brianchitester.comPosts

Introduction

Herbert Alexander Simon (1916–2001) is one of the strangest cases in twentieth-century intellectual history: a single person who made first-rank contributions to economics, political science, public administration, cognitive psychology, computer science, artificial intelligence, and philosophy of science — and who did most of it from the same office, in the same university, with the same wife, over more than a half-century. He won the Turing Award in 1975 (with Allen Newell, for AI) and the Nobel Memorial Prize in Economic Sciences in 1978 (for bounded rationality), making him one of the very few people ever to claim both the highest honor in computing and the highest honor in economics. Along the way he co-founded the field of artificial intelligence, helped invent cognitive science, built the modern theory of organizational decision-making, and wrote a foundational book — The Sciences of the Artificial — about how to do science about designed things rather than natural ones.

Where John von Neumann was a chaotic, party-throwing prodigy who burned out at fifty-three, Simon is the anti-von Neumann: an orderly, married-for-sixty-three-years, lived-in-the-same-house Midwesterner who treated his life as one of his own designs. He even said so explicitly. His autobiography is titled Models of My Life, and the metaphor he chose for his own existence was the maze — a search problem in a branching space, navigated with bounded rationality and satisficing heuristics. This profile examines how that orderly life worked: the daily structure, the partnerships, the marriage, the hobbies, the ruthless prioritization, and the costs.

Time Management & Workflow

Simon’s most distinctive trait — and the one most useful to a reader trying to learn from him — was a kind of quiet, monastic discipline applied to an enormous intellectual estate. He was at his desk early, he stayed there long, and he had a famously direct way of describing the underlying logic. In Models of My Life he wrote: “Time is the tyrant. One cannot be loyal to two occupations any more than one can to two lovers. Whenever I found that one of my hobbies was seriously taking attention from my research, I dropped it.” That sentence does a lot of work. It tells you both that Simon had hobbies — many of them — and that he was willing to kill any of them on contact if they began to compete with the central project. This is the inverse of the fashionable contemporary idea that polymathy means juggling. For Simon, polymathy meant sequencing: full attention to one thing at a time, but with the willingness to swap things in and out as the work demanded.

His workflow at Carnegie Mellon (where he taught from 1949 until his death in 2001  — fifty-two consecutive years on one campus) was built around long, uninterrupted blocks of writing and thinking, punctuated by deep collaborations with a small set of intimate co-workers. He produced an output that is hard to fathom in modern terms: by the end of his life he had authored 27 books and nearly 1,000 papers, making him one of the most cited social scientists of the twentieth century . To put that in perspective, a productive academic might write thirty or forty papers in a career. Simon wrote a thousand — across half a dozen disciplines — while also helping found three departments at his university, raising three children, learning roughly twenty languages well enough to read in them, climbing mountains, and playing competitive chess.

The mechanism that made this possible was not polyphasic sleep or three-watches-on-his-wrist eccentricity (à la Buckminster Fuller); it was simply decades of compounding focus. Simon believed that expertise was a matter of pattern recognition built up over thousands of hours, and he lived as if he believed it. He worked steadily, almost every day, on whatever the current mazes were. His School of Computer Science  office at Carnegie Mellon was famously open — students and colleagues drifted in for impromptu conversations — and much of his thinking happened in dialogue rather than in solitude. But the dialogues were structured around real problems, not socializing. He treated his time at work the way a careful gardener treats soil: as something to be steadily, undramatically improved.

He also worked fast. Like von Neumann, Simon had the kind of mind that compressed long problems into short ones. He famously claimed that expert intuition is just recognition — the chess master sees a board the way you see a familiar face — and his own intuition, after decades of compounding, made his deliberation cycles short. He could read a paper, see what was wrong with it, and dictate a critique faster than most people could summarize the abstract. The combination of speed with steadiness was the engine.

Daily Life Practices & Rituals

Simon’s daily rituals were, for the most part, the rituals of a deeply ordinary American academic of his generation — and that ordinariness is itself worth noticing. He had no Dymaxion sleep schedule, no caveman diet, no tensegrity-model laboratory in his bedroom. He ate normal meals at normal times, slept conventional hours, walked to campus, sat at his desk, and went home. The radical thing was the consistency. For more than fifty years he kept showing up.

