Lightning has already struck twice in Stephen Wolfram's life. Now, he has his kite out in another thunderstorm, hoping it'll strike one more time. Twenty years ago, the British-born Wolfram was the wunderkind of the scientific community, thanks to his work in particle physics and then in a realm even more esoteric: cellular automata. These are simple computer programs that, line by line, construct grids of black and white squares and which, Wolfram posits, symbolically replicate how much of nature works.
Then, while on the University of Illinois faculty in 1988, Wolfram electrified the world again. This time, it was by creating Mathematica, an immensely popular software program that enables scientists to perform exceedingly difficult mathematics.
Most mortals might have been satisfied at that point. If not immediately recognizable, Wolfram had won acclaim -- he was awarded a MacArthur Foundation "genius" grant in 1981, when he was just 21 -- and had become a multimillionaire who owned a high-tech company.
Not Wolfram. Frustrated that his peers lost interest in cellular automata, he dropped out in 1991 so he could devote all of his brainpower to these computer programs and what they implied for traditional sciences, from astronomy to zoology. He maintained his CEO position at Wolfram Research, which now employs 330 people. But otherwise he became a recluse, working through the night alone at his computer keyboard. He now tells of his 11-year effort in a 1,200-page book he just self-published, A New Kind of Science.
AS RANDOM AS PI.It's not light reading for the summer beach. At its core is a computer program Wolfram has dubbed Rule 30. Like the other cellular automata, this one starts with a single black square. The color of the squares in the next line is determined by those immediately above them. Most of these programs result in symmetrical and predictable patterns. Rule 30, however, produces a series of black and white squares as random as pi.
Wolfram asserts that this proves that a simple rule can produce unexpected complexity and that something like it may be responsible for the jumble we call the universe. Heavy. Yet the book isn't just filling storerooms. Since it was released May 14, the first 50,000 copies have sold out -- at $44.50 apiece.
Wolfram took time out recently to spend an afternoon at New York's Museum of Natural History with BusinessWeek Senior Correspondent Michael Arndt. Wolfram, 42, was wearing a gray herringbone jacket, blue pinstripe shirt, black slacks, and a brand new pair of New Balance running shoes. On his right wrist, he wore an analog watch.
He's bald and paunchy, and jabs at his nose to push up the wire-rim glasses that keep slipping down. Their conversation began at a diorama of African wildlife, with Wolfram talking about how the stripes of a zebra are akin to the rows of black and white squares he studied for years on his home computer. Edited excerpts of the discussion follow:
Q: In your book, you argue that Charles Darwin may have had it wrong. The odd characteristics of species, such as a zebra and its stripes, may not have resulted from natural selection, but just came about randomly.
A: The question is, why does a zebra have its stripes? People say it's camouflage. Is it really to a lion? I don't know. But people imagine it's difficult to produce zebra stripes, so they say this must have been something that was produced for the careful optimization of the life of a zebra.
The thing that has been a basic intuition of science is that if you see a complicated phenomenon, it should have a complicated cause. That's a reasonable intuition because when we do things with engineering, that's what we end up doing. In order to make something complicated, we have to have some complicated plans and complicated procedures, etc.
That has led to conclusions like: We look at what we can make as humans. We look at what happens in nature, and what happens in nature is much more complicated than what we as humans can make. Therefore, there has to be some form of higher intelligence responsible for what we see in nature.
But one of the things I've discovered is that you can get complicated patterns by very simple programs, very simple underlying rules which can easily begin and sort of randomly arise, and they're not things that were carefully molded by some elaborate process of natural selection. It turns out that zebra stripes are rather easy to get. It really isn't that hard to make complicated stuff.
Q: Does that mean it wouldn't be that hard to create a human?
A: It depends on what aspect. We're filled with all sorts of organs. A lot isn't known about how things grow. But it will be increasingly found that the actual programs that produce some of the complicated structures in organs are actually very simple programs. When it comes to thinking about brains, we may be proud of our computational ability and the fact that we can figure out stuff about math, for example.... But we may not be special.
There is an aphorism that the weather has a mind of its own. That might be less silly than we imagine because the fluid motions that take place in the weather represent sophisticated computations, and I'm sure they represent ones as sophisticated as the computations that go on in our brains. So in a sense it's like saying there's mind in all these things. I call this computational equivalence.
It's an ultimate Copernican realization. It used to be that we thought the earth was the center of the universe and it was really, really special. Copernicus showed we're not at the center at all. Still, we've maintained this belief that we're special because of out computational ability. But we may not be special, or unique, in this regard either.
Q: Can we reduce the universe then to a simple program?
A: Oh, yes. Oh, yes. I've been doing science for 27 years now, and so I've been able to test whether the things I think are true actually turn out to be true. I've been as bold as that in the past. But I've been right. Usually in doing science, you get an idea of whether you're right or not because you get started, and things either fall into place and they keep on falling into place, or, oops, I can't explain that and, oops, I can't explain that. In this case - a simple program representing the universe -- lots and lots of things that I didn't imagine would fall into place fell into place easily.
