To note from Alvin Plantinga’s lecture at Biola University on the Evolutionary Argument Against Naturalism (EEAN) , Richard Dawkins once leaned over and remarked to A.J. Ayer at one of those elegant, candle-lit, bibulous Oxford college dinners that he couldn’t imagine being an atheist before 1859 (the year Darwin’s Origin of Species was published); “…although atheism might have been logically tenable before Darwin, Darwin made it possible to be an intellectually fulfilled atheist.” However, according to Plantinga’s argument, this statement may seem to lack substance.
First, just what do we mean by Naturalism? In short, according to Ernest Sosa, Naturalism “eschews or rejects appeal to the supernatural, and traces our origins back to blind and uncaring forces” . Thus, physical reality is all that exists; no supernatural or personal forces lie beyond the scope of the universe. Hence, “the universe is all there is, and everything operates by natural law” . Plantinga’s EEAN notes three given areas of consideration:
- (1) Theism: we human beings have been created by a wholly good, all powerful and all knowing person: one who has knowledge, aims and intentions and acts to accomplish them.
- (2) Naturalism: You can consider my definition above, or, as Plantinga says: “The theistic picture minus God.”
- (3) Cognitive Faculties: The powers or faculties of capacities whereby we have knowledge or form belief: memory, perception, reason, and maybe others.
Plantinga’s argument can be stated as such: Given the acceptance of our belief in Naturalism, what is the reliability of that belief given our arisal through successive evolutionary history human qua human? or, as Ernest Sosa further explains: “[T]he probability that our cognitive faculties are reliable must be either quite low or at best inscrutable. This defeats any belief we may have in the reliability of our faculties. Absent such belief, finally, we are deprived also of epistemic warrant (Authority, justification) for all beliefs deriving from such faculties” . The thrust of the argument however, is its devastating conclusion: “But among those beliefs is the very belief in naturalism, which therefore defeats itself” .
Consider the probability function P(R/N&E). N stands for Metaphysical Naturalism – N excludes the existence of God as understood under traditional theism; E stands for the arisal of our cognitive faculties by way of evolution; R is the claim that our cognitive faculties are reliable. Now, what is the probability of R, given N&E? First, consider a few things regarding the matter of conditional probability.
Take for example that we have some given factor H, which means that an individual will live a healthy lifestyle passed the age of 75. So, what is the probability of H on the condition that an individual never exercises, doesn’t eat right, and has high blood pressure? Well, the obvious answer seems to be that the probability of H given those factors would be considerably low. However, the probability of H will be much higher on the condition that if the individual exercises, eats right, and does have a lower blood pressure.
In a likewise fashion, what is the probability of R given metaphysical naturalism and the arisal of our faculties by way of evolution? According to Charles Darwin himself, he might consider the probability of R to be fairly low. As he once wrote in his Autobiography, “But then arises the doubt, can the mind of man, which has, as I fully believe, been developed from a mind as low as that possessed by the lowest animals, be trusted when it draws such grand conclusions?” 
Thus, the thrust of the probability of R packs a weak punch. Since the naturalist contends that our faculties derive from brute forces, it seems that we cannot rationally hold to our “implicit trusts” of the reliability of our faculties. However, even more notable from this point, is that if we cannot trust our faculties, then we cannot rationally trust anything that is a deliverance of those faculties. Naturalism is a deliverance from our faculties, therefore, we cannot rationally hold on to that belief. Consider the thrust of this conclusion:
Consider now any subjects who face the question whether their faculties are reliable, and realize that if they do have reliable faculties, this is a contingent matter, and that they cannot just assume so and let it go at that. Given the contingency of the reliability of their faculties, what assurance is there that though they might be unreliable, in fact they are reliable? Wouldn’t the inability to give a rational, nonarbitrary answer to this question itself constitute a problem? 
The contingent nature of our cognitive faculties constitutes that they are one particular way and could have possibly been another, but aren’t; thus, the naturalist ignoring this question seems to catch himself in a bit of a problem. To end with Plantinga: “The conclusion to be drawn, therefore, is that the conjunction of naturalism with evolutionary theory is self-defeating: it provides for itself an undefeated defeater.”
-  See the outline at http://www.calvin.edu/academic/philosophy/virtual_library/articles/plantinga_alvin/an_evolutionary_argument_against_naturalism.pdf
-  Ernest Sosa, Natural Theology and Natural Atheology in Alvin Plantinga, ed. Deane-Peter Baker (Cambridge University Press: 2007) p. 95
-  Norman Geisler, Introduction to Philosophy (Baker Books: 1980) p. 432
-  Sosa, 95
-  Ibid.
-  Charles Darwin. Autobiography in Portable Atheist, ed. Christoper Hitchens (Da Capo: 2007) p. 96
-  Sosa, 101