Once a physicist: William Poundstone

William Poundstone is a Los Angeles-based author and essayist. His most recent book, Are You Smart Enough to Work at Google?, was published in paperback by Little, Brown in September

William Poundstone

What sparked your interest in physics?
I recall reading a paperback edition of Martin Gardner's Relativity for the Million, probably when I was about eight or nine. From then on I was hooked. I had already been interested in the usual run of scientific topics for a kid – dinosaurs, chemistry, astronomy – but that book clinched my interest in physics.

You studied physics at the Massachusetts Institute of Technology (MIT) – did you enjoy it? 
Yes, especially the lab work and computers. This was the 1970s, so like everyone else, I took a course in FORTRAN, believing that it would be the lingua franca of the digital millennium!

Why did you decide to become a writer?
I always wanted to be a writer because I enjoy books. It did not really occur to me that an aspiring author might seek an English degree.

What skills have helped you to succeed?
After MIT I had a brief career as an editor at a medical journal publisher, and that was useful. I learned about usage and editorial style, things that were not taught in my English classes (or in physics classes, for that matter).

How do you think your physics background has influenced your writing?
I think my physics background has been incredibly valuable in writing about social sciences and other non-physics topics. In physics the evidence is (relatively) clear-cut, and there is intense competition among ways of accounting for it. In the social sciences the evidence is a lot more slippery and you've got theories or schools or trends that, like aestivating salamanders, can survive for long periods without the need to slake their thirst on evidence. The physics background has been useful in discerning what's real in the social sciences. Max Planck said that new scientific truths triumph only when their opponents die off, and I think that is true to an even greater degree in the social sciences.

What are you working on now?
I am writing a book on the predictability of human actions. Its historic point of departure is the outguessing machine that Claude Shannon built at Bell Labs in the early 1950s. It was a simple machine that played the game of matching pennies against a human. All the human had to do was to generate a random sequence. But Shannon knew this was incredibly hard to do: our minds aren't built that way. Dozens of famous scientists and mathematicians visiting Bell Labs were challenged to match wits with the machine, and they all lost. Today the outguessing machine is a milestone of early artificial intelligence, not least because it managed to beat the world's greatest geniuses with just 16 bits (2 bytes) of memory. In the book I show how the outguessing machine's basic precept – that humans beings cannot act in a truly random fashion – has been influential in all sorts of fields, from tennis strategy to Internet marketing to high-frequency trading.

Do you have any advice for today's physics students?
My advice would be to take a variety of courses in physics, mathematics and other fields. Success is often a matter of seeing a connection or an analogy that no-one else has. Your chances of doing that are greatest when you have stocked your mind with a rich assortment of tangential ideas.

This article appeared in the November 2012 issue of Physics World.

Related information

Once a physicist

Other IOP websites

The membership magazine of the Institute of Physics