We often hear repreated Gandhi's famous phrase: "An eye for an eye makes the whole world blind". As popular as it is, it is wrong. It will make the whole world blind only if, upon somone poking out my eye, I go and poke out the eye of a random bystander. However, if I poke out the eye of the one who poked out mine, the result of the "eye for an eye" priciple will be the rapid blinding, and thus neutralization, of the aggressive eye-pokers: the ones who poke out eyes unprovoked.
Gandhi's aforementioned dictum is often used to argue against vengeance, to claim that vengeance is a bad thing. That, too, is wrong, and for the same reason as the literal dictum itself.
Let me shift gears for a moment, and discuss game-theoretic modeling of conflict resolution policies.
Imagine a society which comprises two types of individuals: "doves", who never fight for any reason when confronting another individual, and "hawks", who always fight to the bitter end when they want something. When two individuals have a conflict over some resource, they act in accordance with their type.
- If two doved conflict over a resource, the first one to grab it gets to keep it.
- If two hawks conflict, they fight it out until one gets too grievously hurt, or dies.
- If a dove and a hawk conflict, hawk wins the resource.
The thing to understand is that this is a very unstable arrangement. In the society of all doves, a hawk will always win. In a society of all hawks, a dove will never win, but also never get killed or grievously injured, even as hawks brutalize each other. In either case, the minority gets an advantage, rapidly becoming majority, after which the former majority gets an advantage, etc. Social structure becomes very unstable.
Now imagine two new hybrid agent types: "bully", who acts like hawk with doves, and like dove with hawks, and "relatiator", who acts like dove with doves and like hawk with hawks. A moment's reflection will show the bullies with destabilize the society even further; but retaliators, in contrast, will stabilize it. In fact, a society of all retaliators is the only stable simple arrangement, because neither hawks nor doves have an advantage in it, and bullies are at a decided disadvantage. Relatiators will make it possible for the society to real evolutionary equibrium.
This simple model shows why vengeance exists as a social institution: any society, in order to be stable, must have evolved it.
In fact this is exactly what happened, and not only on a societal level. Sense of vengeance and justice -- which, I would argue, are two sides of the same coin -- is hard-wired into out brains.
There have been multiple studies done on using The Ultimatum Game to study human behavior; this is a game where two individuals, "proposer" and "aceptor", figure out how to divine, say, $20. "Proposer" (who works for the experimenters) makes a proposal allocating various sums to himself and the rest to the "acceptor"; and the acceptor, who is the real subject of the experiment (though he thinks "proposer" is a subject as well) must decide whether to acept or reject it. If the proposal is rejected, neither individual gets the money.
Even though unalloyed selfishness would suggest that the accepter should accept any amount above $0 (because some money is better than no money), in fact accepters tend to reject grossly skewed proposals, thus losing money themselves but also inflicting a greater punishment on the unfair proposer. In a way, these vengeance-driven, or justice-driven, acceptors throw themselves on the metaphorical grenade in order to punish the wrong-doer. In psychology this is knows as "altruistic punishment".
The really interesting thing is that if you use Transcranial Magnetic Stimulation to suppress the prefrontal lobe function (specifically right dorsolateral prefrontal cortex), the accepters suddenly start accepting much more unfair arrangements. They still recognize them as unfair, but their impulse to punish the unfairness, to inflict vengeance on the wrong-doer, is suppressed (1, 2). This suggests that the sense of vengeance, the same sense which is necessary to keep the society fucntional and stable, is determined at biological level, and is not merely a social institution.
What we arrive at, then, is the realization that vengeance is a good thing, and is in fact necessary to keep the society functional.
Of course, like all human institutions, it's imperfect; and when misused, can lead to endless cycles of violence, "Hatfields and McCoys"-style. However, if we refrained from vengeance altogether, we would be inviting abuse by those who are willing to use force aggressively.
Moral of this story: don't start a fight, but always finish it.
From my blog.