A while ago I discussed the Evolutionary Argument Against Naturalism (EAAN) with Wade Tisthammer of the Maverick Christian blog. Here I want to explain the problems I see with his argument. (For his post-discussion thoughts, see here.)

To begin, it’s worth pointing out that there isn’t one “EAAN” but several. In a discussion with Stephen Law on the same channel about a year ago (see here), they identified at least three versions of the argument: the semantic epiphenomenal version; the systematically false beliefs represented by Paul and the tiger version; and Thomas Crisp’s abductive version, explained by Tyler McNabb, that (only) our most abstract metaphysical beliefs would be unreliable on naturalism. That is all too much to get into here; for now we’ll look at Wade’s version.

Wade’s version of the EAAN

Wade’s version hinges on a counterfactual conditional as a key premise. He argues that if it had been the case that any particular brain state was able to give rise to vastly different mental states, our physical behavior would be the same, and thus our beliefs would be blind to natural selection. (This is because NP properties cause physical behavior, so any associated mental state would be causally irrelevant to behavior, and to evolution.) Wade’s conditional statement considers what would be true had some alternative state of affairs been true:

(On naturalism) If a belief’s associated NP properties had any different semantic content associated with it instead, the same outcome in the physical world (e.g. one’s behavior) would result.

The problem with Wade’s appeal to such a hypothetical conditional is that we have no reason to think the antecedent is possible or plausible on naturalism. Wade is proposing that if a form of naturalism with totally unconstrained mental content were true, then some skeptical result would follow. Sure, but who cares? Wade fails to give the naturalist a reason to think that such a scenario is anything more than conceivable. The naturalism of philosophers, from Thomas Nagel to Dan Dennett to David Chalmers, goes untouched.

Moreover, the naturalist can cook up a parallel argument against Wade’s Christian theism to the same effect: it’s conceivable that a creator god could have systematically deceived his creatures (for reasons we’re unaware), and as a result our cognitive faculties would be unreliable. If that had been the case, surely a debilitating skepticism would follow. The counterfactual conditional could read something like the following:

Had God systematically deceived us for reasons we’re unaware, our cognitive faculties would be unreliable.

Following Wade, the naturalist could also parallel the auxiliary arguments: we can imagine a hypothetical alien species who had been created by God to be systematically deceived. For the aliens, R would be low or inscrutable. And, they’d get a defeater just as the person who had taken drug XX that undermines cognitive reliability. Thus, belief in Christian theism results in radical skepticism and is irrational to believe. Or so the parallel argument would go.

The obvious problem is the Christian has no reason to think God would deceives his creatures. The antecedent of the key counterfactual conditional is merely stipulated as conceivable. Nothing more. In my view, these kinds of arguments do nothing to internally critique the actual views of naturalists, and Christians, respectively.

A leap made in the argument

The EAAN is best thought of as an internal critique against the conjunction of materialism and Darwinian evolution. What it’s not, however, is a critique against all atheistic or non-supernatural views; unfortunately Wade muddies the difference between materialism and all other non-supernatural views in his presentation of the argument. He defines naturalism as “the view that the supernatural does not exist,” but then draws from arguments that focus on reductive and non-reductive materialism (i.e. the work of Plantinga). Wade then leaps to the conclusion of a defeater for all non-supernatural accounts of human cognitive faculties.

Wade’s conclusion doesn’t follow for the obvious reason that naturalism (as Wade defines it) doesn’t entail materialism about minds or a specific view of semantic content. A non-supernaturalist who rejects the existence of supernatural persons and forces doesn’t thereby commit themselves to materialism about minds and bodies (reductive or non-reductive) and semantic epiphenomenalism.

There are non-theistic theories of mind and teleological accounts of metaphysics and evolution that avoid the EAAN’s attack against materialism and a mechanistic view of Darwinian evolution. And there is a range of non-supernatural views of mind and cosmos that don’t make it likely that semantic content is causally inefficacious. For instance, Stephen Law discusses a view in which mental content is conceptually constrained on NP properties; that is, one cannot “plug any old belief content into any old neural structure, irrespective of that structure’s behavioral output.” So, it seems to me, if the Christian is able to build into their metaphysical view of God that he wouldn’t systematically deceive his creatures, the naturalist (or atheist) can build into their view that mental content is conceptually constrained to particular brain states.

For Wade’s argument to stick, or more specifically, for Wade’s argument to prove something relevant to the beliefs of most naturalists, he needs to provide a compelling reason to think that conceptual constraints on mental content is unlikely on naturalism and that it’s more likely that evolution would select against conceptually constrained accurate beliefs in favor of unconstrained false beliefs. Without that kind of evidence, Wade’s version of the EAAN doesn’t pack much of a punch.