The History of Mr. Papert
By Martin Boyle
Originally appeared in The Logo Exchange

Abstract This article outlines the life and career of Seymour A. Papert through the mid-1990s. We follow the development of Papert from his early formal and informal education in the South Africa of the 1930's through to his sage era in The Children's Machine of the 1990's. On the way we will trace the formative influence of Papert's work with Jean Piaget in Geneva, his first serious collaborative work with Marvin Minsky, leading to the bombshell, Perceptrons, through to the success of his greatest educational achievement - the development of the computer programming language Logo. We will see how, in the later part of his academic career, Papert has drawn out the constructivist principles of Piaget into his own constructionism at work in the classroom. Anecdote spices the life of Papert with real humor and unexpected actions will give insightful glimpses into the workings of the great man's mind.

illustration by Peter Reynolds

Early Life

Seymour A. Papert was born on March 1, 1928 in Pretoria, South Africa. The man himself seems shy or unwilling to divulge many details of his early life, though in Mindstorms he owns to an interest in gears from early childhood experiences, indeed:

Before I was two years old I had developed an intense involvement with automobiles. (Papert, 1980 p vi)

Had the world then developed as it has now such precocity may even have enabled him to enjoy a childhood sponsored by Henry Ford!

Papert further reveals a love for Daisy who left me with an eleventh commandment: "Thou shalt invent three theories every day before breakfast and throw them away before dinner." (Papert 1993b p 58)

Such is the eclecticism of Papert's subsequent endeavors that one is inclined to accept that he fell, hook with the lot, for that beloved teacher's advice.

Papert's childhood gives every appearance of being outside the norm. His father was an entomologist who spent the best part of Seymour's early childhood roaming the east coast of southern Africa in pursuit of the tsetse fly; a life-style which required all members of the family to turn their hands in whatever direction was necessary. For a young boy, this must have been more than just compensation for an eccentric upbringing.

The Papert family's way of life was straight out of a Hemingway story. (Crevier 1993 p 84)

The story tells of following bush trails, hunting for food and falling in love with the transmission differential of broken-down trucks - strong formative experience for the young mind.

The Paperts were the only white people in the area. This led the young Seymour into trouble when he had to eventually attend school in Johannesburg. Unfamiliar with the strict and convoluted edicts of the political and social consequences of apartheid the ten-year old organized evening lessons for the illiterate black domestic servants of his area. He soon found himself in serious trouble with the authorities for such illegal activities; the consequences were to have interesting future reverberations:

This was just the first of Papert's anti-apartheid activities, activities that would later lead the United States immigration authorities to deny him a visa for many months. (Crevier 1993 p 84)

The logic of the situation was lost on the young boy for

Adults justified their reluctance to let blacks sit at school desks by citing fear of contagious disease. But, reflected Papert, these are the same servants who take care of babies and cook the food in the whites' homes. How can the ruling class think like that? (Crevier 1993 p 85)

Such interest in matters logical and illogical soon earned the talented boy an invitation to attend seminars in philosophy at the University of Johannesburg!

Early career

Before long Papert was a philosophy student in his own right at the University of Witwatersrand where he was awarded his BA in 1949. After this his interests switched to mathematics, leading to a doctoral degree from Witwatersrand in 1952.

In order to widen his horizons Papert then moved overseas. He was awarded a Commonwealth research scholarship to St. John's College, Cambridge, U.K. This eventually enabled him to complete a second doctorate. While at the National Physical Laboratory outside London, Papert ran into one of America’s foremost workers in the emerging field of Artificial Intelligence - Ed Feigenbaum. Feigenbaum was a Fulbright Fellow for the year, and he had a memorable friendship with a young and somewhat eccentric South African scholar named Seymour Papert. (McCorduck, 1979 p 275)

More significantly for the future, at a symposium in London itself, he first met Marvin Minsky. The encounter with Minsky was to prove the genesis of the second great collaborative effort of Papert's professional life; but first he turned his face towards France.

Papert spent the 1956/57 year as a researcher at the Henri Poincare Institute at the University of Paris, to complete the research for his doctorate. Soon after, an opportunity opened and Papert was to spend a fascinating episode of his life as a researcher working under Jean Piaget at the International Centre of Genetic Epistemology, at the University of Geneva.

