Is It Correct to Say There is "No" Evidence For the Supernatural Part 1

I thought it would be helpful to put the argument I made together in one post, since it’s kind of scattered around in different places, to make it easier for others to understand the argument I’m making in the context of the whole argument. I’ll do my best to clarify things to avoid confusion as much as possible.

I plan to spit this into two parts, part 1 to explain my reasoning, and after some discussion of the first part, part 2 to give an example of an abductive argument that uses evidence from science to make a metaphysical claim. And just to clarify, I’m not saying that there’s empirical evidence for the supernatural that would in any way confirm it’s existence. Only that there is empirical evidence to support the claim that it does exists.

To start with I’m going to go by more or less standard dictionary definitions of terms used which I’ll post here. If you want to use a different definition make sure you mention how you are defining a particular term. Otherwise I’ll assume you’re using the definition I’ve posted here.

Scientific evidence: Scientific evidence is evidence which serves to either support or counter a scientific theory or hypothesis. Such evidence is expected to be empirical evidence and interpreted in accordance with scientific method.
(I’m assuming here that if it qualifies as scientific evidence at any time it would still be scientific evidence, or evidence from science, no matter what the context, even in the context of supporting a metaphysical claim.)

Scientific method: a method of procedure that has characterized natural science since the 17th century, consisting in systematic observation, measurement, and experiment, and the formulation, testing, and modification of hypotheses.

Supernatural: of or relating to an order of existence beyond the visible observable universe.

Empirical evidence: Empirical evidence is the information received by means of the senses, particularly by observation and documentation of patterns and behavior through experimentation. The term comes from the Greek word for experience, ἐμπειρία.

Now I’m going to introduce a couple of my own distinctions of two types of empirical evidence.

Direct empirical evidence: evidence from observation of an actual empirically accessible event or entity that has been repeatedly confirmed by the studies of different independent scientists and has been established as a scientific fact.

Indirect empirical evidence: empirical evidence which is used in an inductive or abductive argument to support an inference to a claim to an empirically inaccessible event or entity.

Now my contention is that in the context of the statement in question, to say “no evidence” is an unqualified use of a broad term that can be interpreted in various ways. However, what I find is that as it’s used in regards to the supernatural, most, if not all, naturalists use the term in the more narrow sense of my definition of “direct empirical evidence.” So unless it is qualified as such, I contend that it is incorrect to say there is “no evidence” when in the broad sense of that term it’s not true.

This is my reasoning for that claim. When empirical evidence is used abductively to infer a scientific claim, that evidence would fall into the definition of scientific evidence. Now to allow for that to qualify as scientific evidence in an abductive claim is to admit that evidence can be indirect in nature. So if it’s scientific evidence, and it is allowed in the indirect sense to qualify as scientific evidence in an abductive inference to a scientific claim, why would it no longer qualify as scientific evidence, or evidence from science, in an abductive inference when used to support a metaphysical claim?

Now beyond this there is the issue of whether or not an abductive argument using indirect empirical evidence can even be made to support a metaphysical claim. There is at least one such arguments out there that I’m aware of. And I think it can be demonstrated that there are more that can be made as well. But I’d like to keep that for part 2 after I see what objections there are (including ones that have been raised already so they’re all in one place) for what I’ve presented here so far.


Evidence is data that is more likely on a particular hypothesis than on any competing hypothesis.

To say there’s no evidence for (or against) the supernatural one would first have to define the supernatural and then try to derive predictions from this definition.

What I mean is once we have defined the supernatural we can consider what one would expect to exist if the supernatural is real, and then show that these expectations are more likely to exist if the supernatural is real, than on any competing hypothesis.
It’s important to understand that we are not just talking about showing that the evidence is merely consistent with the hypothesis. The data has to be more likely on the supernaturalism hypothesis, than on any other hypothesis.

This is a double-edged sword of course, because if we derive predictions from the hypothesis that says certain patterns are very likely if the hypothesis is true, yet we don’t find these patterns, this now constitutes evidence against the hypothesis. After all, the hypothesis said this data should be expected, yet we looked where the hypothesis said it should be but didn’t find it.

