A Life of the Mind: Remembering Herb Simon BY DAVID KLAHR AND KENNETH KOTOVSKY

我現在才讀到這篇
寫許多他愛學生與教學的事


April 2001
A P S O B S E R V E R
Vol. 14 No. 4

Features

Psychological Scientists on Advisory Panels

The SAT I and California

I was the PSPI Canary

Remembering Herb Simon

Departments
Front Page
Presidential Column
Student Notebook
Department Profile
Teaching Tips
People
Members in the News
Announcements
Meeting Calendar Employment Ads



A Life of the Mind:
Remembering Herb Simon
BY DAVID KLAHR AND KENNETH KOTOVSKY

As we reported in last month's issue, Herbert A. Simon died on February 9, 2001 at the age of 84. The Observer invited Klahr and Kotovsky, two of Simon's former students and long-time colleagues in the Department of Psychology at Carnegie Mellon, to write a few words about not only the accomplishments, but also the personal and intellectual sides, of this remarkable individual.

Herb Simon's death prompted a remarkable outpouring of expressions of loss from around the world. They ranged from official condolences from The Chinese Academy of Sciences, to an informal "Quaker Meeting" in our Department at which his saddened and tearful colleagues shared their sense of loss, to the spontaneous postings of remembrances and eulogies on the Internet. Thus, it is with some trepidation that we take on this task, for we are fully aware that there are hundreds of others - students, collaborators, disciples, colleagues, and friends - who yearn to honor him by telling the world something about how he enriched their lives.

The official recognition of his contributions is nothing short of astonishing: Let's start with the Nobel Prize in Economic Sciences and the National Medal of Science and membership in the National Academy of Science. In addition, he received honorary degrees from two dozen universities around the world, as well as the top awards from APS, the American Psychological Association, the American Psychological Foundation, the Association for Computing Machinery, the American Economic Association, the American Society of Public Administration, the American Political Science Society, the Operations Research Society of America, and the Institute of Management Science.

A more complete record of Herb's accomplishments and awards is available online. His own description of his remarkable life can be found in his autobiography, Models of My Life (Simon, 1991), and a fascinating account of his seminal role in the "cognitive revolution" is provided in Pamela McCorduck's book Machines Who Think (McCorduck, 1979).

SIMON'S IMPACT ON COGNITIVE SCIENCE
Each of the disciplines listed above could-and do-claim Herb Simon as "theirs," because of the seminal ideas he contributed to their field, and the longevity of the influence of those ideas. Here we will only comment on what he did for, and to, cognitive science. Simply put, Herb changed our field at a time when it was badly in need of change.






Wrong again! Herb loved his teaching as much as his research, for he saw no difference between the two.




In the immediate aftermath of WWII, the engineering and mathematics community was developing sophisticated ways of analyzing information and control processes. In the nascent field of computer science, there was an emerging understanding of computability. In psychology, studies of vigilance and attention were producing theories of human information processing cast in terms of boxes and arrows. But all of this work was catapulted to a new level of rigor by the introduction of the kinds of computational models that Herb and his long-time colleague Allen Newell developed in the late 1950s.

Some of this work represents the foundational work in artificial intelligence (AI). For example, in 1956, Newell and Simon produced the first running AI program - The Logic Theorist - that proved many of the theorems of symbolic logic in Whitehead and Russell's Principia Mathematica. This first demonstration of computers doing symbolic "thought" or information processing was soon followed by an ever-burgeoning array of seminal work in AI that led to the physical symbol system hypothesis, which presented a generalized view of an intelligent system as any physical device - including the human mind - that could manipulate symbols.

Another strand of this work provided the foundation for modern computational models of human thought. These models have one important property that distinguishes them from all other types of theoretical statements: they independently execute the mental processes they represent. That is, rather than leaving it to the reader to interpret a verbal description of such processes as encoding an external stimulus or retrieving a memory or focusing attention, computational models actually do the encoding, searching, retrieving, and focusing. Consequently, the complex implications of multiple processes can be unambiguously derived.

