Showing posts with label prisoner's dilemma. Show all posts
Showing posts with label prisoner's dilemma. Show all posts

Saturday, April 27, 2013

The Evolution of Virtue 6



Former student and friend Miranda has been putting me to the task in a long running exchange appended to this post.  I will address two general questions she poses at various points in this and a subsequent post. 
First, if virtuous behavior is selected for, then why does vicious behavior persist?  Why are some of us still tempted to do immoral things and why are some persons so inclined toward immorality that they become tyrants? 
I reply that selection works by favoring traits that are reproductively fit within a distinct environmental context.  The usual term for this is a niche.  Several distinct sets of organisms may be found at different altitudes on a mountain because temperature, moisture, etc., form distinct niches.  Niches may be geographically separated or they may overlap and more than one may be found inside another.  An environmental niche will usually include other organisms in such roles as predators and prey, mates and partners.  Thus, in adapting to a niche one species will often create a niche for another, as a large animal, in making a home in the forest, makes a home for his load of parasites.  Finally, different individuals within the same species will sometimes occupy slightly different niches.  The reproductive interests of male and female are rarely identical. 
The simplest model for understanding how distinct niches can favor both morality and immorality is the infamous prisoner’s dilemma.  This is a simple game in which two players have each the choice to cooperate or defect.  Here is the payoff matrix:

The game produces a paradox.  In a single round, each player does better by defecting regardless of what the other player does.  In an extended series of rounds, both players do better by cooperating.  That is the logic of justice. 
In a famous tournament involving computer programs that consisted of strategies for playing the game, the most successful strategy was called Tit For Tat.  This program always begins by cooperating and then on each subsequent move it mimics what the other player did on the last move.  TFT wins by forming cooperative partnerships with any player inclined to cooperate and by punishing defectors (as well as limiting they damage they can inflict).  TFT is the simple logic of righteousness in exchanges.  The unrighteous, always-defect hawks fare poorly against a population of honest cooperators. 
TFT also creates a niche for the saintly, always-cooperate program, which prospers when it is playing with TFT.  However, this in turn creates a niche for the occasional hawk who makes out like a bandit when he encounters a saint. 
Replace points with offspring and this provides a model for the evolution of virtue.  Honest cooperators are stronger, collectively, that dishonest defectors.  However, the more successful that an honest community becomes, the more it leaves a niche for the conman who can effectively exploit the trust that the cooperators have achieved.  That is why most people are not very good at cheating and why no human community is free from cheats. 
Christopher Boehm argues in the second chapter of Moral Origins that, while our moral emotions keep us in check most of the time, the temptation to cheat when we think we can get away with it is also selected for.  This seems to me to be right.