There are lot of falsifiable statements over the internet, an it is easy to make up one falsifiable statement.

Such as:

  1. Earh is flat ... Falsifiable through satellite image.

  2. Earth is round ... Falsifiable through satellite image.

  3. Water is made up of atoms ... Falsifiable through chemical experiments.

There are claims which are currently not falsifiable, but maybe possible to falsify with improvement of technology. Such as

  1. Aliens stay on a distant exoplanet 1000 light years apart (falsifiable if teleportation or travel at a speed faster than light)

  2. Elephants always imagine of chocolates (falsifiable if mind reading techniques are possible).

But I want to know, what are some claims, which are truely NOT falsifiable?

  • 2
    My candidate for non-falsifiable statements are the statements of a solipsist or a sceptic.
    – Jo Wehler
    Mar 5 at 19:56
  • 2
    You asked for examples. Solipsist: There is no physical world, all phenomena are spontaneously created by my mind. Scepticist: We cannot obtain knowledge. See also en.wikipedia.org/wiki/Philosophical_skepticism
    – Jo Wehler
    Mar 5 at 20:19
  • 1
    “God does exist” and “god does not exist” - one is falsifiable, the other is not.
    – gnasher729
    Mar 5 at 23:54
  • 1
    Good question, good start! Points for that. Looks as though you can answer your own question. So far as I'm concerned, the realm of pseudoscience is jam-packed with particular instances of unfalsifiable claims. Google/youtube well-known so-called skeptics (alive & deceased). Bonam fortunam homo viator.
    – Hudjefa
    Mar 6 at 2:12
  • 1
    Santa claus is real
    – quant
    Mar 7 at 12:05

6 Answers 6


Drawing inspiration from Descarte's evil demon, the statement "Everything I experience is a hallucination, and I am actually a brain floating through space" cannot be falsified, since any empirical test could be a hallucination.

Another example, similar to yours about aliens, is "There is a completely undetectable substance on this table". One could reasonably argue that a "substance" which is undetectable is no substance at all, but the claim certainly can't be disproved by any measurement or experiment.

  • That are very interesting examples
    – user72899
    Mar 5 at 20:23
  • 1
    I was going to give Descartes' "Cogito, ergo sum" but you beat me to it.
    – How why e
    Mar 5 at 22:35

If I make a claim about the existence of an entity which I define as being beyond the scope of any possible test, then by definition you cannot falsify the claim. That's why you can't disprove the existence of certain gods.

  • 2
    “There are many invisible things, but we don’t know how many, because we cannot see them”.
    – gnasher729
    Mar 5 at 23:55
  • 1
    i.e. Russell's teapot. Mar 6 at 15:57

Falsifiability is not an either/or proposition, but a matter of degree, and even the degree is subject so dispute. It has probably never been the case in science that a useful, accepted theory was abandoned just because some single observation falsified it. There are always hypotheses to save the theory. For example, when it was discovered that Mercury didn't follow Newton's laws of gravitation, astronomers didn't decide that Newton's theory was wrong; instead, they speculated that there was a mysterious tenth planet called Vulcan that would explain the orbit. When Mendelian genetics was discovered by the larger biology community, it contradicted Darwin's theory of descent with modification, which was based on the assumption that variation was random and continuous, but biologists did not abandon Darwin's entire theory; instead, they postulated a new mechanism based on genetic mutations--not natural variation, but failure of the genes to reproduce properly. This new mechanism allowed them to keep the general theory of descent with modification just like Vulcan let astronomers keep Newton's laws.

The history of science is absolutely chock full of such instances, most of which were too obscure to have made it into the history books. Here is the problem with Popper's theory of falsifiability: when some observation O contradicts some theory T, then there are several possibilities for dealing with the problem:

  • Decide that T is incorrect and needs to be thrown out.
  • Decide that O was made in error.
  • Decide that there is some assumption surrounding T that needs to be modified.
  • Ignore it as probably not relevant.

In general if you have a theory that is useful over a broad range of applications, what is so rational about throwing it out over a tiny number of exceptions? So whether a theory is falsifiable or not depends on how valuable it is, how well-established it is, and how important the apparent exception is, among other things.

