I find Alvin Plantinga’s evolutionary argument against naturalism (it should probably be called the ‘evolutionary argument against correct belief formation’, but I’ll get to that in a second) to be perfectly permissible (with certain revisions). This might come as a shock to some, considering that I am an open agnostic atheist about gods. In fact, I am far more of an apathist in spirit, for positing the existence of gods is a non-explanation for, I think, a non-problem. Plantinga’s argument boils down to little more than an attack against assumptions in epistemology, not against ‘naturalism’: evolution and our traditional theories of knowledge are incompatible.
Rephrasing Plantinga’s argument a bit by jettisoning his Bayesian framework and religious assumptions, we have the fact that evolutionary pressures directly select for (approximately) correct behaviors, and do not directly select for (approximately) true content of beliefs or justifiers for our beliefs, undermines the reliability of belief in light of evolution. Some beliefs may be true, others may be false, but we cannot tell them apart through any truth-detecting organ (the problem of the criterion); true beliefs may be the result of improper justifiers (epistemic luck); or correct justifiers for true beliefs may have no causal connection to the belief-formation processes (Gettier counter-examples, ‘demons’, and so on).
(As an aside, I’ve been disappointed in the whole discussion around Plantinga’s argument: while his initial formulation was lacking in some respects, the response has been to attack the argument, rather than to consider that upon reflection, Plantinga’s argument is rather humdrum: most of our beliefs are in fact false. Most children have an intuitive — but wrong — folk physics, and even a great number of adults hold these same folk physics. People often see ghosts where there are none, or think certain numbers lucky or unlucky, or think the stars and planets guide their lives, or see the agency of gods all around them.)
Someone might say that specific ways of cognition has been sexually selected for — in much the same way the peacock’s tail has undergone sexual selection — in the past few thousand years. Therefore, there has been direct selection for (approximately) correct beliefs, rather than behavior. Such an objection does not stand, for this assumes that there has been a direct selection on belief-formation processes and not behavior. But put that aside. While we may assume (quite rightly, I might add) that we have a more robust belief-formation process than before sexual selection, that does not entail that our cognitive states–our beliefs, our attitudes, and desires–under sexual selection are in any way reliable: they’ve just been directly selected for based on their success under a very limited domain, and we still have to figure a way to sort out true from false beliefs, and so on. After all, we still have to deal with the fact that we routinely have false beliefs selected-for by sexual selection that we intuitively take to be true, and are often not shakable even under the most intense pressure.
Another individual might retort that through the development of language, our belief-formation processes were selected for again, but this time not through evolutionary selection, but through a different form of selection: our beliefs, when expressed linguistically, are intersubjectively criticizable. This is analogous to the argument that science is self-correcting through intersubjective criticism (for testing is but a limited form of criticism). Therefore, our belief-formation processes, while not initially reliable or necessarily truth tracking, will over time self-correct in light of criticism. This too cannot stand, for there are many different ways this selective pressure may act. We may choose to protect our most basic beliefs come what may by changing beliefs on the periphery or reject our most basic beliefs outright when confronted with hearsay, or anywhere else in between. In fact, the act of argumentation may not be a self-correcting procedure, but as a self-protecting procedure, as a way of defending any position against criticism. The sociological issues of power-relationships have, after all, historically been full of the powerful using any argument in their arsenal to defend their position of power. This too does not take into account the constant proliferation of stances in philosophy, for while some have fallen by the side of the road, more join the march every day. And what of politics, where argumentation has become the art of lying with a smile? Furthermore, even if we assume the most optimistic position in regard to language and argument, that our cognition evolved in such a way, and then is shaped through the proper ways of going about a discussion, we may still revise our beliefs in all the most improper ways: we may ‘self-correct’ incorrectly.
That is, in a very condensed form, why I take Plantinga’s argument seriously. The human condition leaves us with no way of knowing that satisfies traditional theories of knowledge.