Daniel Kahneman: Adversarial Collaboration

There are different forms of adversarial collaboration. So one form is that you have people who disagree on some theoretical issue trying to conduct experiments that will resolve their differences or reduce their differences.

And that is the first step in adversarial collaboration: let’s see what we agree on and then let’s try to trade our differences. In some cases, they’re friends, so you can easily agree, although you agree that when you write up the results you will write them in two voices, so you do not commit yourself to agreeing on the interpretation of the result.

Sometimes you need an arbiter to run the experiment. That is when relations are more tense. Instead of the reply and rejoinder format, you can agree to write a joint article in which you first settle what you agree on and then what you disagree on.

You have to be willing not to win; that is, you have to be willing to accept a draw and to see that in the interest of science and civility and other things, a draw is better than a win. I happen to find a few people who are willing to do that.

By the way, it is not easy because people who think poorly of your work and of your ideas get on your nerves, and so you have to overcome that. The best of example of adversarial collaboration that I have had was on issues of the boundaries of intuition.

When is intuition marvelous, and when is it flawed? And it was with Gary Klein, who is a well-known author and who was a great proponent of expert intuition. And so we worked for six or seven years together trying to hammer out our differences about whether intuition is valid or not, and we came out with an article that was actually—well, it was a bit in two voices—but on most things we agreed.

And that was the best adversarial collaboration of all...