects continuously thereafter if the other defects even once, and a strategy where one chooses to defect from start to finish. If both cooperate, each receives a payoff of 1; if only one defects, the defector receives a payoff of 2; and if both defect, neither receives any payoff. When two players who have chosen conditional cooperation strategies face each other, they begin by cooperating and continue to do so thereafter. On the other hand, when a conditional cooperation strategy meets a defection strategy, in the first game only the defector receives a payoff of 2, and thereafter both parties continue to defect, resulting in no additional gains. According to this analysis, when the likelihood of the game being repeated is sufficiently high, the expected payoff of the conditional cooperation strategy exceeds that of the defection strategy. Therefore, in an environment where repeated interactions occur, defection can no longer be considered the best strategy.
In this way, the repetition-reciprocity hypothesis offers a very clear answer to the puzzle of altruistic behavior. In short, it posits that the attitude of “I’ll treat you the way you treat me” is the root of altruism. The fact that betraying a partner with whom one frequently interacts or meets repeatedly leads to long-term losses is easy to understand intuitively and can also be explained through mathematical analysis. Of course, reality is not this simple, and the Repeated Reciprocity Hypothesis also has several limitations. The most notable limitation is that it assumes repeated encounters. For example, altruistic behavior that occurs in situations where there is little chance of a repeat encounter—such as leaving a tip at a tourist destination where you are unlikely to return—is difficult to fully explain with this hypothesis alone. Nevertheless, the Repeated Reciprocity Hypothesis remains a powerful starting point for understanding human cooperation and altruism. Ultimately, uncovering how altruistic humans emerged is closely tied to the process of understanding human nature, the structure of society and norms, and who we are as beings. For this reason, the repeated reciprocity hypothesis will continue to serve as an important theoretical foundation for research on human cooperation and social behavior.