He did, however, have a few practices that gave the days texture. Long mountain walks were one of them. Simon was a passionate hiker and amateur mountaineer; he and Dorothea camped, canoed, and hiked extensively in Wisconsin and the Sierras when they were young, and he carried a love of the outdoors throughout his life. Hiking served the same function for him that pacing served for Aristotle or that walking served for Nietzsche: a way to think while moving. He noted in Models of My Life that some of his best ideas surfaced on long, slow ascents.

He played classical piano — seriously enough that it counted as an art rather than a diversion. He played chess competitively, and his interest in chess was not casual: it became one of his most important research domains, since chess for Simon was a controlled laboratory for the study of human cognition. The Newell-Simon work on chess programs in the 1950s and 60s, and his later collaborations with the Dutch psychologist Adriaan de Groot on chess expertise, produced some of the foundational findings of cognitive science — the famous result that masters store roughly 50,000 chunks of board patterns, recognized at a glance. So when Simon played chess in the evening, it was not exactly leisure; it was the same project, in a different mode.

He collected languages the way other people collect stamps. He set himself the rule, only half-jokingly, that he would learn the language of any country he visited well enough to read its newspaper. By the end of his life he had working reading knowledge of roughly twenty languages, ancient and modern. His autobiography  describes a man for whom learning was not effortful so much as constant. He picked up Mandarin late in life when he began collaborating with Chinese cognitive scientists; he read Russian in order to follow Soviet psychology; he read German because his father had brought it from Darmstadt.

He was also, more obscurely, an amateur entomologist — specifically a beetle enthusiast — and an amateur painter. None of these were performance hobbies. They were forms of attention. The unifying thread, whether the object was a beetle or a chess board or a Mandarin character, was Simon’s belief that the mind is a pattern-recognition device whose proper diet is interesting structure. He fed it constantly.

The one ritual he did not practice was idleness. There is no record of Simon meditating, no record of him taking long sabbaticals to “decompress,” no record of his ever going on a vacation that did not also involve research. The cost of this is real and we will return to it. But the daily texture of his life was the texture of a man who had decided, decades earlier, that there was nothing more interesting than work.

Domains of Pursuit

What separates Simon from a merely productive scholar is that his thousand papers were not in one field. They were spread across a half-dozen, each of which would, in another person, have constituted a full career. Below is a partial map.

The thing to notice is that these domains are not unrelated. Simon worked them as a single problem with many faces. The unifying question — how do real, finite, biological minds and organizations make decisions in a complex world? — drove all of it. The chess studies fed the AI work, the AI work fed the cognitive psychology, the cognitive psychology fed the economics, the economics fed the public administration, and the philosophy of science gave the whole structure a name. He was a polymath the way Darwin was: not by hopping between unrelated curiosities but by taking one big question and pursuing it down every corridor it opened.

Employment & Economic Model

Simon’s economic model was as orderly as his daily routine: a lifelong tenured professorship at a single university, with the security and the autonomy that arrangement produced. After studying political science at the University of Chicago (BA 1936, PhD 1943) and brief stints at Berkeley and the Illinois Institute of Technology, he joined the faculty of what was then the Carnegie Institute of Technology in 1949 — and stayed for fifty-two years until his death. He was the Richard King Mellon University Professor of Computer Science and Psychology from 1966 onward, a chair he held for the last thirty-five years of his life.

This is not the story of a polymath who lived on the financial margins to preserve creative freedom (Fuller) or who made a fortune in finance and then walked away (Taleb) or who funded chaos with government money (von Neumann). It is the story of a man who used the classical American research-university model with maximal effectiveness. A salaried position at a serious university — one with strong colleagues, good students, modest teaching loads, and almost unlimited intellectual freedom — turned out to be sufficient infrastructure for nearly any project Simon wanted to pursue.

He was also a heavy beneficiary of the postwar U.S. funding ecosystem. He consulted for the RAND Corporation in the 1950s, where the Logic Theorist work was done. He held research grants from the Office of Naval Research, the National Science Foundation, the Carnegie Corporation, and various other public and private bodies. These grants funded graduate students, computer time (which was extraordinarily expensive in the 1950s and 60s), travel, and the ability to assemble small interdisciplinary teams around specific questions. Simon was an effective grant-writer in part because he was unusually clear about what he was trying to do.