...the Parisian discovery that had the biggest impact on my life was Jean Piaget, who at that time was giving a course at the Sorbonne. I got to know him and was invited to work in his center in Geneva, where I spent the next four years and became passionately interested in children's thinking. (Papert, 1993b p 33)

Although Papert was to deviate from the gospel of the master in terms of the rigidity of the stages of child cognitive development,the Piagetian influence pervades the remainder of his work. Piaget, fundamentally and overtly, contributing to such diverse concepts as the establishment and efficacy of computer-rich microworlds through to the layering development of societies of mind

Allow Papert, with metaphor as ever his telling servant, make his own eloquent point:

When Piaget is poured into a new decade, much will change. Whether one has conservation of Piaget will depend on what one perceives as most important in the thinking of the great master. My own view is that the essential aspects of his work have not fallen by the wayside. On the contrary they are stronger and more relevant than ever. (Papert, 1988 p 3)

And later:

In my Piaget, stages and even most stages of "active learning" are quite secondary. I focus instead on his constructivism and structuralism. (Papert, 1988 p 4) 

The growing influence, within the constructionist movement in education of Microworlds, StarLogo, Lego robots and magic bricks, a decade on, strengthen that assessment of Piaget's worth.

The artificial intelligentsia and perceptron sagas

Following the four years in Geneva, I became a professor of mathematics at MIT. Many factors made the move attractive. There was the prospect of access to computers and of working with Marvin Minsky and Warren McCulloch, as well as a wonderful sense of playfulness that I had experienced there on brief visits. When I finally arrived, all this came together in all-night sessions around a PDP-1 computer that had been given to Minsky. It was pure play. We were finding out what could be done with a computer, and anything interesting was worthwhile. Nobody yet knew enough to decree that some things were more serious than others. We were like infants discovering the world. (Papert, 1993b p 33)

One wonders how many of us have shared that heady experience with Papert when computers are new, computers are an unknown quantity? With computers anything goes!

In 1958 Marvin Minsky and John McCarthy founded the Artificial Intelligence Group at MIT. Minsky had not forgotten the young Papert who had made such an impression on him at their London meeting.

Papert remembers his first meeting with Minsky in a recent London speech.

…for it was at the 1960 London Symposium on Information Theory organized by Colin Cherry that an event happened which changed my career path and made me follow the course that brought me here. I came to that meeting as a mathematician interested in computational ideas and Information Theory. I came there with a paper in which I had a little theorem. And what happened was the worst nightmare of somebody coming to a meeting with a theorem.

The speaker before me announced exactly the same theorem and proved it at least as well as I did, not quite the same, but you can’t get much credit for just having a slightly different proof. Now that could have been a nightmare; in fact, it turned into a great gift. That person was Marvin Minsky. Marvin and I came to that meeting with essentially the same paper and this led to a collaboration that continued for many years and is responsible for almost everything we did in the next decade and has certainly colored everything I have done since then. (Papert, 1998)

After the previously mentioned visa problems were sorted out with the US immigration authorities Papert traveled to MIT. He strode into Minsky's office, sat down, and they were away - never to look back!

Rarely had co-operation between two researchers been so productive: colleagues no longer said "Minsky-and-McCarthy," but "Minsky-and-Papert." The two soon initiated new research programs in the theory of computation, robotics, human perception and child psychology. When the Artificial Intelligence Group formally became the MIT AI Laboratory in 1968, Minsky and Papert acted as co-directors. (Crevier, 1993 p 86) 

It was an exhilarating time in AI. Sane men and women with PhDs from highly respected seats of learning were claiming that the office thermostat was intelligent! Perhaps with hindsight and the passage of time they deserve to retain their now found anonymity!

Alan Newell and Herb Simon back in 1956 had produced software which could churn out proofs of theorems from Russell and Whitehead's Principia Mathematica

In fact, the Logical Theorist discovered a shorter and more satisfying proof to Theorem 2.85 than Whitehead and Russell had used. Simon wrote this news to Lord Russell, who responded with delight. However the Journal Of Symbolic Logic declined to publish an article co-authored by the Logical Theorist describing this proof. (McCorduck, 1979 p 142, footnote)

Minsky was building robots; Minsky and Papert were designing vision machines; and then there was chess...

The argument ebbed and flowed on the matter of intelligent machines. The philosopher Hubert Dreyfus published, in 1965, a report for RAND called Alchemy and AI (later expanded into the book What Computers Can't Do.) generally panning what Dreyfus termed the artificial intelligentsia. The ensuing battle was bloody and is, indeed, far from over.

Papert wrote a paper to refute Dreyfus, also for RAND, but the attorneys would not touch it. It was eventually brought out as a Project MAC report with no lawsuits ensuing. (McCorduck, 1979 p 196)

There's still great debate about who did what and who said what in the great chess debate. The fact is that a chess playing software package which, we will call MacHack throughout for simplicity, had been beaten by a ten-year old boy to the delight of Dreyfus who was alleged to have claimed that chess playing computers could never beat any human player.

The program was improved (in fact it was a different program) and somehow Dreyfus, who was a poor chess player, was persuaded to play it.

