Plantinga's Evolutionary Argument Against Naturalism

December 11, 2009

One of today's most prominent Christian apologists is the philosopher Alvin Plantinga of Notre Dame University. He has written a trilogy setting forth a new epistemology by which Christians can justify claiming to know that their beliefs are true. The books in the series are Warrant: The Current Debate, Warrant and Proper Function, and Warranted Christian Belief. A thorough critique would probably take at least another book if not a trilogy of its own, I'm not going to even try to begin it here. I'm going to address just one point he mentions briefly and almost offhandedly in the second and third volumes.

In the final volume of the series, Plantinga claims that under any naturalistic assumption, there is no good reason to have any confidence that anything we believe is true. In particular, he argues, the process of evolution, absent any supernatural intervention, was extremely unlikely to have produced brains that could effectively distinguish truth from falsehood. Many rebuttals have been published since the book came out, and he addresses some of them in an online article, "Naturalism Defeated" (Plantinga 2004). Among the objections that he addresses is what he calls the Maximal Warrant Objection. This paper will review and critique his response to it, concluding that it does not overcome the objection. From this it will follow that Plantinga's argument fails to present a real epistemological difficulty for adherents of naturalism.

Plantinga compares two hypotheses about human origins, both of which assume the truth of the current theory of evolution by natural selection and assume that our cognitive faculties are reliable. On one hypothesis, which we may be label theistic evolution (TE), God, as understood by the major monotheistic religions, was in some way involved in the evolutionary process so as to ensure that humans acquired reliable cognitive faculties. On the other, naturalistic evolution (NE), there is no god or other supernatural agency that could have affected the course of human evolution. According to Plantinga, the assumption of reliable cognition (R) cannot be justified under NE and therefore is defeated. But without R, the theory of evolution loses its own warrant and so the believer in naturalistic evolution is caught in an epistemological quandary. Theistic evolution, on the other hand, does support R. That being so, Plantinga concludes, naturalistic evolution "can't rationally be accepted" (1994).

Plantinga summarizes the Maximal Warrant Objection thus: "According to this objection, R has a great deal of intrinsic warrant for us. . . . It has so much intrinsic warrant, in fact, that it can't be defeated—or at any rate can't be defeated by the fact that P(R/N & E) is low or inscrutable" (2004). "Warrant" here is Plantinga's preferred term for that which distinguishes knowledge from mere true belief. A proper critique of his proper-function theory of warrant (Plantinga 1993) would take us far beyond the scope of this paper. Without getting sidetracked by debates over the nature of knowledge, we'll say that some person S ought to believe a proposition P if and only if P is warranted for S.

The maximal warrant objection claims that R is properly basic—what many epistemologists call a foundational belief. It gets its warrant not by inference from other propositions but in some other way that appropriately terminates the chain of justification. An appropriate termination has to be non-arbitrary, but it is widely supposed that the reliability of at least some of our cognitive faculties can be non-arbitrarily justified. Plantinga seems to agree with this, or at least not to wish to challenge it: "Let's also agree that R does has [sic] warrant and perhaps a great deal of warrant, when it is taken as basic. Still further, we can add that R plays a unique and crucial role in our noetic structures: if we are reflective and come to doubt R, we will be in serious epistemic trouble" (2004).

This does not imply, he argues, that it cannot be defeated. And that is quite so, but the question is whether naturalistic evolution does defeat R if R is taken to be properly basic. It does not seem to. If we are justified, antecedently of our accepting either naturalism or evolution, in trusting our cognitive faculties, then naturalistic evolution cannot itself be a defeater for R. Plantinga's case depends on his observation that, conjoined with naturalism, evolutionary theory gives us no reason to expect reliable cognitive faculties. But that is irrelevant even if it is so. As Plantinga himself points out, our belief in evolution itself depends on R because the entire enterprise of science presupposes that, in general, our cognitive faculties work the way we think they ought to work. And, according to his own "First Principle of Defeat," "If S rationally believes that the warrant a belief B has for him is derivative from the warrant a belief A has for him, then B is not a defeater, for him, of A" (2004).

Now, what he says is not that evolution itself defeats R, but the conjunction of evolution with naturalism. But naturalism does not add anything to evolutionary theory that makes it inconsistent with R. It simply denies a proposition that theism adds to evolutionary theory that, according to Plantinga, makes the theory imply R. Theism could not do that, though, if evolution were antecedently inconsistent with R, or in any case, Plantinga does not argue that theism is neutralizing any inconsistencies between evolution simpliciter and R. His argument implies rather that theism itself is what justifies R, and the only role evolution serves is to describe the means by which God made our cognitive faculties reliable (2004).

A portion of his response to the Maximal Warrant Objection seems intended to demonstrate that under naturalism, R should not be regarded as properly basic. If he succeeds at all, it is only by incorporating various elements of his proper-function theory, which appeals to design plans, cognitive environments, and various intents and purposes, among other things (Plantinga 1993, 19). A naturalist accepting an alternative epistemology incorporating some more conventional version of foundationalism could disregard Plantinga's comments on this particular point. At the very least, an argument to the effect that, with or without evolution, we could never justify trusting our cognitive faculties unless we believe in God would appear to be assuming Plantinga's conclusion.

