Verified by 
This week came news that scientists tested the prisoner’s dilemma with actual prisoners. They found much greater levels of cooperation than in non-prisoner subjects. Most people who have discussed it are surprised.
But should we be?
In case you need a reminder, the prisoner’s dilemma is a fascinating and theoretically influential thought experiment from game theory.
From Wikipedia: "Two members of a criminal gang are arrested and imprisoned. Each prisoner is in solitary confinement with no means of speaking to or exchanging messages with the other. The police admit they don't have enough evidence to convict the pair on the principal charge. They plan to sentence both to a year in prison on a lesser charge. Simultaneously, the police offer each prisoner a Faustian bargain. If he testifies against his partner, he will go free while the partner will get three years in prison on the main charge. Oh, yes, there is a catch ... If both prisoners testify against each other, both will be sentenced to two years in jail."
If both refrain from confessing, game theorists call that cooperation. That behavior is taken as a laboratory model for beneficial cooperation in a variety of circumstances. Not cooperating is called defection.
It’s been important because of its role in studying cooperation, the evolution of cooperation, altruism, trust, and social bonding.
But there’s one thing that almost everyone misunderstands about it. We assume that cooperation is morally good. So we root for cooperation to win out. We even think that cooperation is a fragile state that needs to be nourished and supported, and it’s the milk of human kindness. And we think defection is morally bad.
All false!
The prisoner’s dilemma is just a bunch of numbers. They resemble some behaviors in the real world, and in those situations, there are certainly good and bad behaviors. But the equations in the prisoner’s dilemma can apply to just as many situations in which defection is good and cooperation is bad.
One of those situations is anti-trust regulation. The movie The Informant provides a sense of how easy it is for companies to perniciously cooperate with each other and how much benefit they can exploit the rest of us from doing so. The government, representing the consumer, has a huge incentive to stop this cooperation, and yet evidently, has a difficult time rooting it out completely.
Another example is a bunch of students sharing answers on an exam. Or any group of people helping each other cheat. In The Insider, we the audience are rooting for the main character to defect, and turn on the cheating tobacco industry.
Another example is the Mafia. The Sopranos does a great job of showing how just a little bit of cooperation can give mafia members the ability to commit terrible crimes in broad daylight in front of the police with almost no danger of being tried. In that case, police and government worked for decades to devise strategies that would result in defection and still have not fully succeeded.
Point being, cooperation is not morally superior to defection. Either one can be better. It just depends on the circumstances.
Maybe the problem is the terminology. Instead of calling them ‘cooperation’ and ‘defection’, we should sometimes call them ‘collusion’ and ‘honesty.’
And maybe we shouldn't be surprised that criminals are more likely to cooperate. They live in a world where cooperation is greatly prized.
And come to think of it, going back to the original prisoner's dilemma story, as law-abiding citizens, should we really be rooting for criminals to cooperate with each other and exploit the justice system to get less time in jail?
