Oct 28 2008, 08:38:29 AM EDT in category [Instructor Posts: Design in the News]
One part of collaboration (not the only one) is trusting your co-conspirators. A classic puzzle that explores this is a game called the Prisoner's Dilemma.

Imagine that you and a partner have been caught committing a crime. There's not enough evidence to convict either of you of a major offense, so if you both stick to your alibis, you will both get a light sentence (say, six months). If you rat out your partner, and he says nothing, you will go free, and he will get ten years; vice versa if he rats you out. (If you rat each other out, you both go to jail for five years). It's possible to work together in this situation (by not betraying your partner), but it's more likely that you won't collaborate. It looks like a better deal for you to betray your partner, no matter what he does, because you will get less jail time.
That's an interesting game, and a good thing to remember if you find yourself on the wrong side of the law. There's a slightly more complicated version, however, that's more relevant to life on the outside: the Iterated Prisoner's Dilemma (IPD). The original game is a one-off: you and your partner play a single round, making a single choice to cheat or not. In the iterated version, you play over and over again, which gives you the chance to punish or reward your partner's past behavior. You have a chance of establishing a relationship and consequences. So what do you do now, cheat or play fair?
In his 1984 book, The Evolution of Cooperation, Robert Axelrod described one winning answer to that question. Axelrod launched a contest in which people entered computer programs that played IPD against each other. Each program played the part of a single prisoner in the dilemma. Mathematical psychologist Anatol Rapoport submitted a program with four lines of code that won the contest. He called it Tit-for-Tat, and its strategy was this: start out playing fair, and then do whatever your partner did last time.
Axelrod described Tit-for-Tat's success as follows:
- Be nice. Start out by collaborating with your partner. Tit-for-Tat's first move is always to play fair.
- Be strict. Don't reward your partner's bad behavior. If your partner cheats you, retaliate on the next turn.
- Be forgiving. If your partner later decides to collaborate, don't hold a grudge: play fair again.
- Don't be envious. Don't try to outscore your partner.
The flaw in this winning strategy is the danger of the retaliatory spiral. This is the stuff of wars and feuds. The solution is to inject a bit of random kindness into the mix. This is enough to prevent the death-trap.
Another set of 'rules' for negotiation and collaboration underscore this tough fairness. In collaboration, one works on a project and the relationship with fellow collaborators. In the best collaboration scenarios, one places a high value on both the relationship and the project. When one values the project more than the relationship, one becomes domineering and argumentative. When one values the relationship more than the project, one tends to over-accommodate the other's wishes. When one values neither the project nor the relationship, resignation and apathy result.
Collaboration = high value relationship + high value project
Domination = low value relationship + high value project
Accommodation = high value relationship + low value project
Resignation = low value relationship + low value project
Does this help explain past experiences you have had in collaboration and negotiation? Can you see patterns in your own or other's behavior? How can you work with someone you dislike or disagree with? Should we bother? What is the value in negotiation and collaboration?