Construing miracles as \textquotedblleft{}violations,\textquotedblright I argue that a law of nature must specify some kind of possibility. But we must have here a sense of possibility for which the ancient rule of logic---ab esse ad posse valet consequentia---does not hold. We already have one example associated with the concept of statute law, a law which specifies what is legally possible but which is not destroyed by a violation. If laws of nature are construed as specifying some analogous sense of what (...) is naturally possible, then they need not be invalidated by a (rare) violation, and Humean miracles remain a genuine possibility. (shrink)

TAKING ONE STANDARD DEFINITION OF ’MIRACLES’ AS ’VIOLATIONS OF LAWS OF NATURE, BY A VOLITION OF GOD’, I ARGUE THAT NO REPORT ASSERTING THE OCCURRENCE OF A MIRACLE CAN BE TRUE. WHATEVER IS INCOMPATIBLE WITH A TRUTH MUST ITSELF BE FALSE, AND NO STATEMENT OF A GENUINE LAW OF NATURE CAN BE OTHER THAN TRUE. OBJECTIONS TO THE ARGUMENT, INCLUDING THOSE BY MACKIE AND SWINBURNE, ARE REBUTTED.