Science Slip-ups

kw: book reviews, nonfiction, science, scientists, errors

I couldn't pass up the title: Brilliant Blunders: From Darwin to Einstein, Colossal Mistakes by Great Scientists That Changed Our Understanding of Life and the Universe, by Mario Livio. I was hoping it would be something better than one-upmanship ("See how much more we know now; aren't we good?"), and I was happily right. I've read an article or two by Dr. Livio—anyone has who has read publications by the STSCI—but none of his books. This one is a pleasure.

There isn't a way to review this book in detail without giving too much away. The concept, however, is clear. Those scientists we call truly great had such useful intuition that even when they were wrong, their errors pointed the way for others who'd never have gotten a better theory otherwise.

To take one example of the five in the book, Lord Kelvin (né William Thomson) calculated an age for the earth, based on temperatures in deep mines, and certain assumptions of thermal conductivity and initial temperature, of 100 million years. He was wise enough to do a bit of sensitivity analysis, and stated that the actual value would probably fall in the range 30-400 my. When others later showed that his assumption of solid conduction may not be correct, and that the earth could be much older, he stubbornly denied it. Yet, he was but one who had said the earth might be older if a source of energy other than remnant heat were involved. He died in 1907, so he didn't learn that radioactive decay is just such an energy source.

I wonder, is Kelvin's earlier underestimate really a blunder? Under the assumptions he made, it was quite valid. And his aim was not quite what we think. There were, at that time, two opposed camps. Biblical literalists maintained the dogma that the Earth is no more than some 6-8 thousand years old. Bishop Ussher's calculations putting Genesis 1:1 as the evening preceding October 23, 4004 BC was well known, but not without competitors. Yet few theologians, whether Christian or Jewish, were amenable to the geologists' contention that, as James Hutton had written in 1788, the Earth showed "no vestige of a beginning, no prospect of an end." Kelvin wished to correct both errors. He showed first that the rise in temperature with depth in mines indicated an Earth that had not always been as it is now. This should give pause to a geologist who might surmise there is infinite time for modern landforms to develop. Yet Kelvin then showed that deep time was still pretty deep, at least some tens of millions of years, not a few thousands. I find it astonishing that so many today (mostly in America) still deny a "geological" age for the Earth.

In the last section, the author explores the Cosmological Constant, Λ, which, the folklore tells us, Einstein called his "biggest blunder". Then he digs further, and determines that the "biggest blunder" statement probably originated with George Gamow, not Einstein. Einstein himself thought it might be an error once cosmological expansion was discovered. But he was unsure if it ought to be eliminated. Considering that Λ is now called "dark energy", and contributes about 75% of the gravity in the universe, his "error" or "blunder" (pick your favorite) makes up 3/4 of everything!

Oh, and just by the way, I am not certain if this interpretation of Λ is true. It is based on finicky measurements of supernova brightnesses, particularly in the 6-7 ga range of ages. I have yet to learn of a proper study of the effect of metallicity on peak luminosity of Type 1a supernovae, particularly the C/O composition of the original star. The general metallicity of the universe some 7 gy ago was about half what it is today, and was even less some half a billion years earlier when the stars that went "boom" at 7 ga were formed. Do we have another blunder in the making?

It has been occasionally said that some errors are so bad they are "not even wrong." The "blunders" limned in this book might be said to be "not quite wrong". We now have a better way to explain (theorize) each, but will future scientists look back at today's best theories as "not quite wrong, yet not quite right either"? Count on it.