In the preceding paper, ~ Hector-Neri Castafieda presented a variety of objections to a system of deontic logic sketched in my book Doing the Best We Can. 2 In this paper, I attempt to answer some of the most important of Castafieda's objections. First, however, it will be useful to say a few words about the DBWC system.
My fundamental thesis is that we ought to do the best we can. I take this to mean that what a person ought to do, as of some time, is whatever he does in the best of the possible worlds then accessible to him. But, of course, we often fail to do the best we can. My view is that if someone fails to do what he should have done in the first place, then he ought to do the best he can consistent with that failure. More generally, consider any condition, q. To say that as of some time, someone ought to see to p, given that q occurs, is to say that he does see to the occurrence of p in the best of the q-worlds then accessible to him.
For present purposes the most important sorts of obligation are two sorts of individual moral obligation. A full-blown statement of absolute individual moral obligation takes this form: MOs, t, p. 'MO' is the moral obligation operator; 's' is to be replaced by the name of some person; 't' is to be replaced by the name of some time; 'p' is to be replaced by some sentence (or an abbreviation, such as a sentence letter). The sentence as a whole is read: 'as of t, s morally ought to see to it that p occurs', and it is true (roughly) iff in all the best worlds accessible to s at t, p occurs?
A statement of conditional individual moral obligation takes this form: MOs, t, p/q. It is read as follows: 'as of t, s morally ought to see to it that p occurs, given that q occurs.' According to the account proposed in DBWC, 'MOs, t, p/q' is true (roughly) iff p is true in all the best q-worlds accessible to s at to 4