I would claim that there’s no evidence for the supernatural because the supernatural has not been meaningfully defined in a way where it leads to clear expectations. I haven’t seen a model of the supernatural that says a particular data pattern should exist.

1 Like

You still haven’t explained how/why you infer a supernatural cause from your indirect evidence as opposed to just a natural but as yet unknown explanation.

By my count you dodged the question at least 4 times on the other threads.


I would like to know how people in different disciplines react to this. I always thought of evidence as being neutral and that it could be used to support one hypothesis or another, or to disprove one or another. That said, the evidence is typically the same, and often used both ways. This would be the case in a court of law. The evidence is gathered by both sides, shared and then either side presents the evidence framing a narrative that supports one side or the other.

Do you (@Rumraket) see evidence in the way that you presented it? I would suggest that your definition above is better suited for “proof” rather than “evidence.” But it is possible that this is the clinical use of the word, and it appropriate for this situation. Either way, though, this is an instance, I believe, where understanding how the terms are being used and agreeing upon them up front is helpful. I think that there is plenty of “evidence” for the supernatural, but I think that there is very little “proof.”


I think supernatural defined as a cause that exists outside of the natural realm with power to create and responsible for the initial creation of certain aspects of the natural world, e.g., time, space, matter, and energy, would lead to the prediction that science would reveal certain aspects of the natural world for which it cannot account for with natural causes. And if there are aspects of nature for which after significant investigation the evidence keeps pointing to something that cannot be accounted for by known natural causes, that would be evidence in support of the claim.

1 Like

I will give a few suggestions, but they fall into my own definitions (above) of evidence and not necessarily proof. I think that the sheer number of species that had to develop at just the right time in order to form a healthy ecology points to the existence of an intelligent creator. I don’t believe that an unguided process of evolution alone is “fast enough” to allow for the food chain, end to end, to have evolved (many times over) on its own. It is much more likely that the system would collapse due to too much or too little food or predation.

I think that the number of just-so aspects that apply to the design of the universe and the solar system point to an intelligent creator. I do not believe that there is proof, per se, for either, nor do I believe that I can know (or one can tell through experiment) how, when or where this intervention may have occurred or if it was like George’s “cosmic poolshot” that he often discusses.

I think this definition is neither clear nor rigorous enough. What does “visible observable” mean? What sort of observations? Do personal visions count?

We’ve discussed this before:

Many people would say that if a controlled experiment shows up evidence (direct or indirect - those terms are not rigorously defined either) for the supernatural, then the “supernatural” becomes part of the natural. The evidence becomes regular scientific evidence. This goes back to the issue of defining “supernatural”. One could argue that quantum entanglement is as weird as anything proposed by supernaturalists, yet we regard it as a perfectly natural phenomenon.


What do you say is the smallest number of species which would constitute an ecology? We have environments like black smokers on the deep ocean floor with only a handful of extremeophile microbes doing just fine. Why couldn’t the first self replicators have an extremely simple ecology which increased in complexity over time? We’re talking at least 3.8 billion years here.

Then how fast is too fast, and how did you determine that number? Remember food chains we see now didn’t have to be present 3.8 billion years ago. Almost certainly much simpler ones were.

No, that is data, or “facts”. There is a difference between data and evidence. The data only becomes evidence (for, or against) in light of a particular hypothesis that makes the evidence more, or less expected, or unexpected.

This would be the case in a court of law. The evidence is gathered by both sides, shared and then either side presents the evidence framing a narrative that supports one side or the other.

Both sides gather data (typically just called the “facts”), and then they present hypotheses to explain the data. Typically the better hypothesis is the one that best, or most plausibly accounts for all the facts. Most plausibly is just another way of saying the explanation which, if real, is most likely to have yielded the particular set of facts.