When Herb's contribution of verbal protocol analysis as an empirical constraint on computational models was added to the mix, psychology had a vehicle for systematically and empirically investigating mental operations. These contributions helped turn our field back to the study of mind - a focus that had been largely lost in the behaviorist era - with a methodology that reflected the rigor of empirical methodology contributed by behaviorism. Herb Simon's contributions moved psychology into the mainstream of American science just as it was blossoming in the early post-war decades, and played a major role in the advent of cognitive science as we know it today.

A LIFE OF THE MIND
Rather than further elaborate on the public record of Herb's formal accomplishments, we will now attempt to convey our reply to many who have, in the past few weeks, asked us: "What was he like?" "What kind of a guy was he?" "Who was this man?"

These are harder to answer than might be expected. In some ways, we knew Herb well: He was advisor to each of us during our graduate careers four decades ago. Via different paths, we wound up as his colleagues at Carnegie Mellon for most of our careers. Our collaborative work with him brackets nearly the full gamut of his years and his interests in psychology, from the computer simulation of cognitive processes to the psychology of scientific discovery, and as colleagues we have spent hundreds of hours with him in scientific, administrative, and social discourse.

But to claim to know Herb Simon well is beyond any single person, except perhaps his loving wife Dorothea, for whom he was friend, colleague, scientific collaborator, and husband of 64 years. Whatever we say here will be an inadequate depiction of this remarkable person, but we hope that, at the least, our words and those of others we cite will provide a glimpse of him.

Herb lived a simple life. He walked to work from his home a mile from Carnegie Mellon. He hated air conditioning, refused to move his office into the renovated wings of our building, and for years after the dissemination of word-processors, continued to type his manuscripts on a manual typewriter. His home was warm and inviting but not in the least pretentious.

His life was a life of the mind. He inhabited his office for long hours on weekdays and weekends as well. Entering that office was an intellectual adventure. Whatever the topic was, you could be sure that you would engage a mind that was relentlessly seeking to understand some aspect of the world. It was a rare meeting that didn't involve Herb jumping out of his seat and pulling a book off the shelf to consult about some issue that came up. Following his curiosity was what his life was about - and it led to wondrous places.

As one of his former students wrote a few days after Herb's death:


Many graduate advisors in the sciences teach their students how to write. Mine taught me what to read and how to read it. My first and every conversation with Herb turned out to be the kind I had read about in C.P. Snow's novels about life at Cambridge: Civil conversations with a tutor who gave you a great deal of stuff to read, discuss and write about week after week, year after year. I was nudged to read Theory of Oscillations by Andronow and Chaikin ("Have you had any nonlinear differential equations?") and the Federalist Papers ("They were modeling and building a whole new society. Can you imagine what they would have done with computer simulations?").

Over the next three years, I was encouraged to read the classics in an astonishing range of subjects: General equilibrium theory in economics, real analysis, operations research, mathematical logic, the infant science of computer networks, cognitive psychology, and the philosophy of knowledge. I was made to read. Few human beings, including my classmates, have received the quantity and quality of educational attention I received in those years. I simply enjoyed myself, insufficiently realizing the responsibility that comes with such an education. (Ramamoorthi Bhaskar, personal communication)

While Bhaskar's remembrance is from the 1970s, Herb's level of energy, excitement and enthusiasm never waned. A current collaborator writes of his last interaction with Herb, just a few months ago. Remember, as you read it that Herb was 84 years old at the time:


Late one afternoon this past November, Herb and I were meeting to discuss how to present a very complex set of findings on human category learning. Our research group had been struggling to extract these findings from a massive body of qualitative and quantitative data, truly labyrinthine in its complexity. Herb had saved the day, of course, having invested possibly more than 80 hours alone in analysis, after our group had spent easily 10 times that without success. After he summarized the findings, we paused for moment as his presentation and my comprehension took all available capacity. He smiled and remarked with his eyes twinkling brightly that he "had not had as much fun in a long time."

