Extraordinary Claims Require Extraordinary Evidence

[robvalue]

When considering any kind of anecdote, it's not simply a matter of determining if they are telling the truth. They may be telling what they think is the truth, but that doesn't stop them being mistaken. If someone tells me they "saw a ghost", I don't think to myself that they are lying to me. I think, given no further evidence to look at, that the probability they are mistaken is virtually 100%. People are mistaken all the time, it's a very simple explanation. The chances of them actually having experienced and correctly identified some phenomenon unknown to science is miniscule by comparison.

They probably did see something. But their instant conclusion that it was a ghost is not reliable. Such a thing has never been shown to exist, and crucially, we have nothing to compare a potential ghost to besides fiction and other anecdotes for them to draw such a conclusion. Upon questioning, the response is almost always the argument from ignorance fallacy: "What else could it have been?"

Of course, you can't discount the possibility that someone really did see something that actually was "a ghost", whatever that may mean. But there's no need to discount the possibility. All that you need to do is consider the likelihoods. And since anecdotes provide no actual evidence to examine, it's entirely arbitrary to believe one over another when the claims are fantastical like this.