The evolutionary argument against naturalism (EAAN), by Alvin Plantinga is perhaps the most original and tangled argument for God to have arose since the middle ages. To fairly consider it, it should be poised as a dichotomy: the reader is presented with two basic options,

1) Metaphysical Naturalism and evolution by natural selection are both true, or

2) Theism and evolution by natural selection are both true.

The argument is then two pronged - on one hand, the EAAN proper intends to establish that believing in N&E (Naturalism and Evolution) is an irrational position because their combination entails that our reasoning faculties are unreliable; on the other, a complimentary (often implicit) argument is that G&E (God and Evolution) are compatible, as God provides a foundation for our rationality that is maintained regardless of the truth of evolution.

In this page we'll attack both propositions. In the first section we will critique the EAAN proper, showing that it is derived from armchair philosophizing that misunderstands evolutionary theory and reason itself. We will distinguish between naturalist Reason (RN) and the Reason posited by theology (RG). We will show that P(RN|N&E), the probability of naturalist Reason given metaphysical naturalism and evolution, is high. In the second part we will show that indeed P(RG|G&E), the probability of theistic Reason given God [regardless of evolution] is high, but that since in practice we have RN (naturalistic Reason) this actually is an argument against God's existence.

The EAAN[]

The core article of the EAAN is[1] that "P(R|N&E) is low", that is that the probability (P) that our cognitive faculties - memory, perception, reason, and so on - are reliable (R) given (|) metaphysical naturalism (N) and evolution by natural selection (E) is low. Reliable here is meant to be defined in terms of the efficiency of it functioning well, "a cognitive faculty--memory, perception, reason--is reliable if the great bulk of its deliverances are true." It is important to note that reliability is a function of the actual truth of the deliverence, the correspondence between what it determines and what reality is like; it is not a function of how useful is their determination. The crux of the EAAN revolves around the difference between being useful and being true.

Plantinga employs a lot of armchair arguments to establish this core premise. For example, he argues that evolution selects for actions, not beliefs, so that it is possible for a false belief to be selected for if it is conductive to life, especially if it is coupled to a counterintuitive desire so that the desire + belief combination will lead to a desired action. The example given is a desire to befriend tigers combined with the belief that you can accomplish this by running away from them.

It seems to elude Plantinga that P(R|N&E) is an empirical question. In order to answer it, we need to look at the way mental faculties actually evolve under evolution. This shows that animals have consistently and independently evolved the same mental faculties and that these faculties all function with high fidelity, despite occasional differences in their implementations[2]. Should the reader insist that this, too, is merely another example of Theistic Evolution in which God is directing animals towards the Truth, model simulations can also be constructed to demonstrate this feature under controlled conditions[3].

More importantly, however, a look at the biological record shows that these faculties are not absolutely reliable. Rather, they are reliable in the natural environment of the organism. The most telling example is why moths are drawn to a flame - they actually desire to fly in a straight line, but their spatial faculties are not reliable[4].

This is the type of Reason[5] that is to be expected under metaphysical naturalism, a Reason that is adaptive within the ancestral environment. Does this mean that we will find our Reason to be false? It is more complicated than that. Our mental faculties form the shapes of our thoughts, and we cannot escape them. They act as Kantian categories shaping our perceptions and thought patterns. We cannot escape them. Rather, it would mean that as we stray away from our ancestral environment into further and stranger reaches we will generally find that we need to conduct a lot of mental 'gymnastics' to build up a coherent picture of the world - our mental faculties will fail us, as we have strayed from the ancestral environment in which they were valid, and we will need to circumvent them.

The world is, intuitively, flat. If our faculties were truly Reliable, that would be the real shape of the world. But it isn't. Thus, children need to be taught the world is a sphere[6]. They still intuitively think of the earth is flat, however; we all do. In the same way we think of time as flowing uniformly, despite Einstein's theory of special relativity, and think of space as static despite Einstein's theory of general relativity, and think of reality as independent of observation despite quantum mechanics, and so on. The further we stray from our ancestral environment the stranger the universe appears - just like Naturalism entails (as there is no mechanism within Naturalism to endow us with faculties that are reliable beyond the evolutionary ancestral environment). Our faculties are not at all Reliable.