He also took an active role in building institutions, which is its own kind of economic strategy. He helped found the Graduate School of Industrial Administration at Carnegie Tech (now the Tepper School of Business), the School of Computer Science, and the cognitive science core of the Psychology Department. Each of these became internationally important. By the time he died, the institutions he had built were sustaining the next generation of work — a kind of compound interest in scholarship.

He did not become rich. Academic salaries in the 1950s and 60s were modest by professional standards, and Simon never started a company, never took outside directorships of consequence, never wrote a popular book aimed at the bestseller list. The Nobel money, which arrived in 1978, was not transformative. By all accounts he and Dorothea lived modestly in the same Pittsburgh house for decades. The arrangement gave him what he needed: time, autonomy, colleagues, students, and a stable place to put his attention.

Family & Personal Relationships

Simon married Dorothea Isabel Pye on December 25, 1937 — Christmas Day — and they remained married for sixty-three years until his death. Dorothea outlived him by a year, dying in 2002. They had three children: Katherine, Peter, and Barbara. By any conventional measure, the marriage was a success: long, devoted, intellectually shared, and warmly described in his autobiography. Dorothea earned a master’s in political science from the University of Chicago, did her own research at Carnegie Mellon and the University of Pittsburgh, and co-published with Herbert in two distinct fields  — public administration and cognitive psychology. The papers under “Simon and Simon” are not a vanity project; they are real collaborative work.

That said, an honest accounting requires noticing what such a productive life cost the family. Simon was a workaholic in the most literal sense. He was at his desk almost every day. The famous quote — “Whenever I found that one of my hobbies was seriously taking attention from my research, I dropped it” — implicitly applies to a great deal more than hobbies. Models of My Life candidly discusses an unconsummated love affair, a long stretch of marital strain, and Simon’s awareness that his children sometimes felt like background characters in a life dominated by work. He included in the book letters he had written to his children, which is touching but also, read between the lines, slightly defensive. He was not absent in the literal sense; he was present and devoted. But he was also, always, working.

Dorothea seems to have been a genuinely equal intellectual partner, which made the arrangement workable in a way it otherwise might not have been. She did not have to wait for him to come home from the office to have an interesting conversation; the office and the home were continuous with each other. She also handled, as was conventional for her generation, the bulk of household and child-rearing labor that allowed Simon to keep his fifty-year focus. The marriage was a system, in something close to Simon’s own technical sense of that word — a set of mutually supporting parts that produced an output neither component could have produced alone.

His other most important personal relationship was professional: his fifty-year partnership with Allen Newell, beginning at RAND in 1955 and continuing until Newell’s death in 1992. Simon repeatedly said that the work he did with Newell was the best work of his life. They met when Newell was a young researcher at RAND and Simon was an established economist visiting in the summers; within months they had decided to try to make a computer think. The collaboration produced the Logic Theorist, GPS, Soar , the unified theory of cognition, and the Turing Award. Reading their joint papers, it is often impossible to tell which sentences are Simon’s and which are Newell’s — a sign of one of those rare partnerships in which two minds genuinely fuse on a problem. When Newell died of cancer at sixty-five, Simon delivered the eulogy, and it is by far the most emotionally exposed thing he ever wrote.

Philosophy of Life

Simon’s philosophy of life rested on four ideas that, taken together, explain almost everything about how he lived.

The first was bounded rationality. The mind is finite. Information is costly. The world is complex. Therefore, every real decision is made under conditions of profound ignorance, and the appropriate response is not to pretend otherwise but to design good heuristics — rules of thumb that work well enough, often enough, to be worth following. This is not a counsel of despair; it is a counsel of humility. Simon really did believe that his own life should be lived this way. He picked a small number of important questions, designed his daily structure around them, and stopped second-guessing.

The second was satisficing. There is no globally optimal life. There is only the next decision, evaluated against criteria, accepted when good enough. Simon famously said that the rational decision-maker does not choose the best option in some absolute sense; she chooses the first option that crosses her aspiration threshold. He lived this. He did not agonize over alternative careers. He stayed at one university because the university was good enough — and because the cost of switching was higher than the marginal benefit. He stayed in one marriage for the same kind of reason. He treated his time as a finite resource to be allocated, not as a problem to be optimized.

