Agnostic theists and atheists believe themselves to be hopelessly ignorant concerning the existence of the supernatural or paranormal. To them, gnostic atheism isn't adequately supported by evidence, and because "absence of evidence isn't evidence of absence," and "all you know is that you know nothing," they suspend judgment on whether or not there is a god, soul, ghost, demon, or afterlife

"There might be a god that doesn't interfere with the world it created, you could never know," they say

But can we know? Can philosophy prove that there is no such thing as supernatural or paranormal? How can we argue against every interpretation of god, soul, ghost, demon, or afterlife, if their existence can't be proved or disproved?

  • 7
    No, we can't prove the non-existence of anything.
    – Frank
    Commented Mar 2, 2023 at 18:55
  • 5
    Start by proving something banal does not exist. For example, count your co-workers, then prove you don't have one more. stargate.fandom.com/wiki/The_Fifth_Man
    – Boba Fit
    Commented Mar 2, 2023 at 19:01
  • 2
    You will find Zombies interesting. How can you prove you're not surrounded by them?
    – user64708
    Commented Mar 2, 2023 at 19:13
  • 3
    Does this answer your question? Asserting that Atheism and Theism are both faith based positions
    – user64708
    Commented Mar 2, 2023 at 19:16
  • 2
    @eirene Now that I think about it, and re-read the comments, they aren't that different. Thank you for the link
    – ActualCry
    Commented Mar 2, 2023 at 19:31

10 Answers 10


You cannot prove it, and that is why the debate is endless.

What you can prove is that it is impossible for all religions to be true, since they make mutually exclusive claims, and you can point out that religions are a form of indoctrination, but those arguments have been made by Dawkins and other for years with little measurable impact.

Of course, a handicap one faces in trying to muster scientific arguments to challenge ideas about the supernatural etc is that science can't yet explain the thing we believe most- consciousness. So if science can't encompass consciousness properly, what else is it missing?

  • 7
    The scientific question behind consciousness is not whether it exists, but how it exists, how it came to be, how it works, and so forth. Just like we didn't need the theory of relativity to know that objects (that are denser than air) fall when dropped (i.e. gravity exists). There is enough clear evidence to support that conclusion above any other. For supernatural questions, on the contrary, the question is indeed whether they exist, and not how they exist.
    – NotThatGuy
    Commented Mar 3, 2023 at 9:43
  • 1
    @NotThatGuy but that is precisely my point. We know that consciousness exists through our experience of it. We admit that there must be some unknown mechanism that causes it. Physics does not predict the existence of consciousness. So there is at least one phenomenon that exists for which we have no physical explanation. How then, can we be sure there are not others? I should say that I am playing devil's advocate here- I am not a proponent of the supernatural. Commented Mar 3, 2023 at 10:01
  • 4
    But the existence of consciousness has met its burden of proof, regardless of whether we understand the underlying mechanism.
    – NotThatGuy
    Commented Mar 3, 2023 at 10:12
  • 2
    @NotThatGuy why is that relevant to my point? I am not doubting the existence of consciousness. I am simply saying that you can't appeal to physics to rule-out the supernatural. Commented Mar 3, 2023 at 10:32
  • 1
    @raygrant It is impossible to disprove a 12 year old's imaginary friend because he's sitting right there eating breakfast. Everyone else knows it. You're just not looking right. Commented Mar 4, 2023 at 15:18

There is a lot of poor thinking one find in theism/atheism discussions, and you have a few as assumptions in this question.

Many falsely think one cannot “show a negative”. But this is how most hypothesis testing is done. Come up with a hypothesis, such as that your neighbor makes their money dealing drugs, and one can make predictions based on this. If the hypothesis is true, the neighbor would have to have a stock of drugs, spend a lot of their time in one on one meetings, in which money and objects change hands, and would need a supply themselves of the drugs. So, check these out. If you search the neighbor’s dwelling and there are no drugs, they don’t spend time in one on one’s with furtive exchanges, etc, then one has shown the negative — your neighbor isn’t actually a drug dealer. This is how all of science operates — “proving” negatives over and over. But as noted above, all of these demonstrations fall short of “logical proof” because EVERYTHING empirical can only be provisionally shown to be true.

A major misunderstanding is the possibility of, and necessity for “proofs”. Whether a particular god exists is an empirical question about our world and logic proofs are not applicable to supporting evidence for or against an empirical question.

This leads into the impossible standard fallacy and false dichotomy of gnostic vs agnostic atheism. Since god questions are empirical questions, and nobody can ever be certain on an empirical question, there isn’t anybody who ever meets the certainty criteria for “gnostic” atheist or theist.

