Is Moral Cooperation Rooted in Nature’s Productivity?
Game theory and the progressive revolt against natural norms
Is morality anomalous or does it repurpose the mathematical fabric of nature?
According to Jonathan Maas’s reading of game theory, the latter is the case. He says that in so far as it’s an expression of the urge to cooperate, at least, morality is rational so that even a learning algorithm will eventually seize on this social strategy.
Supposedly, the political scientist Robert Axelrod established as much by hosting a competition for algorithms to play out the prisoner’s dilemma. Some well-known game theorists designed different algorithms, effectively betting on one or another strategy, such as whether to betray the other prisoner or cooperate and risk being betrayed.
What the tournament showed is that strategic betrayal succeeded when the games were one-off, when the fellow prisoner lacked the ability to punish the betrayer in a series of iterated prisoner dilemmas. But in such a series, cooperation came out ahead of betrayal as a strategy. The winning option was to cooperate unless you were betrayed, in which case the winning move was to punish the betrayer by betraying that selfish algorithm in the next round, and then to forgive and revert to a cooperative mode.
