This is the original essay in which Frankfurt introduced his counterexamples. But what seems troubling to me is that the very force of those counterexamples is based on a lack of causal connection between the controller and the action. What insures that we cannot do otherwise is something that doesn't cause the action in question, therefore we are more inclined to jump to the "responsible" verdict. Or some people are, others, who might be inclined to accept the central argument for open theism, may disagree. But whatever the case is, Frankfurt even points out the lack of causal connection as the reason for giving the "responsible" verdict, then the argument jumps the tracks and argues for the compatiblity of moral responsibility with causal determination.
You change the very thing that makes the counterexamples work when you go to the case of causal determination, yet these counterexamples are supposed to undergird the compatibility of moral responsiblity with causal determination. This seems just wrong.