Contemporary epistemology of religion may conveniently be treated as a debate over whether evidentialism applies to religious beliefs, or whether we should instead adopt a more permissive epistemology. Here evidentialism is the initially plausible position that a belief is justified only if “it is proportioned to the evidence”. For example, suppose a local weather forecaster has noticed that over the two hundred years since records began a wetter than average Winter is followed in 85% of cases by a hotter than average Summer. Then, assuming for simplicity that the records are reliable, the forecaster is justified in believing with less than full confidence that this Winter, which is wetter than average, will be followed by a hotter than average Summer. But evidentialism implies that it would not be justified to have full belief, that is belief with 100% confidence. Again, consider someone who has a hunch that this Summer will be hotter than average but cannot justify that hunch further. Hunches are not considered evidence, so the belief is not considered justified. If, however, the huncher can cite a good track record of hunches about the weather that have turned out correct then the belief would be considered justified. For although hunches are not considered evidence, memories about past hunches are, as are the observations that corroborated the past hunches.
Evidentialism implies that full religious belief is justified only if there is conclusive evidence for it. It follows that if the arguments for there being a God, including any arguments from religious experience, are at best probable ones, no one would be justified in having a full belief that there is a God. And the same holds for other religious beliefs, such as the belief that God is not just good in a utilitarian fashion but loving, or the belief that there is an afterlife. Likewise it would be unjustified to believe even with less than full confidence that, say, Krishna is divine or that Mohammed is the last and most authoritative of the prophets, unless a good case can be made for these claims from the evidence.