I cannot think of any reasonable definition of “natural”, “physical” or “material” which is concordant with our everyday use of the terms. It is remarkable how much time is spent arguing over theses like “naturalism” when most of the participants are not even able to give reasonable definitions of these terms. Of course, there is an easy answer: naturalism is the view that nature is all that there is. Similarly, physicalism might be the view that only physical things exist. That people find this satisfying is remarkable to me. Is it not obvious that this pushes the question back to what is meant by “natural”, “physical”, etc? And when I ask for definitions of these, all the answers I get are different in content and similar in inadequacy.
One common tactic is to attempt a definition by giving examples of ostensibly natural and supernatural things. Not only is this a peculiar way to answer the question, it almost always gives rise to an absurd way of reasoning. Many, for example, argue that theism is supernatural, and then give examples of supernatural things like ghosts, fairies, magic, and so on – essentially, anything that is implausible or in some sense disconfirmed by scientific inquiry. It is then an easy induction from the fact that all other supernatural things are implausible or disconfirmed to the conclusion that, therefore, theism is implausible or disconfirmed, and requires such a huge shift in our ontology that the explanatory benefit is outweighed by these considerations. But this is obviously absurd if we can’t give a useful unifying definition of “natural” which explains why all these other entities/theories, including God, are implausible.
What really seems to be happening is that people are making a set of all implausible things, adding God, giving it an artificial name, and then declaring that the inductive inference from all other members having a property (i.e. being implausible) to God having the property. But this is obviously absurd. I could just as easily give a stipulative definition of a new term “falseism”: a proposition is a falseism if and only if it is either false, or it is the proposition, “there are no Gods” (call this proposition A for expedience). Then, I could easily construct any number of parodies of arguments typically used against theism on account of its being supernatural. For example, I could make a simple inductive argument from the falsehood of every single other falseism to the falsehood of A. Or perhaps we could pull a Hume: we’ve never observed a falseism to be true, and so no amount of evidence or reasoning could reasonably convince us that any new falseism is true (for example, A): “There must, therefore, be a uniform experience against every [falseism], otherwise the [proposition] would not merit that appellation. And as a uniform experience amounts to a proof, there is here a direct and full proof, from the nature of the fact, against the [truth] of any [falseism]; nor can such a proof be destroyed, or the [falseism] rendered credible, but by an opposite proof, which is superior.”
And, of course, we can come up with any number of these arguments. Adding a falseism to our epistemology would be so outrageously different such that we would need extraordinary warrant, and certainly no warrant which we can practically attain. There’s no evidence of any other falseism being true – so we don’t even know if it’s possible that this falseism (A) could be true. And so on. Most arguments against theism (on grounds of supernaturalism) can be parodied this way.
One more thought: don’t these terms seem a bit ad hoc? Surely things that we now accept as natural or physical would have, at some point, been deemed supernatural or non-physical? What about certain concepts entertained in contemporary physics – string theory? Wormholes? Special relativity and the implications for time? Energy? Dark energy? Dark matter? Virtual particles? It’s easy to imagine these being seen as non-natural by people before they were taken seriously within physics, but it seems very much like serious consideration by the scientific community is enough to deem them supernatural. It simply doesn’t seem like “naturalism” has the semantic consistency to be a useful label.
And so that’s why, in short, I don’t buy the natural-supernatural distinction. It’s much easier to talk more specifically about ideas than to frame these debates as being about these two labels, with ever-changing meanings. I’d rather someone tell me why theism is false than why supernatural phenomena generally are false. And I’d rather argue simply for the truth of theism (or falsehood of atheism) rather than weigh myself down with gratuitous burdens like demonstrating the existence of things beyond nature (whatever that means).
Of course, I may be wrong. And I know that some attempts have been made to define it more rigorously (though usually, in my experience, at the expense of our intuitions about what we normally mean by “natural”). I’m happy to entertain such attempts, and this is no guarantee that I won’t eventually call myself a naturalist (or supernaturalist) if the terms become relatively fixed. But this is why, for now, I can’t see any merit in appealing to those labels.
(For those so inclined, atheist philosopher Gregory Dawes has an excellent discussion of this whole issue in his “Theism and Explanation”, in the first chapter. Much of the rest of the book is unpersuasive, but his first chapter is excellent).