The following remarks are based on the first two sections of Chapter Four, "Deontological Desiderata," of William P. Alston's Beyond "Justification": Dimensions of Epistemic Evaluation (Cornell UP, 2005), pp. 58-67.
1. It makes sense to apply deontological predicates to actions. Thus it makes sense to say of a voluntary action that it is obligatory or permissible or impermissible. But does it make sense to apply such predicates to beliefs and related propositional attitudes? If I withhold my assent to proposition p, does it make sense to say that the withholding is obligatory or permissible or impermissible? Suppose someone passes on a nasty unsubstantiated rumor concerning a mutual acquaintance. Is believing it blameworthy? Is suspending judgment required? Or is deontological evaluation simply out of place in a case like this?
3. This brings us to the issue of doxastic voluntarism: Are any of our believings under our direct voluntary control as regards their coming into existence? To introduce some terminology of my own:
Extreme doxastic voluntarism: ALL beliefs are such that their formation is under one's direct voluntary control.
Limited doxastic voluntarism: There are only SOME beliefs over the formation of which one has direct voluntary control.
Doxastic involuntarism: There are are NO beliefs over the formation of which one has direct voluntary control.
Note that the issue concerns the formation of beliefs, not their maintenance, and note the contrast between direct and indirect formation of beliefs. Roughly, I form a belief directly by just forming it, not by doing something else as a means to forming it.
4. Alston's project in the pages I am covering is, first, to show that we do not have "effective voluntary control over believings." (p. 60), and second, to infer that the deontological triad "required, permitted, or forbidden" has no application to believings and such related propositional attitudes as withholdings. Alston, then, is a doxastic involuntarist.
Alston's thesis, to be precise, is not just that we do not, but that we cannot, have direct voluntary control over belief formation. He means the 'cannot' psychologically, not logically. His point, then, is that it is psychologically impossible to believe at will.
You can see why his book is entitled Beyond "Justification." Alston doesn't think that there is any place in epistemology for justification if this term is taken deontologically. The sneer quotes suggest that there is no such thing as epistemic justification. Thus, if I understand Alston, it makes no sense on his view to say that a belief or a believing is justified since that is tantamount to saying that it is permissible. But no believing is permissible or impermissible since no believing is under the direct control of the will. Therefore, for Alston, justification interpreted deontologically has no place in epistemology. There is no such thing as epistemic justification.
5. Alston's examples in support of his thesis fall into three categories.
a) One cannot believe at will what is obviously false. It is obviously false that the Romans continue to exercise hegemony over Western Europe, and it is hard to disagree with Alston when he concludes that no one has the power to believe this falsehood at will, just by deciding to believe it. (p. 63)
b) One cannot not believe what is obviously the case. It is obviously the case that this thing in front of me is a computer monitor. Can I disbelieve this perceptual deliverance? No. Seeing is believing. It is a more subtle question whether I can suspend judgment in the manner of Husserl's phenomenological epoche. But this is a topic for a separate post. For now I am happy to concede to Alston that one cannot disbelieve at will what is obviously the case.
c) The matter becomes much more difficult when we turn to propositions from religion, philosophy, science and elsewhere that are neither obviously true nor obviously false. It is not obviously true that God exists, but neither is it obviously true that God does not exist. It is not obviously true that doxastic voluntarism is true, but neither is it obviously true that it is not true.
Suppose I am concerned with the freedom of the will, study the issue thoroughly, but am torn between libertarianism and compatibilism. It is surely not obvious that one or the other is true. If the positions strike me as equally well-supported, then nothing at the level of intellect inclines me one way or the other. Must not will come in to decide the matter? Or consider the weightier question of the existence of God. Suppose the arguments pro et contra strike me as equally probative so that, at the level of intellect, I am not inclined one way or the other. If the issue is to be resolved, must I not simply decide to believe one way or the other? But Alston will have none of this: "How could we do that any more than, lacking any reasons at all for one alternative rather than another, we decide to believe that the number of ultimate particles in the universe is even rather than odd?" (p. 65)
This response packaged in a rhetorical question strikes me as very weak. No one cares what the number of particles is let alone whether it is odd or even. Indeed, it is not clear that the question even makes sense. (How could one possibly count them?) The God question is toto caelo different. In Jamesian terms, the God question is live, forced, momentous, and not intellectually decidable. A live issue is one that matters to us and seems to need deciding. Whether the number of ultimate particles is odd or even is certainly not live. A forced issue is one that is compulsory in the sense that we cannot not take a stand on it: to remain agnostic or uncommitted on the God question is practically to live as an atheist. There is nothing forced about the particles question. A momentous issue is one about which it matters greatly which position we adopt. The particles question is clearly not momentous. An intellectually undecidable question is on which, if it is to be decided, must be decided by an act of will.
So what I would say to Alston at this point in my study of his book is that in some cases — those that fit the Jamesian criteria — the will does in fact come into play in the formation of beliefs and indeed legitimately comes into play. To the extent that it does, a limited doxastic voluntarism is true.