June 21, 2007
Evolution-2: The lack of evidence for perfect design
In the first post in this series, I showed with the example of a soap spray nozzle how natural design could come up with systems whose intricacy and complexity is such that it was superior to the efforts of intelligent human designers. But what about the argument that a god-like designer would be able to come up with an even better nozzle design? It is true that if we allow for the existence of such a designer, we could get the best possible design for a nozzle. The catch is that assuming that god is a perfect designer opens up a whole set of new problems, not the least of which is why if god is so powerful he would need any kind of nozzle at all and not simply create any kind of spray he/she needed.
Let me start with a limitation of natural selection. There is a well known result about any method of solving a problem that starts (like natural selection does) with some state, tries out small variations, selects the one that shows the greatest improvement over the starting point, tries out variations based on that new state, selects the best one again, and so on, which is exactly the way that natural selection works. The problem is that while you will end up with a better result than the one with which you started, it may not be the very best solution that is conceivable. Such algorithms result in finding a locally optimal solution but not a globally optimal one.
As an example, suppose you are in open ground and totally in the dark. For some reason, you need to get to the highest point in the ground, say because flooding is occurring and you know the water is rising very slowly. (The specific reasons are not important. The point is to have some kind of external pressure that drives the selection process in one direction.) You could gingerly take small steps in every direction, see which way went up the most, and move one step in that direction. Then you again take tiny steps in all directions and select the one direction that moved up most, and move to that position. And so on. By repeatedly doing this, you are guaranteed to arrive at a peak.
(This is how natural selection works, though to be a more accurate analogy, we need to start with many people at the starting point, have couples move in each direction, have only the couples that get to higher ground survive while the others drown, have those successful couples produce lots of children at that location, who then move as couples in different directions, and so on.)
The catch is that the peak you arrive at may not be the highest peak in the vicinity. If a yet higher peak were to be separated from your initial starting point by even a small dip in the ground, you would miss it using this algorithm, since it does not allow you to make a short-term disadvantageous change in anticipation of future benefits. Natural selection is not guaranteed to produce the very best or the most perfect solution or design. It instead works on a 'just good enough for now' basis. This means that biological systems do not necessarily make progress towards perfection even though they do become more complex over time.
Now a god-like designer would presumably be able to see all the possible solutions (even in the dark) and pick the one that is best overall and guide you to that point. But the interesting thing is that the results of nature are more consistent with the 'just good enough for now' strategy of natural selection than that of a perfect designer. After all, we know that while nature's designs (by which I mean designs arrived at by natural selection) are marvelously adapted and successful for many things, they are by no means perfect.
As Sean B. Carroll says in his book The Making of the Fittest (2006) which examines the DNA evidence for natural selection:
Modern species are not better equipped than their ancestors, they are mostly just different. They have often gained some coding information in their DNA and, as I have shown throughout this chapter, they have often lost some, or even many, genes and capabilities along the way.
The fossilization and loss of genes are powerful arguments against notions of "design" or intent in the making of species. In the evolution of the leprosy bacterium, for example, we don't see evidence that this pathogen was designed. Rather, we see that the organism is a stripped-down version of a mycobacterium, which still carries around over a thousand useless, broken genes that are vestiges of its ancestry. Similarly, we carry around the genetic vestiges of an olfactory system that was once much more acute than what we have today.
The patterns of gain and loss seen species' DNA are exactly what we should expect if natural selection acts only in the present, and not as an engineer or designer would. Natural selection cannot preserve what is not being used and it cannot plan for the future. (p. 136)
The very fact that it is estimated that over 99% of all the species that ever existed are now extinct is powerful evidence against perfect creation. The only way out of this for the religious believer is to think that god, although perfect, is somehow holding back and deliberately creating imperfections and thus making it merely look like something like natural selection is at work. Or god does not interfere at all, ever in the natural selection process once it began way back at the beginning of life. Or is simply careless and produces sloppy designs.
Darwin himself, based on his careful study of plants and animals, found it hard to believe in the idea of an intelligent designer. His biographer David Quammen in the book The Reluctant Mr. Darwin (2006, p. 120) highlights the kinds of questions that troubled Darwin, and which he expressed in letters to the Harvard botanist Asa Gray, who believed in the idea of special creation of humans.
I cannot see, as plainly as others do, evidence of design & beneficence on all sides of us. There seems to me too much misery in the world.
. . .
Why would a benevolent God design ichneumon wasps, for instance, with the habit of laying eggs inside living caterpillars, so that the wasp larvae hatch and devour their hosts from inside out? Why would such a God design cats that torture mice for amusement? Why would a child be born with brain damage, facing a life of idiocy?
. . .
An innocent & good man stands under [a] tree & is killed by [a] flash of lightning. Do you believe (& I really shd like to hear) that God designedly killed this man? Many or most persons do believe this; I can't and don't.
The question of pointless suffering and loss were not hypothetical issues for Darwin. He had been devastated when his own beloved daughter Annie had, at the age of ten, died after a long and mysterious and undiagnosed wasting illness. Darwin seemed to feel that such things were incompatible with a benevolent deity. As Quammen writes, "Any god who controlled events on Earth closely enough to preordain such an occurrence – or to permit it, if permission was necessary – wasn't one that Darwin could take seriously."
Darwin's theory of evolution by natural selection, although not aimed at doing so, ultimately provided the basis on which belief in a designer god, and thus god itself, could be abandoned.