It's hard to find things that John Searle says that I would accept. Case in point:
"when one has several reasons for performing an action, one may act on only one of them; one may select which reason one acts on"
(Thanks to Neil Sinhababu for pointing to this passage.)
There must be some user manual for the mind that John Searle has that I don't. I don't know how someone can select which reason to act on. Acting for one reason rather than another happens, sure, but why think it something you do or choose to do?
Suppose we say that:
(i) What you do/choose to do has moral worth only if it is done for the right reasons; and,
(ii) A-ing for reason R1 rather than R2 is itself something you do.
If 'A-ing for R1' denotes an action, say, B-ing, _it_ has moral worth only if B-ing is done for the right reason. (If it is something you do and you can select the reason, surely you could choose to do it for the wrong sort of reason (e.g., you choose to act for the moral reason rather than the prudential reason for the reason that your mind-reading girlfriend who examines only the first layer of your motivating reasons is likely to be deceived into thinking you are a good person when in fact you are a cad). Seems like a regress looms given (i). Denying (ii) looks good. (I think this is essentially an argument from W.D. Ross and I think Alvarez and Hyman run a version of it in their paper where they argue that actions aren't events.)
Now, it's true (I think) that you can choose not to act for some reason by not doing the act that would be motivated by that reason. So, if I choose to sleep with someone knowing both that we're in love and that I'll get a free lunch out of the deal, I don't think I can choose to sleep with this someone for the lovey reason rather than the foody reason and choose not to act on the foody reason (or choose to sleep with someone from the other reason). Of course, I can determine that I won't act from the foody reason at all by simply refusing to sleep with this someone. But, that's another matter entirely.
Is that wrong? The problem isn't just a regress. I have no experience I'd describe as the experience in which I choose or decide to act for one reason rather than some other one. I've never thought, "Well, now that the question as to whether to do that is settled, what shall my reason be for doing it?"