Stephen Law vs Alvin Plantinga on the Unbelievable podcast on the topic of the "The Evolutionary Argument Against Naturalism".
Basic argument, from Wikipedia on Evolutionary argument against naturalism
I haven't read the responses, so that my response will be fresh and untainted.
- N as naturalism, which he defined as "the idea that there is no such person as God or anything like God; we might think of it as high-octane atheism or perhaps atheism-plus."
- E as the belief that human beings have evolved in conformity with current evolutionary theory
- R as the proposition that our faculties are "reliable", where, roughly, a cognitive faculty is "reliable" if the great bulk of its deliverances are true. He specifically cited the example of a thermometer stuck at 72 °F (22 °C) degrees placed in an environment which happened to be at 72 °F as an example of something that is not "reliable" in this sense
and suggested that the conditional probability of R given N and E, or P(R|N&E), is low or inscrutable.
Plantinga's argument began with the observation that our beliefs can only have evolutionary consequences if they affect behaviour. To put this another way, natural selection does not directly select for true beliefs, but rather for advantageous behaviours. Plantinga distinguished the various theories of mind-body interaction into four jointly exhaustive categories:
- epiphenomenalism, where behaviour is not caused by beliefs. "if this way of thinking is right, beliefs would be invisible to evolution" so P(R/N&E) would be low or inscrutable
- Semantic epiphenomenalism, where beliefs have a causative link to behaviour but not by virtue of their semantic content. Under this theory, a belief would be some form of long-term neuronal event. However, on this view P(R|N&E) would be low because the semantic content of beliefs would be invisible to natural selection, and it is semantic content that determines truth-value.
- Beliefs are causally efficacious with respect to behaviour, but maladaptive, in which case P(R|N&E) would be low, as R would be selected against.
- Beliefs are causally efficacious with respect to behaviour and also adaptive, but they may still be false. Since behaviour is caused by both belief and desire, and desire can lead to false belief, natural selection would have no reason for selecting true but non-adaptive beliefs over false but adaptive beliefs. Thus P(R|N&E) in this case would also be low. Plantinga pointed out that innumerable belief-desire pairs could account for a given behaviour; for example, that of a prehistoric hominid fleeing a tiger:
Perhaps Paul very much likes the idea of being eaten, but when he sees a tiger, always runs off looking for a better prospect, because he thinks it unlikely the tiger he sees will eat him. This will get his body parts in the right place so far as survival is concerned, without involving much by way of true belief. ... Or perhaps he thinks the tiger is a large, friendly, cuddly pussycat and wants to pet it; but he also believes that the best way to pet it is to run away from it. ... Clearly there are any number of belief-cum-desire systems that equally fit a given bit of behaviour.
Thus, Plantinga argued, the probability that our minds are reliable under a conjunction of philosophical naturalism and naturalistic evolution is low or inscrutable. Therefore, to assert that naturalistic evolution is true also asserts that one has a low or unknown probability of being right. This, Plantinga argued, epistemically defeats the belief that naturalistic evolution is true and that ascribing truth to naturalism and evolution is internally dubious or inconsistent.
There are certainly several things in this argument that need to be thought out. Point (1) is certainly true: any beliefs that do not affect behavior in any way would be invisible to evolution. I doubt there are many beliefs in this category. Even something vague, like the belief in a deist god, would at some point affect behavior in some way. Point (3) is also uncontroversial. Point (2) I'm not sure I understand entirely, so I'll focus on point (4).
There are two parts - "behaviour is caused by both belief and desire," and "innumerable belief-desire pairs could account for a given behaviour" with the example of running from tigers because of various false beliefs. The final piece of this, however, is the limited resources of the brain. Our brain takes many computational short-cuts, because of efficiency and limited resources. If you have a number of different belief/desire pairs that lead to behavior, the brain will settle on those belief/desire pairs that generalize to the largest number of things. This will certainly be in cases where the result is closer to reality than not. We run away not only from tigers, but also bears, snakes, and spiders. The desire to pet or the mistaken belief that running away is the best way to get towards an animal would not generalize quite so well to these other creatures.
Plantinga uses this argument from rationality to bolster his theism - if rationality is unlikely given naturalism and evolution, then theism is more likely. However, he fails to notice that rationality isn't truly the norm. Yes, we can make some astounding feats of rationality as a collection, but it takes serious work to overcome the natural biases, and failures of reasoning inherent in the human condition. The cognitive short-cuts leads us to make many errors of observation, deduction, and induction. He also fails to point out the many beliefs that people have that are irrational but have a benefit, or the beliefs that people have which are irrational and have no effect and thus can get carried along without being selected against. In fact, the situation is very much like what you'd expect from evolution - some wrong beliefs with adaptive benefits, some wrong beliefs with no penalty, some correct beliefs by accident, etc… Finally, even if our beliefs were 90% wrong, we'd still be able to trust rationality as a collection - which is exactly what science does. It forces fallible people to not believe things until it can be replicated, it downplays anecdotal evidence, arguments from authority, and recognizes the cognitive failures of humans in order to overcome them.