I've been on vacation for a couple of weeks, which has given me some down-time to do some reading and thinking.
Of course, what I have thought about may not always be what the author intended for me to think about, but it tends to be the books with unexpected inspiration that I will remember the longest.
One lesson I really hope I remember: don't use any authority you may or may not have earned on one subject to push assertions on another subject you actually don't know quite enough about.
I picked this little tidbit up from The Math Instinct: Why You're a Mathematical Genius (Along with Lobsters, Birds, Cats, and Dogs). Now, overall, it was an entertaining read, with some good tidbits to think about. The author earned my trust over the course of the book, to the point where I would begin to take for granted some of his assertions.
And then he had to go and make an assertion about something he apparently knows a little bit less about: biology.
Clearly, the author is passionate about his belief in the fact of evolution. But he made the following assertion: "That's far too short a time frame for there to have been any major structural changes in the brain - evolution occurs over hundreds of thousands if not millions of years."
Evolution simply isn't time-bounded. This statement is false. Look at the evolution of the peppered moth. That has happened since the industrial revolution - far fewer than hundreds of thousands of years ago. It is the rapid evolution of bacteria that causes problems with preventing disease. With fewer necessary structures to worry about, higher mutation rates in viruses accelerate this even further.
Major structural changes in the brain are generally prevented because there is extensive error-correction for genes in humans - unbounded mutation would lead to far more ways of creating a human that simply don't work at all, so humans with error correction will be favored over humans without error correction. Furthermore, there are fewer selective pressures on brains - a person with a mutation that causes him or her to have some unique capability may leverage that to, perhaps, make more money, but the person without this advantageous mutation is going to have exactly the same opportunity in most cases to reproduce and pass on his or her genes.
These arguments spun around my head for a bit, and I considered trying to track down the author to vent my frustration for spoiling a perfectly good read with something that was so clearly wrong. But then I realized that, instead, I should find Keith Devlin and thank him instead. This is a blunder that is all too easy to make, and it served as a wake up call to pay attention. I know a little bit about Windows application compatibility, but I could always know more. I know a little bit about Shims, but the folks who write them know more. I respond to a lot of questions, and on more than one occasion I have had to send out a mea culpa owning up to something I was wrong about (which I feel is critical to do, and I hope I haven't forgotten one). And that's with subjects I genuinely am considered somewhat of an expert on. There are plenty of other subjects about which I am genuinely interested and will write about that are more of me thinking aloud than me speaking with authority. I just hope I never use the wrong tone so people quote me as authoritative when I am not. Particularly if I happen to be wrong. Nevertheless, I suspect I probably inadvertently will.
So, thanks, Keith, for giving me a few hours of truly pleasurable reading. And thanks even more for an otherwise innocuous oversimplification that made me realize just how important earning trust and authority is, and how much more important it is not to take that for granted or over-extend it, intentional or not. And for the reader, there is really only one antidote. Just keep reading. Trust but verify. Round out your perspectives. Because you really can't help it - it's a psychological shortcut to take assumptions into account that you are likely not even aware of. Me? I plan to be especially vigilant when discussing topics that may have a blur with subjects about which I actually (mostly) know what I'm talking about. Beyond that, I'm pretty casual, so I'm sure I'll make a few slips.
I'm not sure exactly how to thank Keith, since I don't actually know him, so I figured I'd just link to his books. They are genuinely entertaining and informative. Plus, they'll remind you of just how beautiful mathematics can be. I think we forget that sometimes.
Have a happy new year, and a very compatible 2008!