Let me begin today’s newsletter by saying that I don’t have a good sense of who Stephen Wolfram is. I’m not a mathematician, statistician, computer sciences professional, or other scientific expert. I’m a Patrick Gunkel scholar. To the extent Stephen Wolfram had some interactions with Gunkel, that’s extremely interesting to me. Otherwise I don’t have the time to go out of my way and learn much about Wolfram other than what I can gather from Wikipedia, which suggests to me he’s a computer scientist, businessman, and overall polymath genius.

I’ve written about Wolfram tangentially before, mostly noting that Gunkel is mentioned in Wolfram’s eulogy for Gunkel’s long-time friend and supporter Edward Fredkin.

As documented by Wolfram, in July 2003, Fredkin emailed Wolfram hoping to find work for Gunkel. This email was sent after a meeting between Wolfram and Gunkel in Woods Hole, Massachusetts.

Wolfram mentions this meeting in the Fredkin eulogy but does not provide any details.

Fortuitously, however, an email from Gunkel’s archives in the MIT Distinctive Collection describes the meeting in detail from Gunkel’s perspective.

In a July 24, 2003, email to one of Gunkel’s friends, Gunkel states that Wolfram was visiting Woods Hole to talk about his recent book, A New Kind of Science (ANKS), at the Woods Hole Oceanographic Institute. After the talk, Wolfram walked over to the large house where Gunkel was renting a room in the attic.

“Now and again people would mention my name [to Wolfram], or refer to me or my work… this had made [Wolfram] curious about me,” Gunkel wrote.

Whenever Wolfram shows up at a place where an interesting person lives, Gunkel reported, he sets about trying to meet that individual.

And as the below email excerpts show, Gunkel was delighted by the meeting with Wolfram and had only good things to say about his encounter:

Word about the meeting quickly reached Fredkin, who then emailed Wolfram pleading Gunkel’s case.

The problem was that Gunkel had been in a financial crisis for at least a year by that point, and as usual Fredkin wanted to try to help him out.

“You might be surprised at the number of wise and intelligent people who really appreciate Pat and his works,” Fredkin wrote, according to the email excerpt Wolfram provided on his website. “Yet after more than 30 years of fitful, diverse yet nearly continuous support, Pat has come to a situation that, to him, looks like the end of the line. There is no doubt that Pat can be a difficult person to help, but I notice that he has mellowed with age. Of course, Wolfram Research is not a charitable institution. But I believe that Pat’s ideas on ideonomy are really important and that those ideas may form the basis of interesting future applications.”

Alas, Fredkin’s push to place Gunkel with Wolfram did not succeed.

Wolfram says that he might have hired Gunkel but couldn’t figure out what Gunkel could do for his business interests.

Wolfram’s assessment was probably very accurate: Gunkel lacked the technical skills of a computer programmer, mathematician, or statistician—he was a mainly a philosopher-theorist, was not interested in conducting scientifically valid experiments, and was biased by his own intuitive conclusions about the universe. Plus he was almost impossible to work with and had a terrible, disruptive personality. Just imagine—someone who was so eccentric and difficult that he was kicked out of MIT, a veritable hornet’s nest of eccentric and difficult people!

Gunkel’s encounter with Wolfram was part of an enduring pattern.

Gunkel had encounters like this over and over again in his life—people who heard about him, were impressed or intrigued by him, and tried to figure out how to work with him.

Nearly all such encounters ended in disaster.

The meeting with Wolfram was no exception to Gunkel’s lifelong pattern.

According to Wolfram, Gunkel followed up the meeting in absolutely classic fashion—by sending Wolfram a critique of ANKS, along with a critique of Ed Fredkin’s own ideas.

As Wolfram notes, this didn’t really factor into his decision about whether to hire Gunkel. It was mainly the fact that he had no clue what Gunkel could do for him.

But let’s pause a moment nevertheless to consider what kind of person Gunkel must have been, and what this moment says about him.

