Alvin Plantinga's Argument Isn't Against Evolution. It Never Was.
Evolution is a premise. Naturalism is the problem.
There is a line in Where the Conflict Really Lies I want you to hold on to before we go any further. Alvin Plantinga, building what is possibly his most famous philosophical argument,1writes this: "I will be assuming for purposes of this argument that evolution is true."2
Read that slowly.
I will be assuming for purposes of this argument that evolution is true.
This is the man behind the Evolutionary Argument Against Naturalism — the EAAN, as it’s known in the philosophy literature. And the argument he’s constructing begins by granting evolutionary theory. If you had Plantinga filed somewhere under “Christian intellectual who argues against Darwin,” that filing cabinet needs reorganizing.
What the Argument Actually Is
The EAAN is a conditional argument. Its form is: if evolutionary theory is true and naturalism is true, then we have a significant problem. It takes no position on whether evolution is actually true. It simply asks what follows if it is.
Naturalism, in the philosophical sense Plantinga is targeting, is not a vague cultural attitude. It is the specific claim that there is no God and no supernatural reality of any kind — that the natural world, matter and energy operating by physical law, is all there is. Evolution, on this account, is the unguided mechanism by which we got here: random mutation and natural selection acting over deep time, with no intention behind it.
Natural selection selects for survival. Organisms that are better at surviving and reproducing pass on their traits. That’s the mechanism.
Here is the question the EAAN presses: does survival require accurate beliefs?
Sometimes, obviously, yes. If you believe there is no tiger behind that bush, and there is a tiger behind that bush, your inaccurate belief is likely to end your participation in the gene pool. Some connection between true belief and survival seems clear enough.
But the connection is not tight. What natural selection produces are behaviors — behaviors that aid survival. Beliefs can produce behaviors, but many different beliefs, including wildly false ones, can produce the same survival-enhancing behavior. A creature that runs from tigers survives whether it runs because it correctly believes tigers are dangerous, or because it holds some bizarre false belief about tigers that, by coincidence, produces the right running behavior, or because it doesn’t represent tigers at all but has an instinct that manifests as running. Natural selection cannot reach in and check which of these is operating. It selects for the running.
What natural selection cannot do — because it has no mechanism for it — is select directly for truth. It selects for survival. If truth-tracking beliefs happen to aid survival, they'll tend to get passed on. But there is nothing in the mechanism that guarantees alignment between "what our cognitive faculties produce" and "what is actually the case."3
Which means: on naturalism plus evolution, we have significant reason to doubt that our cognitive faculties are reliable truth-trackers. They were shaped to keep us alive, not to give us accurate metaphysics.
The problem — and this is where the argument gets elegant — is that this doubt includes the belief in naturalism itself. If we cannot trust our cognitive faculties, shaped by blind, unguided processes with no interest in truth, then we cannot trust the reasoning that led us to naturalism. The naturalist has produced a worldview that, if true, undermines the epistemic foundations of the faculties that produced it.
Plantinga calls this "self-defeating." Not a logical contradiction — it doesn't refute itself the way "this sentence is false" does. But it gives the reflective naturalist an undefeated defeater: a reason to doubt her beliefs that she has no good resources, within her own worldview, to answer. If my faculties are unreliable truth-trackers by design, why should I trust the belief that my faculties are unreliable truth-trackers? Why should I trust anything, including naturalism? The floor gives out.4
Why Evolution Isn’t the Target
Now notice what’s doing the work in that argument. It’s not evolution. It’s naturalism.
Evolution, in the EAAN, functions as a premise about the mechanism by which cognitive faculties were produced. The problem arises because, on naturalism, that mechanism is unguided — aimed at nothing, designed by nothing, answerable to nothing except survival. Remove naturalism from the picture and you change the character of the mechanism entirely.
If you are a theistic evolutionist — someone who accepts evolutionary biology but holds that God was the agent behind it, guiding or instantiating the process — the EAAN does not apply to you. On theistic evolution, there is a principled explanation for why faculties shaped by natural processes might also be reliable truth-trackers: God, in creating human beings through evolutionary processes, can plausibly intend the result. The self-defeating loop never gets started, because the process was not unguided. It was aimed, at some level, at producing creatures capable of knowing both the world and their Creator.