Papert, smiling recalls, "I organised the famous chess match. That was beautiful. He was - well, it wasn't all pathetic and sad because he was quite convincing. He was going to beat it very easily. And that also said something about him, almost naive. We didn't know. About halfway through we all thought Dreyfus was going to win." (McCorduck, 1979 p 198)

Herb Simon had this to say:

“Dreyfus thought that MacHack would play bad, mechanical, non-human chess. But it was a wonderful game - a cliff-hanger between two woodpushers with bursts of insights and fiendish plans...great moments of drama and disaster that go on in such games.” (McCorduck, 1979 p 199)

Dreyfus was soundly defeated . Revenge for the artificial intelligentsia was very sweet; but, as throughout history, successful houses soon turn upon themselves.

Frank Rosenblatt had been a classmate of Minsky at high school in the early forties. In 1962 he introduced the perceptron to the press with, it must be said, some fanfare. The perceptron was a simple neural network, a model of artificial intelligence at odds with the then fashionable symbol manipulation models. Minsky had toyed with neural networks, in fact his PhD dissertation concerned them, and had dismissed their worth at that time. Thus the claims made by Rosenblatt purporting to demonstrate the learning powers of the perceptron were viewed right from the start with skepticism in the Papert camp.

The MIT faction did take the situation seriously though. David Waltz, a graduate student recalls:

“Marvin and Seymour really were interested in Perceptrons. I and a bunch of other students took a seminar from them, where the goal was to figure out as much about Perceptrons as possible. We were merely to explore in a methodical sense what they were capable of and what they weren't, and try to characterize them in some way.” (Crevier, 1993 p 107)

The result of all of this activity was the publication of a book by Papert and Minsky in 1969 which they called Perceptrons, and which demonstrated mathematically and most ably that the simple Perceptron was totally incapable of solving a wide range of important mathematical problems.

The repercussions of their book were immediate and dramatic. Rosenblatt and his collaborators were totally unable to rebut the arguments. Neural network research was dead in the water amid claims of deliberate sabotage to divert federal funding away from networks and into symbol manipulation programs. No self-respecting researcher would dare touch neural network research for a decade until the connectionist movement of the eighties which has proved greater potential for fruitful results. Connectionist researchers in AI blame Papert and Minsky to this day for the decade of neglect!

In the 1972 printing of Perceptrons there is a handwritten dedication to the memory of Frank Rosenblatt who was lost in a boating accident, by all accounts a broken man, shortly after the Perceptron affair.

We will leave the last word on the matter to Papert; he wrote in 1988:

“Did Minsky and I try to kill connectionism, and how do we feel about its resurrection? Something more complex than a plea is needed. Yes there was some hostility in the energy behind the research reported in Perceptrons, and there is some degree of annoyance at the way the new movement has developed; part of our drive came, as we quite plainly acknowledged in our book, from the fact that funding and research energy were being dissipated on what still appear to me (since the story of new, powerful network mechanisms is seriously exaggerated) to be misleading attempts to use connectionist methods in practical applications. But most of the motivation for Perceptrons came from more fundamental concerns many of which cut cleanly across the division between networkers and programmers.” (Papert, 1992 p 346)

Mindstorms and logo

We turn now to that most famous episode of Seymour Papert's contribution to educational life - the development of the computer programming language Logo. Matters of common knowledge will be mentioned briefly. Suffice to say that the language had its beginnings in Papert's thoughts of the late sixties and culminated in the publication of his seminal work Mindstorms: Children, Computers and Powerful Ideas in 1980.

The Logo language was a development of the list processing language known as Lisp, and its genesis from lists gave rise to its name - the Greek tern for word or knowledge. The famous turtle probably finds its antecedents in a similar living model which used to roam the Bristol UK laboratory of brain physiologist Grey Walter in search of food by way of power outlets! The rest, as most would say, is history.

A second edition of Papert's book Mindstorms was published in 1993 and it is to the introduction of that edition to which we turn our attention.

As is right and fitting Papert uses the introduction to the second edition as a debugging exercise for the first edition; practising his preaching on not getting things right the first time!

Papert acknowledges that he failed to anticipate the number of educators who would read Mindstorms - especially elementary teachers. He expressed disappointment that many of his exemplary microworlds are classically physical and an impediment to such readers going beyond the early part of the work where the field is rich in Newtonian physics! Papert suggests, in retrospect, that he would reorganize the book to present the Images of a Learning Society chapter first in order to entice a broader audience further into the plot.

Papert is also concerned with the notion, which he suggests some have read into the work, that computers will cause changes in the way children think.

“What I was saying, and still say, is something slightly more subtle: I see Logo as a means that can, in principle, be used by educators to support the development of new ways of thinking and learning. However, Logo does not in itself produce good learning any more than paint produces good art.” (Papert, 1993a p xiv)

The emphasis in the first edition on structured programming is also regretted. Papert would much have preferred to introduce the tinkering, bricolour as an alternative programming style much earlier in the piece and he stresses his subsequent work in this field with his then wife, the techno-psychologist Sherry Turkle. (Turkle, 1984).