Another is the widely made claim that theism is a faith based view. If one spends much time actually talking with theists about why they believe what they do, and they will cite things like personal prayer experience (first person empiricism), experienced and historical miracles, and the reliability of historical testimony. This is an evidenced, empirical justification.

Atheists likewise, when asked why they think atheism is true, will cite things like the repeated refutations of scripture claims, the LACK of miracles and direct experiences of god in the world, the incoherence or immorality of religious doctrine, and the relative utility of a secular worldview. Actual theists and atheists are not “agnostic” either. Nobody actually falls into any of the four boxes of that silly graphic.

Back to another part of your question: how to refute the “supernatural”. This is a poor term, as one common meaning is “that which is beyond evidence and reasoning”, and if you use that definition then you CAN’T provide either evidence or reasoning against the supernatural, by definition. A far better term would be spiritual.

For an atheist to provide high confidence of the falsity of all spiritual beliefs, the best method to do so would be to show that we can have very high confidence in the truth of materialism. IF materialism is reasonably taken as true, then there isn’t any spirit, and can’t be any Gods.

  • 5
    In your example of the drug-dealing neighbour you can prove he doesn’t deal drugs if you can prove the various factors listed but really it moves the problem of proving a negative down the line, since you would need to prove he does not have a stockpile, does not have one-on-one meetings, etc
    – 11684
    Commented Mar 3, 2023 at 8:21
  • 1
    "If one spends much time actually talking with theists about why they believe what they do, and they will cite things like..." - and if one digs deeper and points out the epistemological problems with those, they will very often resort to some sort of "I just have faith" defence. Well-known apologist, William Lane Craig, infamously presents arguments for God's existence despite saying he will continue to believe "even if all the arguments for [God's] existence were refuted".
    – NotThatGuy
    Commented Mar 3, 2023 at 9:31
  • 6
    Irrespective of the rest of the answer (which I have some sympathy for), your description of hypothesis testing is incorrect. Hypothesis tests never prove anything, they can only disprove things (proof is the wrong word of course, but thats by-the-by) - but the very way a hypothesis is construct means that even if you disprove the positive you are not proving the negative. That is in a hypothesis test rejecting the null does not mean accepting the alternative, and not rejecting the null does not mean accepting the null. Commented Mar 3, 2023 at 9:52
  • 1
    "For an atheist to provide high confidence of the falsity of all spiritual beliefs, the best method to do so would be to show that we can have very high confidence in the truth of materialism" - materialism is the non-existence of the supernatural, so you seem to be saying the best method to show that all spiritual beliefs are false, is to show that all spiritual beliefs are false. Consider "For non-unicorn-believers to provide high confidence of the non-existence of unicorns, the best method to do so would be to show that we can have very high confidence in the truth of unicorn-non-existence"
    – NotThatGuy
    Commented Mar 3, 2023 at 9:53
  • 3
    What you described with the drug dealer example looks a lot like what we call in Math "Demonstration by asurd" (direct translation from french, you may know it under a different name). What most people forget about this type of proof is that one must start by showing there is only a limited number of options. If you don't prove this first, showing that alternatives lead to the absurd won't prove anything. In your example the reasoning is not very rigorous
    – Kaddath
    Commented Mar 3, 2023 at 10:50

No phenomenon in reality is supernatural or paranormal by definition.

Only explanations can be supernatural or paranormal, i.e. outside of known physics.

Gods, ghosts and souls are not existing things, they are just attempted supernatural explanations for natural phenomena. They are not testable scientific theories, they are pure fiction.

  • Probably we would just expand our physics to cover anything new we found.
    – Scott Rowe
    Commented Mar 3, 2023 at 11:32
  • 3
    Physics has always been about things we have found. We cannot study things we have not found. Commented Mar 3, 2023 at 11:46
  • 1
    Souls aren't even properly speaking supernatural entities, they are metaphysical constructs.
    – jaredad7
    Commented Mar 3, 2023 at 17:19

Gnosticism is probably best defined not as complete and total certainty, but rather as being as sure as we can be about (religious) claims.

Pretty much no-one would claim to be uncertain about the existence of their own family members, even though their existence are empirical questions and we can't (justifiably) have complete certainty on that.

Pretty much no-one would claim to be uncertain about the non-existence of an invisible unicorn in their back yard, even though we can't have complete certainty on that.

To put religious gnosticism on a pedestal of complete certainty with irrefutable proof, that no existence or non-existence claim can meet, renders the idea of gnosticism entirely pointless.

