In the course of the interview Collins is asked about scientists who insist that religion is irrational. To this he replies:
Certainly this has been one of the more troubling developments in the last several decades. I think that commits an enormous act of hubris, to say -- because we're now so wise about evolution and how life forms are related to each other -- that we have no more need of God. Science investigates the natural world. If God has any meaning at all, God is outside of the natural world. It is a complete misuse of the tools of science to apply them to this discussion.
Myers sums this up by saying, "Note that he's saying two things clearly here: God is outside the natural world, and that you can't apply science to the issue." Actually, though, it isn't clear he is saying this latter part. If we are not to go about arbitrarily prooftexting, we have to read what is said in context. And when we do we find that, although there's a seriously unfortunate ambiguity in Collins's "this discussion" (one not corrected in Myers's "the issue"), what Collins is saying seems a bit more sophisticated than Myers assumes. In essence the structure of the answer is the following:
(1) Claim: It's troubling that scientists are saying that religion itself is irrational.
Reason: Saying that knowledge of evolution of itself eliminates the need for God is an act of hubris.
(2) Claim: It is a misuse of the tools of science to apply them to 'this discussion'.
Reason: Science investigates the natural world. God does not fall within the scope of this investigation.
There are two ambiguities that have to be decided upon in the interpretation of this passage. The first is how (2) relates to (1). The most plausible interpretation is that (2) is supposed to be a reason for (1-Reason); i.e., it's an explanation of why the claim is really an act of hubris. This would explain (among other things) why Collins builds up to (2-Claim) rather than starting with it, and also why he sees (2) as relevant to (1). The more difficult question is what is meant by 'this discussion'. Myers takes it more precisely as "the question of the existence of gods." While this is possible, it would be odd in context. I think a more reasonable reading would take 'this discussion' as the topic of the question itself, i.e., whether religion is irrational. If taken in that way, Collins's argument is that, since God does not fall within the scope of scientific investigation, it is a misuse of the tools of science to try to use them to argue that religion is irrational. The response would then be a direct answer to the question.
Collins does go on to say, in response to the next question, "And God is certainly outside of nature. So for a scientist to say, 'I know for sure there is no God,' seems to commit a very serious logical fallacy." But this is not surprising, because it follows directly from (2-Reason). This is not enough to accuse Collins of inconsistency, as Myers does, because in nondeductive reasoning inability to prove only converges on being itself a proof/disproof to the extent that all relevant evidence is taken into account. Suppose, for example, that I take a random sampling consisting of a thousand birth certificates. This body of evidence is not of the right kind for proving that Myers does not exist, because there is relevant evidence that it does not take into account. It might, however, include evidence that Myers exists (e.g., Myers's birth certificate might be one of the certificates in the sample). To say 'It is not possible on the type of evidence in evidence-set E to prove that X does not exist' is not in general inconsistent with 'Some of the evidence in evidence-set E supports the claim that X exists'. The latter claim is weakened by the failure to account for all relevant information; but it is not inconsistent with the former. What Myers needs in order to make his charge of inconsistency stick is the stronger claim: that science is not relevant to the existence of God. But Collins appears to be committed only to the claim that science by its nature is necessarily not able to cover all the relevant evidence required to prove that God does not exist. I have no doubt that Myers would disagree with this claim; but there's nothing inconsistent about its use in Collins's argument. If scientific inquiry is necessarily limited in scope, and something falls outside that scope, it is impossible to argue that it doesn't exist merely because it falls outside that scope. This does not rule out, however, that the results of scientific inquiry may be used in an inquiry of larger scope that does include that thing. When this is recognized, one sees clearly that the real dispute here is not a matter of logical consistency, but a disagreement about the scope of scientific inquiry.
