Gale Tales

Abstract

My first encounter with David Gale was indirect. My seatmate, (in 1966, I believe), in George Dantzig’s linear programming class at the University of California Berkeley was Richard Sutherland, a graduate student of mathematics that David brought with him when he moved to Berkeley from Brown University. Sutherland was always scratching around on the most interesting looking mathematics while Dantzig was lecturing on linear programming. I suspected that Sutherland had already known much of the material, and was perhaps just auditing the class and would pay attention only when Dantzig was saying something that Sutherland didn’t already know. Actually I found sitting next to Sutherland rather intimidating since I had to pay attention to Dantzig in order to grasp the material. While Dantzig lectured on, I asked Dick one day: ‘What is that stuff you’re working on?’ He replied, ‘Multi-sectoral growth theory. I’m trying to prove that the optimum solution to this infinite horizon convex programming problem, which generates a dynamic system on Rⁿ has a steady state and converges to that steady state as T → ∞’. I didn’t understand what he said at the time but I was fascinated. I whispered, ‘Who, on the UC mathematics faculty, does that sort of thing?’ Dick replied, ‘David Gale’. ‘Is David Gale here? I thought he was at Brown. That’s great news. I’d like to work on the kind of stuff you’re doing’, I said. As Dantzig looked at us rather sternly, Sutherland wrote on his tablet: ‘Gale just moved here. Go talk to him about becoming one of his Ph.D. students’. Since, at the time, I was only a first year graduate student in mathematics at the University of California Berkeley I didn’t have the nerve to go and talk to the Great Gale.