I've been reading Plantinga's third volume and am puzzled by his account. As I understand it, Plantinga wants to block merely epistemic complaints about Christian belief--that is, he wants to block the kind of complaint where you bracket the truth of the claims in question and claim that whether or not the beliefs are true, there is something wrong, on purely intellectual grounds, with holding them.
The method used by Plantinga seems to be this: describe a model on which, together with his account of warrant, the beliefs would be warranted if the model is true. This is the perplexing part for me.
Here's why it's perplexing. Try to describe a belief that is a paradigm case of an epistemically problematic belief. I submit that I can always construct a model on which, together with Plantinga's account of warrant, the belief in question would be warranted.
If that's right, then the Plantingian method is the hired gun on behalf of the death of epistemology movement, isn't it? If my jaded account of his approach is correct, then the only issues we can responsibly discuss are issues about the truth or falsity of a belief. That's certainly worth doing, but my concern remains that there ought to be cases where we can criticize a belief on epistemic grounds independent of the question of truth or falsity.
So my question is this: how can we understand Plantinga's appeal to models without it having this untoward implication?
So, roughly the problem is this: Start with some broad form of reliabilism. Then add that there is this mechanism x which is reliable. That x exists and is reliable is not open to dispute. And x happens to produce a belief that p. So,the belief that p is warranted.
I think there are two things to say, which are related. First, there still are concerns of rationality and the like that need to be addressed with respect to p. Second, the plausibility of the entire enterprise will depend on the plausibility of the propositions from Christian theism it incorporates. So, we will have questions like: How plausible is Christian theism and how plausible is it to think that, given Christian theism, there is the sensus divinitatis. Though Plantinga isn't so concerned with giving evidence for Christian theism, I think that there is (supposed to be) a difference between Al's proposal and some other proposal where you take his account of warrant, tack on some crazy mechanism, say that it's reliable, and say that therefore beliefs it produces are warranted (when clearly they're not). Christian theism (together with the sensus divinitatis) as a world view is supposed to be reasonable in a way that your tacked-on mechanism isn't. Furthermore, with Christianity we don't have an easy reductio (or many would say we don't) of the following sort: The sensus divinitatis gives rise to a belief that p (or better, a whole class of beliefs, p1, ...,pn)). But there's no way p1-pn are true, so we have reason to question the reliability of the mechanism (or, perhaps, its existence). But if you posit some mechanism and say it is reliable (as part of your model), and it is spitting out nutty beliefs, we have reason to question the model--either the reliability of the mechanism or the existence of the mechanism.
It wasn't clear to me after reading Plantinga whether or not warrant entails truth on his account. If it does, then it is easy to see how answers to questions about the warrant of Christian beliefs are not logically independent of answer to questions of the truth of Christian belief. I think it was his goal to establish this and that his account of warrant is epistemically possible.
So far as I can tell, fallibilism is true. It is possible to have a warranted false belief. So, if it is a consequence of his view that warrant entails truth, then I have a reason to reject his view. So, it isn't a consequence of his view by my lights. (This is all from memory so please forgive error)
It seems that there is still many places to challenge Plantinga. First, to the truth of theism and Christianity. This is where he is going. Second, to the epistemic possibility of the model. This is where I am going. His model, if actually true, requires the existence of a certain kind of mental process, a sensus divinatus. Is there such a thing and is it reasonable to believe in it without empirical support? Second, can't we just be reliabilist and deny the theistic baggage for resons of simplicity (suitably stated)? Third, aren't their better theories of warrant that if true, would entail his is necessarily false? If we believed one of these, say evidentialism, wouldn't this show his theory epistemically impossible?
Being an athiest I reject the truth of theism and being an evidentialist I reject the epistemic possibility of the model. Wanting empirical evidence for the existence of the sensus divinatus I find belief in that unreasonable too, but I feel like I am getting off the boat too easily.
Christian, I think you are making philosophical life a little too easy. By "evidentialist" I think you mean "empiricist", a view hard to articulate and harder to defend. I think the right way to go here is Matt's: ask Plantinga to defend the plausibility of the model, and then we'll get some real good epistemology out of the defense!
There are several quite interesting comments on this issue over at Certain Doubts, if any are interested.
Jon,
You wrote:
Here's why it's perplexing. Try to describe a belief that is a paradigm case of an epistemically problematic belief. I submit that I can always construct a model on which, together with Plantinga's account of warrant, the belief in question would be warranted.
I reply:
But that's not true. There are a lot of beliefs for which a model of the sort plantinga offers for theism and christianity cannot be offered. There are many models which, if true, imply that the beliefs they entail *do not* have warrant. For instance, Plantinga would cite 'naturalism' as an example. If naturalism is true, then our belief that it is true, even if right, has no warrant.
Jon,
You wrote:
Here's why it's perplexing. Try to describe a belief that is a paradigm case of an epistemically problematic belief. I submit that I can always construct a model on which, together with Plantinga's account of warrant, the belief in question would be warranted.
I reply:
But that's not true. There are a lot of beliefs for which a model of the sort plantinga offers for theism and christianity cannot be offered. There are many models which, if true, imply that the beliefs they entail *do not* have warrant. For instance, Plantinga would cite 'naturalism' as an example. If naturalism is true, then our belief that it is true, even if right, has no warrant.
Newman--you wrote:
reply:
First, your claim about "many models" is irrelevant. There are many models on which Christian belief is unwarranted, but that doesn't undermine Plantinga's argument.
Second, naturalism may be an exception, but it can't be one without granting the soundness of Plantinga's objection, and that can't be taken for granted. There are very few who've carefully examined the argument and think it works.
Third, your remarks about beliefs entailed by models has to be wrong, unless it's a necessary truth that the particular belief you're thinking of is necessarily false. If you've got a model that entails that p is true, that's the best case you could ask for when wishing for a model to legitimate the belief that p.