To pick an example often invoked by Christians, they claim that the hypothesis that Jesus really did rise from the dead is the best (aka most likely) explanation for the “facts” of the resurrection (that Jesus was crucified, died from it, was later buried and his tomb discovered to be empty, and that multiple putatively independent witnesses saw him alive days after his crucifixion and death). And that alternative hypotheses are less likely to yield that same set of facts, because then auxiliary hypotheses have to be invoked to explain the same set of facts (why was the Tomb empty if he died? Why did people later claim to have seen him alive? And so on). Therefore, the argument goes, (supposing these really are facts), Jesus crucifixion, death, burial, the empty tomb, and later post-mortem appearances to his followers really are evidence for his resurrection.

Do you (@Rumraket) see evidence in the way that you presented it?

Yes I do. It is essentially the Bayesian definition of evidence, and used everywhere both in science and in law. It is not proof of course, since anything but facts that would be impossible on a given hypothesis always has some non-zero probability of obtaining.

Either way, though, this is an instance, I believe, where understanding how the terms are being used and agreeing upon them up front is helpful.

We certainly agree on that.


That’s not what that word means.

1 Like

Neither is that.

1 Like

Wrong again. Thru most of human history, the movement of the sun across the sky every day could not be “accounted for by known natural causes.” Supernatural causes were suggested, such as gods pulling the sun w/ a chariot. We have relatively recently learned there is a natural explanation, which now even young children know.

1 Like

Yeah that really doesn’t make sense. After all the universe itself and it’s laws could extend beyond the horizon of visibility (if space expands faster than light), but that would imply there is some distance away from us where space and atoms “on the other side” becomes supernatural?

I don’t think that’s what people who argue for the supernatural mean. They usually speak about things like Gods, ghosts, souls, divine intentions, and afterlives. What do these things have in common that make them “supernatural”?

1 Like

I think this whole discussion is misguided. Why bother trying to demonstrate the existence of “the supernatural” as a whole?

Just demonstrate the existence of one or more things that people often consider to be “supernatural”: Ghosts, angels, gods, souls, whatever. Then we can get into an ontological discussion of whether these things are “supernatural.” Otherwise, really, what is the point?


OK. So that is a helpful distinction which I wasn’t clear on. So data, or facts, become evidence when they support or undermine a claim. Am I understanding that correctly?

I beg to differ. The example you’re using is what I understand as a typical god of the gaps argument, we don’t know, therefor god. What I’m arguing from is scientific data that provides evidence that significantly undermines the possibility in a natural cause for a specific event or entity. So I’m arguing from what science does know as evidence to support my claim which is perfectly justifiable as far as I can tell.

Which you’ve completely failed to provide while still dodging this critical question:

How/why do you infer a supernatural cause from your indirect evidence as opposed to just a natural but as yet unknown explanation?

Your dodge count is up to 5 now .


So I think maybe where the conversation breaks down here is whether or not ideas are primary or secondary in determining reality. If we go by present empirical data there is a beginning to all space, time, matter, and energy.

That would logically lead to a cause of those aspects of reality in their realm of existence that would transcend that realm. Call it what you like, but I would argue that that transcendent realm of existence is what is meant by the supernatural.

Now if you want to say that ideas are primary then you can pretty much imagine any scenario you want to for explaining data that conflicts with your views regardless of whether it makes sense of the reality we experience or not. So where does that leave us? Seems to me unless one of us shifts our position on whether or not ideas are primary, we’ll just have to agree to disagree. :slight_smile:

Where conversation really breaks down is when you duck those tough questions about your claims you can’t answer.


Does it? Scientists have explored a number of potential pathways of an event that occurred billions of years ago. We know that amino acids and other molecules are building blocks of life. Has the work scientists have done to this point, looking through a lens back billions of years, “seriously undermined” all possibilities of their natural synthesis and replication?

At a minimum, if it’s valid to make an abductive argument that various unexplained phenomena lead to the possibility of a conclusion of supernatural causes, how is it any less valid for scientists to make an abductive argument that the presence of simple chemicals as a building block of life lead to the possibility of a conclusion of synthesis in nature? I don’t see why the metaphysically minded can’t explore their own possibilities, while scientists explore theirs.

By the way, came across this article. Pretty excellent for a newb like me