August 06, 2007
Evolution-19: The Boeing 747 in the junkyard
(Please see here for previous posts in this series.)
As I have emphasized repeatedly in this series, the hardest thing to appreciate about evolution is how a cumulative sequence of very tiny changes can lead to big changes. The problem is that our senses can only detect gross differences between organisms and our minds can only comprehend short time scales and to appreciate evolution requires us to overcome those limitations. This is why skeptics need to actually study the details and convince themselves that it works.
I have the same problems when it comes to teaching modern physics topics like quantum mechanics or special relativity. Our senses and intuition have evolved to enable us to deal with objects that are on a human (or 'classical') size scale and traveling at speeds that are not too great. But the effects of quantum mechanics only become manifest when describing the very small, subatomic level of particles that we cannot see, and there our intuition completely breaks down. Similarly, the effects of special relativity become manifest only for objects traveling close to the speed of light, which we do not encounter in everyday life and again our intuition is incapable of dealing with it. So when physicists talk about a single electron simultaneously traveling by many different paths from a single initial starting point to a final point, or twins aging at different rates depending on their speed of travel, these ideas initially seem preposterous.
When teaching these subjects, I warn my students that their intuition is quite likely to lead them astray, that what their gut feelings tell them is reasonable or unreasonable is undependable, and that they have to constantly check those intuitive reactions by doing calculations to convince themselves that these counter-intuitive results drop out naturally from a coherent theory
The same thing is true for evolution. Mutations are too small to be visible and time scales are too long to comprehend, so one should not depend upon what seems reasonable to make judgments. Steven Pinker (How the Mind Works 1997, p. 163) points out that: "A hypothetical mouse subjected to a selection pressure so weak that it cannot be measured could nonetheless evolve to the size of an elephant in only twelve thousand generations." This is quite an amazing result. It is not at all intuitive and is hard to convince oneself that this could be possible unless one does the calculations, or trusts those who do the calculations.
But people who want to throw doubt on evolution exploit this breakdown of intuition by making statements of broad generality. For example, one often hears that the evolution of life as described by natural selection is as likely as a tornado sweeping through a junkyard and spontaneously assembling a Boeing 747 airplane. This analogy was initially proposed by astrophysicist Fred Hoyle in his 1983 book The Intelligent Universe. Hoyle and his co-worker Chandra Wickremasinghe used this example to support their alternative theory of panspermia, that life originated elsewhere in the universe and arrived on Earth from outer space via meteors.
Neither Hoyle nor Wickremasinghe are creationists and have their own reasons to want to discredit natural selection, but intelligent design creationists seized on this vivid image of the 747 in the junkyard and exploit it heavily in their anti-Darwin crusade, and Wickremasinghe has even appeared as a witness for them at some court trials.
To counter this analogy, one needs to look at exactly what natural selection says and compare it with what its opponents portray it as. Jerry Coyne (a professor in the Department of Ecology and Evolution at the University of Chicago) in a devastating review of intelligent design creationist Michael Behe's new book gives a nice example using the familiar example of throwing dice.
Take for example, some adaptation of a gene that, starting from the original organism, requires twenty mutations at twenty different locations for the desirable new feature of the organism to appear, with the mutations occurring in a specific order so that each mutation confers a slight selection advantage to the organism. Suppose that the random mutations are represented by the throw of a die and the required mutation at a particular site occurs when you throw a six. This means that it will take an average of six throws for the first mutation to occur. Recall that evolution is a step-by-step process that builds on past successes and I have already described how even a slight selection advantage is sufficient for a single mutation to become universal in the population, so this mutation will be stable. It will then take another six throws for the second advantageous mutation to occur, and so on, so that it will take an average of 120 throws for all twenty mutations to occur. If the dice is thrown at the rate of one a second, that means it will take about two minutes for all twenty mutation to have gone into effect.
What the Boeing 747 analogy does is to assume that you have twenty dice and throw them all at once and that all twenty must come up six simultaneously for the new feature to appear. The odds against this are astronomically high. At the same rate of one toss per second, this would take more than one hundred million years. As Coyne says, "This sequential way of getting twenty sixes is infinitely faster than Behe's method. And this is the way natural selection and mutation really work, not by the ludicrous scenario presented by Behe."
Arguing by analogy and example is often necessary when trying to explain esoteric points, but is also tricky and has to be done with care. No analogy is a perfect replica of the actual process and you have to make sure that the analogy you select corresponds accurately to the phenomenon being analogized as far as the crucial elements are concerned. In the case of evolution, the key point to bear in mind is that a sequential series of changes, each of which is beneficial and stable, takes much less time (i.e. is far more likely) to occur than for them to occur simultaneously. This is why intelligent design creationists try to desperately find examples of systems that (they argue) could not have occurred by sequential changes. But they have failed.
POST SCRIPT: If FDR had been like George Bush. . .
Jacob Sager Weinstein says that he "got tired of right-wingers saying, "If the media had been as hard on FDR as they are on Bush, we'd have lost World War II." So I started wondering. . . What if FDR had run his war like GWB?"
Here is the video that resulted from his musings.