Frequently, when problems with a particular account of God are presented by the atheist, those that are more sympathetic with belief will respond by pointing out that narrow objections like that don’t prove that there is no God. Perhaps all they show is that one particular account of God doesn’t make sense. For the non-believer this perpetual moving of the goal posts can be exasperating. Every earnest attempt to get clear on just what is meant by this mystifying term God meets with evasions and sidesteps.
But let’s not forget the nature of the situation here. The burden of proof for the believer is not merely to change the story about God until they land upon one that doesn’t seem to have the problems pointed out by the atheist. Suppose that the atheist’s objections drive the believer back to some sufficiently vague, general, or abstract account of God that seems to be in less acute conceptual crisis. Has the believer now be vindicated? Is believing now epistemically inculpable? No. What this exercise may have produced is an account of what God could possibly be that is motivated by several arguments about what he cannot be. But redefining or reconceiving God in some fashion that lessens the blow of the problem of evil, or reduces the cognitive dissonance surrounding omnipotence, doesn’t give us reasons to think that that being actually exists. All that might show is that if there were a God then the sorts of being he could be are narrower than the believer held at the start. It’s no more acceptable to conclude that God is actual on the basis that some description is possible than it is reasonable to conclude that there are weapons of mass destruction in Iraq because it is possible that there are. Belief in God doesn’t become reasonable merely by being free of obvious inconsistencies. There’s still the question of evidence—what are the grounds for believing that such a being is real? Unless we can answer that question, having a viable story of what God might be isn’t any more rationally convincing than the possibility that there are invisible, undetectable elves in my garden. Sure, their existence is compatible with the evidence—they can’t be seen or detected. But the mere possibility that something is real isn’t sufficient to make believing that it is actually real supported by the evidence. An account of God needs to make sense and it has to have some substantial evidence in its favor. We shouldn’t confuse far-fetched possibilities with justifications for realities.
In this shuffle, an important point often gets lost. There are some remarkably sophisticated and carefully articulated accounts of God out there being presented by philosophers. Consider this recent definition of omnipotence from Flint and Fredoso:
S is omnipotent at t in W if and only if for any state of affairs p and world-type-for-S Ls such that p is not a member of Ls, if there is a world W* such that Ls is true in both W and W*, and W* shares the same history with W at t, and at t in W* someone actualizes p, then S has the power at t in W to actualize p (Flint & Freddoso 1983, p. 99).
Other discussions of God’s knowledge, God’s consciousness, and God’s goodness are similarly technical and arcane. Recent explanations of omniscience have included substantial efforts to define God’s knowledge in light of the restrictions that Cantor’s theorem and Godel’s Incompleteness Theorem cast on truth.
But in the history of the origins of human religions, none of the accounts of God started out reflecting attempts to evade counter examples the way the modern accounts do. Originally, God was a simple, anthropomorphic figure with various character flaws, and little philosophical sophistication. Richard Gale portrays this dialogue with excess charity. Doubter’s questions and challenges to the conception of God have helped clarify the nature of God. One has to wonder, if these modern accounts of God’s nature are closer to accurate, then why is there no hint of any of them in any of the original accounts of God that founded the major religious traditions? Why did it take thousands of years to uncover God’s nature, particularly since by most accounts he desires that his nature and existence be known? And why is it only with non-believers persistent objections that God’s true nature slowly comes into focus?
No doubt it is a useful and challenging project to perpetually devise new accounts of God and his properties that resist more and more inventive counter examples. But we must keep in mind that doing so only addresses the hypothetical: if there were a God, then this is what he’d be like, not if we can give a sophisticated version of God’s description that isn’t obviously problematic, then God is real.
The same goes for discussions about God outside of the philosophy journals. If the non-believer is troubled by the seeming incompatibility of God’s existence with pointless suffering, and the believer responds, “God’s plan, God’s goodness, and God’s nature are beyond our comprehension,” that doesn’t give us any reason to think there actually is one. In fact, these defense-by-ignorance responses seem to give us even more reason to think that believing is not reasonable or supported by our evidence.
So the question remains: there’s not much point to constructing some seemingly internally consistent account of God that answers complicated counter examples unless we have some substantial reason to think that enterprise is going somewhere. Are we working out the details of something that we know to be true? If so, then on what grounds? (And will those grounds indicate that God's knowledge is defined in part by Cantor's and Godel's theorems?)
Or are we just exploring the conceptual details of a belief that’s going to be procured by faith in the end? If believing by faith is an acceptable route, then why not abandon internal consistency along with any need for evidence?