So to call oneself a gnostic atheist is roughly to say "I'm as sure that a god doesn't exist as I am that there isn't an invisible unicorn in my back yard".

Although atheism is typically more a question of whether to reject existing/known god claims (not any god claim).

Some atheists may concede the possibility of the existence of some unknown god, but this is not a useful question. Given the very fact that they're unknown, we know nothing about them, we don't know what they want or whether there's any consequence to not doing what they want, and we don't know whether they even know that we exist. We would need to wait for their existence to become known before it should reasonably influence our lives in any way.

So that leaves existing god claims, and for that it would be up to theists to provide evidence before you'd become convinced that their claim is true (much like if someone wants to claim there's a unicorn in their back yard). And if you've evaluated their evidence and found it to be lacking, you may be convinced that their claim is false.

  • 2
    So, it is a lot like a court of law: someone makes a claim / charge, then there is evidence etc and then a decision. Not so hard to understand. God is inexistent until proven otherwise, beyond a reasonable doubt. We just have to pick our Solomon whom both sides would accept an answer from. That's the actual problem: what people are we going to trust?
    – Scott Rowe
    Commented Mar 3, 2023 at 11:26
  • @ScottRowe The further point is that we only need a Solomon if the invisible-unicorn-owners expect the rest of society to change to support their unicorn. Until then, do what you like. But from that point on, the burden is on the theists to provide evidence. That's the interesting thing about atheism - most societies have theist laws grandfathered-in which never had this test applied, otherwise there'd be no reason to give tax breaks to a church and not to, say, a fishing club.
    – Graham
    Commented Mar 3, 2023 at 11:59
  • @Graham the Solomon is the person that the unicorn-believers would accept the answer: "No. Doesn't exist" from. That's the sticky point: who, as Mansur Al-Hallaj said, would you trust to kill you (or at least your unicorn)?
    – Scott Rowe
    Commented Mar 3, 2023 at 17:34
  • @ScottRowe I wouldn't intend to say whether the unicorn does or doesn't exist. Rather, I'd say whether care and feeding of the unicorn is something which should put a burden on the rest of society.
    – Graham
    Commented Mar 4, 2023 at 18:39
  • @Graham I had a psychology professor who said that a person's values are shown by: 1. how they spend their time, 2. how they spend their money. So, yes, money is important. Society can't pay for all that it arguably should already, so paying for things that don't exist is simply not acceptable. It interests me, because I recently saw the Rouen cathedral and many other religious sites. Upkeep is expensive, the sites bring tourists who bring money to the cities... But the main 'users' are fewer in number, too few to justify a cathedral. The buildings exist, surely. Do we pay for those?
    – Scott Rowe
    Commented Mar 4, 2023 at 20:10

You can't prove that anything doesn't exist as long as that thing is coherent. However, you can't prove an infinite number of things. You can argue against it in the sense that you can say you have no reason to believe in them. And ultimately, that is all that matters.

  • How many gods can dance on the headland of a planet? Hindus would say: an infinite number. Ding! Ding! Ding! Correct!
    – Scott Rowe
    Commented Mar 3, 2023 at 11:45

Statistical: For example, because we all have smartphones now, people are 10x more likely to have a camera with them at any given time, so there should be approximately 10x more serious claims of ghost and UFO photographs - but there aren't anything like that number. That's a "good enough" proof that photos of UFOs and ghosts aren't real.

  • 4
    Maybe digital cameras are unable to capture images of ghosts and UFOs the way film can? Then we would have a sharp decrease in the amount of such photos!
    – Scott Rowe
    Commented Mar 3, 2023 at 11:31

To claim the nonexistence of a metaphysical entity, you must show that it would be logically impossible for it to exist. You cannot use observation because your senses are part of the physical, so the metaphysical entity would not be subject to them even if it did exist.

Successful examples of this include mathematical and geometrical proofs, such as the non-existence of the square root of 2 or a squared circle.

For gods or demons, it is unlikely you will succeed, because these beings are generally speaking not all that illogical. I am not aware of any logical argument for the non-existence of God. There are, however, many arguments that God can exist and even must exist, such as those from Medieval philosophers.

Generally the existence of God is not a matter of logical proof but faith. You must have faith that God exists and loves His creation, that if you accept His authority and entrust your life to Him, He will provide that your life will take the appropriate course. This divine guidance would likely not happen in some overt and easily observable way, because by its nature it is a metaphysical phenomenon, and it is ultimately your faith that is being tested and not your senses or your reason. Therefore, believers in God are rarely interested in logical or empirical evidence of His existence which would either fail to promote or possibly even diminish their faith.

