University of California: In Memoriam, 2001
John C. Harsanyi, Economics: Berkeley
| 1920-2000 | |
| Professor Emeritus |
John C. Harsanyi, Professor Emeritus in the Walter A. Haas School of Business and the Department of Economics at the University of California, Berkeley, and winner of the Alfred Nobel Memorial Prize in Economic Sciences in 1994, died on August 9, 2000 at the age of 80. He had served on the Berkeley faculty since 1964.
In Harsanyi's remarkable journey to eminence there was danger, courage, and a stubborn pursuit of elusive truth. His work has profoundly influenced our understanding of rationality in human affairs.
He was born in Budapest, Hungary on May 29, 1920, the son of Jewish parents, Charles Harsanyi, a pharmacist, and Alice Gombos Harsanyi. He graduated from the Lutheran High School in 1937, after winning first prize in a national mathematics competition. Harsanyi then pursued studies in pharmacy, which permitted him to evade forced labor in the early months of the Nazi occupation. When Nazi persecution peaked, he went into hiding (in November 1944) with the aid of Jesuits, until the arrival of the Russians in January 1945. After the war, Harsanyi joined his father as a pharmacist until it became clear to him that his interests and his future lay elsewhere.
Harsanyi enrolled in the University of Budapest to study philosophy, sociology, and psychology. He earned a doctoral degree in 1947 and started a teaching career. Harsanyi's views, however, differed sharply from those approved by the Communist regime. His principal professor and senior colleague, the sociologist Alexander Szalai, fearing that Harsanyi's honesty and outspokenness
His escape from Hungary in April 1950 was a joint venture with Anne Klauber, a psychology student who became Harsanyi's wife, and her parents. With a hired guide they made a three-night crossing of minefields and swamps, successfully reaching Austria. After six months in refugee camps in Austria, they made their way to Australia. In Sydney, Anne Harsanyi became a dressmaker while John held a varied series of odd jobs and enrolled in evening courses at the University of Sydney. He received an M.A. from the University of Sydney in 1953, writing a thesis entitled "The Research Policy of the Firm" (published in the Australian journal, Economic Record). Shortly after, Harsanyi produced two papers on utilitarian ethics, reopening a debate that many were content to view as settled, and asking whether comparing utilities between persons can lead to reasonable social choices if we introduce uncertainty. The papers were published in a major journal, The Journal of Political Economy, and have become classics. Although he moved on to other problems, utilitarian ethics continued to intrigue Harsanyi throughout his career, and near the end of his life, in a book project that sadly remains unfinished, he returned to the subject of his first major papers.
The publication of these papers led to a lectureship in economics at the University of Queensland in Brisbane. It was here that Harsanyi discovered a copy of von Neumann and Morgenstern's 1944 book, The Theory of Games and Economic Behavior and began his lifelong explorations into the theory of games. In 1956 he applied for a Rockefeller Foundation Fellowship, proposing to do graduate study at Stanford University.
He was awarded the fellowship and upon arriving at Stanford, enrolled as a regular student in the doctoral program. Kenneth Arrow, who became Harsanyi's mentor at Stanford (and a Nobel laureate 16 years later) has written that he had read the utilitarian-ethics papers, expected to meet an established scholar, and was very surprised when Harsanyi asked him to supervise a dissertation. "A little conversation established clearly enough that he was the fully developed scholar I had thought he was. I asked him why he was bothering to take a Ph.D.; it was unlikely that he would learn anything he didn't already know." Harsanyi's answer
After visiting semesters at Yale and at Carnegie, he returned to Australia as Senior Fellow at the Australian National University in Canberra. In 1961 he came back to the United States to join the Economics Department at Wayne State University. He visited the Haas School of Business at UC Berkeley briefly in 1963 and again for the first half of the 1964-65 academic year. In January 1965, he accepted a permanent appointment as Professor of Business Administration and later became Flood Professor and Professor in the Department of Economics. Just before his retirement in 1990 he received the Berkeley Citation, the highest honor awarded by the Berkeley campus to its faculty.
He was elected to the National Academy of Sciences and to the American Academy of Arts and Sciences. He received an honorary Doctor of Science degree from Northwestern University and was named a Distinguished Fellow of the American Economic Association.
