G.G.: No, they may both be rational. But suppose you and your theist friend are equally adept at reasoning, equally informed about relevant evidence, equally honest and fair-minded — suppose, that is, you are what philosophers call epistemic peers: equally reliable as knowers. Then shouldn’t each of you recognize that you’re no more likely to be right than your peer is, and so both retreat to an agnostic position?
L.A.: Yes, this is an interesting puzzle in the abstract: How could two epistemic peers — two equally rational, equally well-informed thinkers — fail to converge on the same opinions? But it is not a problem in the real world. In the real world, there are no epistemic peers — no matter how similar our experiences and our psychological capacities, no two of us are exactly alike, and any difference in either of these respects can be rationally relevant to what we believe.
G.G.: So is your point that we always have reason to think that people who disagree are not epistemic peers?
L.A.: It’s worse than that. The whole notion of epistemic peers belongs only to the abstract study of knowledge, and has no role to play in real life. Take the notion of “equal cognitive powers”: speaking in terms of real human minds, we have no idea how to seriously compare the cognitive powers of two people.
Very much agreed with Antony on this; I discussed something in the broad vicinity of Antony's argument in another post on epistemic peers. In any case, it's nice to see a top-notch epistemologist pointing it out.
ADDED LATER
Helen De Cruz has an interesting counterargument on the point. It seems to me that the argument conflates epistemic peerage with another kind of peerage, which I call intellectual peerage in the post at the link in the last paragraph above. But this is a purely practical assessment. And contrary to her suggestion, I see no way in which it can salvage Gutting's argument; why would one think that disagreement of informed people or even experts should lead to agnosticism, given that the differences will in principle allow you to assess which one is right, or more likely to be right? All we seem to get is that each person should continue to develop their views, taking into account the arguments and reasons of the other, not retreat to agnosticism.