The third was a deep commitment to logical positivism and the empirical method. As an undergraduate at Chicago in the late 1930s, Simon adopted the worldview of Rudolf Carnap and the Vienna Circle  (via A. J. Ayer’s Language, Truth and Logic), and he never substantially revised it. A statement was meaningful only insofar as it was empirically testable; metaphysics was largely confusion; the proper task of intellectual work was to build operational models of phenomena and check them against data. This is why he liked computer simulations so much: a working program is, in a sense, a forced operationalization. You cannot wave your hands inside a computer.

The fourth was the conviction — articulated most fully in The Sciences of the Artificial — that designed things deserve their own science. Most of the world that humans inhabit is artificial: cities, organizations, languages, algorithms, institutions. These things follow their own kind of regularity, distinct from the regularities of physics and biology. A science of the artificial is normative as well as descriptive: it studies not only how things are but how they might be, given goals and constraints. This was, in a quiet way, Simon’s deepest philosophical commitment. He thought of his own life as an artifact — something he was designing, with bounded rationality, against criteria. The autobiography is called Models of My Life because that is literally what he believed he was building.

There is something striking about how modest this philosophy is. Simon did not claim to know what the good life was. He claimed only that, given finite cognition and an interesting world, the best one could do was pick a small number of mazes and explore them carefully.

Tools, Environment & Infrastructure

Simon’s tools were unglamorous. He was a typewriter-and-yellow-legal-pad person for most of his career, switching to word processors only when they became reliable. His office at Carnegie Mellon was famous for being packed with books and papers , with a chalkboard for working through arguments, and a steady stream of students and collaborators passing through. The CMU Archives now hold his papers — roughly 172 linear feet of material — which gives a sense of the documentary footprint of his thinking.

The most important tool he built, with Newell and Shaw, was the computer as an instrument for studying mind. The early AI programs — Logic Theorist, GPS, the chess programs — were not just engineering achievements. They were theoretical instruments, in the same sense that a particle accelerator is a theoretical instrument. By forcing a theory of cognition to run as a program, you forced it to be specific enough to be wrong. This was Simon’s deepest methodological move: making theories run.

The most important environment he built was Carnegie Mellon itself. The Graduate School of Industrial Administration, the School of Computer Science, the cognitive science core of the Psychology Department — these are infrastructure. They are the labs in which his ideas could be developed by the next generation. Simon understood, in a way many brilliant individualists do not, that a good university department is a kind of long-running experiment whose output lasts longer than any single career. Much of his time, especially in the 1950s and 60s, was spent in committee work, hiring, curriculum design — institutional plumbing. He thought it was worth it because it was, in his sense, design.

He was also an early and committed user of computers themselves. The IBM 704 at RAND, on which the Logic Theorist ran, was effectively Simon’s first major laboratory instrument. He spent decades after that working with successive generations of mainframes, then minicomputers, then workstations. By the end of his life he was using personal computers like everyone else, but he had been thinking about cognition-as-computation since the year ENIAC was unveiled, which gives a sense of how early he was.

Tradeoffs & Costs

The honest accounting of Simon’s life has to include what the work cost. He was, by his own quiet admission, a workaholic in a generation that did not yet have the word. He was at his desk most days. His hobbies — chess, languages, hiking, piano, painting, beetle-collecting — were real but, as he said, expendable. If a hobby began to compete with the research, the hobby went.

The clearest cost was family time. Simon’s three children grew up with a father who was loving but very, very busy. Models of My Life is unusually candid about this. He includes letters he wrote to his children and reflections on what he wishes he had done differently. He does not pretend the trade-off was costless. The marriage to Dorothea survived and flourished, but it survived in part because Dorothea was herself a researcher who shared the intellectual life. A more conventional marriage might not have absorbed the strain. The autobiography also discusses a long-running, unconsummated emotional attachment to a colleague — one of the few places where the orderly Simon admits that the orderliness had pressure points.

