OK oddly enough I actually took the atheists side against a Christian in epistemology, of all things. But that's because he was arguing for Van Until and prosper stuff.
That doesn't mean, to any degree, that the atheist are off the hook epistemically. Atheist epistemology is also entirely a crass approach to understanding Hue. Most of the one's I run into try a 1x1 correspondence between what they see and the universe is. They take a kind of "what you see is what get" universe.
Science knew that empiricism was not enough for an epistemology or a science that's they invented parsimony. But the atheists seem unconvinced. They demand that unless a belief system is based directly upon empirical observation it must be doubts; of course they miss how much of their own system (naturalism and natural sciences) are based upon extrapolations from the correspondence theory.
There are two major problems, or kinds of problems with this failed Human approach. Before I go into that, I want to stress that it is a failed an approach. Hue is often quoted for one thing or another, but few philosophers today claim to be Humans and few follow Hue to the letter precisely because his way was beaten a long time ago.
IN the development of philosophy there is an order to way things progressed. Each philopsher in turn was defeated and his system shown to be in error, up to the time of Ant. Descartes brought the rationalist approach, reason could do it all. Hue showed that reason couldn't do it all, we had base our approach upon what we observe and what we extrapolate from that. Ant in turn showed that Hum's empiricism was inadequate because empiricism by itself, while questioning everything, can't actually establish anything.
The two areas of problem are these:
(1) Can't get at all the data
(2) The data you do get could be part of the illusion.
As for no 1, its obvious that our take on the universe is just a small and unrepresentative sample. We dot' have nearly enough observations of the universe as a whole to rule out things the atheists don't like, such as God and the SND. So the atheist wind up creation [i]er zots[/i] criteria which are really nothing more than circular reasoning; such as the atheist take on miracles. Like Hue says "it doesn't' happen often enough" of course when it does, you rule it out because it's not supposed to happen according to the atheist guideline based upon their limited sample.
As for the second problem, there is no way that this kind of observation can get under the inherent problematic of epistemology; that of the epistemologist fallacy. The Epistemologist fallacy is the idea that I can get outside of my own perceptions and check them. Any check I take on my perceptions depend for its validity upon my perceptions. This is a catch 22 that cannot be overcome.
What this amounts to is we have to make a judgment. Our epistemic standards must always and forever turn upon a leap of faith, a judgment a decision arrived at through arbitrary means and through educated guess work.
To prevent the problem in the no. 1 category, we should adopt a phenomenological approach. that would mean that we say what can happen int he universe is open ended we allow the sense data itself to suggest its own categories.