There he was in an absolute financial crisis; a wealthy businessman with an entourage of people had shown up at his house to discuss his work; there was a potential opening to establish a relationship…and what does he do, but send—instead of a resume, I assume, because Gunkel didn’t have one—a critique of the book that the wealthy businessman had just come to town to promote!

On some level Gunkel probably thought he was being logical.

If Wolfram was going to hire Gunkel, it would probably be a good idea to know where his thoughts lay regarding ANKS.

But on a deeper level, Gunkel was likely very triggered by Wolfram.

It’s easy to imagine Gunkel suffering from a case of sour grapes when confronted with a figure of Wolfram’s stature, especially given Wolfram’s attempt to found “a new kind of science” with his book on cellular automata.

The reason why is obvious: Gunkel felt that if anyone was going to found a new kind of science it should be him, and the name of that science should be ideonomy.

As it turns out, Gunkel didn’t write only one review of ANKS.

In fact, my archival research has found that Gunkel wrote three different reviews of ANKS, and sent these reviews not just to Wolfram but to his other correspondents including renowned Princeton physicist Freeman Dyson and philosopher and cognitive scientist Daniel Dennett.

Gunkel knew Dyson, and he surely saw in him an intellectual ally, as Dyson had absolutely slammed ANKS publicly, calling it “worthless” and a “case of style over substance.”

As for Gunkel’s relationship with Sennett, this archival email is the only evidence I have found that they were even correspondents; much more work can be done here to understand what kind of relationship they had.

Now, I’ve never read Wolfram’s book on cellular automata and I’m not sure I will. 

There are only so many hours in the day.

I know Gunkel felt there was a conceptual intersection between cellular automata and ideas, as I wrote about here, but I’m not laboring under the illusion that I can develop a comprehensive map of everything that ideonomy touches. Many other people are going to have to do that.

Instead, for the benefit of future Gunkel scholars (if there are any), and at the risk of irritating Wolfram, I will offer Gunkel’s definitive review of Wolfram’s book merged from three different source documents:

The Amazon.com review of Stephan Wolfram’s book that is only available via Y Combinator:

And the forwarded email commentary from Gunkel to Freeman Dyson and Daniel Dennett:

THE DEFINITIVE PATRICK GUNKEL REVIEW

of Stephen Wolfram’s 2001 book

“A New Kind of Science”

Stephen Wolfram's magnum opus is extremely brilliant and absolutely fascinating. Or at least I can say this about those parts of the book that seem to interest me the most: the Notes, in 349 pages of (overly) small print, and the invaluable Index, as well as certain figures. The Notes remind me not a little of the unique sweep and precision—although not of the abiding wisdom—of the eminent physicist John Wheeler in his books, possibly owing to the sheer energy and power of both minds, and the overspreading play of ideas and interests.

The rest of the volume, or main text, has the virtue of simplicity, and is essential as an overview of the author's research and "new science," and for its explanations and direct assertions, but it is on the whole of inferior interest and value.

Even if Wolfram's key beliefs, those which underlie his system, should turn out to be wrong, his book will long go remain what it is now: enormously challenging, useful, and informative, and a masterpiece of interdisciplinary scholarship and universal thought and creativity.

The main value of his methods and system, I believe, is in their correction of certain inexcusable neglects in theoretical and experimental physics, and science generally. The argument of the book, however, exaggerates the absolute importance of these methods and principles, and of certain kinds of natural phenomena and behavior; and, in a cardinal error, it mistakes them for physical reality itself.

Ironically, however, Wolfram has not sufficiently generalized his ideas, for they can be put in a still higher, or given a more fundamental, form, and extended to a larger class of related things. He has wrongly grasped and praised the mere tail of a great animal, of which science itself, until now, has held a shank.

"Worthless," Freeman Dyson is quoted by Wired magazine as saying about Wolfram's A New Kind of Science. That, too, is a possibility! And lovable as the man 'is', it is a tempting assessment of his self-trumpeting masterpiece. In my remarks above, I was trying to be generous in defiance of precisely that temptation.

