In the Alastair McKinnon paper (“Miracle” and “Paradox”) I cited in a recent comment, it is argued that miracles of a certain kind are impossible. The impossible ones are the ones that would provide any evidence of supernatural objects or occurrences.
We can call any of those most wondrous miracles a “miracle1.” Any other miracle, which would be more in the nature of an amazing coincidence, we can designate as a “miracle2.” Note that there’s nothing about a coincidence that provides evidence of the supernatural (something beyond nature). And what does it mean to be “supernatural”? It means to violate a natural or physical law.
OK, so why does McKinnon say there can’t be any miracle1s? He says there are two ways of looking at natural laws. The first is to make them absolute. An absolute law would be of the following form:
Whenever events of type X happen, events of type Y must happen.
But many people think such absolute laws are really not to be found. Such people would say that physical laws should really be taken to be of this form:
99.999% of the time that events of type X happen, events of type Y will also happen.
So, for example we may think that If you throw a ball in the air (under normal circumstances) either that it must fall, or that in an extremely high percentage of instances it will do so.
Ok, let’s take an event that is standardly called a miracle, say, Jesus walking on water. To be a miracle1, it would have to violate a natural law. If the supposed law is
Whenever a something of the density of human being tries to walk on water s/he will sink.
then, if Jesus really walked on water, that supposed law was actually false, because it says that EVERY SINGLE TIME something with the density of a human being attempts to walk on water, there will be an epic fail. Thus, what Jesus would have done by walking on water would have been to show that something that seemed to be a physical law wasn’t really one after all.
But what about if we construe laws as statistical rather than absolute?
This time, the walking on water would be consistent with a law that doesn’t require that a particular result occurs EVERY SINGLE TIME.
So, McKinnon concludes, in neither case is a natural law violated, which means that in neither case have we been given anything that provides evidence of the occurrence of anything supernatural (outside of physical laws). We have either disconfirmed something we wrongly believed was a physical law, or we have found an amazing coincidence (like a bee stinging a locomotive engineer just in time to wake him up before the train hits a car–amazing, but no proof of anything supernatural).
FWIW, I think this is a very clever argument. And what I think it does is suggest that being able to bring out “amazing coincidences” should be enough for theists. If somebody or something can make my prayers come true, I don’t care if this somebody or something is, strictly, supernatural or not.
As I’ve said, though, Christians aren’t any better at getting what they want by wishing/praying than anybody else, AFAIK.