Tuesday, March 2, 2010

Prisoner's Dilemma

Prisoner's Dilemma describes a situation where two people (or groups of people) have the opportunity to cooperate with one another or try to take advantage of one another. The Wikipedia article on Prisoner's Dilemma gives one formulation of the story:

"Two suspects are arrested by the police. The police have insufficient evidence for a conviction, and, having separated both prisoners, visit each of them to offer the same deal. If one testifies (defects from the other) for the prosecution against the other and the other remains silent (cooperates with the other), the betrayer goes free and the silent accomplice receives the full 10-year sentence. If both remain silent, both prisoners are sentenced to only six months in jail for a minor charge. If each betrays the other, each receives a five-year sentence. Each prisoner must choose to betray the other or to remain silent. Each one is assured that the other would not know about the betrayal before the end of the investigation. How should the prisoners act?"

While I'm certainly not suggesting that I hope criminals will cooperate and avoid justice, this scenerio is the basis for a 'game' which looks at cooperation vs competition as well as individual benefit vs collective benefit and trust vs suspicion. The Stanford Encyclopedia of Philosophy has a long, complex, and detailed article on Prisoner's Dilemma (PD) which points out that "PD is a game in which a 'cooperative' outcome obtainable only when every player violates rational self-interest is unanimously preferred to the 'selfish' outcome obtained when every player adheres to rational self-interest." Their idea of 'rational self-interest' is that you look out for yourself regardless of what that means for the other person.

In a very early post (Two Basic Principles, 6/30/08), I looked at the difference between self interest that only sees immediate gain vs self interest that sees that we are all connected and thus what benefits everyone benefits us. Prisoner's Dilemma interests me because it resembles a lot of social situations that I can think of (although some of these situations more resemble the Game of Chicken). The question that PD most raises for me is how can we increase the likelihood of cooperation?

In some versions of PD, it's played multiple times, which I think more approximates life situations where you are involved with someone else on an on-going basis. The Concise Encyclopedia of Economics suggests punishing someone who 'cheats' as a deterrent. They point out several problems with this, including: "The cheater’s reward comes at once, while the loss from punishment lies in the future." and "Punishment will not work unless cheating can be detected and punished." They also mention, as an example, a game strategy called 'Tit for Tat'.

I mentioned 'Tit for Tat' (although not by name) in my post on Complexity Theory (7/16/08) and talked more about Prisoner's Dilemma and Tit for Tat in my post on Original Virtue (9/14/08). Basically, Tit for Tat cooperates in the first round of PD, and thereafter does whatever its opponent did to it on the last round. Robert Axelrod has written an entire book on Prisoner's Dilemma, Tit for Tat, and what supports cooperation entitled, The Evolution of Cooperation.

The Stanford folks describe Tit for Tat as "...nice, meaning that it is never the first to defect. ... retaliatory, making it difficult for it to be exploited by the rules that were not nice. ... forgiving, in the sense of being willing to cooperate even with those who have defected against it ... And ... clear, presumably making it easier for other strategies to predict its behavior so as to facilitate mutually beneficial interaction." Wikipedia's version is 'Nice, Retaliating, Forgiving, and Non-envious.' (Non-envious meaning "not striving to score more than the opponent".) In my post on Complexity I referred to it as a computer program that was "nice, forgiving, tough, and clear". Being nice, forgiving, clear, and non-envious, all seem like great qualities to me. My problem is with the 'retalitory' piece--the 'tough' part and the 'punishment' that the Economics people are advocating.

Here is the rub. There are lots of people and institutions that will take advantage of anyone who is nice and does not retaliate. Ironically, this 'nice' behavior can encourage aggressive behaviors. It's referred to sometimes as 'enabling' or in Chogyam Trungpa's memorable phrase, 'idiot compassion'. But I am not sure that 'retaliation' or 'punishment' is the answer. I think that we need to find ways of stopping violent, oppressive, or exploitive behavior without retaliation or punishment. I'm thinking about things like non-violent resistance. But this a place I think we are going to need to look at more in the future.

Wikipedia points out that Tit for Tat has parallels to the concept in evolutionary biology called "Reciprocal Altruism". Here biologists look at how cooperation may have evolved and what the advantages of it are. (Some of this is covered in my post on Original Virtue.)

What is clear to me from reading some of the studies done with Prisoner's Dilemma is that cooperation is more likely in a small group where people know one another and there is some long term commitment. In these cases, trust is more likely to build and people are more unlikely to do something to hurt someone they will be dealing with in the future. This is a good case for building community. It also helps to surround yourself with cooperators. (The Stanford site looks at this possibility toward the end in the section entitled 'Spatial PDs'.)

Looking at Prisoner's Dilemma for me is about trying to see what stops us from being more cooperative and looking at ways to get beyond that. Turil has a very nice post that I just discovered as I was about to post this on 'escaping from the prisoner's dilemma'. She has a much better take on PD and Tit for Tat than I put here.

I see all this as part of the bigger question of educating (see my post on 'Creating Social Change', 7/2/08) and creating a paradigm shift or a 'Shift in Consciousness' as Joanna Macy puts it (see my post of 11/15/09). I know I've said that I want to look at that--it's just pretty daunting for me right now. Maybe at some point I'll be inspired to take that on.

Quote of the Day: "What makes it possible for cooperation to emerge is the fact that the players might meet again. This possibility means that the choices make today not only determine the outcome of this move, but can also influence the later choices of the players." - Robert Axelrod


Turil said...

From what I have observed, enabling, in the harmful pop psychology sense, only happens when you actively help someone get something that is harmful to them - something toxic, mentally or physically, such as addictive drugs, junk food, negative media, etc, or when you lie to them or others about what's going on. Punishment is, of course, also harmful as it harms someone either directly through violence or indirectly by depriving them of something they need to be healthy. Thus, enabling and punishment are equally damaging, as you already pointed out. But cooperating with others in general isn't the same as enabling, which is only cooperating with others to help them harm themselves.

Cooperation to help people heal themselves is exactly what our goal is. So the best path I've found is to offer people something they need - something that will help them heal and grow - while letting them be free to take your offer or not, so that they are allowed to grow and heal at the pace that is appropriate for them.

I'm not sure why this middle path of compassion (between the idiot compassion of helping people harm themselves and the punishment/violence of harming people directly) is not so easy to see, but after many years of looking, I've finally found it, and it's made relationships SOOOOO much easier! :-)

Also, thanks for the compliment and link! I've been really appreciating your blog as well, and like to use your ideas to bounce my ideas off of.

MoonRaven said...

Thanks so much for your comment, Turil. I love what you said: "Cooperation to help people heal themselves is exactly what our goal is." Wow. Perfect.

I would also add besides not helping someone harm themselves, we need to stop people from harming others--without being hurtful ourselves. I think this, too, is part of the middle path of compassion.