On the other hand, it is very unhealthy to stay up all night, as Wolfram describes himself as doing, in his monologic book. For one thing, it can blight one's view of mankind on topside. For another, if I may speak here as a neuroscientist, it can warp one's judgment, or ability to distinguish between reality and a fantasy, a thing in one's mind and that greater thing—that most stubborn fact—which is the world itself.

Cellular automata are so simple that they can suggest everything in nothing, and act as a powerful generator of illusions, particularly because of their novelty of form and behavior. The physicist who is first confronted with the mechanism of life must be knocked head over heels in a fool's amazement.

And Stephen Wolfram suffered, and clearly suffers even now, from the additional, most grievous and unkind, disadvantage of having been an infant prodigy.

For me there is nothing new about Wolfram's primary ideas. I have toyed with such notions for a quarter of a century, and so have a great many other people. Such ideas have power, but it is a very different question whether they have physical plausibility at the most fundamental level, and a necessity that goes beyond redundancy in offering more than do scores of other competing 'explanations' of the physical world, at the tiniest, intermediate, and largest scales.

A passage in Montaigne's Essays (book One, Chapter 50: "On Democritus and Heraclitus") is pertinent: "Things in themselves perhaps have their own weights, measures, and states; but inwardly, when they enter into us, the mind cuts them to its own conceptions."

I think the mind of Wolfram cuts them short! Moreover, I feel that, for the moment at least, the attractions of cellular automata—like a carnival that snags eye, ear, and leg through its tricks, bustle, and strangeness—are meretricious.

But my greatest objection to cellular automata lies in my belief that little of what is thought essential to them is, and what is is not peculiar to them. In short, they are merely little mirrors of a much greater idea, or a deeper and more comprehensive truth; one which, to begin with, is neither cellular nor automatic, and which is discrete (or 'binary') only as a process of transformation, when it is discrete at all.

As a neuroscientist, I have seen, over 30 years, not dissimilar errors—errors, for example, of misplaced emphasis, extrinsic formalization, and naively truncated research into physical detail and parallel mechanisms—plague abortive theories of the brain.

I am anxious that this excessively clever fellow not be allowed to become, through an initial absence of penetrating criticism of his ideas, and his megalomanical and almost autistic obsessions, another Pied Piper of Hamelin, in an age which is already blurring the distinction between the real (and natural) and the artificial (and fanciful), to a degree that could soon become pathological.

Incidentally, in Browning's homonymous poem, which he wrote in 1824, the Pied Piper of Hamelin "is a musician who, according to an old German legend, came, in 1284, fantastically dressed, to Hamelin in Prussia, and offered for a sum of money to rid Hamelin of its rats. He charmed the rats by his piping into the river Weser, where they drowned." So at first the Piper was a heroic figure.

Later, however, "as the reward was withheld, he in revenge by his magic piping drew the one hundred and thirty children of the town to a cavern in a hill, which closed after them forever."

In real life, most heroes ARE piebald figures.

At bottom, I find Wolfram's youthful but embarrassingly perseverative love affair with cellular automata deeply amusing:

Two decades ago, to caution the ardor of a friend of mine, Prof. Edward Fredkin—whose pioneering work upon, and advocacy of, cellular automata, in fact later led to Wolfram's own bewitchment—I constructed an instructive list of "250 Single Things People Have Tried To Reduce All of Nature To".

Some of the more conventional rival candidates have been matter, mass, energy, time, space, number or mathematics, points, lines, spirit, mind, ideas, God, machines, motion, vortices, nothing, and life (biology), apart from the 233 OTHER 'sole fundamental entities' I was able to recall on the spur of a moment, while confronting a Plenum of ulterior possibilities.

The monoideistic advocates of these '1-substance' metaphysics (or cosmologies) were often just as bright, confident, and fanatical as their latest fashionable successor, or avatar, S. Wolfram.

The latter's faith in his own uniqueness, at least, is anything BUT unique!

There is power in simplicity, but not matter. Cellular automata may be a valuable new formalizing tool in physics, added to mathematics, but I would bet all the world that they are not the world.

— Patrick Gunkel

I used Claude Opus to help deconflict the three reviews. These are Claude’s editorial notes: