QED: The Strange Theory of Light and Matter by Richard P. Feynman

Somehow or another, I have managed never to read Feynman's famous book on Quantum Electro-Dynamics. It always seemed a little too much like work, but having found myself in the position of writing a pop-science book about quantum physics that includes a chapter on QED and Feynman diagrams, it seemed like it would probably be a good idea to get and read a copy.

An odd side effect of the mythologization of Feynman-- partly his own doing, and partly the work of hero-worshipping nerds-- is that it's easy to forget just how good he was at doing this sort of thing. So much time is spent on the skirt-chasing, lock-picking, and bongo-playing that you can lose sight of the fact that he had a real gift for presenting science. The Feynman Lectures on Physics are a little dated now, but loaded with really good ideas, and this book is an excellent treatment of the subject that made his name.

This is an unusual book in a lot of ways. It's an edited transcription of a set of four lectures he gave at UCLA in the early 80's, and as such contains a lot of little asides and exclamations that almost certainly work better out loud than they do on the page. It's also an uncompromising treatment of the subject, in that Feynman sets out to present QED to a general audience in some detail, and not just as a set of mad hand-waves and bold assertions. He works through the rationale for the path integral formulation of quantum mechanics almost from first princples, and goes into a surprising amount of detail about how to calculate the g-factor for an electron (way more than I was willing to attempt, that's for sure).

And it works. On the surface, at least. It still fits in the general category of what one of my grad school professors said about the Feynman Lectures:

The thing about Feynman is that you read it, and you say "Yes! I understand! I'm doing physics!" Then you try to solve a problem, and you find that, well, that you're not Feynman.

He's obviously not trying to teach anybody to do real QED calculations here-- he says that explicitly every five pages or so-- but there's a similar effect at work. The presentation here is so elegant and seamless that it's a little hard to imagine someone who isn't Feynman using this book as the basis to explain the theory to someone else. If you already know a little bit about it, this can illuminate some other aspects, and provide some tricks that you could combine with pre-existing knowledge to make a good presentation, but the book alone conveys more of a sense of understanding than actual understanding.

Lest you think I'm damning with faint praise, though, I should say that this is no mean feat. The material he presents is in many ways at a level of weird and abstract that's comparable to anything else you'll find in modern science, and yet he presents it in such plain language, and such a no-nonsense manner that it's surprisingly easy to follow along, and you leave with a clearer sense of the subject than in almost any other popular book I've read. It's really a masterful performance.

I highly recommend this book for anyone who is interested in the subject.

(I'll note, though, that if you get the 2006 Princeton University Press edition (as I did), you should feel free to skip the remarkably pompous introduction by A. Zee that's touted on the cover. Its tone is almost nothing like the book itself, and could easily create a false and negative impression of what reading it will be like.)