Manipulation arguments and the explanatory nature of moral responsibility Manipulation arguments against moral responsibility (or against compatibilism about moral responsibility) rely on the following assumption: (M) Manipulation reduces responsibility because of features shared with causation (or with deterministic causation). Consequently, they would be undermined if the reduction were due specifically to the agent’s being manipulated—intentionally caused to act in ways not relying on the agent’s rational cooperation. To strengthen (M), Pereboom has argued that responsibility is equally reduced by versions of his manipulation cases where analogous natural events are substituted for manipulators. Significantly, however, these versions seem less intuitively compelling, suggesting that (M) is mistaken. In this talk, I suggest that manipulation arguments seem uniquely compelling because manipulation provides particularly straightforward cases of actions caused by conditions outside the agent’s control. Moreover, equally straightforward non-manipulative cases are possible: manipulation arguments are merely a species of arguments from straightforward causation by external factors. Such arguments rely on the following assumption: (S) Straightforward causation by external factors reduces responsibility because of features shared with causation (or with deterministic causation). (S) might seem more plausible than (M): although the intentional and social nature of manipulation might be especially responsibility undermining, whether causation is straightforward in the relevant sense depends on explanatory interests and perspectives, and it might seem that moral responsibility must rest on a more secure basis than that. Before closing, however, I provide evidence that our everyday notion of moral responsibility is essentially guided by certain explanatory interests, suggesting a way for defenders of moral responsibility to reject (S).