Review: Six Not-So-Easy Pieces, by Richard P. Feynman
I have, over the years, spent a lot of time on the internet, and the sort of argument styles you find online have started to grate after awhile. Not to say it's surprising, you'll meet ten people a day whose self estimation exceeds their capability, but the digital fields have propagated a certain strain, the self-styled Vulcans or the Slans finally given their opportunity to shine in a purely intellectual medium, and okay, a few of them do shine, even in political arenas, but I'm increasingly convinced that quality thinking is a rare flower even among the Comic Book Guy set, that intelligence and a penchant for declarative sentences may not be sufficient to overcome the various self-deceptions, inappropriate generalizations, faulty logic, bad weighting of counterargument, inability to distinguish hindsight from foresight, and all the rest of the bestiary of cognitive curlicues that obstruct most minds from grasping the nature of things in a broad but generally accurate and consistent sense.* Now and then, a great or even exceptional thinker will emerge, and rarer still will that thinker be able to communicate with accuracy, humor, and humility. Feynman was a gem.
Of course, one of the side effects of reading Feynman is that once that high of uncut understanding wears off, your own brain feels a little muddier afterwards. I think that the audience that evolved to love this guy is primarily composed of scientists looking for deeper meaning into what they already know. (I suppose I could put myself into this group, although without doubt, relativity isn't something I already know in a meaningful way.) These lectures were great at communicating insight, but they're actually lean on facts. I not at all certain that I'd have wanted to learn freshman physics from this guy. I wouldn't have been left with the tools to do the problem sets.
What I didn't realize when I picked this up, is that they're six connected pieces. The progression here, examining the laws of physics through geometrical transformations, holds together cleverly and proceeds naturally to relativity. I further appreciated the discussion of the contemporary setting of Einstein's theories, what he was specifically trying to address with them, and that the Lorentz transformations (which in special relativity, Einstein explained the validity of) had already been thrown out there as fudge factors to explain the discrepancies between predictions of Maxwell's equations (where it falls out as a constant in the simplest EM wave equation) and what's expected by conventional expectations of relative motion, as well as the confusing experiments that were going on when the great man was busy theorizing. The context makes relativity seem a lot less weird, frankly.
I recall reading that one of Einstein's proudest achievments was his index notation for vectors. Basically, he simplified the math symbols so that you don't have to write big brackets, or unit vectors, or summations. It's pretty intuitive and handy, occasionally adopted formally by fluid mechanics nerds (which Einstein dabbled in), and frequently informally by any hack engineer who's tried his** hand at primitive excuses for number crunching. If you ask a mechanics guy and a computer guy, both inclined to use index notation, what's the meaning of a "vector," you'll probably get two different answers, and thinking this way, I'd have rather seen Feynman move down from the general, in which a vector is a column of numbers of arbitrary length, which is merely limited to three when it needs to describe Euclidian space, or, presumably, four when describing Minkowski space. (How dare he present this subject forty years ago!) It's not immediately clear that the innovation couldn't have been constrained to new algebraic operations specified to describe relativistic effects in 4-D space, rather than specifying a new sort of space in which the fourth dimension has separate properties. I will say that the geometrical analyses that Feynman described struck me as very clever. ("Clever." I mean it's Einstein here--this is like a mosquito calling a jetliner sort of large.) I suppose I'd have to do problem sets to get the real scoop on it all (and I've no interest in that).
In one of the lectures, Feynman had a great bit about using measurement as a tool for explaining the universe, and took an entertaining poke at fly-by know-it-alls who'd pretend it's intuitive (he's favoring empiricism over pure rationalism I guess, but he didn't frame it that way). That discussion of how the geometry of space time could be deduced from within the system was the most animated, and, I think the most successful application of his grand insight, and it occurs to me I've seen a dozen explanations along the same lines (usually with graphics of marbles making dimples in a computer-generated 2-D grid) that, unlike Feynman's analogies, were not terribly intelligible at all. I've caught the bug's-eye thing before too and the jumpy speaking style too (I don't usually "hear" when I read, but I could feel Feynman speaking in this one), and it occured to me that many of my professors were among the generation of (inevitably lesser) Feynman copycats.
* Alternate theory: I'm projecting.
** For a possible theory about my use of gendered pronouns, see note 1.



