Trumpets of the Sky

R. B. Laughlin
Korea Advanced Institute for Science and Technology
373 Guseong-dong, Yuseong-gu, Daejeon 305-711, Republic of Korea

Chosun Ilbo, January 17, 2005

It snowed briefly the other day while I was stuck in traffic, and I found myself thinking about snowflakes. There wasn't anything better to do, so I was not wasting time. You may remember learning that snowflakes have six sides, are beautifully symmetrical, come in many fantastic shapes, and are never exactly alike. (The latter never made any sense to me, incidentally. If no two flakes are alike, why are the six sides of a given flake alike?) As often happens in off-and-on snowing, the stuff wasn't coming down as flakes at all but fluffy blobs that descended cautiously and indirectly, like mosquitoes do. Before each blob committed suicide on my windshield you could see it was a lacy cotton ball---a society of flakes clumped together out of convenience, perhaps, although they melted too fast for me to tell for sure. The clumps were all different, however, like people.

When scientists start bragging that they know everything, just ask them to explain snow. They can't. As a kid you also learn that snowflakes have the shapes they do because ice crystals have six sides. This is ridiculous. It is like saying that the pyramids of Egypt have square footprints because they are built with rectangular blocks. Something causes an effect if modifying it makes an important change while modifying something else doesn't. As it turns out, we have much experimental evidence that the shape trends of snowflakes are wildly sensitive to environmental factors such as humidity, wind speed, temperature of the air in which they form, and so forth. Just below freezing, for example, you tend to get mostly hexagonal plates. A few degrees below that you get mostly needles. A few degrees below that you get mostly hollow columns. A few degrees below that you get mostly hexagons with regular indentations. Only when the air is extremely cold do you get the beautiful branching dendrites of the familiar snowflake icon. Even then there are complex trends in the pattern that are sensitive to other factors. Also, snow won't form if the air is cleansed of the microscopic bits of dust and sea salt it normally contains. Sometimes you hear very aggressive computer people claim they can explain snow, but it's just not true.

After thinking about this a while you realize that snow formation has a lot in common with economic life: it is very beautiful, it rests on simple fundamentals, it is too sophisticated to anticipate from these fundamentals, it has a few important broad-brush characteristics that are easy to identify and measure, it can be manipulated through primitive environmental controls but not otherwise, and it happens all by itself without micromanagement.

Science and technology have been immensely important in creating modern economies, in particular the Korean economy. Koreans often go so far to argue that science and technology have caused their economic miracle, and that declining interest in these things among the young must be fought at all costs if the country is to have a future. I am inspired and deeply touched by this attitude, but I have a big problem with the part about ignoring costs. Technical things are extremely important in all advanced countries, but I don't think they explain the economy any more than ice crystals explain snowflakes. Economies are like the many primitive things in nature that happen by themselves for reasons nobody completely understands. They are not very amenable to planning. I therefore think ignoring the costs of science and technology is exactly the wrong thing to do. It's a practice that inevitably winds up burdening the economy to facilitate science and technology, rather than the reverse. That helps no one.

At any rate, it came to me in the car that the economy is, in fact, like this snowstorm: in the words of Ralph Waldo Emerson, it is fanciful and savage---announced by all the trumpets of the sky. It is delightful and sophisticated. The whole idea of explaining it with science is foolishness.


[Copyright 2005 R. B. Laughlin. The author grants permission to copy, distribute, display, and perform this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.]