There is a tension at the heart of naturalism that its most honest proponents have noticed. The claim is that human beings are entirely the products of unguided physical processes — that the brain is an organ shaped by natural selection for survival and reproduction, with no access to truth as such, only to adaptive behaviour.
But if this is true, the question arises: why should we trust our brains’ conclusions about anything beyond immediate survival? Why should the outputs of a survival machine be reliable guides to the deep structure of reality? And in particular, why should we trust the conclusion that naturalism itself is true?
The problem stated precisely
Natural selection is a process that selects for survival value, not for truth. A belief that leads to adaptive behaviour will be selected for even if it is false — even if it is wildly, systematically false. In many documented cases, false beliefs are more adaptive than true ones. Overestimating threat leads to fewer fatal errors. Overconfidence increases risk-taking in ways that can be reproductively advantageous. The evolutionary history of the human brain is not the history of a truth-tracking instrument. It is the history of a survival instrument.
A geneticist and committed atheist — one of the architects of modern evolutionary biology — put this precisely a century ago: “If my mental processes are determined wholly by the motions of atoms in my brain, I have no reason to suppose that my beliefs are true — and hence no reason for supposing my brain to be composed of atoms.”
The self-undermining quality of this observation is intentional. If pure naturalism is true, then the reasoning capacity that led to the conclusion “pure naturalism is true” is itself an unreliable guide to truth. The conclusion undermines the means by which it was reached.
What this is not saying
The argument is not saying that atheists are irrational, or that evolution-shaped brains cannot produce true beliefs. They clearly can. The argument is about the epistemic foundation — about what justifies trusting our cognitive faculties on matters far removed from survival, including mathematics, logic, and metaphysics.
On pure naturalism, the answer is: nothing. There is no guarantee that our faculties track truth on these matters. We might happen to get things right, but we have no principled reason to expect that we do, and no principled way to distinguish reliable from unreliable reasoning on the deep questions.
The theistic alternative
On theism, the situation is different. If human cognitive faculties were designed by a rational being to track truth — or were, in some meaningful sense, oriented toward the real — then their reliability on the deep questions is not a brute accident. It is explained by their origin. The match between the rationality of the cosmos and the rationality of the human mind is not a coincidence. It is a consequence of their common source.
This does not prove theism. But it establishes that theism has explanatory resources that pure naturalism lacks — specifically, an account of why the minds that are asking these questions can be trusted to reach reliable conclusions about them. The universe appears to be rational. Human minds are capable of apprehending that rationality. The theist has a reason to expect this correspondence. The naturalist has to hope for it.