Email Link

The Daniel Ellsberg interview

...continued from page 3

We were there for six months during the Suez Crisis. We evacuated all of the Americans from Alexandria at one point while the British and French were attacking Egypt. Then I came back to the Society of Fellows.

To answer your question, I had written this thesis on game theory. I wrote several articles based on that, and I remember I was correcting proofs by torchlight, flashlight and moonlight in a foxhole in Vieques during several months of maneuvers in the Marines.

When I was back in the Society of Fellows, I got interested in an off-shoot of game theory. I have been a critic of game theory, almost the first one, and one of the few up to this day to criticize the foundations of game theory. But that spun off into a field called decision theory, which deals with decisionmaking under uncertainty not only against an advisory, like game theory, but against nature, in effect. Decisions like, do we evacuate Houston in the face of uncertain warning of a hurricane coming along? A hurricane isn't a conscious adversary, except in the case of about a third of our country who can only understand that in terms of divine will. But just any kind of decision under uncertainty.

I wrote my PhD thesis at Rand (Corporation). I took six months off from my work at Rand and did a PhD theory on decisionmaking under uncertainty.

I concluded in that thesis that the reigning theory at that time, which was that people should act as if they assign precise probabilities to events; like say, it's 35 percent likely that the hurricane in Houston will make landfall on a certain day, or something like that I felt that was unrealistic, not only in terms of their expectations, but everybody admits they were more vague than that. They even act as if they assign post-probabilities to events, and I had a different theory of how that worked, which I called 'situation ambiguity,' which is where you don't know enough to assign even close probabilities to events.

When we initially spoke, we thought we would talk about the Pentagon Papers, of course that was before the presidential election and the completion of your book "The Doomsday Machine." In light of that, we wanted to ask you how from your own life the importance of the Pentagon Papers and your work as a defense analyst at the Rand Corporation led you to write a book about the country's nuclear war strategy from the 1960s to today, what that strategy involved then and where you believe we are today?

I went to Rand one summer when I was in the Society of Fellows. It was sort of the Vatican of decision theory. So, I was drawn to Rand not because of their defense work for the Air Force, but because their mathematics department, in particular, and their economics department, had done a lot of work in decision theory and game theory. I went as an economist, which was my field at that point. The economics department was working on nuclear strategy a great deal for historical reasons I won't go into.

...continued on page 5
Pages 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15


Comments ()