I will forever cherish his words and these moments - and so many more. I feel immeasurably grateful to have been so incredibly privileged to work with this wonderful man. (Jim Staszewski, personal communication)

COSMOPOLITAN AND LOCAL
One of us once went to Herb for advice about how to make the difficult choice between two excellent opportunities for a beginning academic job. One offer was at a top-notch, well-established, but "conventional" university. The other was at an exciting, innovative, unconventional, and fledgling university. Herb said that, according to the then-current thinking in sociology, there were two kinds of academics: "cosmopolitans" and "locals." Under this dichotomy, cosmopolitans make their contributions and achieve their fame through their research, as well as through their activity in national and international professional organizations. Locals, by contrast, tend to be known for their contributions to their own university, through outstanding teaching, program and institutional development, high-level administrative service, and so on. As he put it, one simply had to choose the kind of career one wanted - cosmopolitan or local - and then pursue the consequences.


No matter how famous or how honored, he kept at his intellectual work throughout his long and productive career. He was the consumate intellectual and academic and never wavered from that no matter what formal adminstrative or external opportunities were proffered to him.


But here, at last, we reveal that even Herb Simon had a major flaw: He failed to take his own advice. The record at Carnegie Mellon is clear. During the same extended period over which he published nearly 1000 articles, he also exerted a profound, long-lasting, and intentional influence on his home institution. Herb was not only a University Professor - our highest academic honor - he was also a member of the Board of Trustees!

In the pursuit of his intellectual goals, Herb created, rearranged, sacked, supported, and integrated many organizational elements within the university. As a result, the list of departments that bear his irrevocable and indelible stamp include, in addition to the Department of Psychology, the Department of Philosophy, the Department of Social and Decision Sciences, the Graduate School of Industrial Administration, the Heinz School of Public Policy, the School of Computer Science, and, yes, even the Physics Department. Not only does each of these individual units bear the influence of Herb Simon's creative vision and administrative determination; the ethos and nature of the entire university also reflect his vision and commitment.

How did he manage to do this while he never was an official Head, or Dean, or Provost? First, he never left Pittsburgh. Not even to accept the oft-proffered invitation to spend a year at the Institute for Advance Study in the Behavioral Sciences at Stanford or accept a Chair at just about any place in the world that he fancied. Second, he believed that good science can produce effects in the world, and he did, after all, start his career in political science and management science. He understood how organizations worked, and he used that knowledge to unprecedented effect at Carnegie Mellon.

No, Herb never made the choice between cosmopolitan and local. He could do it all. And we mean "all." For one might expect that the superb scientist and the master politician would have to slack off somewhere, and that it perhaps would be in his teaching. Wrong again! Herb loved his teaching as much as his research, for he saw no difference between the two.

A LOVE OF TEACHING
Herb's teaching assistant for last semester's course on cognition and problem-solving related how Herb got so excited by the projects proposed by the wonderful group of graduates and undergraduates in the course (characteristically drawn from a wide variety of disciplines and programs) that he insisted on doing much of the TA's job himself - things like reading each revision of the students' progress reports on their term projects. This love of teaching characterized Herb's entire career. (Indeed, on that morning in 1978 when word of his Nobel Prize brought reporters and TV equipment cramming into his office, Herb glanced at his watch and suddenly terminated the interviews. The reporters protested, but Herb astonished them all as he continued his exit: "Gentlemen", he said with an impish grin, "I have to go teach. After all, that's what they pay me for.")

Another telling anecdote is his presentation to a group of first-year students in their dormitory. After the talk, he got down on the floor and the discussion went on and on as he and the students ate their "elegant" submarine sandwich dinners. When the host (one of the authors of this remembrance) tried to close things by inquiring if Herb was ready for his ride home, he said, "No, we're doing fine here. Why don't you run along. I'll just walk home when we're finished." There were, after all, minds to enrich and ideas to explore.

His consummate devotion to his teaching was also exemplified by the nurturing gentleness he demonstrated as he guided his students' intellectual development in weekly meetings. These discussions were often wide-ranging, as he shared his seemingly limitless intellect. He had the endearing (but disconcerting) habit of charitably assuming that the listener understood what he was talking about and could follow his thinking from discipline to discipline as he developed an idea.

