In the last few years, there has been a growing philosophical interest in the problem of moral responsibility for omissions. Like actions, however, omissions are not all-of-a-kind. Recently, most of the research effort in this field has been devoted to the so-called unwitting omissions. However, in some cases, people make clear-eyed, or quasi-clear-eyed, decisions about not interfering with a given course of action potentially having unethical consequences (let’s call these decisions witting omissions). In this paper, we abstract away from the epistemic concerns that typically refer to unwitting omissions to discuss the problem of moral responsibility for omissions as omissions, i.e., as non-events that may contribute to the occurrence of a state of affairs without necessarily being their primary cause. In particular, we call attention to how to define the set of omissions we are accountable for. Indeed, even narrowing the scope to witting omissions, there is an awful lot of morally undesirable events that we could contribute to preventing if we just wanted to do so. Thus, the question is: in which cases are we responsible for our witting omissions? In this perspective, we first consider the proposals of referring to derivative, role, or vicarious responsibility for arbitrating between the relevant cases. Although not mistaken, these proposals are helpful only in a limited subset of situations. Employing the example of a witness witnessing a crime by chance, we discuss a more encompassing strategy. Siding with those who see omissions as causes, we defend a counterfactual approach based on identifying when people could intervene and are normatively required to do so.