Yes, so, um, well, i mean, i think we should give a fair hearing to objections to semi compatible sm of any sort. And maybe we can do that in just a minute, you gentlemen have the time. But i will just say beforehand, i do think that, well, there are two kinds of problems. I can just state them very quickly. One is, it looks like it really is difficult to identify what counts as ficiently responsive. Range of reasons to which an agent is responsible that renders her free, rather than unfree. So it's very natural to think that, if you want to discount certain cases of unfree action, you say, all the agent isn't reasons responsive. But look, take somebody who's clearly unfree, sy she has an extreme pobiao o compulsive disordera merly has the example of a anagora phobic. And you want to say, ye a standard way, areasos responsive, theorist would say that the agoraphobic isn't free, as when she stays in her house. Um, a. She's not responsive to the many good reasons to leave her house. But does that mean sheis not responsive to any reasons whill if you put it on fire, that would give the agorfobga reason to leave. And suppose they would leave, the mere fact that they would flee in fear terror is not a reason to think that on their they have sufficient reasons responsiveness to be free. Well, what the hell is sufficient? It's very hard for the reasons responsive theorist to nail that down. So that's a problem. Let me say, forgive me one more thing before i rap this chunk of our discussion. Op just because i alluded to it earlier and it almost slipped my mind. I i would like to go back, though, to the case of the willing ad it and use it as a test case to think about different judgments that would be issued by the frank ferdianor the mesh or mesh theories generally say, and a reason's responsive theorist. You see, i'm one of the reasons that i think despite the problems or the difficulties or the hurdles of advancing properly, a reason's responsive theory, i think it does better with cases than, for instance, frank approach. Lik frankfort high lights the willing adect as an illustrative case of an agent who acts of her own free will because of the satisfaction of the masked mesh, and is free t o relevant sense. But i've always found the case of the willing adoct to be a really problematic case, as an example of an agent who retains a kind of free of the the agent is suffering from a severe addiction. And on a reasons responsive approach, the agent might act in accord with what she most desires. But if the addiction is severe enough, she is not adequately reasons responsive to be free, even on the basis of just semi compatible or source compatible considerations. And so i think those kinds of cases are cases that you could use to sut a way, the the differences between the the judgments that you get from a reason's responsive theory and mesh theories like forts. And my intuitions at least a line with thinking that the reason's responsive theory has won up one nesh theories in cases like those. Well,

one attractive feature of semi compatibleism, or source compatibleis is that, as you mentioned earlier, its consistent with the soundness of the consequence argument, which we devoted a whole episode to in season one. But for any kind of compatiblist, including source compatiblists, the manipulation argument remains a challenge. We had an episode on that with dirk peraboom in season one, and in that episode, he referred to your response to the manipulation argument said, we should have you on the show, which we were already planning to do, but could you briefly explain the challenge from manipulation and then how you think compatible should respond?