1. A gene favoring the psychological tendency, “Be kind and generous to every other animal, regardless of how they treat you” would be too vulnerable to defectors to survive for very long in the gene pool. Defectors could easily take advantage of individuals who tended to be kind and generous to everyone; they could continuously derive benefits from these individuals and never have to pay anything in return. Even after being betrayed by defectors many times, universally altruistic individuals would continue to waste energy behaving generously towards them. Operating under this advantage, the defectors would tend to produce more offspring than generous individuals. The defector gene would become more common in the population with each passing generation, until eventually all copies of the indiscriminate generosity gene would disappear.
It is interesting that in many human moral systems, the epitome of virtue is to behave with kindness and generosity to all, regardless of reciprocal treatment.1 Examples include Singer’s utilitarianism or Christianity’s, “But I say unto you, That ye resist not evil: but whosoever shall smite thee on thy right cheek, turn to him the other also.” (Matthew 5:39) The same idea is found in many other places as well. This common theme is likely the result of a mixture of our reasoning abilities and our instinct for reciprocal altruism. While universal altruism is too susceptible to defectors to be beneficial, humans did evolve an instinct to perform acts of altruism they trust will be reciprocated in the future. Reasoning and selective altruism evolved separately, but it seems that reasoning, when applied to the basic instinct for reciprocal altruism, often leads us to the conclusion that universal altruism is the ultimate moral action.
2. To avoid the problems faced by the indiscriminately generous individuals from question one, one evolutionarily stable strategy is to be a “grudger,” or someone who behaves altruistically towards all individuals at first only. Then, as grudgers are betrayed by cheats, they learn who the defectors are, and in the future they refuse to behave altruistically towards known defectors. Grudgers get some of the benefits of altruistic behavior, but avoid many of the costs of universal altruism (at least if there are enough of them in the population). Once they reach a critical percentage of the population, they do better than both universal altruists and universal defectors. (Although I am unclear on how this critical mass of grudgers is reached in the first place.)2
Over time, selection for grudgers over universal altruists and defectors acted on the population of evolving humans. As a result, we are wired to be watchful for defectors, so that we know to avoid future altruistic behavior towards them. The empty tank of gasoline is evidence of a possible defector, and the full tank of gasoline is evidence of altruistic trustworthiness. The negative and positive psychological experiences are our evolved methods of choosing the better person to interact with in the future. We tend to seek out positive psychological experiences (and avoid negative ones)3, leading us to interact repeatedly with the reciprocal altruist but not the defector.
3. Facial recognition ability is essential if humans are to be effective grudgers. Throughout human evolution, natural selection favored individuals who avoided altruism towards non-reciprocating individuals and who sought out interactions with reciprocating individuals. It is impossible to do this effectively if one cannot tell other individuals apart. Faces, because they are easily seen and vary significantly between individuals, are the easiest feature to use for identification. Historically, individuals with good facial recognition skills wasted less energy helping defectors, and they benefited from more true reciprocal relationships. This was evolutionarily advantageous, so facial recognition genes became more and more common in the human gene pool over time.
1. Rather than evolving a tendency for universal altruism, it makes sense that an extremely social species would evolve the tendency to express a desire for universal altruism. The more we convince others to be altruistic, the better our own lives will be. One of the best ways to convince someone that they should be altruistic is to assure them we will be altruistic ourselves (and one of the best ways to assure them of this is to actually believe we will be altruistic when the situation arises). However, when that situation does arise, the optimization programming in our minds may switch gears and steer us towards selfish behavior.
2. I don’t know enough about the mathematical models of how novel genes motivating altruism gain a foothold in an inhospitable social environment, but I know they exist, for example, Sobel and Wilson’s Unto Others.
3. It’s interesting that sentences like these have an inherent structure which suggests we have such-and-such psychological traits rather than we are such-and-such traits. It may seem like a trivial distinction, but it’s not when considering issues of personal identity.