Q: Is this something you're thinking about all the time? It's not like you go home at night, and your mind starts wandering and before you know it, you're thinking about a rerun of a TV show?
A: No, this is what I think about. It's really fun. There are a lot of things that I've figured out that have been big mysteries for a long time. It's exciting to see that it's not really such a mystery. It's something that one can understand. This is what I like to do, figure stuff out. And I don't watch television, so that frees up a great many hours.
Q: Most noted scientists aren't businessmen. And most business executives don't have a science background. They seem to be mutually exclusive, in fact. Yet you're both a scientist and a successful CEO. How do you manage to bridge this gap?
A: The business I'm in is one where ideas count. I hire lots of creative people at my company. In terms of running a business, I've always thought most business stuff is common sense. It's thinking things through with clarity. If I don't understand something, I don't believe it. My company is probably painfully straightforward. I've never particularly liked, nor been very good at, the political life.
It's good that I make my living as a CEO because I wouldn't make my living as an employee very successfully. I find it easier to make a decision myself. Sometimes it will be wrong. Usually, it will be right. What I don't like doing is getting a consensus. I just get very frustrated. I usually say, this is what I think, and I don't really care what you think.
My main interest in science, as in business, is building things. I am not a very competitive character. I always feel that if I'm competing on something, my sentiment is, why I am doing this because someone else is doing it too? What's the point? I like to feel that the things I'm doing are unique.
Q: Let me ask you about your book: The title, A New Kind of Science, is kind of audacious, isn't it?
A: I don't know. I think it's a good representation of what it is.
Q: No offense, but it does take quite a bit of, shall we say, self-assurance to make that declaration.
A: One of the biggest challenges in doing a project like this is having the confidence to believe you can do something that is meaningful. In science, I've become ever-more ambitious in the things I've attempted. I've been able to do that because I succeeded at earlier things. This kind of science is more ambitious -- it's off-scale ambitious.
At this point, I'm confident enough to say that if I go into a field of science, I will be able to solve foundational problems in limited amounts of time. It's like finding bugs in software: I know where the foundational weaknesses are. One has to make sure one doesn't overdrive one's intuition, so to speak. But one cannot be afraid.
I get in the mail all the time people's mad theories of the universe. One of the things that is so often true is people think that to figure things out, they don't need to know all this other technical stuff.
Q: Like it just came to them in a flash?
A: Right. That's not the way it works. It'd be great if I could just say it occurred to me one afternoon. But it's a lot of work. When I look at this book, I'm very pleased it only took me 10 years. There's a lot of stuff in it. It's beyond my capabilities to be able to get them all in a flash. Big intellectual structures like the one I'm trying to build don't come quickly. They typically take years.
Q: What kind of response do you hope your book will generate?
A: I'm trying to do things at the highest leverage point, which is the most abstract point. So the technological frontiers are downstream. I'm brashly predicting that within 50 years, there will be more new technology being created with the kind of science I'm building than traditional science.
In traditional technology, we know of wheels and pulleys and levers and various mechanisms. What I'm providing is a new collection of mechanisms one can use for technology. The only thing I will find frustrating is if nobody actually reads the damn thing.
Q: What has motivated you?
A: Many scientists are motivated by wanting to show the world that they're cleverer than everybody else. That certainly was a component of my motivation in my younger years. But as time has gone by, who am I trying to convince? Because I started doing science when I was young, most of my peers were in mid-career when I was a teenager. Now this stratum, most of them have retired or died. There are no grand old people of science to impress anymore.
Also, the more confident one is, the less one tends to care about what others think. For me, the most important thing is doing the science. I like the fact that I figured things out. In a sense, that's my piece of philanthropy for the world.
Q: When you were doing this research, did you bounce ideas off of peers?
A: I really don't find that useful. To get people to the point where they could really debate what I'm talking about, they would have to know a lot of stuff about what I had done. It really wasn't terribly worthwhile. I had various research assistants. But we mostly talked about details. I used to collaborate with people when I was younger, more for social reasons. A lot of people find it is useful to be interactive with people. But I haven't found that. I kind of like to rely on my own judgment.
Q: Obviously, you've gained a lot by spending all these years on this project. Do you think you've given up something as well?
A: I could have built a much bigger company over the last 10 years if I had wanted to. There have been lots of small projects, where I could see that I could figure out something really nifty in two weeks, and I've had to take a pass on them. I have not been able in anyway to be opportunistic for a long time.
Q: You're married now and have three young children. Did you have to sacrifice some family time too?
A: I don't know. I would always like to spend more time with my family. But one of the things is that I ended up working at home for these years. There's no point in me going to an office to work on a computer.
Q: Are any of your kids showing a proclivity toward science?
A: They're smart kids. One of my exercises is to see if I can explain complicated scientific stuff to young kids. It's a good test.