The natural and the supernatural

What it is

The supernatural is whatever lies outside or above nature — not merely unexplained, but of a different order from the things science studies. Gods, spirits, souls, magic and miracles are the standard inhabitants of the category.

The category has an awkward property. Ask what would happen if a supernatural force turned out to act reliably, measurably and repeatably — if prayer produced a consistent, replicable effect under controlled conditions — and the answer is that it would become part of nature. We would have discovered a new regularity in the world, added it to physics, and stopped calling it supernatural. That is what happened to magnetism, to contagion, to lightning.

So "supernatural" turns out not to name a kind of thing at all. It names a relationship between a phenomenon and our current understanding — and it is defined partly by not being tractable. A supernatural claim that became well-evidenced would cease, by that very fact, to be a supernatural claim.

Where it came from

The distinction is not universal and it is not ancient. Most of the traditions this site covers did not draw it, and many still do not.

In animist and folk cosmologies, spirits are simply part of the furniture of the world — features of the landscape with their own dispositions, no more outside nature than weather. Anthropologists working across a range of societies have repeatedly found that the categories "natural" and "supernatural" do not map onto local distinctions, and that imposing them distorts what is being described. The same holds for much of classical Chinese thought, where qi is a natural principle rather than an intrusion into nature.

The sharp version is a product of medieval Christian scholasticism, where a stable order of nature was distinguished from the realm of grace acting upon it. It hardened further in early modern Europe alongside the idea of fixed, universal laws of nature — because you cannot have a category of exceptions until you have a rule for them to be exceptions to. The concept of the supernatural, in other words, is a by-product of the scientific revolution rather than something it inherited.

How it is used

Science operates under methodological naturalism: look for natural causes, because those are what the methods can find. This is a working constraint on inquiry, not a claim that nothing else exists, and the distinction matters — it is the reason a religious scientist is not being inconsistent. It is also routinely conflated with metaphysical naturalism, the much stronger claim that the natural is all there is, by people on both sides of the argument who find the conflation useful.

The category also does damage through the God of the gaps. If the supernatural is defined as what science cannot explain, then it necessarily shrinks every time science explains something. Theologians have warned against this framing for over a century, on the grounds that it makes God a temporary hypothesis with a poor track record — a warning generally attributed in its sharpest form to Dietrich Bonhoeffer, who objected to a God found only at the boundaries of knowledge rather than at its centre.

What we can and cannot say

We can say the category is doing far less work than its prominence suggests, and that most arguments framed as natural-versus-supernatural are really about something else: whether a phenomenon is regular or capricious, testable or untestable, mechanistically describable or not. Those are better questions, because answers to them would actually change what you do next.

We can say the boundary has moved constantly and in one direction. A very long list of things once held to be outside nature are now inside it, and no item has travelled the other way. That is a real asymmetry and it deserves the weight it carries — though it is a track record rather than a proof, and track records can end.

We can say that if something in this site's territory — say, a well-established psi effect — were ever robustly demonstrated, the correct response would not be "science has been refuted". It would be "nature is larger than we thought", followed by a great deal of work figuring out the mechanism. Arthur C. Clarke's remark that sufficiently advanced technology is indistinguishable from magic runs in this direction too: the distinction may say more about the observer than about the thing observed.

What we cannot say is that the category is meaningless. It marks something real about how people organise their expectations, and for anyone inside a tradition that uses it, it does genuine work. But it is a poor tool for settling arguments, and this site will generally prefer to ask whether a specific claim is testable, and what happened when it was tested.

Further reading

• Robert Larmer, "Is There Anything Wrong with 'God of the Gaps' Reasoning?" (2002) — a defence, worth reading against the grain of the usual dismissal.
• Barbara Forrest, "Methodological Naturalism and Philosophical Naturalism" (2000) — careful on the distinction that most of the argument runs aground on.