I'm interested in this:
You ought to see to it that: if you believe you ought to \phi, you \phi (Broome's Principle).
(1) You ought to see to it that: if you believe that you ought to \phi, then you \phi.
(2) You ought to see to it that you believe that you ought to \phi.
(C) You ought to see to it that you \phi.
Broome once thought of “ought” as expressing a relation between an agent and a proposition. On his view, “You ought to believe p” and “You ought to see to it that you believe p” have the same meaning (1999: 79). Suppose you are handed a gun and a pill. If you take the pill, you will believe that you ought to shoot one person with your gun. If you take neither, you know that hundreds will be killed. It is tempting to say that in the case described, you ought to see to it that you believe you ought to shoot one person with your gun. Presumably, this requires taking the pill. So, I think you should take the pill. As a result of taking the pill, you believe that you ought to shoot one person with your gun. I do not think this is what you ought to do. Either Broome's Principle is false, “ought” does not pick out a relation between agents and propositions, or Broome's Principle is not to be understood in terms of combinations of actions and beliefs an agent ought to bring about.
So, what does Broome's Principle mean? I know that we're supposed to talk about rational requirements now rather than wide-scope "ought", but is that because everyone has repudiated the idea that "ought" takes wide-scope in the way it does in Broome's Principle?