1. Indiscriminate attempts to kill rivals are not beneficial, because there is usually a limitless supply of other individuals waiting to fill the dead rivalís place. Therefore, killing one rival represents a great personal cost, but the benefit goes almost entirely to whatever other individual is then able to fill the newly empty niche. Instead of general violence ruling populations, evolutionarily stable strategies usually establish themselves. An evolutionary stable strategy is one which, once it is adopted by the majority of the population, is superior to any other potential rival strategy.† Natural selection then makes it impossible for any initially rare rival strategy to successfully take over and become predominant.1
2. People are generally more aggressive and violent when faced with one (or a few) specific threats, rather than with a generally threatening situation. For example, a starving person probably would not strike out randomly at everyone around them. However, if the starving person saw one other person in possession of some food, they would quickly become aggressive and attempt to steal the food. This makes sense, because generalized aggression would likely result in net harm to the starving person, since it would waste energy, antagonize rivals, and not result in any gain in food. More focused aggressive behavior, on the other hand, has the potential to be beneficial, since the starving person could plausibly overpower one rival. Of course, the situations in which humans are actually violent and aggressive are seldom this simplistic and obvious. Modern society is complex enough that struggles are usually more complicated than a direct battle for sustenance. However, most cases of human aggression can be traced back to some similar origin.2
3. In some long-ago mammalian ancestor, a gene appeared which predisposed parents to love their children. Parents who loved their children were more likely to provide their children with food and keep them safe, even when this put the parents at a personal disadvantage. Ordinarily, altruism would not be favored by natural selection, but in this case some degree of altruism actually was favorable. That is because parents and children share fifty percent of their genes, and because children are likely to outlive their parents. Genes do not care where they exist. A copy of a certain gene in a parentís body has no particular loyalty to the parent, and it would just as soon preserve a copy of itself in a child as in the parent. In fact, it is preferable to exist in younger children (who have the potential to reproduce in the future) than in an older adults.3 Because of all this, the genes of parents who loved their children were passed on more than were the genes of unloving parents, and the child-loving gene gradually became widespread in the population.
4. A gene causing its bearers to care about themselves would have better evolutionary success than any other similar gene, for example one causing love for siblings or offspring. At first glance, the instinct for self-preservation would be no more advantageous (on the level of gene success) than an instinct to sacrifice oneself to preserve the lives of two or more siblings. Practically, however, individuals have no foolproof way to tell who their siblings are. A sibling-saving gene would sometimes result in mistakes; individuals would sometimes sacrifice themselves to save individuals who did not actually share many of their genes. In the long run, natural selection will favor the self-concern gene over the sibling-concern gene, because it is impossible for the self-concern gene to make this type of mistake. The self-concern gene is good at preserving copies of itself, so individuals with this gene survive long enough to reproduce, thereby passing on their genes to the next generation. (Although since individuals are not entirely self-centeredóparents, children, siblings, and friends do care for each other--the ideal balance of instincts appears to be a blend of self-preservation and the preservation of others)4.
1. However, evolution should allow for similar, and more nuanced strategies to evolve, such as: always follow the current ESS unless conditions abc arise, then follow program xyz.
2. There is, at this intersection, the interesting and fairly new field of evolutionary criminology.
3. I donít know exactly how finely-tuned evolution can turn this dial, but there is also the factor to consider of further possible children produced by the parents.
4. Given that the others would likely perform an act of reciprocal altruism.