Old-earth creationist who accepts deep time and common ancestry? Same result. Young-earth creationist who rejects common descent entirely? Also no self-defeating loop. The argument’s grip requires the specific conjunction: evolution as the explanation plus naturalism as the metaphysics. Adjust either element, and the argument has nothing to hold.
This is why Plantinga titled the book Where the Conflict Really Lies. He is making a point about misidentified battlegrounds. The conflict is not between science and Christian faith — Christianity, he argues, is not in conflict with anything in contemporary science. The conflict is between naturalism and its own preferred science of human origins, because naturalism cannot borrow evolution’s explanatory power without paying a philosophical price it has no resources to cover.
What This Means If You Accept Evolution
I want to make this implication explicit, because I think it gets missed in the usual ways the EAAN gets described.
If you accept evolutionary biology and you are a Christian, Plantinga’s argument is not your problem. It is your tool. He is not arguing against your science. He is arguing that naturalism cannot coherently complete itself as a worldview — that its metaphysics and its science are in tension with each other in a way that theism, including theistic evolution, simply is not.
The theist who accepts evolution can say: yes, natural selection explains the emergence of our cognitive faculties, and the God who stands behind the process can ground their reliability. The naturalist who accepts evolution faces a harder road: she must explain why faculties shaped by a blind, survival-oriented process with no interest in truth should be trusted to deliver accurate metaphysics — including the metaphysics of naturalism.
Plantinga’s contribution is to stand at the point where evolutionary naturalism tries to complete itself as a worldview and ask whether it can. The answer he gives, carefully argued across several books, is: not without self-defeat.
Is the argument finally decisive? Philosophers have pushed back on various steps — Patricia Churchland argues that selection for survival is sufficiently "truth-tropic" to underwrite reliability; Dennett and others have questioned whether the probability of reliable faculties is really as low as Plantinga claims.5Plantinga has replied at length. The literature is substantial and worth engaging. But that longer conversation presupposes getting clear on what the argument actually is.
The conflict really lies elsewhere. Plantinga has been saying so for thirty years. Given how often the EAAN gets misread as an attack on Darwin rather than on the metaphysics of his devoted followers, it might be worth taking him at his word.
Possibly — there are serious contenders. Plantinga's free will defense against the logical problem of evil, developed in God and Free Will (1967) and The Nature of Necessity (1974), is at least as well known in philosophy of religion circles, and is widely credited with effectively dissolving the logical version of the problem of evil as a challenge to theism. The modal ontological argument in The Nature of Necessity — which uses possible worlds semantics to give the ontological argument genuine philosophical traction — has its own substantial literature. His God and Other Minds (1967) launched the Reformed epistemology tradition by arguing that belief in God is epistemically on par with belief in other minds. And his three-volume work on warrant — culminating in Warranted Christian Belief (2000) — makes the case that Christian belief can be fully rational and warranted without requiring prior philosophical argument for theism. Plantinga is one of those philosophers where the question "what is he most famous for?" depends heavily on who you ask.
Alvin Plantinga, Where the Conflict Really Lies: Science, Religion, and Naturalism (Oxford University Press, 2011), 313. The EAAN was first developed at length in Warrant and Proper Function (Oxford University Press, 1993), ch. 12, and has been refined across several subsequent works.
Plantinga develops this point in some detail by considering the relationship between neurophysiology, belief content, and behavior. The key move is that natural selection operates on behavior via neurophysiology, but the content of beliefs — what they are about — is not directly visible to selection in the relevant way. See Where the Conflict Really Lies, 315–322.
The technical structure here is Plantinga's argument that N&E (naturalism-and-evolution) gives the naturalist a defeater for the reliability of her faculties, R, and that this defeater cannot be defeated — because any attempt to defeat it would itself employ the unreliable faculties in question. See Where the Conflict Really Lies, 338–344.
Patricia Churchland's "truth-tropic" response appears in "Epistemology in the Age of Neuroscience," Journal of Philosophy 84 (1987): 544–553. Dennett's engagement is scattered across several works; see especially his response essays in Alvin Plantinga, ed. Deane-Peter Baker (Cambridge University Press, 2007). Elliott Sober offers a more technical treatment in "Evolutionary Theory and the Reality of Macro-Probabilities," in Ellery Eells and James Fetzer, eds., The Place of Probability in Science, Boston Studies in the Philosophy of Science, vol. 284 (Springer, Dordrecht, 2010).