[NB: A lot will need to be edited to take account of material at the end of the paper.]
I'm reading Schroeder's "Means-End Coherence, Stringency, and Subjective Reasons" as I'm interested in Ewing's Problem and his take on it. As someone who is somewhat inclined to try to accommodate the data that led Ewing to say that there is a special sort of subjective 'ought' that captures the kernel of truth in the idea that you ought to let your conscience be your guide in terms of some sort of wide-scope 'ought' statement, I feel obliged to address the arguments of this paper. S raises three objections to the wide-scope solution to Ewing's Problem. First, consider:
(Con) If you believe you ought to A, you ought to A.
There's got to be some sort of normative relation between the belief and the action if one conforms to (Con) (or the belief and failure to act if one fails to conform), and the wide-scoper suggests that (Con) misrepresents that normative relation. Really, we ought to replace (Con) with:
(Wide-Con) You ought to: do A or do not believe you ought to do A.
First, S objects that duties of conscience are not 'symmetric' in the way that the duty (Wide-Con) imposes upon us is. My reaction is to say that if there wasn't some independent reasons that bore on whether to act or whether to believe, the duty would be symmetric. However, if there are independent reasons that bear on whether to act or believe, we could appeal to those reasons to capture the sense in which it seems that the duty is not symmetric. S anticipates this sort of response and says this:
A naive response to this objection holds that though changing his beliefs is not
ruled out by the instrumental principle, it is ruled out by some other principle
governing theoretical reason, which says not to change your beliefs about what you
ought to do, or some such thing. But this is short-sighted. In general, if Yves ought
to do either A or B, and ought not to do B, then it follows that Yves ought to do
A. This means that if there really is a principle according to which Yves ought not to
change his belief, then in any circumstance in which that principle applies, the only
way for him to fulfill his Wide-Scope requirement to either-do-A-or-not-believethat-
he-ought is for him to do A. And so it follows that in those cases, he ought to do
A. But that just means that in any case in which Yves ought not to change his belief,
it follows from the Wide-Scope view that his belief must be true. So the Wide-
Scope view is still committed to holding that Yves’s belief about what he ought to
do is infallible in any case in which he is believing rationally. And this looks like a
second bad problem for the Wide-Scope view (pp. 227).
Part of me thinks this isn't a big deal. First, there are some who think that you cannot rationally fail to identify what you ought to do or believe. They'd embrace the conclusion S thinks spells doom for the wide-scoper. Second, those who do not think that you can always rationally identify what you should do would likely say that it doesn't follow from the fact that the relevant ought-belief is held rationally that it is held permissibly, in which case they needn't say that any rationally held belief about what you ought to do is true. (For example, suppose you think knowledge is the norm of belief. In cases where you believe you should A but you shouldn't A you can't know you should A (b/c of the factivity of 'knows'), so the wide-scoper who opts for the knowledge account is not committed to the unfortunate view S thinks sinks the wide-scoper's view.
I think it's interesting to note that S seems to be committed to a kind of infallibility problem of his own if he still sticks by his view that says that when someone judges that they ought to A there is an objective reason for them to A since that suggests that the subject has infallible judgments about what kinds of considerations would count in favor of A-ing. We're just not that good at identifying favorers.
Later in the paper, S defends the idea that if someone intends to do S, he subjectively ought to do A. His defense involves the idea that if you intend to A you have some beliefs such that, if they are true, you objectively ought to A. This places an odd constraint on intention: you can intend only what it is possible to be such that you ought to do it. (If only it was that easy (for minimally intelligent people) to quit smoking!) It seems an obvious problem with this suggestion can be found in aisle 7. Yes, that's where they keep the soup. Seeing the array of cans, I know I ought to get Campbell's Tomato soup but no can is jumping out as the can to buy. If I don't judge that there is some particular can that I ought to buy, how can I intend to take a can from the shelf? S's reply is wicked clever:
I think that what cases of picking really show is that the intention to do A needn’t
require the antecedent belief that one ought to do A. But I don’t think they show that
someone who intends to do A need not believe that she ought to do A. Before Zach
makes any decision, of course, the carton on the left and the carton on the right are
perfectly on a par. But after he has made his decision, the carton on the left comes
out on top. The fact that he has picked it is now a relevant difference between the
two. So if he ought to take one of them, the one on the left is the one that he ought to take. (238)
It's clever, but I think it's wrong. Suppose that as you reach for the can you intend to pick and dutifully form the belief that you ought to pick that can, you are momentarily distracted and fail to notice that you grabbed an equally good can sitting next to the one you intended to pick. First problem. There's no sense in which you failed to do what you should but there's a perfectly good sense in which you failed to do what you intended. Thus, what you intended to do is not something that you subjectively ought to do because what you subjectively ought to do is (roughly) what you objectively ought to do when certain beliefs of yours are true. Second problem. To comply with the requirement that you only intend to do what you believe you ought to do (in some sense of ought), don't you have to form false beliefs? Aren't these beliefs the subject knows full well would be false?
_End Transmission. To be resumed later.