A second cost was breadth-versus-depth in the conventional sense. Simon spread himself across so many fields that some specialists in each have argued, occasionally, that he was insufficiently deep. The pure mathematicians did not always take his AI work as seriously as the AI community did; the neoclassical economists never fully accepted bounded rationality as a replacement for the rational-actor model rather than a footnote to it; some experimental psychologists felt the symbol-system hypothesis ignored the messiness of biological cognition. Simon was aware of all of this. He thought the cost was worth paying because the integration across fields was where the interesting work was. But it is true that, had he stayed in any single field, he might have built an even larger monument inside it.

A third cost, harder to see but real, was the quietness of the public legacy. Simon is genuinely famous within academia and almost unknown outside it. He never courted public attention, never wrote a trade book, never gave the kind of TED-talk-shaped speeches that build modern intellectual celebrity. He was content with the slow compounding of citations and the building of institutions. The trade-off there was that ideas like bounded rationality and satisficing — concepts that are arguably as important as anything in twentieth-century social science — entered popular culture only later, and often via second-generation popularizers (Kahneman, Thaler) who got the public credit. Simon does not seem to have minded. He had what he wanted: the work, the partnership with Newell, the marriage, the university, the mazes.

A fourth cost was physical. Simon was not a man who tended his body the way Taleb tends his. He worked long hours sitting down, and he carried, by mid-life, the mild posture and softness of a career academic. The hiking and mountain-climbing of his younger years tapered off. He died at eighty-four, which is a long good run, but one suspects a more athletic version of the same life might have given him five or ten more years to keep working.

He was, in his own bounded-rational way, aware of all of these trade-offs and willing to make them. He had picked his mazes.

Legacy & Influence

Simon’s legacy is difficult to summarize because he is one of those rare figures whose ideas have become so absorbed into the intellectual atmosphere that we no longer notice them. Bounded rationality is now the default framework in behavioral economics, cognitive psychology, public policy, organizational theory, and large parts of computer science. Satisficing is in every undergraduate decision-theory textbook. The physical symbol system hypothesis — Newell and Simon’s claim that cognition is symbol manipulation — has been challenged by connectionists and embodied-cognition theorists, but it set the terms of the debate, and it remains one of the few candidate theories of mind that is detailed enough to be either right or wrong.

In artificial intelligence, Simon and Newell are foundational figures. The Logic Theorist is generally regarded as the first true AI program . The General Problem Solver introduced means-ends analysis. The chess programs and protocol-analysis methods set the agenda for symbolic AI for decades. The current deep-learning era is a swing away from symbolic methods and toward statistical ones, but the questions of AI — what is intelligence, can a machine have it, how would we know — were largely framed by Simon and his collaborators in the 1950s. The recent return of neuro-symbolic systems is, in effect, a partial return to ground Simon and Newell broke seventy years ago.

In economics, Simon’s most direct heirs are the behavioral economists — Daniel Kahneman, Amos Tversky, Richard Thaler, Cass Sunstein — whose work won at least three subsequent Nobel Prizes and reshaped policy worldwide. The whole modern apparatus of “nudges,” default-setting, choice architecture, and prospect theory traces back to Simon’s insistence that real human decision-making must be studied empirically rather than assumed.

In cognitive science, he is one of the founders. The annual Herbert A. Simon Lecture  at the Cognitive Science Society is named for him. The whole methodology of computational modeling of cognition — building a program, running it, comparing its behavior to human data — descends from the Newell-Simon program.

His institutional legacy at Carnegie Mellon is enormous. The School of Computer Science at CMU is now one of the top three or four in the world; the Tepper School is one of the top business schools; the cognitive science work in the Psychology Department continues to be internationally important. The Simon Initiative at CMU continues his work on evidence-based learning science. He is, in a literal sense, still hiring people forty years after his most active period. This is what it looks like to build institutions rather than just careers.

And finally, Simon’s life as a model — to use his own metaphor — is itself a legacy. He showed that one could be a polymath without being chaotic; that one could spend fifty years in one office and still touch a half-dozen disciplines; that the answer to “how do you do so much?” is not a productivity hack but a few decades of consistent attention to a small number of well-chosen questions. He believed that a life, like an organization, is something designed under bounded rationality — and he designed his with unusual clarity about what he wanted it to do. Most of us could do worse than to take that as the lesson.

2026 © Brian Chitester.