At the time Harsanyi first encountered it, the theory of games was new and rich with promise for the understanding of economic, social, and political behavior--in markets, firms, legislatures, international negotiations, and elsewhere. But no one yet saw how to use it to build useful models that capture key aspects of complex real phenomena. One major failing was the assumption that all players have the same knowledge of the game that they are playing; each player knows the strategy sets of all the players as well as the payoff that every player receives for every choice of strategies that the players can make. That is highly unrealistic when (to cite just one example) the players are profit-seeking firms, since a firm's information about its costs is typically private and costs affect profits.
Repairing this defect is perhaps the best known of Harsanyi's many major contributions, which helped shape the theory of games that we know today. In three 1967-68 papers, he developed the essential concept used today in analyzing games of incomplete information. Put simply, the game starts with "nature" choosing each player's "type." All players know the probabilities of nature's possible choices, but each player's own type is known
Another basic innovation was Harsanyi's reinterpretation of mixed (randomized) strategies, developed by von Neumann and Morgenstern to resolve games that would otherwise have no equilibria. (At an equilibrium, each player's choice is best for that player, given the other players' choices.) Again realism raises a sharp challenge: neither generals nor managers toss coins to make their choices. Harsanyi showed that in a mixed-strategy equilibrium of the game, each player's mixture has a natural interpretation as a belief held by the opponent as to what the player will do. No player tosses a coin before choosing.
Harsanyi took major steps towards reconciling the cooperative theory (where any group of players can make an explicit side agreement) with the non-cooperative theory (where each player chooses independently of the others). He showed that several key concepts in the cooperative theory can be given a noncooperative interpretation.
Finally, Harsanyi faced the difficult question of choosing among equilibria. In a great many games, including games that arise in widely used models of competing firms, there are several equilibria and it is not at all clear which one is most compelling. We now have tools for selecting one of them, thanks to the joint work of John Harsanyi and Reinhart Selten, who (together with John Nash) shared the 1994 Nobel Prize. The Harsanyi-Selten collaboration began in earnest during several long visits that Selten made to Berkeley (one of them as a Visiting Professor in the Haas School of Business). It included a year that Harsanyi spent in Germany and concluded with a joint 1988 book, A General Theory of Equilibrium Selection in Games. But Harsanyi was still not satisfied. Persistent and creative to the last, he produced a completely new selection criterion near the end of his career. Some scholars prefer it to the criterion developed in the joint book, a preference shared, it appears, by Harsanyi himself.
The preceding sketch overlooks, needless to say, many other contributions that are found in his six books and 97 articles.
As a teacher at the Haas School of Business at Berkeley, Harsanyi had two main assignments: teaching Intermediate Microeconomics
John Harsanyi was unfailingly courteous and warm, gentle and witty. He is remembered at the Haas School of Business as a good citizen, who served willingly on committees, and performed the institutional roles essential for faculty governance.
He enthusiastically shared his many interests and his strongly held views on current political dramas, on American and European institutions and cultures, and on the lessons that he drew from his hard earlier experiences. As a scholar he listened to others. He never wavered from his own program of discovery, but was always open to the new ideas of others in pursuing it. He will be long remembered.
John Harsanyi is survived by his wife Anne, and by his son Tom, who teaches at Harvard University.
Courtesy of University Archives, The Bancroft Library, University of California at Berkeley, Berkeley, CA 94720-6000; http://bancroft.berkeley.edu/info
http://content.cdlib.org/view?docId=hb987008v1&brand=calisphere
Title: 2001, University of California: In Memoriam
By: University of California (System) Academic Senate, Author
Date: 2001
Contributing Institution: University Archives, The Bancroft Library, University of California at Berkeley, Berkeley, CA 94720-6000; http://bancroft.berkeley.edu/info
Copyright Note:
Transmission or reproduction of materials protected by copyright beyond that allowed by fair use requires the written permission of the copyright owners. Works not in the public domain cannot be commericially exploited without permission of the copyright owner. Responsibility for any use rests exclusively with the user
University of California Regents
Academic Senate-Berkeley Division, University of California, 320 Stephens Hall, Berkeley, CA 94720-5842