Myers, to his great credit, does appear to recognize that this is the core dispute, but doesn't adequately distinguish this from the inconsistency charge. Myers's inconsistency charge is not well-founded in terms of the textual evidence; but this does not affect one way or another his criticism that intervention places God within the scope of scientific inquiry, contrary to what Collins seems to assume. That dispute can only be resolved by determining whether there is a type of inquiry with broader scope than scientific inquiry that is capable of examining the question of whether a natural event is caused by something 'outside of the natural world'. If Collins's argument in 2-Reason is accepted, then Myers's criticism on this point can't stand, at least in the form he gives it -- if A falls outside of the scope of a given type of inquiry, then the mere fact that A has effects that fall inside the scope of that inquiry doesn't bring A within the scope of the inquiry; it just means that there are things within the scope of our inquiry that are effects of things outside it, and cannot be recognized as such without a form of inquiry that has a large scope. For an analogy: a researcher into finch physiology might study the physiological forms of finches, but not the effect of climate on the physiology of finches. To recognize those things falling within the scope of his inquiry as the effects of climate, the researcher has to appeal to an inquiry with broader scope -- one that can say something about climate. Collins can be seen as suggesting that there is an inquiry of broader scope than scientific inquiry, one that can recognize certain things within the scope of scientific inquiry as effects of things outside it. If there is no such inquiry then an argument can be made that all relevant evidence has been considered, and the inability to prove that X exists becomes a strong (albeit defeasible) argument that X does not exist. This is not quite the argument Myers makes; but it is, I suggest, closer to the argument he should be making, because it yields (I take it) the same point, while avoiding the error made in the argument he actually gives (namely, the confusion between having effects of A that fall within in the scope of inquiry and having effects of A that fall within the scope of inquiry precisely as effects of A).
Amanda Marcotte also has some comments. I can't really say much about them, because I'm having more difficulty following the threads of her argument. For instance, her comment on the (1-Claim) is "as if it’s not the height of arrogance to believe the entire universe was constructed for our moral well-being". It isn't clear how this is relevant, however. Collins's claim is that it's arrogant to claim to be able to prove something that, by the very nature of the inquiry, you can't; but taken straightforwardly, Marcotte's criticism appears to attribute arrogance to the content of the belief, and tis is not a coherent notion. (For example, no matter what p is, if I have good reason to believe p, my believing p can't be considered 'arrogant'.) Since Marcotte's phrasing suggests that she sees are response as turning the tables on Collins, perhaps the best way to interpret it is this: Collins attributes arrogance to Dawkins because, he says, he claims to be able to prove something that can't be proved given the type of inquiry he appeals to; likewise with Collins, who claims to be able to prove something that can't be proven given the type of inquiry he appeals to. The claim attributed to Collins, that "the entire universe was constructed for our moral well-being" doesn't appear (unless I missed it) to be Collins's claim. Rather, he makes the much weaker claim that "God did have an interest in the appearance, somewhere in the universe, of creatures with intelligence, with free will, with the Moral Law, with the desire to seek Him." However, with development, one might still be able to argue that this involves a claim to prove something that can't be proven given the type of inquiry appealed to. I'm not sure how that argument would work though, since Marcotte doesn't expand much on this point. I have a similar problem with most of her other claims, since she passes over them quickly, and so I find them difficult to interpret in a way that would make sense. For example, she seems in one place to imply that the fact that religious belief tends to be geographically located and passed down in families suggests that it is not true; apparently forgetting that acceptance of scientific claims also tends to be geographically located. It's an interesting question whether it is passed down in families, but it does seem at first glance likely, given that parents teach their children. Other beliefs appear to be similar, e.g., feminist ones. While the covariation with location and family is probably not the same for all beliefs, it seems unlikely, at least without further evidence, that there is any general type of belief that is not affected, because teaching is local and affected by family membership. So she must mean something else, but I don't know what it would be. She also says, "I always thought that it was interesting that the very people who argue that god is omniscient and omnipotent refuse to believe that such a being could have just planned it all out, farted it out one day and never had to monkey with it because he planned it all out and his plans are perfect like he is." But at first glance this seems to confuse 'could' and 'did' -- people who believe that God is omniscient and omnipotent by that very fact are committed to the claim that God could have done this. (Assuming it does not involve a logical contradiction; the particular way Marcotte formulates does appear to conflict with other attributes commonly believed to belong to God, like eternity, but such formulations are usually due only to the manner of speaking, and are not inconsistencies in the proper sense. I think it's fairly clear what Marcotte means.) Most of them explicitly believe it, although they would formulate it somewhat differently. But it doesn't follow from this that this is what in fact happened. So, again, Marcotte must mean something else; but I don't know what it would be.
In any case, what people need to remember, and what bloggers too often forget, is that rational criticism doesn't consist in taking the most convenient interpretation of one's opponents and criticizing that, but in taking the most charitable interpretation the evidence admits, and subjecting that to criticism. This is tricky, and no one does it perfectly, since it takes a solid analysis of the evidence, a fertile imagination in good working order, and a steady will to hold the latter accountable to the former. But some serious effort will usually take us much of the way. It doesn't seem to me that Myers and Marcotte are putting much effort in, and their criticisms seem to suffer for it. I think Alejandro's criticism of Collins, at least when he talks about Collins's comments about the Big Bang, is closer to the sort of model one should take in a case like this, and I think he identifies some clear weaknesses in Collins's argument.
