Some kinds of altruism are easier to explain. Kin altruism is the clearest example. If you have siblings or children, then there are other beings with whom you share genes. By sacrificing yourself for your close relatives, you might be insuring the propagation of your shared genes. Thus, parental sacrifice or sacrifice for siblings could confer an evolutionary advantage.
But what about altruism towards non-relatives? This clearly exists, both in humans and other animals. Given selfish genes, why would anyone make a sacrifice, of any kind, to a non-related person?
One possible explanation is reciprocity. We might be willing to do a favor for another individual, if the cost is low and the probability of receiving a favor in return is high enough. In other words, we might be willing to do a good turn, if there is a reasonable chance that person would reciprocate in the future.
Altruism is often defined narrowly as behavior that benefits the recipient at a cost to the actor. But note here that most evolutionary theorists define altruism more broadly to include any behavior where an actor creates benefit for a recipient. That action may or may not benefit the donor (or the donor’s genes). In case of reciprocal altruism both actor and recipient benefit, although those benefits may not occur simultaneously. Note that reciprocity does not depend on shared genes, indeed, there are examples of reciprocity across species. Symbiosis would be a good example. You provide a good home for the bacteria in your gut, your microbiome. In turn your microbiome reciprocates by helping you digest food.
For larger animals to participate in reciprocal exchanges they generally need certain cognitive characteristics. First they must live in close social contact with other animals, so that the opportunity for receiving a favor in return is high. They must be able to distinguish between different individuals and be able to remember who reciprocates and who does not. Finally, there must be the ability to identify and punish cheaters.
And, in fact, social animals with these cognitive prerequisites often do practice reciprocity. How such a system of reciprocity can arise has been well studied by game theorists.
But are these explanations, kin altruism and reciprocity, sufficient to account for human altruism? No they are not. Human altruism seems too extreme. While other animals may engage in reciprocity, humans practice what is called “strong reciprocity” – not only do we reciprocate, we reward cooperative behavior and impose costs for violations of cooperative norms. Indeed, we often do so even at a cost to ourselves. Humans will incur loss to punish norm violators, even when they are not the victims of violation itself.
Moreover, humans often act altruistically even with the awareness that there is no possibility for reciprocation. Take the example of kidney donation, here the cost to the donor is high, the recipient is frequently unrelated, and the possibility of any significant reciprocation is vanishingly small. Yet people do donate kidneys, sometimes to complete strangers. This is, of course, an extreme example, and most people do not give up their kidneys. But all of us participate in a complex society with its vast social networks, such as global markets and nation states, demanding extensive cooperation with non-kin. The benefits of such global cooperation may be obvious to us now, but how did we get here? Can we explain that leap?