The paranormal is an entirely separate thing. It means outside of normal experience and usually refers to apparitions, UFOs, cryptids, psychics and the like. There is nothing inherently metaphysical about these things and paranormal enthusiasts will readily produce naturalistic explanations for their favorite phenomena. To prove that UFOs or Bigfoot or telekinesis don't exist would be the exact same process as proving that Willy Wonka's chocolate factory or your morning cup of coffee doesn't exist. These methods are well documented in many places so I won't reproduce them here, I refer you to the works of Karl Popper and others.

  • You are correct in general. Note that there is a classical proof that omnipotent beings do not exist, and there is also a lemma in linear algebra which says that if quantum mechanics is accurate, then omniscient beings do not exist. So, for example, Jehovah (as described in the Torah, Bible, Quran, etc.) does not exist.
    – Corbin
    Commented Mar 5, 2023 at 18:57
  • @Jessica - BELIEVERS ARE RARELY INTERESTED IN LOGICAL OR EMPIRICAL EVIDENCE...NOT! The Medieval believing philosophers engaged in logical proofs and syllogisms. And many believing scholars have been very much concerned with empirical evidence ,especially that which was repeatedly demonstrated by the rabbi from Nazareth in front of eye-witnesses both friendly and hostile. The error often expressed today is that "believers have faith in faith." While some succumb to this out of ignorance, it misses the true definition of faith: FAITH IS BELIEF THAT IS A RESPONSE TO A REVOLUTIONARY FACT.
    – user64825
    Commented Mar 10, 2023 at 23:14
  • 1
    @raygrant What's with the shouting case? Lay off the caps lock, man.
    – Jessica
    Commented Mar 30, 2023 at 17:10

tl;dr It might not make sense to assert that external concepts do-or-don't exist. Below is a story to showcase the ambiguity.

Thought experiment: Intertwined simulated worlds.

Alice and Bob are friends.

Alice and Bob both decide to play SimWorld, which is based on an older game called "The Sims". Basically, Alice and Bob each get to create their own virtual-world filled with artificially-intelligent people called "sims".

Alice and Bob each go home and make their own worlds individually:

  1. Alice tells SimWorld to start with a model like the world she's already in. Partial to stories about magical users who're secretive about their abilities, Alice adds in a magic system that the sims can use.

    Alice chooses to focus on her in-simulation counterpart, Sim-Alice, who she grants powerful magical abilities.

  2. Bob likewise tells SimWorld to start with a model like the world he's already in. Partial to stories about eccentric inventors who're secretive about their abilities, Bob allows sims to have technical skills that allow them to build amazing devices with ease.

    Bob chooses to focus on his in-simulation counterpart, Sim-Bob, who he grants powerful mad-scientist abilities.

The next day, Alice and Bob talk to each-other about their new sim-worlds. Each is amused by the other's fantasy, deciding to have the other's in-simulation counterpart to experience delusions of such abilities in their own worlds. This is:

  1. In Alice's sim-world, Sim-Bob thinks that he has powerful mad-scientist abilities, but doesn't.

  2. In Bob's sim-world, Sim-Alice thinks that she has powerful magical abilities, but doesn't.

Alice and Bob both agree that Sim-Alice and Sim-Bob should meet up in both of their sim-worlds to go flying through the air on a flying-carpet.

  1. In Alice's sim-world:

    • the flying-carpet can fly because Sim-Alice cast a magic-spell causing it to fly;

    • Sim-Bob thinks that he caused the carpet to fly, but didn't.

  2. In Bob's sim-world:

    • the flying-carpet can fly because Sim-Bob instructed his helper-nanobots to fuse anti-gravitational particles to it;

    • Sim-Alice thinks that she caused the carpet to fly, but didn't.

Alice and Bob both submit their simulations to be run on a cloud-server. The same server receives both, and due to its advanced evaluation-engine recognizing an opportunity to fold the work into a single evaluation, runs the same evaluation to effect both Alice's sim-world and Bob's sim-world. The cloud-server reports the result back to both Alice and Bob, having satisfied the evaluation-constraints for both of their simulations.

Reviewing the simulations, both Alice and Bob realize that, in both of their sim-worlds, Sim-Alice and Sim-Bob argued about if magic was real and if their flying-carpet worked magically. Which of them was correct?


In order to prove the supernatural doesn't exist a person would have to have traversed the whole of the universe...and beyond. But to prove the supernatural (supranatural) exists, all the deity has to do is show up once!

