For certain kinds of claims, it seems obvious in a way to not believe in them due to their lack of evidence or unfalsifiability. For example, one may not believe in undetectable goblins or fairies.

But for other kinds of claims, I also fail to see “no evidence” and yet many would still believe them.

For example, how reasonable is it to believe that the earth is a sphere? Is it reasonable to believe that the earth is a sphere? What is the degree of belief you should have that the earth is a sphere?

For all three of these claims, how would you test that you’re wrong? An ardent skeptic can look at all the evidence, state “well, I could be being tricked” and claim that it’s not reasonable to believe in a spherical earth. Why exactly would he be wrong? You can’t even show that he’s wrong to an nth degree! Even intuitive considerations of the form “well I can’t put an exact probability to it, but surely it’s above 80% likely for the earth to be a sphere” can’t be falsified.

Why is “there are invisible ghosts” any less falsifiable than “it is reasonable to believe the earth is a sphere with X% or X-Y% probability”? If not, why believe in either of them?

Of course, for pragmatic reasons, one would ultimately always have to make a choice and make decisions. No one is arguing against this. But that arguably says more about our actions and what we would do as a matter of fact than belief (unless one just defines belief to just be how we would act in given scenarios).

  • How does the dogmatic skeptic know that it's possible to be wrong in the first place? What justifies believing in the possibility of being tricked? (This is not a response to Pyrrhonian skepticism, incidentally.) Quine did contend that the web of our beliefs is too vast and intricate for naive falsfiability to be a stable criterion for adding to or taking away (at least almost) any/all of the strands in the web. Commented Aug 29, 2023 at 18:07
  • He doesn’t know that it’s possible to be wrong. But he doesn’t know that it’s impossible to be wrong either. Hence, he doesn’t have to believe, although pragmatically, he can simply not consider that hypothesis. Arguably, he has no choice to
    – user62907
    Commented Aug 29, 2023 at 18:24
  • How does he know that he doesn't know that it's impossible, though? If he doesn't know what knowledge per se is, might he not know things without knowing that he knows them? Commented Aug 29, 2023 at 18:42
  • It depends on how you define “know”. I suppose some would say you don’t know something if you can doubt it.
    – user62907
    Commented Aug 29, 2023 at 19:03
  • 1
    Yeah I concur. I have a problem with the notion of possibility in the first place. There’s no such thing. I believe in one mind independent reality where things happen or have happened or they don’t.
    – user62907
    Commented Aug 30, 2023 at 0:27

2 Answers 2


The two claims are different in an important way. The existence of invisible fairies (I'll assume they are also unhearable, untouchable, etc., to the point they have no detectable impact on the world) is inherently unfalsifiable. There is no possible difference it would make in the world. The proposition that the earth is a sphere--or flat, or a cube--is in principle falsifiable. The truth of the matter makes a detectable difference in the world.

However, you are apparently coming from a standpoint of personal falsifiability. You cannot personally tell the difference between a flat earth and a round one--it's too big for you to have direct experience of it.

For most people, the answer would be that the round earth is attested to by a variety of things external , but that I can see--photos, historical records, scientific experiments. All those things could be faked, but it would take a concerted conspiracy. Therefore, I am willing to extend personal credulity to the round earth, but not the invisible fairies.

You are free to be radically skeptical about anything you don't have direct personal experience of, but it would make for a very limited set of things you could be said to know.


There are plenty of propositions that are both falsifiable and rational to believe. This is because falsifiability concerns evidence, but there are other forms of epistemic justification other than evidentiary ones. So there are beliefs that can be both justified without being empirically justified; and there can be beliefs that are both justified and cannot in principle be empirically justified.

For example, consider a priori truths (truths we can justify through reason alone), such as the claim that '2 + 2 = 4'. This claim is unfalsifiable, because there is no possible state of affairs that could prove it false (consider: if we found that two objects and two objects, when bundled together, made five objects, we would not think that 2 + 2 = 5; we would think that an extra object had been created). Yet no one would seriously suggest that it is irrational to believe that 2 + 2 = 4. This general pattern applies to many if not all a priori propositions, such as mathematical or logical rules of inference or the truths that follow from them.

You must log in to answer this question.