Thomas Schelling is the Lucius N. Littauer Professor of Political Economy, Emeritus, Harvard University, and Distinguished University Professor, Emeritus, Department of Economics and School of Public Affairs, University of Maryland.
The Royal Swedish Academy of Sciences awarded the 2005 Nobel Prize in Economic Sciences to Robert J. Aumann and Thomas C. Schelling "for having enhanced our understanding of conflict and cooperation through game-theory analysis."
Quotes from Thomas Schelling’s April 2007 lecture at Trinity University:
I spent the better part of three months trying to convert pages into paragraphs, paragraphs into sentences, and finding simple examples that illustrated complex ideas, and in the end had a manuscript that, as far as I could tell, had sacrificed nothing in the compression to what was eventually about 30 printed pages. I called it “An Essay on Bargaining” and it appeared in the June, 1956 American Economic Review. Ever since then I’ve had, as a motto for writing, an observation of Antoine de St. Exupery, from Wind, Sand, and Stars. He was an early French aviator who pioneered airlines in North Africa, Chile, and Argentina. It was the design of aircraft that led him to the generalization, “In anything at all, perfection is finally achieved not when there is nothing more to add, but when nothing more can be taken away.” When I write I try to honor that maxim.
Let me digress here with a comment I’ve often enjoyed making. I begin by posing the question, what is the greatest invention in all of mathematics? I answer =. (The equals sign.) It supports the equation, which allows one to manipulate data in orderly fashion, preserving the equality. What is the greatest invention in all of business history? Double-entry bookkeeping (the equivalent of an equals sign). What is the greatest invention of macroeconomics? National accounts—quadruple entry bookkeeping. What is the greatest invention of game theory? The payoff matrix. These are all accounting frameworks. They allow an orderly presentation of quantitative data and a framework for manipulating the data. There would be no way to explore how many relations there could be between two parties without matrices, no way to identify all those situations in which it is an advantage to go first, or to go second, situations in which one party, or both parties, or neither, can usefully make a threat, or a promise, or when a threat will work only when coupled with a promise. I don’t expect real game theorists or mathematicians to agree with me, but I like the concept of an “accounting framework,” which the equation, or the matrix, or the double-entry or quadruple-entry system is.
In 1961 John J. McCloy was named White House Adviser on Disarmament. He set up eight committees; I was named chairman of one on “War by Accident, Miscalculation, and Surprise.” …. We made several recommendations, of which one was destined to become famous. It was here that I learned that there was no provision for any reliable contact between the President and the Kremlin. At a time when I could direct-dial my mother three thousand miles away to wish her happy birthday, Kennedy had no way to get in touch with Khrushchev. We recommended what later become known as the “hot line.” Henry Owen, a member of my committee, was in the State Department and resolved to make the hot line happen. Eventually he persuaded Soviet diplomats that the hot line was their idea, and so it came about. Finally, in 1963 Cyrillic alphabet teletypewriters were delivered to the State and Defense Departments with direct land-line connection to their counterparts in the Kremlin, and every day something like “the quick red fox jumped over the lazy dog” got transmitted in both directions to make sure everything was working.
Here I interject a brief note. In 1959 I was persuaded to write an article on the subject of “accidental war,” with the suggestion that I introduce it with a review of recent fictional accounts of nuclear war. I did, and reviewed On The Beach by Nevil Shute, Alas Babylon by Pat Frank, and Red Alert by Peter Bryant. I was so impressed by the latter as superior to any professional analysis I’d seen, that when Stanley Kubrik saw my article reprinted in The Observer of London he approached the author, whose real name was Peter George, and brought him to the States to do a screenplay. They spent an afternoon and an evening with me, trying to bring the story into the missile age—there were no missiles in the original Red Alert, published in 1958—and in the end Kubrik decided to do it as a “nightmare fantasy,” and called it Dr. Strangelove.
The question arises, was that book, Micromotives and Macrobehavior, game theory? If so, what makes it game theory? I didn’t think so at the time—it just didn’t occur to me—but there were hints in the Nobel Committee’s remarks that the Committee thought it was. And I see now that everything in that book had to do with individuals’ adjustment to, or anticipation of, the choices of others, and with needs and opportunities for coordination, formal cooperation, or submission to direction or regulation. …. I’ve now concluded that this is game theory. Game theory, I find (or choose to define), is very much about situations. What is “prisoners’ dilemma” but a situation? The logic of choice is central to game theory, but what game theory—again I’m thinking about the invention of matrices—is often critical for is identifying situations in which choice is somehow puzzling, problematic, or challenging, and in identifying how many distinct situations of a certain kind there may be.
Game theory is usually defined as concerned with how individuals should choose rationally in situations in which the optimal choice for each depends on the choice of another or the choices of others. I have been, I believe, more concerned with how to select a behavior that will influence another, or others--how to influence others by constraining their expectation of how oneself will act. This has led me to consider the institutional, legal, cultural, sometimes physical circumstances that make it possible to commit and reliably signal promises, threats, negotiating positions, and other kinds of commitments, and to appreciate when inability or disability, even deliberately incurred, may induce the favorable behavior of others. This is still game theory, but with a different slant.