  • Yes, ideally we would take a Bayesian perspective where evidence simply increases or decreases the probability of a hypothesis, never 100% ruling it out nor 100% confirming it.
    – causative
    Mar 5 at 21:28
  • I think this is answer is technically wrong in its first sentence - there is an either/or and a difficult/easy. But I also think the answer is useful and everything after the first sentence is correct, so +1 from me.
    – g s
    Mar 5 at 22:29
  • @gs, I would address your complaint but I'm not sure what you mean. Are you talking about the fact that every proposition that has empirical implications is falsifiable in the sense that some empirical observation may contradict the implications? Mar 5 at 23:57
  • 1
    @JimmyJames: See also miasma theory, the luminiferous aether, and a good three or four different models of the atom.
    – Kevin
    Mar 6 at 18:25
  • 1
    While this is very interesting, it does not seem to be related to the question?
    – AnoE
    Mar 7 at 15:18

Basically any statement that starts with one day there will be is unfalsifiable unless it involves a contradiction. You cannot check all the one days to find out.

For example:

"One day there will be a person who can breathe hydrogen gas," is unfalsifiable.

"There is now no person who can breathe hydrogen gas," is in principle falsifiable. (We could, in principle, check all the people and see.) Likewise "There is a person who can breathe hydrogen gas."

"One day there will be a person who can breathe lamp posts," is neither falsifiable nor unfalsifiable because it is logically false: breathing isn't a thing that can be done to lamp posts. Likewise, "There never will be a person who can breathe lamp posts," is not falsifiable or unfalsifiable, it's logically true.

Similarly, any statement that starts with somewhere there is is unfalsifiable unless it involves a contradiction. You cannot check all the somewheres to find out.

  • 1
    "There will never be" is falsifiable though, simply by providing the counterexample. "One day there will be" is debatable, it could be falsifiable if it can in principle be ensured there can be no more candidates in the future at some point, due to extinction.
    – kutschkem
    Mar 6 at 9:00
  • Similarly, nowhere there is is falsifiable by providing the counterexample.
    – kutschkem
    Mar 6 at 9:00
  • @kutschkem You're right. I will edit.
    – g s
    Mar 6 at 9:33
  • It's absolutely possible to breathe hydrogen and survive. I think you mean something more like 'can a person breathe hydrogen in lieu of oxygen and survive' which I think might be a little unwieldy.
    – JimmyJames
    Mar 6 at 22:54
  • is in principle falsifiable the "in principle" is superfluous. It does not matter how hard it is to do the falsification; the only thing required is that there exists a recipe on how to do it, even if it is very hard or even too hard for our capabilities. You can with confidence simply write it s "is falsifiable". The difference to your "one day" example is that the "one day" hides an infinite amount of days, and checking an infinite amount of data is, indeed, not generally possible, no matter how much capacity the one doing the checking has available.
    – AnoE
    Mar 7 at 15:22

I am going to go a bit further than Marco Ocram's answer about gods. Any question about the existence of something is not falsifiable unless the "something" is well defined, and the term "existence" is also given a clear definition.

It may seem that "existence" requires no definition for logical discourse, because "we all know what it means". However, I challenge that notion. To give a simple example, one might ask whether the species homo sapiens exists. One person might use the term "exists" in such a way that individuals of the species homo sapiens exist, but that a species, per se, does not. To such a person, a species might be a "set of organisms (possibly empty if the species is extinct) with a certain genetic commonality", and to that person, sets are simply constructs of the mind, and do not "exist". But to someone else, the species homo sapiens does exist. Another example might be socially constructed "things" such as money, or borders or laws. It might be argued by someone that the boundary between Brooklyn and Queens does not "exist", that it is an imaginary line, and consequently does not exist. A similar argument might be made that neither Brooklyn, nor Queens, exists, that they are purely constructs of the mind.

And so it is with statements such as "God, defined as X, does (or does not) exist". Such statements are in principle not falsifiable unless the ambiguities of the term "exists" have been ironed out.

  • 1
    However, I challenge that notion. Not only you. plato.stanford.edu/entries/existence is one of the longer pages on SEP, with dozens of references. It is not an easy topic at all.
    – AnoE
    Mar 7 at 15:24

Consider the Silurian Hypothesis:

The Silurian hypothesis is a thought experiment[1] which assesses modern science's ability to detect evidence of a prior advanced civilization, perhaps several million years ago.

While we might be able to detect some evidence of such a civilization, there's really no way to prove that one never existed in the history of earth. The evidence for it would most likely have been destroyed by geological processes.

You can find a lot of examples of unfalsifiable statements in the extensive history of witch trials e.g. the Trier witch trials.

You must log in to answer this question.