The recent developments in Logo, in particular Microworlds, is seen as a move towards the encouragement of a richer and wider epistemological range and, we may say, a wider interpretation of what activities constitute computer programming.

Constructionism and the Children's Machine

In recent years the writings of Seymour Papert return ever more strongly to his Piagetian roots. In the late seventies his collaborative work with Marvin Minsky on the development of Society of Mind concepts is significant; though noticeably unreported on Papert's part. Such societal metaphors show, in their layering of different bands of cognition in the developing mind, an echo, of Piaget's stages of readiness. In the eighties Papert's work with Sherry Turkle on epistemological pluralism and with Idit Harel (Harel, 1991) on children as software designers carries forward the constructivist notions of building knowledge structures into constructionist principles which are driving children's activity in this decade.

Papert would, I suspect, consider a definition of constructionism as an oxymoronic concept. We will have to do with the banal, flat and constrained learning by doing - especially physically making.

“Constructionism shares constructivism’s connotation of learning as "building knowledge structures" irrespective of the circumstances of the learning. It then add the idea that this happens especially felicitously in a context where the learner is consciously engaged in constructing a public entity, whether it's a sand castle on the beach or a theory of the universe.” (Papert, 1991 p 1)

Constructionism finds a true home in a computer rich culture and herein lies the heart of Papert's objection to current educational practice. He is not as might have appeared from Mindstorms anti-teacher rather he is anti the prevailing school culture which constrains children, physically and epistemologically, in the pathway of its own liking.

In his closing address to the 1990 World Conference on Computers and Education he appealed, in the spirit of those times as a comparison with the political structure of the then USSR, for perestroika in epistemological politics.

“His [Mikhail Gorbachev] slogan of perestroika (which literally means "restructuring") became synonymous with a policy of struggling to reform a system in serious crisis without calling into question the foundations on which it was built . It should be clear by now that I see most of those who talk loudly about "restructuring" in education in much the same light - though few of them have the courage to carry the reforms as far in their realm as Gorbachev did in his.” (Papert, 1993b p 206)

Perhaps the computer is The Children's Machine and the vehicle for freeing thought.

Endnote

We end with this vision of Papert:

Absent-minded like many driven intellectuals, Papert is said to have once realized, mid-way across the Atlantic, that he had left his wife behind in a New York airport. Colleagues report that he sometimes forgets to show up at lectures and, when he does, tends to get carried away into whatever topic fascinates him at the moment. A man of dramatic personal magnetism, he is likely to startle interviewers with juggling demonstrations at airport terminals or by stopping his car in the middle of a U-turn to formulate a thought. Papert's aphorisms, like Minsky's, tend to stick. One of his favorites is that we are to thinking as the Victorians were to sex. (Crevier, 1993 p 86)

I will leave you to unwrap that saying of Papert's for yourselves!

References

Crevier, D. (1993) Artificial Intelligence. New York: Basic Books.

Harel, I. (1991) Children Designers. Ablex: Norwood, NJ.

McCorduck, P. (1979) Machines Who Think: A personal Inquiry into the History and Prospects of Artificial Intelligence, San Francisco: W. H. Freeman & Company.

Papert, S. (1980) Mindstorms - Children, Computers and Powerful Ideas. New York: Basic Books.

Papert, S. (1988) The Conservation of Piaget: The Computer as grist to the Constructivist Mill in Forman, G. & Pufall, P. (Eds.) Constructivism in the Computer Age. Hillsdale NJ: Lawrence Erlbaum.

Papert, S. (1991) Situating Constructionism in Harel, I. & Papert, S. (Eds.) Constructionism. Norwood, NJ: Ablex.

Papert, S. (1992) One AI or Many? in Beakley, B. & Ludlow, P (Eds.) The Philosophy of Mind. Cambridge MA: the MIT Press.

Papert, S. (1993a) Mindstorms - Children, Computers and Powerful Ideas. Second Edition, New York: Basic Books.

Papert, S. (1993b) The Children's Machine - Rethinking School in the Age of the Computer. New York: Basic Books.  

Papert, S. (1998) Transcript of Child Power: Keys to the New Learning of the Digital Century at the 11th Colin Cherry Memorial Lecture on Communication, Imperial College, London. http://www.connectedfamily.com/main_alt.html

Turkle, S. (1984) The Second Self: Computers and the Human Spirit. London: Granada.

About the author

Martin Boyle is a member of the Faculty of Education at Monash University, Victoria, Australia. He presented a version of this paper at conferences organized by the Computer Education Group of Victoria. Martin Boyle may be reached at: martin.boyle@education.monash.edu.au.