1. The Ontological Problem of Miracles
The ontological problem of miracles is the problem of explaining what miracles are and how they are possible. These questions are logically prior to the questions of whether any miracles have occurred or whether such-and-such an event is a miracle. You may believe, for example, that miracles have occurred, and you may cite as an example of a miracle Therese Neumann's subsistence for decades on no food except a daily communion wafer. The philosopher of religion, without necessarily denying either the general occurrence of miracles or this particular instance, will then ask: what is it that makes this supposedly miraculous event miraculous and how is the existence of miracles rationally integratable into the rest of what we know and believe about the world? In short: What are miracles? How are they possible? The philosopher of religion needn't be arguing for miracles or against them; he may simply be trying to understand them, both in themeselves, and in relation to everything else.
In this respect the philosopher of religion may comport himself like the typical philosopher of science. It is rare for a philosopher of science to argue against any scientific procedure or result; for the most part, philosophers of science simply try to understand science: the nature of scientific explanation, the status of laws, etc. They do not question the truth of scientific results.
The ontological problem of miracles is also logically prior to the various epistemological questions that arise, questions about the reliability of the testimony of eye witnesses, etc.
2. The Problem of Mind-Body Interaction
This problem arises in the philosophy of mind for the substance dualist. If one is a substance dualist and one takes some mental events to have physical effects, and some physical events to have mental effects, then one is faced with the question: How is this causal interaction possible? There is reason to think that the physical realm is causally closed, which implies that everything that happens within it has a sufficient causal explanation in terms of other physical events, and that nothing but a physical event could have a physical effect.
3. The Miracles Problem as a Special Case of the Interaction Problem
On a standard definition of 'miracle' traceable to Hume, a miracle is (i) a transgression (violation, contravention) of a law of nature, which is (ii) brought about by a particular volition of the Deity. Now my mind, even if a Cartesian res cogitans, is not divine but it is a supernatural agent, assuming we mean by 'nature' the actual system of space-time-matter. (We could also mean by 'nature' the actual system of created entities, and if we mean this, it would follow that garden variety Cartesian minds are not supernatural agents.) But if I am a res cogitans, and I will that my hand grab hold of a cigar, and this grabbing-hold takes place, then the problem arises as to how this is possible given that laws of nature (conservation laws) would appear to be contravened. But this is just one of the ontological problems of miracles generalized. To put it the other way around, the problem of miracles is a special case of the interaction problem, the special case in which the supernatural agent is God.
Of course, the physical universe is not God's body, so that is a point of disanalogy. But it seems to be the case that, if we define 'miracle' as above in the Humean manner, then the problem of how God can intervene in, or interfere with, the physical world by acts of will is a special case of the problem of how any supernatural agent can intervene in, or interefere with, the physical world by acts of will.
4. Solving Both Problems in One Fell Swoop?
What I just wrote can be questioned in several ways. For example, one can question the Humean definition of 'miracle.' But let's assume that the miracles problem is a special case of the interaction problem, the case in which the supernatural agent is God. The question I want to raise is whether a certain solution to the miracles problem is also a solution to the interaction problem.
The solution concedes that laws of nature, if there are any, cannot (logically) be subject to counterexample. After all, a law that admitted of counterexamples would not be a law! So if there are laws, they are exceptionless. A law is not to be confused with a law-proposition. A law is the truthmaker of a law-proposition. The latter are 'bivalent' entities, either true or false; thus it makes sense to say of a law-proposition that it is false. But laws are not 'bivalent': it makes no sense to say of a law that it is false or nonobtaining. A false law is no law at all. Every law obtains.
But the solution goes on to say that every law of nature has the form:
(x)(Fx –>Gx) unless a supernatural agent by an act of will brings about something incompatible with Gx.*
This seems to do the trick. If this is the deep form of laws, then laws of this form are exceptionless. They are rendered exceptionless because the possibility of exceptions is built into them. There is something fishy about this, but let's hold our noses for the nonce.
If the solution works for miracles, however, it should also work for the general interaction problem. And contrapositively: if the solution does not work for the general interaction problem, then it should also not work for the problem of miracles. So why not say the following: When a volition, which is a state of my Cartesian mind-substance, directly causes some event in my brain that issues in the grabbing of a cigar, there is no violation of any natural law since all such laws hold only on condition that no supernatural agent berings about anything incompatible with the laws!
If so, we have just solved the interaction problem. But surely this is too quick. No one who takes the interaction problem seriously will consider this to be a solution to it. Since the solution does not work for the interaction problem, it does not work for the miracles problem either. We are left with our problem of how miracles, Humeanly defined, are possible.
_______________
*Cf. Joshua Hoffman's comments on George Mavrodes' paper, "Miracles and the Laws of Nature," Faith and Philosophy, vol. 2, no. 4 (October 1985), p. 349. Hoffman endorses something like the solution I have sketched.