So, if we regard R as axiomatic—which is just another way of saying it is properly basic or foundational—then it does not matter if naturalistic evolution fails to justify it, and so the naturalist is in the epistemic clear unless, somehow, NE entails that R is either impossible or extremely improbable. Plantinga does not demonstrate such an entailment, and so his argument fails even if nothing else is wrong with it. In addition, though, his implied claim that we should be surprised if naturalistic evolution had produced reliable cognitive faculties is questionable.

He invites us to imagine "Paul," a hominid like us inhabiting a world similar to our own. Paul has desires that in some cases, if fulfilled, would result in his demise, and his beliefs are formed independently of the reality in which he lives. Just by chance, though (Plantinga proposes nothing else to account for it), the behavior resulting from his conjoined desires and beliefs happen to be what he needs to do in order to survive (2004). We can stipulate that, under NE, this could possibly have happened. In the actual world, though, it did not happen, and this ought not to surprise us. A rigorous defense of this proposition cannot be attempted here. Papers addressing some of the issues relevant to the discussion include Churchland (1994), Churchland and Sejnowski (1990), Dennett (1978), and Rosenberg (1999). What follows are some very condensed comments that the author hopes are pertinently suggestive.

Churchland's "four F's: feeding, fleeing, fighting, and reproducing" (Churchland 1987, 548) are the fundamental desires that we should expect evolution to have hard-wired into our cognitive systems. We should also expect any other desires we might have acquired to have been favored insofar as their satisfaction would help us satisfy these fundamentals. What we desire should correspond at least occasionally to what we need. Our cognitive faculties seem to produce, among other things, moment-by-moment decisions about whether and how to undertake behaviors that will satisfy those desires. This is a kind of computation. The brain might or might not do some things that no computer can do, but we don't have to settle that argument here to affirm that the brain is at least a computer, i.e. that among the things it does is to compute. At least some and perhaps all of our beliefs are the output of computational processes to which the inputs include sensory data or previously formed beliefs. This kind of computing power is biologically very expensive. While this does not imply that it must have been adaptive, it justifies a supposition, absent evidence to the contrary, that it was in fact adaptive.

There is only one actual world, but we can imagine an infinity of other possible worlds, and a random belief generator could produce propositions corresponding to any of them. Beliefs generated with only a random connection to reality are thus virtually guaranteed to be false, and so there is a high probability that any particular behavior based on randomly generated beliefs will fail to be that which would fulfill any given desire the behavior is meant to accommodate, including those four fundamental desires. It is therefore reasonable to suppose that natural selection would favor any cognitive mechanism that managed in some way to track sensory input about the organism's real environment and process it with something approaching logical consistency, and to suppose that sensory input to itself be reliable.

The only hope of defending a disconnect between beliefs and behavior seems to lie in maintaining an ontological distinction between the syntax of neurophysiology and the semantics that permeate our conscious thoughts (Plantinga 1994). Churchland argues, cogently in the author's opinion, that while a conceptual distinction can be rhetorically useful, nothing compels us to deny that "awareness just is some pattern of activity in neurons" (1994, 30-31), and if that is true of awareness, we should expect it to be true of beliefs as well.

If this is so, then it is not the case that reliable cognitive faculties would be an improbable outcome of naturalistic evolution; but, even if it were the case, naturalistic evolution still would give us no reason to question the prima facie presumption of our having reliable faculties. Naturalists are therefore justified in dismissing Plantinga's argument and continuing to hold their belief that our cognitive faculties are indeed reliable and were produced by natural selection without any supernatural assistance.

(This essay is based on a paper I wrote for a class in the philosophy of religion, taught by Dr. Matthew Davidson at California State University, San Bernardino, in 2009.)


Churchland, Patricia Smith. 1994. Can neurobiology teach us anything about consciousness? Proceedings and Addresses of the American Philosophical Association 67, no. 4:23-40. JSTOR. (accessed December 4, 2009).

Churchland, Patricia Smith. 1987. Epistemology in the age of neuroscience. The Journal of Philosophy 84, 10:544-553. JSTOR. (accessed November 26, 2009).

Churchland, Patricia Smith, and Terrence J. Sejnowski. 1990. Neural representation and neural computation. Philosophical Perspectives, Vol. 4, Action Theory and Philosophy of Mind. 343-382. JSTOR. (accessed December 4, 2009).

Dennett, Daniel C. 1978. Why you can't make a computer that feels pain. Synthese 38, no. 3:415-456. JSTOR. (accessed December 4, 2009).

Plantinga, Alvin. 2004. Naturalism defeated. philosophy/virtual_library/articles/plantinga_alvin/naturalism_defeated.pdf (accessed December 9, 2009).

Plantinga, Alvin. 1993. Warrant: The Current Debate. Oxford: Clarendon Press. Oxford Scholarship Online. philosophy/0195078624/toc.html (accessed September 27, 2005).

Rosenberg, Alex. 1999. Naturalistic epistemology for eliminative materialists. Philosophy and Phenomenological Research 59, no. 2: 335-358. JSTOR. http:// (accessed December 4, 2009).

Site home

Philosophy index

(This page was last updated on August 6, 2010.)