Atheists and agnostics don’t like miracles (though ironically they need them to justify their evolutionary worldview: Five Atheist miracles and A miracle by any other name would be … called science?). They often claim that miracles are somehow impossible, or inherently improbable, or unprovable—although their proofs become circular, as explained in Miracles and science. The idea is that miracles can be safely ignored as an option before the evidence is considered (see Defining arguments away—the distorted language of secularism).
Skeptics often use the limits of science to justify ruling out miracle claims. For instance, we can’t use science to describe how miracles happen. After all, we have no empirical access to how non-physical beings might interact with the physical world to produce effects in it. But if there’s no describable ‘mechanism’ by which non-physical beings can produce effects in the physical world, can we justifiably believe that non-physical beings like God can affect the physical world?
Now, miracles should normally be considered not violations of ‘laws of nature’ but additions to them. Since scientific research is confined to ‘natural law’, it could not by definition investigate events outside it. Moreover, there are many things in science that are accepted though we do not understand their mechanisms. Examples abound in the properties of subatomic particles. As such, even within science, just because because we don’t understand something does not mean that it can’t be.
Nonetheless, there is a certain level at which the ‘mechanism’ by which a miracle is caused is indescribable. For instance, terms for describing spiritual realities (e.g. ‘non-physical’, ‘immaterial’, and ‘incorporeal’) are usually negative concepts—they tell us what such things are not, rather than telling us what they are. They don’t give us a readily available positive account of what non-physical beings are made of and how they make stuff happen, especially one we can test empirically.
But does it matter that the ‘mechanisms’ of miracles are beyond empirical description? Not really. Why? Well, how do we know a miracle has happened? It usually involves two aspects: an event not explainable by physical or biological causes, and a context for the event that warrants seeing it as the action of a supernatural agent.1
Jesus’ resurrection is a perfect example. There is powerful evidence that Jesus was seen alive by many people after He had died by crucifixion. That implies an event not explainable by physical or biological causes. After all, no biological or physical cause back then (or today!) could bring a man who died by crucifixion back to life in such a way as to convince people that the resurrected man was the conqueror of death. Moreover, Jesus claimed to be the Divine Messiah, and that His resurrection from death would be the ultimate vindication of His claims. This clearly provides a context that warrants seeing Jesus’ resurrection as a divinely caused event.
However, though we are inferring an adequate cause (God) based on the context and the effects we see, we have no way to investigate how Jesus’ body came back to life. Does this mean that the miracle conclusion is too fast? No. It’s based in part on what we know to be empirical fact—no biological or physical cause can spontaneously reanimate a corpse. This gives us two options with respect to Jesus’ post-mortem fate. We can limit the causes we’re willing to admit into the investigation to physical and biological causes, and say that we can’t know what happened. Or we can admit that we know too little about how putative non-physical causes work to be able to rule them out, and that by a process of simple elimination, a non-physical cause is the only plausible explanation left for the evidence surrounding Jesus’ post-mortem fate. As such, we arrive at the conclusion that a non-physical cause obtained without knowing how He did it. The only thing holding the skeptic back is his naturalistic bias.
There are also parallels between miracles and intelligent design. Both appeal to the inadequacy of natural causes, and both invoke an intelligent agent as the only plausible adequate cause of the effects we observe. The difference is that with design we don’t so much look for a religious context but for features common to designed objects (see David Hume and divine design and Who designed the Designer?). Even though the designer may be indescribable, it doesn’t mean we can’t infer a designer from the effects we observe. ‘Intelligent Design’ proponent Stephen Meyer, addressing questions regarding the origin of the major animal body plans, put the matter in a helpful way:
“There is a different context in which someone might want to ask about a mechanism. He or she may wish to know by what means the information, once originated, is transmitted into the world of matter. In our experience, intelligent agents, after generating information, often use material means to transmit that information. A teacher may write on a chalkboard with a piece of chalk or an ancient scribe may have chiselled an inscription in a piece of rock with a metal implement. Often, those who want to know about the mechanism of intelligent design are not necessarily challenging the idea that information ultimately originates in thought. They want to know how, or by what material means, the intelligent agent responsible for the information in living systems transmitted that information to a material entity such as a strand of DNA. To use a term from philosophy, they want to know about “the efficient cause” at work.
“The answer is: We simply don’t know. We don’t have enough evidence or information about what happened … to answer questions about what exactly happened, even though we can establish from the clues left behind that an intelligent designer played a causal role in the origin of living forms.”2
Meyer goes on to provide an analogy from archaeology in the statues on Easter Island. It was easy enough to see that they were designed, but we still don’t know how they were designed. Another interesting example is the famous portolan charts.3 They are ‘too accurate’ for what we know of ancient cartography, and even medieval cartography, and yet they exist, but we don’t know how the maps were made. In the same way, we don’t need to know how a miracle happened to know that it happened.
Consider also biomimetics. It treats biological systems as designed systems that are so optimized that they actually teach us about how to apply science to technology, as well as giving us new technologies. Of course, biomimetics researchers often give the obligatory nod to ‘how well evolved’ the functional feature they’re copying is. But it’s empty praise (and committing the fallacy of “advantage proves adaptation”). And in reverse engineering functional biology, they are not reproducing the process (evolutionary or otherwise) that supposedly produced the feature they are reverse engineering. They are just reverse engineering the ‘final product’ with no reference to the processes involved in how the ‘original’ it came to be. But it’s the very fact that they are reverse engineering the feature of functional biology that strongly suggests these features look engineered because they were engineered.
But this also explains why the engineer’s inability to reproduce the process a divine designer used to produce something is utterly irrelevant. We don’t need to be able reverse engineer the process by which God produced something to be able to reverse engineer what God produced (not that we could reverse engineer everything God produced—we can’t create mass-energy from nothing, for example—but we might eventually be able to reproduce some things He has produced). For instance, say that Jesus miraculously healed a man blind from cataracts. He just spoke the word, and the man was healed! Now, consider Fred Hollows4 (who was an atheist) healing a man blind from cataracts … by means of eye surgery. And let’s say that the effects were the same in both people—their sight is ultimately fully restored. Would it be right to say that Fred Hollows has reproduced the effects of Jesus miracle? I think there’s a sense in which it would be right to say that. The change in circumstance brought about—the restoration of sight from cataract blindness—is the same in both scenarios. The only difference is how it happened (and likely Hollows’ patient’s recovery time).
The indescribability of the ‘mechanism’ by which a miracle happens is irrelevant to whether miracles are possible or knowable. We not only know too little about putative non-physical causes to describe them mechanistically, but we also know too little about them to rule them out. But we often do know enough of what is physically possible to rule out physical causes by processes of elimination and comparison.
... there can be a role for empirical anomalies to tell us at what points, and to a certain extent in what ways, God may have directly manipulated matter to produce His intended effects—in the context of events likely to involve miraculous elements for which we have independent testimony of in Scripture. In other words, we only expect such ‘empirical anomalies’ relating to physical effects produced in events like Creation Week and Noah’s Flood.We have an observable body of evidence in the sky. It gives us certain information about the objects up there, and also about their histories. How do we fit that in to the context of Genesis 1:14-19? That is the question of creation cosmology. This doesn't call into question the reliability of Genesis 1, but seeks to understand how the observable evidence of astronomy is consistent with it. Please see Modern science in creationist thinking and Distant starlight and the days of Genesis 1 for more information.