Enzymind Labs

Comments on "A New Kind of Science", S. Wolfram

(Mark Dow, 4/26/07)

"It is a safe rule to apply that, when a mathematical or philosophical author writes with a misty profundity, he is talking nonsense."
-- Alfred North Whitehead, An Introduction to Mathematics, 1948

Over the several years since publication (2002) I have had several brief opportunities to browse S. Wolfram's book "A New Kind of Science", and my only impression was that of curiosity. The images are interesting to an abstract repetative image geek (me), and I understood, and appreciated, the general focus of cataloguing the range of behavior of simplest possible cellular automata (CA).

Recently I had reason to flip through the book more than usual: I've been tinkering with graphical representations of simple recursive rules (e.g. Thue-Morse tilings). There are only a small number of simple recursive rules so it was inevitable that I'd run across the well structured and popular online book.

Intellectual alarms went off. The first alarm bell was due to running across this interesting letter (third image), from Richard Feynman to Wolfram in the 1980's), from Wolfram's on-line scrapbook. Soon I was reading the book linearly, from the beginning. The first chapter is outrageous, from a scientific honesty perspective. I made it through Chapter 6, but with an increasing sense of bemusement and disappointment.

There's a lot to like about the book; thoroughness in demonstrating a range of simple local rules, the popular presentation of math concepts, the beautiful use of illustrations, and hints about what these systems have to do with the rest of math.

That said, there is more not to like. There is no new kind of science present; I had always assumed the title was simple marketing, a benign reference to using computational exploration. A quick search for critiques of the book yeilded a wealth of illuminating, and often amusing, slams and analyses. I should have done this earlier than chapter 6. The rest of the book appears to contain no testable hypotheses, only speculation about potential applications.

Here are the best reviews -- those that I agree with most. They say it better than I can.

Cosma Shalizi's "A Rare Blend of Monster Raving Egomania and Utter Batshit Insanity "
All of this review is accurate, as far as I know. I don't agree with, or don't know enough to agree with, Shalizi's refutation of Jaynes' work. I'm a big fan of maximum entropy methods that Jaynes, in many ways, originated. The review is particularly apropos, as Cosma himself is a learned arrogant prick, but a self-effacing one: also see his "Comments from My Students, Physics 208, Fall 1994 "

Ilya Nemenman's slides, "A New Kind of Science?" reviews content, which "leave aside the majority of social, stylistic, and political correctness questions".

Humor related to A New Kind of Science. The essay in this collection "Smithers, release the monkeys." by Mark Fisher is also a serious challenge.

Ray Kerzweil's "Reflections on Stephen Wolfram's 'A New Kind of Science ' "

Ralph E. Griswold's A New Kind of Science: Book Review