Abstract
We can safely infer that a proposition is possible if p is the case. But, I argue, this inference from the actual to the possible is merely explicative in nature, though we employ it at times as if it were ampliative. To make this inference ampliative, we need to include an inference to the best explanation. Specifically, we can draw a substantive conclusion as to whether p is possible from the fact that p is the case, if via our best explanation we can explain how p could occur again in the complete and coherent set of propositions that describes the actual world.