Although Herb was gentle with his students, he had a passion for ideas that often led to intense debates. As the discussion heated up he would often, at such times, reach a point where he was about to deliver a potentially "lethal" blow to his opponents' arguments. When this happened, this ex-amateur boxer would soften the impact by the use of the word "friend", starting off his argument with his trademark "Look, friend..." as he went on to demolish the opposition. It was as if he was sensitive to the impact he could have by bringing to bear his massive intellect, and desired to assure the interlocutor that it was only the intellectual argument that was at issue and not the friendship. Nonetheless, it was a very intimidating front end - humans are after all, subject to classical conditioning, and the "Look, friend" was, to say the least, an ominous conditioned stimulus! (We never received the "Look, friend" warning while we were still students - it was reserved for those who could take it, and we remember vividly, that when it finally was first used with one of us it felt like our bar mitzvah - it marked the not entirely painless entry into adulthood!)

Ideas and the pursuit of knowledge drove this man. While he was not averse to attaining power and influence, he did so only in support of higher level goals-either to make a contribution to one of his many communities-this life-long "New-Dealer" had a strong social conscience and commitment-or to position himself to insure that the research environment would become and continue to be a good and supportive one. He never accepted a position that took him away from his research and teaching. His day to day and year to year and decade to decade behavior indicated that working on research, poring over data, and the wonderful sharing of his intellect in weekly research meetings in his office with his colleagues and students was his highest calling. No matter how famous or how honored, he kept at his intellectual work throughout his long and productive career. He was the consummate intellectual and academic and never wavered from that no matter what formal administrative or external opportunities were proffered to him.

Nearly 40 years ago, in a paper that attempted to predict the influence of automation on the shape of organizations, Herb reflected on the nature of human needs and satisfactions. He wrote: Man is a problem-solving, skill-using, social animal. Once he has satisfied his hunger, two main kinds of experiences are significant to him. One of his deepest needs is to apply his skills, whatever they be, to challenging tasks - to feel the exhilaration of the well-struck ball or the well-solved problem. The other need is to find meaningful and warm relations with a few other human beings - to love and be loved, to share experience, to respect and be respected, to work in common tasks. (Simon, 1965, p.110)

Herb is gone now. We are forever grateful to him not only for setting the "common tasks", but also for creating the context in which so many of us have found the exhilaration, meaning, respect, friendship, and love of which he spoke, but will speak no more. We will miss our teacher and colleague, but surely his teaching will go on in our lives. His intellect and integrity provide a model to live by. With each memory his lessons continue.

How can we honor him? Perhaps, as we enter the millennium of intelligent systems, it is worth considering something Herb wrote nearly 40 years ago:


"The definition of man's uniqueness has always formed the kernel of his cosmological and ethical systems. With Copernicus and Galileo, he ceased to be the species located at the center of the universe, attended by sun and stars. With Darwin, he ceased to be the species created and specially endowed by God with soul and reason. With Freud, he ceased to be the species whose behavior was - potentially - governable by rational mind. As we begin to produce mechanisms that think and learn, he has ceased to be the species uniquely capable of complex, intelligent manipulation of his environment."

"I am confident that man will, as he has in the past, find a new way of describing his place in the universe - a way that will satisfy his needs for dignity and for purpose. But it will be a way as different from the present one as was the Copernican from the Ptolemaic." (Simon, 1965)

Perhaps the time has come add another name to the Pantheon of those who have profoundly and irrevocably changed the way we understand ourselves: Herbert A. Simon.




REFERENCES

McCorduck, P. (1979) Machines who think. San Francisco, CA: W. H. Freeman
Simon, H. A. (1965). The shape of automation for men and management. New York: Harper & Row.
Simon, H.A. (1991). Models of my life. New York, NY: Basic Books.



ENDNOTES

1. For comments by Ryan Tweney, a historian of science, see: www.people.virginia.edu/~apk5t/
STweb/HerbSimon.html; from various members of the American Political Science Association, see http://www.apsanet.org/new/simon/.
2. See: www.psy.cmu.edu/psy/faculty/hsimon/hsimon.html

DAVID KLAHR is Professor of Psychology and director of the graduate program in psychology at Carnegie Mellon University, where he was Head of the Department of Psychology from 1983 -1993. His research interests include the development of scientific reasoning processes and the intersection of cognition and instruction. He is a Fellow and Charter Member of APS.

KENNETH KOTOVSKY is Professor of Psychology at Carnegie Mellon University. His major research focus is cognition, and in particular how people solve problems and what determines problem difficulty. He is a Charter Member of APS.