I've just watched this video of Alvin Plantinga giving a talk on the conflict between science and religion:
In it he argues that science, particularly evolution, isn't in conflict with theistic belief, but in fact the opposite is true: naturalism and evolution are incompatible. His argument deals with three propositions:
N: Naturalism (there's no such thing as God or anything like a god)
E: Evolution (human beings have come to be by virtue of the processes described in current evolutionary theory)
R: Our cognitive faculties are reliable (i.e. they produce beliefs that are for the most part true)
The argument goes as follows:
The probability of R given N and E is low
Evolution selects for behaviour, not beliefs; it doesn't make a difference to evolution whether your beliefs are true or not
If you accept N and E, and also premise 1, then you have a defeater for R
This defeater can't be defeated
- If you have a defeater for R, then you have a defeater for any belief produced by your cognitive faculties
- One of the beliefs produced by your cognitive faculties is N and E
Now obviously the crux of this argument is premise 1; if this is demonstrated to be true, then the rest of the argument follows logically. Interestingly, this premise takes the form of an argument from ignorance: Plantinga can't see a way for evolution to select for true beliefs, so it must be impossible for evolution to select for true beliefs.
To be fair, I don't think it is a true argument from ignorance, as I think the burden of proof is on the evolutionist to demonstrate that evolution does select for true beliefs. However, it is in the same vein, in the sense that it is dependent on the current state of scientific research - if at some point in the future science can demonstrate how evolution can select for true beliefs, then the argument fizzles away into nothingness. In order to make the argument watertight, Plantinga must demonstrate that it is impossible for evolution to select for true beliefs, but he can only get as far as "I don't see how it is possible".
One of the questions in the Q&A session was: "Isn't it true that false beliefs can lead to maladaptive behaviour?" This I think touches on something important. Maybe we can't see how evolution can necessarily select for true beliefs, but it seems to me plausible that holding false beliefs in general would be maladaptive. For example, if a creature evolved a neurostructure that as a side effect generated the belief "Crocodiles are nice friendly creatures", this neurostructure (and consequently the dependent belief) would be selected out of existence.
I think some of the error (as I would see it) comes into Plantinga's argument by his distinguishing between behaviour and belief, when he claims that evolution only selects for behaviour and not beliefs. He seems to think that in the evolutionary model, beliefs are a mere side-effect of selecting for a neurostructure that produces certain behaviours. Now with the simplest organisms, it is certainly true that evolution selects for certain behaviours (moving towards light, moving away from sudden shadows etc), but it doesn't make sense to talk about these organisms having any sort of beliefs.
Now there are some very complex behaviours out there that appear to be purely reflex actions, for example I remember watching a David Attenborough documentary about a burrowing wasp that goes through a complex series of actions to check its burrow is clear. The actions are completely hard-wired; if the wasp is interrupted in the middle of the process, it will start again from the beginning and will keep doing so as long as it is interrupted, until it dies.
However, at some point there is a limit to the complexity of hard-wired actions that evolution can select for. In order to react in real-time to a complex and rapidly changing environment, an organism must have a rudimentary model of the environment, at a level of complexity appropriate to the organism. To a greater or lesser extent, this model will be built up by the organism through interaction with the environment itself, and through trial and error. Hypotheses will be created and tested for, and either retained or discarded. The actions and beliefs of the organism are thus interdependent, beliefs informing actions, and actions modifying beliefs.
With such a model, it seems to me more likely then that the opposite of premise 1 is true: given N and E, for a complex organism in a complex and rapidly changing environment, the probability of R is quite high.
Now I think there's another problem with the argument, and it's quite subtle. If N and E are true, then according to Plantinga's argument, the cognitive faculties of any creature that have arisen by virtue of N and E are inherently untrustworthy. However, even if you and I belong to that set of creatures, and our cognitive faculties cannot be trusted, that doesn't affect the truth of N and E in any way, it simply affects our ability to know it.
Plantinga uses the following analogy to support premise 2:
Suppose there is a substance XX which is such that you believe that if you take that substance, the probability is very high that your cognitive faculties will be unreliable. Now suppose you also believe that you have taken it. Well, then you've got a defeater for the belief that your cognitive faculties are reliable. Even though you recognise it's possible that they could be, you've got a very good reason not to think that they are.
This is all very well, but it seems to me the reasoning is faulty somewhere. If you believe you have taken substance XX, then you have a defeater for the belief that your cognitive faculties are realiable. But then you also have a defeater for the belief that you have taken substance XX! In fact, as soon as you even believe that substance XX exists, instantly you have a defeater for the belief that your cognitive faculties are reliable, as you may have already taken it but believe not to have done!
I think the problem here may be to do with failing to distinguish from belief about a proposition and the proposition itself, especially since the proposition itself is about what beliefs are possible. I think there's a category error here somewhere, and I think it's possibly in step 3; what do you think?