As we head into Evolution Sunday, I offer this second installment in a series of reviews of Alvin Plantinga's long-awaited book, Where the Conflict Really Lies: Science, Religion, and Naturalism. (For the first installment, see here.)

Plantinga's goal in this book is to show that there is only "superficial conflict but deep concord between science and theistic religion." So it makes sense that he would start with biological evolution, since this is the part of science where many perceive a conflict. The way Plantinga handles the meaning of "random" and "Darwinism" is complicated, so I'm going to postpone discussion of these matters until the next installment and focus here on his broader argument.

Plantinga gets quickly and clearly to the central point of the alleged conflict between evolutionary theory and Christian theism. It reduces almost entirely to the question of whether the origin and history of life is guided or unguided. A teleological view of evolution can be reconciled with Christian theism. "What is not consistent with Christian belief," he writes, "is the claim that this process of evolution is unguided -- that no personal agent, not even God, has guided, directed, orchestrated, or shaped it." According to Plantinga, this is not a claim or a finding of science per se (more on that in the next installment); nevertheless, "there is a veritable choir of extremely distinguished experts insisting that this process is unguided" (p. 12).

Therefore, Plantinga spends ample time responding to the arguments of a few prominent soloists in the choir. First up: Richard Dawkins. However one judges evolutionary theory in general, Dawkins unambiguously claims, as the subtitle of The Blind Watchmaker puts it, that "the evidence of evolution reveals a universe without design." If Dawkins's claim is true, Plantinga notes, then the Christian belief that God "has created human beings, and created them in his own image," is false. So, what of Dawkins's argument?

Despite the promise of the subtitle to deal with "evidence," Dawkins's arguments in The Blind Watchmaker aren't dense with data from biological research. They are largely of the Darwinian storytelling variety. Dawkins spends a lot of time describing how something like the mammalian eye could have evolved in an unguided Darwinian fashion from an earlier organism that lacks such an eye. The series must be continuous and each step, of course, must either confer on its possessor some survival advantage, or at least not exact a large cost in terms of survival. If one has an active imagination and is content with unrealistically Lego-like, bottom-up treatments of organisms, it's easy to conjure up a series of such steps.

Plantinga does an admirable job of analyzing and summarizing Dawkins's basic argument, and even manages to compress several of Dawkins's key questions into one Big Question:

(BQ) Is there a path through organic space connecting, say, some ancient population of unicellular life with the human eye, where each point on the path could plausibly have come from a preceding point by way of a heritable random genetic mutation that was adaptively useful, and that could plausibly then have spread through the appropriate population by way of unguided natural selection?

Dawkins answers this question, in effect, by saying that he feels it's quite plausible. To which the obvious response is: So what? Of course Dawkins feels it's plausible. But that's not much of an argument. Others, such as Michael Behe, argue that at least some such pathways for some organs or molecular machines are quite implausible and unlikely to have been traversed, if one excludes the possibility of intelligent design. So Plantinga rightly concludes that all Dawkins's argument shows, at best, is that "given a couple of assumptions, . . . it is not astronomically improbable that the living world was produced by unguided evolution and hence without design."

There's nothing wrong with an argument that comes to a modest conclusion; but Dawkins claims to have shown that the evidence of evolution reveals a universe without design. As a result, he's guilty of severe overreach. This is especially obvious when Plantinga reduces Dawkins's larger argument to its logical core. It ain't pretty. In The Blind Watchmaker, Dawkins seems to be arguing

p is not astronomically improbable

therefore

P.

That argument form is, Plantinga observes, "a bit unprepossessing" (p. 25). Normally, we don't think that if we can show that some event is not astronomically improbable, then we've established it.

Since Dawkins is a smart guy, though, surely he wouldn't offer such a terrible argument, would he?

Well, smart people make terrible arguments all the time. Still, Plantinga pursues the possibility that Dawkins is drawing on some unstated premises that, when stated, might make his argument a little less bad. I think this is the correct approach, since these premises are very much in evidence throughout Dawkins's writings.

The first is Dawkins's refrain that since we're seeking to explain organized complexity, we can't just postulate it, as we would do with a designer. As Dawkins says, "Invoking a supernatural Designer is to explain precisely nothing."

This is similar to the old "Who designed the designer?" retort. It is, in my opinion, one of the silliest arguments around. And Plantinga shows just that. First of all, even if X would need to be designed by something more complex than X, that doesn't disqualify one from inferring that X is designed. If that were a reliable rule of reasoning, then we could never invoke design. The possibility of a follow up question to an answer doesn't show that the answer is a bad one, especially when one is simply trying to offer a proximate rather than an ultimate explanation.

At some point, of course, the regress of explanation must stop at an ultimate explanation. That's true for everyone. The theist thinks God is the necessarily existing thing on which everything else depends. The materialist has an alternative candidate for ultimate explanation. But Dawkins doesn't come close to considering these sorts of possibilities and defending the materialist alternative. He really seems to think that the very possibility of a follow up question disqualifies design as a real explanation.

The second hidden premise in Dawkins's argument could be his claim that the existence of God is stupendously unlikely. If we add that assertion to Dawkins's argument above -- p is not astronomically improbable, therefore P -- then it makes more sense. If God (or intelligent design generally) is extremely unlikely, and the organized complexity in biology is extremely unlikely give chance alone, then something like Darwinism has to be true. It's the only show in town. If there are three options, and two are extremely unlikely, then perhaps all one needs to show is that the third option is not astronomically improbable. And Dawkins has done that.

Plantinga observes, briefly, that if this is Dawkins's argument, then it depends very little, if at all, on scientific evidence. It's an eliminative argument based on Dawkins's intuitions about what is probable and plausible. We're left with Dawkins's personal intuitions because his arguments for God's improbability are weak. His main argument depends, again, on this idea that anything that designed a biological object would have to be more complex than that biological object. I think what Dawkins has in mind is something like this: If it's unlikely that a bacterial flagellum could have arisen by chance or the Darwinian mechanism, then any agent that designed the flagellum would be even less likely.

Plantinga finds a fatal problem here. Dawkins defines complexity as the property of something that has parts "arranged in a way that is unlikely to have arisen by chance alone." But God is immaterial and so doesn't have parts in this sense. According to Dawkins's own definition of complexity, therefore, God is not complex. One can make a similar point without invoking God. It doesn't follow that because an agent can produce organized complexity, that the agent is complex. (Frankly, I don't think it makes sense to refer to any agent as "complex.") Organized complexity might very well be a reliable sign of an intelligent agent. So Dawkins's argument against the improbability of God's existence, and, a fortiori, the improbability of intelligent design, fails.

The take-home lesson, Plantinga concludes, is "that Dawkins gives us no reason whatever to think that current biological science is in conflict with Christian belief." Plantinga, in contrast, gives us plenty of reasons to agree.