Following Fashion

Today, I received the proofs for what will likely either be my last or second to last published science paper. For those curious, you can find it here. I have another idea for a paper beyond those, but likely won't ever get to it because I'm horrible at multi-tasking. Stick me with a fork. I'm done.

What happened to me in science is that I started to get very bored with the arcane minutiae of what a typical academic scientist gets paid to do and started thinking about broader problems in my field that I thought were much more intellectually interesting.

However I found out that when you do that you can't get funding. The scientific community - at least my little room of it anyway - tends to be geared toward minutiae. Write a proposal to look broadly and sweepingly at a problem and reviewers - other people in your field - often don't even understand what you're getting after. Speaking metaphorically here, they're studying one leaf on a tree or worse yet, one stomata on a leaf. You want to study the whole forest. And their mindset - the reductionist mindset - is that you can't possibly do anything of value thinking so broadly.

My view is that reductionist work is fine when a field is young. The pieces in the puzzle are still fairly big and when you put them together they still can make a compelling picture. But for mature fields, reductionism leads to even finer scale work. At some point the scale becomes so fine that ultimately the work doesn't matter any more. And that's probably true for much of the Earth sciences today including geophysics. It's just work for work's sake.

My proposals were interesting to me. But they weren't following the fashion of reductionism. I wasn't going to get funding anymore unless I went back to the leaf scale. And that stuff was boring beyond belief for me. You can't be a scientist in an academic setting without funding. It was time to leave.

This last or next to last paper followed my anti-reductionist leanings. I wanted to examine just why the average permeability (a measure of how quickly water and other liquids can flow through cracks and pores) of the Earth was 10^(-14) m^2. It's somewhat analogous to trying to understand why human body temperature is 98.6 degrees F and not some other number. What's magic about this number for permeability? And it turns out that the reason this is so is that the average permeability of the Earth is tuned to the average slope of the Earth and average rainfall of the Earth. If it rained less, the permeability of the Earth would be smaller. If there were less mountains, the permeability of the Earth would be greater. There's a real reason that the average is what it is; the number isn't chosen out of a hat.

Now I know that sounds like a very arcane result for the non-Earth scientist, but for a geophysicist, this is a very sweeping and broad statement. Getting funding for such an analysis would be next to impossible. And even though I think it's a very important statement, it's likely that given the reductionist tendencies of my scientific community, it will be ignored. Such is life. If you aren't going to be fashionable, you are going to have problems with acceptance. You just hope fashions change in twenty years and people somehow uncover your work then, scratch their heads and wonder, why didn't anyone else notice this before?

That said, this paper is going to get published. And the reason for this is that I didn't completely ignore fashion. At first I did. Several years ago, we tried to publish this paper with techniques that are considered to be archaic. I'm basically a very old school scientist who believes that if you can't solve something with a paper and pencil - if you can't reduce things down to their essence - you aren't thinking hard enough. I tend to use mathematical techniques that were devised at the turn of the twentieth century. I pound my head against the wall trying to simplify and simplify until I get it down to something compact and elegant. Then I'm happy.

Hardly anyone uses this approach anymore, virtually no one in my generation. No one even knows how to do it. I taught it to myself by spending month after month in the Math Library reading musty textbooks. Instead, when almost everyone solves a problem nowadays, they just throw a bunch of numbers into a computer model and look at the results. It's the current fashion. I find it as ugly and abhorrent as a polyester leisure suit. But people seem to find comfort in these computer models. I don't quite understand why, but they do.

In the first version of this paper, there was no computer model. It was just paper and pencil kind of stuff. I thought it was beautiful. But reviewers thought, "What the hell is this?" And it was bounced. It was bounced from Science. It was bounced from Nature. Clearly, something had to be changed.

My co-author said, "We need a computer model." I cringed. I know how to do computer modeling - it's what I was trained to do as a Ph.D. student - but I have no stomach for it; for me, it's an ugly and tacky way of doing science. But my co-author was right. We needed to follow fashion. So he and another guy crunched a bunch of numbers in a computer model; the paper got a lot bigger and had colorful pictures based on the computer simulations. It now followed fashion. And it was accepted for publication.

Is the paper any better for this work? I'm skeptical, but I'll defer to my co-author who has a better bead on this sort of thing and says it is. The bottom line is that science is like any other endeavor in that you can't just be good or even right; you have to play well with others. If they use computer models, you better use one too.

Fashion is of course a funny thing. Just the other day, my sweetie and I were at the airport and there was this twenty-something blonde girl wearing a pink hoodie with matching pink sweatpants. On the butt of the sweatpants, the word "Juicy" was stitched in big letters. I laughed to myself and noted to my sweetie that this was the tackiest piece of clothing I'd seen in quite some time.

My sweetie informed me that, tacky though it may be, it's fashionable stuff that you buy in high end department stores. You can see it here. The thing must cost less than 10 bucks to make and retails for over $170. Damn, why didn't I get in on the ground floor of this phenomenon? It looks like a...um...leisure suit except a little more tasteless. Everything old is new again.

Maybe paper and pencil based mathematical analysis will become fashionable again some day. I'm not holding my breath.