Plantinga, Defeaters and Reasonable Belief

I have been mulling over this particular issue for a long time and, as it popped up on Stephen Law's blog, I decided to get my thoughts down.

Alvin Plantinga's argument from evolution against naturalism is an ingenious and controversial argument. Plantings argues that evolution without the intervention of a supernatural being makes our beliefs very unlikely to have been reliably formed. One of those beliefs would be evolution without the intervention of a supernatural being. Evolutionary naturalism is therefore "self defeating". Do panic at the word "evolution", the argument is not some idiot creationist nonsense. It is very well formed, very well argued and examining it brings interesting insights into defeaters and what is means to have a rational belief. "Undercutting" defeaters are themselves self-defeating.


A brief summary of the argument
The argument starts with pointing out that if there was no intervention from a supernatural being in the evolution of man then we are entirely reliant on the operations of evolution for the production of reliable belief-forming mechanisms, reliable cognitive processes. Plantinga then questions whether evolution gives two hoots about our cognitive processes. If we behave right then we will be evolutionarily successful, whether or not we believe right. If I have sex (behaviour) then it advantages my genes whether I believe I am having sex, cleaning the windows, praying, drinking beer or whatever. Now, unless we can show direct causation of behaviour by belief (such that I can only have sex if I believe I'm having sex) we are stuck without a selection mechanism for belief. As there are very, very many false beliefs that are possible for any situation and only the one true one the chance of us having a true belief is small. Thus naturalism and evolution should lead us to believe that our cognitive processes are not reliably formed. As this covers all beliefs we form on the basis of those cognitive processes it covers evolutionary naturalism. Evolutionary naturalism confounds the reliability of its own formulation’. Evolutionary naturalism cannot have been reliably formed, it has a "defeater".

Undercutting Defeaters
Take a subject S who believes in naturalism, N, for reasons Rn (perhaps “the arguments on Stephen Law's blog”). She also believes in evolution, E, for entirely independent reasons Re (perhaps “the arguments in 'The Blind Watchmaker'”).

Now, Plantinga's “defeater” is “undercutting”, it does not suggest that either N or E is false. Nor does it suggest that “N and E” is inconsistent. The defeater acts on the reasons for accepting N and E, Rn and Re. Not that the defeater suggests that these are false, rather that there is no good reason to suppose that are true.

Are these good reasons necessary? Allowing good reasons as necessary for Plantinga's conception of knowledge together with Plantinga's anti-naturalist argument leads to the conclusion that neither Rn nor Re can be known and, by extension neither can “N and E”. But Plantinga's argument is not that “N and E” cannot be known (as an absolute sceptic I would have no issue with that conclusion) but that “N and E” is “unreasonable”. If good reasons are necessary for reasonable belief then, for S to be reasonable in disbelieving in either Rn or Re there would have to be good reasons for believing in their negations. Any arguments for the truth of their negations would be arguments for their falsity, which Plantinga has not given. (I am ignoring issues of agnosticism. This is partly because I have powerful intuitions that widespread enforced agnosticism would cause problems. But it is mostly because I haven't figured out how to handle the issues of agnosticism).

Thus Plantinga fails to support his case. But worse, Plantinga's insistence that S rejects “N and E” commits her to an irrational belief. Given that S reasonably believes Rn and Re she is not only reasonable in believing R and E she would be unreasonable if she accepted Plantingas call to reject R and E.

S believes:

1.Rn
2.Re

Plantinga offers no arguments against the entailment of naturalism and evolution by the reasons S has and they do, in fact, entail. Thus S also believes:

3.If Rn then N
4.If Re then E

From 1. and 3., by modus ponens, S is committed to believing:

5.N

and, from 2. and 4., by modus ponens, S is committed to believing:

6.E

Finally, from 5. and 6. by conjunction, S is committed to believing:

7.N and E
If S believes either not-N or not-E then she believes a contradiction.