Again, I stress that this does not mean that our mental faculties are "wrong". The same faculties that insist the earth is flat are the foundation on which Euclidean, and later curved, geometry is built. The physical intuitions on the persistence of objects are part of the foundation for the mathematical formulation of quantum mechanics. And so on. Our faculties cannot strictly be 'wrong'; they define how we think. It is only the correspondence between them and the world that can be mistaken, and thus as we learn more and more about the world we create ever more elaborate models of how it works. These models are not wrong just because we are constructing them based on un-Reliable faculties. Rather, we have no choice but to construct models conforming to the way we think; models constructed on conflicting principles of reasoning would be self-contradicting and invalid.

The Counterargument[]

Along with the EAAN itself, another argument is made in a half-hearted manner. Thus Plantinga says:

"Now according to traditional Christian (and Jewish and Muslim) thought, we human beings have been created in the image of God. This means, among other things, that he created us with the capacity for achieving knowledge—knowledge of our environment by way of perception, of other people by way of something like what Thomas Reid calls sympathy, of the past by memory and testimony, of mathematics and logic by reason..."

This is the argument that P(RG|E&G) is high, or in human language that the probability for reliable mental faculties given that god exists [and evolution is true] is high. One can only agree[7], but one has to be careful in interpreting what Reliable means here.

It is unbecoming of God to implant in us false ideas, even if they are reliable in our local evolutionary environment. These are not the actions of a wise and benevolent God, falsehood cannot be the image of God that we are created in. Thus Thomas Aquinas writes:

"Whatever is put into the disciple's mind by the teacher is contained in the knowledge of the teacher, unless the teacher is teaching dishonestly, which would be a wicked thing to say of God. But the knowledge of principles naturally known is put into us by God, seeing that God Himself is the author of our nature. Therefore these principles also are contained in the Divine Wisdom. Whatever therefore is contrary to these principles is contrary to Divine Wisdom, and cannot be of God." - Thomas Aquinas, Summa Contra Gentiles I, 7 2 [8]

Aquinas of course didn't know that in addition to our basic logical faculties, which he rightly notes "it is impossible to think of their being otherwise", we are also endowed naturally with folk physics that says wrong things about the world, like that the earth is flat, and a host of other mental foibles and erroneous innate natural beliefs and modes of thought (including, embarrassingly enough, logical fobiles).

The traditional theistic position not only entails reliable mental faculties. Consistency demands that all natural beliefs in us will be Strongly Reliable, that is conforming to the truth or at least not misleading. It is, perhaps, permissible to have faulty, limited reasoning and senses. But not deceitful ones. A God that endows us with a false belief is not honest, as Aquinas says, and a Reason that is based on ideas that only apply in the ancestral environment is not one that is in the image of God.

Thus we are led to the conclusion that in a world with God, Reason must be perfectly Reliable, or at least not systematically false. As we explore more and more of Nature, we should expect to find that our basic intuitions reflect the world more and more, not less and less. This is in direct contrast to the experimental data, which shows that our reason includes faulty faculties of all sorts and that we need more and more mental gymnastics in order to construct a description of Nature in circumstances far removed from our ancestral environment. Thus, the evidence points to the conclusion that there is no God.


  1. Naturalism Defeated, by Alvin Plantinga. All Plantinga quotations are from this source.
  2. references needed: evolved reason and differences in algorithms, high fidelity
  3. reference needed: models of Reason evolution. Insisting that these experiments, too, are under divine interference (a Cartesian demon argument) robs us of any ability to reason empirically about the world whatsoever, and is really besides the point of the EAAN.
  4. reference needed: moth attracted to flame. Note that this is a far better example than Plantinga's tiger example, simply because it is a real-world example.
  5. As a side-note, our advanced logical reasoning, our Reason with a capital 'R', was not forged by evolution. Rather, it was forged by recursively using our reasoning faculties on themselves and through language and millennia of civilization abstracting general thought patterns from them. Thus it is that pre-literate societies lack even our sense of logic (reference needed: anthropology of logic in pre-literate societies).
  6. reference needed: children and flat earth
  7. Well, one can certainly object for many reasons, but for the sake of argument let us indeed suppose that for independent reasons we can deduce that a God exists whose nature entails this, or at least that it makes sense to seriously consider the option.
  8. See here, translation by Joseph Rickaby, S.J., M.A. Lond: B.Sc. Oxon, 1905.