The Evolution of Cooperation

The Evolution of Cooperation (1984) is a nonfiction book by American political scientist Robert Axelrod and English evolutionary biologist W. D. Hamilton. Expanded from a 1981 paper on cooperation by Axelrod and Hamilton, the most cited scientific paper ever in the field of political science, its theories are developed by looking at cooperative efforts during World War I, as well as various game theory thought experiments, such as the Prisoner's Dilemma.

The authors begin by setting out their goal for the book: "to develop a theory of cooperation that can be used to discover what is necessary for cooperation to emerge. By understanding the conditions that allow it to emerge, appropriate actions can be taken to foster the development of cooperation in a specific setting." They seek to establish motivations and incentives for people to work together in the absence of a central authority that forces them to work together.

"The Prisoner's Dilemma" is the thought experiment at the center of the book's analysis. In this thought experiment, two criminal conspirators are captured and interrogated in separate rooms. The police have enough evidence to convict them on a lesser charge but require a confession to prosecute the pair to the full extent of the law. The prisoner who betrays his friend will be set free, but only if he himself is not betrayed in turn. There are three possible outcomes to the dilemma: Both prisoners betray one another, in which case each of them receives a harsh sentence of ten years; only one prisoner chooses to betray the other, in which case the betrayer goes free and the betrayed goes to jail for ten years; both prisoners remain silent and each receives a sentence of one year. If each prisoner chooses to pursue rational self-interest, they are both hurting themselves. They would be far better off protecting one another. This is not a real scenario, but the authors, nevertheless, use it as a helpful jumping off point for identifying future scenarios where cooperation is ideal.

To further establish scenarios where cooperation is ideal, the authors created a computer tournament in which cooperation was clearly advertised as a useful strategy. They found that players still tried to exploit the system for their own gain and that doing so threatened the continued existence of the game's framework. For example, one could refuse to cooperate and make short-term gains, but doing so led to the "breaking" of the rules that were necessary for the game to continue to exist. In other words, "winning" isn't always the desired outcome as here, the desired outcome is "continuing to play the game."

The authors call this state of affairs "collective stability." Under these conditions, strategies that enrich the individual tend to threaten collective stability. What's the point of ruling the world if there's no world left to rule? The greater challenge, the authors write, is in determining what scenarios meet these conditions where collective stability is the most desired outcome. Even more problematic, is determining how large groups of individuals come to their own collective consensus about when these conditions are in play.

Here, the authors move away from computer games and hypothetical situations to identify real-life scenarios where principles of cooperation theory bear out. One such example they provide was during World War I in the trenches. If the armies on both sides of the conflict tried to invade the other's territory, both sides would suffer great losses. If each side remained where they were, however, the losses on both sides would be greatly diminished. Doing so would not necessarily bring about a faster end to the war, but it would remove the prospect of almost-certain annihilation from the equation. This, the authors write, is an example of the Prisoner's Dilemma in action. In fact, the trench warfare example is an even more compelling example of cooperative theory than the Prisoner's Dilemma because, in the Prisoner's Dilemma, we are to assume that the prisoners are friends (or at least on the same side of the law). In trench warfare, however, the two sides "cooperate" even though they are sworn enemies.

In addition to social experiments, the writers go on to examine a number of examples in wildlife biology where a population survives due to cooperation. Crucially, animals cooperate without the existence of a coercive government that exists to enforce social contracts and dole out brutal punishments for those who opt out of the social contracts. These coercive governments are not the most effective ways of ensuring cooperation, the authors write. Rather, true cooperation happens because of a shared sense of community among our fellow humans.

The Evolution of Cooperation is heady reading, relying on scientific and mathematical evidence as opposed to mere anecdotal evidence. As a result, its arguments about why cooperation matters and why people cooperate in the first place are all the more convincing.