Over on my personal blog, I have, for the last six weeks or so, been reflecting on Jordan Howard Sobel's 2003 Logic and Theism: Arguments For and Against the Existence of God, and I have been invited to continue my series here at Prosblogion. Earlier posts have discussed, among other things, a variety of ontological and cosmological arguments. We join the discussion in chapter 7, Sobel's critique of teleological (design) arguments.
Much of this chapter is devoted to Hume interpretation and to explaining Bayesianism. The latter seems to be one of several places where Sobel has not decided whether he is writing a textbook or a monograph. As for the former, the 'analogical' version of the teleological argument is, I think, not the strongest version and, although I haven't conducted a survey of the various treatments, I would be surprised if Hume's version turned out to be the best. After all, Hume is at most a half-hearted supporter of the argument; even he doesn't think his argument is all that compelling. (Because the argument is contained in the Dialogues Concerning Natural Religion, there are even some who doubt whether Hume means to endorse it at all.)
The first really interesting thing in this chapter is the discussion of whether the appearance of design in the biological world, or other facts about biology, might manage to make theistic evolution more probable than unguided evolution (pp. 272-277). Sobel makes essentially two points: first, with the possible exception of pre-biotic evolution (the development of the first life forms) there aren't really any 'gaps' left for a God to plug, and, second, that given what we now know, evolution really doesn't look planned or, at least, whoever was doing the planning could've done a better job of it.
The first point, I think, is completely misguided, but I am not inclined to blame Sobel because so many of his opponents are misguided in this way. Hume (according to Sobel) believed that some kind of indefinite and probably imperfect designer was needed to bring about life. 'Intelligent Design' advocates frequently claim that there is some feature of the world that must have happened by a supernatural entity interfering with the course of nature. Hume didn't mean to be defending the religious tradition, but many of the ID folks are trying to do just that. Now, a frequently cited problem with 'God-of-the-gaps' arguments is that history shows that 'gaps' have a tendency to get plugged with perfectly naturalistic solutions. Some ID folks have tried to solve this by giving some kind of reason for thinking that some of the 'gaps' are special and unlikely to be plugged. For those who are trying to defend the religious tradition, however, there is a bigger problem: the doctrine of divine sovereignty. The God of western monotheism can never be a 'God-of-the-gaps': either he is Lord of all creation, or he does not exist. This is not, in itself, an argument against law-breaking miracles (though I've got some of those); it is just to say that, from the perspective of the religious tradition, we must attribute the whole natural order to God, rather than only crediting God with deviations from the natural order. In my view, then, the plugging of 'gaps' should not be troubling to traditional theists, though it might be troubling for non-traditional theists/deists such as Hume might have been. This, let it be stressed, is because even if there were unfillable 'gaps,' this would not help to support theism. I would even go so far as to say that such 'gaps' would be evidence against the existence of God, as traditionally conceived. (In addition to my paper, see Christine Overall, "Miracles as Evidence Against the Existence of God," The Southern Journal of Philosophy 23 (1985): 347-353. Also, I recently stumbled upon this short news item which quotes Guy Consolmagno, the Vatican astronomer, comparing ID to Paganism, on the same grounds I've mentioned.)
Sobel's second point is more interesting. Although Sobel doesn't consider a theory that has God accomplishing his purposes through natural evolutionary processes without law-breaking interventions (this is the account I favor), he does point to some issues that should trouble evolutionary theists. The evolutionary process is brutal and seems to proceed by fits and starts. Many species die out; many animals have useless organs of various kinds; the system depends crucially on death and suffering. Wouldn't we expect God to do better?
On the other hand, from an engineering/design principles perspective, evolution is really quite pretty: it's a self-improving system. And not just self-improving like Bayesian learning for artificial intelligences; self-improving like going from ooze to the human brain. That's quite an improvement! There are problems about a benevolent God accomplishing his purposes through death and suffering, and I don't mean to minimize those. But they may be counter-balanced, at least to some degree, by the sheer impressiveness of the system. Furthermore, since Sobel is interested in considering non-traditional gods (p. 259), we might consider a designer who doesn't care about pain and suffering and just wants to generate sophisticated and intelligent creatures from the simplest basic principles possible. Such a designer would, it seems, be very likely to choose a process like evolution.
It seems to me, then, that evolutionary theory has two effects on the debate at this point: (1) it rules out some, but by no means all, non-traditional gods, and (2) it introduces some new complexity to our treatment of the problem of evil. However, contrary to Sobel's assertions (pp. 272-274), it has not undermined any argument for the traditional God which was any good to begin with.