And this is what many eye-witnesses claimed happened in the first century when Jesus showed up! (Christmas comin' down) Of course, anyone can claim to be a supranatural being, but someone who worked miracles that can only be characterized as being done by a deity, would qualify as this deity. Creative miracles (nihil sine deo) beyond the ability of Nature (physical laws), cannot be dismissed when witnessed by credible observers who could not only verify, but if necessary falsify them.

As to the credibility of the various documented witness accounts concerning Jesus, Simon Greenleaf (expert on judicial evidence from Harvard) and other experts have researched and declared them credible! The witnesses were contemporary with Jesus, and were from various walks of life: tax accountant, doctor/historian, businessman, companion for three and a half years, members of the Sanhedrin, etc.

So it is impossible to disprove the existence of the supernatural (supranatural). God has a lot to say about this!


In Philosophy, there's the concept of Russell's Teapot: an example which acts on the example of "disproving the existence of a tea kettle floating around Saturn"

It's a reference to the fact that some things that are ridiculously unlikely but are also ridiculously hard to prove.

The thing is, though, that this doesn't mean they are impossible to prove.

However, there are two ways I'm aware of to disprove Russell's Teapot, and a third way that doesn't apply to the teapot itself, but some similar claims.

The first is the long way - to disprove at each possible instance. In the case of a tea kettle, take the space of a tea kettle, and simultaneously search a grid of the entire orbit around Saturn with a grid resolution of that size, and if there are no tea kettles, then it is disproven.

The second is the induced way... to find some claim about about the subject in question that doesn't match what is possible, and it requires some impossible aspect of the claim. In the case of the tea kettle, it would be specifically designating a tea-kettle as a human-made object, and going through all human space flights to show that none have put a kettle in orbit around Saturn. Although not as absolute, it's still pretty solid proof.

The third way is if the definition of the "teapot" itself is somehow fully accessible by definition. For example, if instead of a Teapot, Russell's proof said, "a magical lamp that instantly creates a copy of itself in the hands of anyone who dons cream colored pants and a hat." At which point you could disprove the existence of this particular lamp that floats around Saturn using any person that puts on cream colored pants and a hat and seeing that the kettle doesn't suddenly appear in their hands.

All cases of a "Russell's teapot" scenario require at least one full coverage in some aspect to prove or disprove...

  1. Full coverage of Saturn in the first example (Coverage of affected)
  2. Full coverage of teapots in space second example (Coverage of concept)
  3. The Definition including a disprovable full coverage in the third (Coverage of Definition)

Each case, however, will require having a very firm definition of the subject matter. If you're trying to philosophically debate someone on the topic of the supernatural or paranormal, unless it's a purely academic philosophical debate (such as an online religion debate), they're likely going to be highly resistant to actually making a concrete definition and following it to its logical conclusions.

However, for a philosophical debate, the above proofs should be sufficient, even if 1 & 2 are generally impractical for most debates.

In non-philosophical debates, sadly, the solid philosophical proof usually doesn't "prove to people" (in no small part because many people don't fully grasp logic), and frequently people just want to be right without going through the effort to become right.

The times I've had this particular debate in regards to the paranormal with someone, the best I got was "whatever the definition is in the dictionary"... this was an "easy" philosophical solution, as the dictionary I looked it called the supernatural 'outside of what is natural', and then I looked up natural and it was 'not man-made', thereby meaning all man-made things are supernatural by definition... this did not sit well with the person I was debating with.

However, not a philosophical proof against the supernatural, but a physics one making use of full coverage (proof method 2), is CERN's experiments with particle physics. If you consider the supernatural being active non-physical things that influence the physical world, CERN has tracked every particle capable of interacting with fermions (basically, what we're made of), so we are fully aware of all physical interactions with humans - something a soul or ghost would have to do... aannnndd... none of them were potential methods for human interactions by ghosts/souls/etc.

Another argument is that, if you consider humans natural, as "in nature or deriving from nature" is another definition of natural, then anything that exists is, by definition, natural. Therefore, by definition, the supernatural simply doesn't exist, as supernatural becomes a synonym for imaginary or non-existing.

  • 1
    Russell's teapot is the more well-known version of "tea-related object in space of questionable existence".
    – NotThatGuy
    Commented Mar 6, 2023 at 10:56
  • I actually didn't know the name, @NotThatGuy, so thankyou! (I remembered it being brought up in Philosophy class, but not it's name. I'll update the post with it!)
    – lilHar
    Commented Mar 10, 2023 at 19:55
  • 1
    It makes me smile and laugh a little inside... the hypothesized gods are more disprovable than the hypothetical teapot. Commented Feb 24 at 12:23

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .