GOW wrote:I still don't get it. If they saw ghost/faeries/elemental that is observable evidence
No, that’s not observable evidence. That’s an observation. There’s a big difference. Evidence implies a logical connection. Merely observing something does not ensure a logical connection between the observation and your interpretation/explanation.
For instance. I observe a ball rolling down a hill. I can make any number of interpretations/explanations of that observation. The empirical, logical interpretation is that the force of gravity is causing the ball to roll from a point of high potential energy to a point of low potential energy. We can make this logical interpretation because of a scientific theory, based on many different designed empirical experiments, which allows us to make predictions. We can test our interpretation (a hypothesis) by doing scientific experiments such as putting the ball at the bottom of the hill, and observing the results. When we make a good hypothesis and design good experiments, we know a priori what we will observe if the hypothesis is correct. These observations are then “evidence” to support our original interpretation of gravity as the stimulus for the ball rolling down the hill. After these observations are made, we can work at understanding HOW gravity causes balls to roll down hills, and increase our knowledge. Note that evidence really doesn’t prove anything – evidence is something that helps us eliminate other possible interpretations of an observation, and narrow down the possible interpretations that might be right.
That is not the only interpretation/explanation of the ball rolling down the hill, though. Someone else may just as easily say that the ball rolled down the hill (an observation) because a spirit pushed the ball down the hill. However, it is inappropriate to then say that the ball rolling down the hill is evidence for the existence of a spirit standing at that point of space and time. This is because there is no logical reason to suppose that there is a spirit standing on the hill, and that that spirit is the cause of the observation. You cannot design experiments to test for the existence of spirits – in the affirmative or in the negative - and so you cannot gather evidence to support this interpretation of the observation. Now, you can make many observations of balls rolling down hills, and you can interpret each one of these as due to spirits pushing balls, but none of them are evidence, because you’ve designed no experiments to make any logical deductions about your interpretation. You can not narrow down your possible interpretations. All you have are observations, and observations in themselves do not constitute evidence.
For the same reason, if you enter a room and observe a dead body on the floor, that observation is not evidence to support the theory that the person was murdered. There are many possible interpretations of a dead body in a room, of which one is that the person is murdered. Others include: the person fell, the person had a heart attack, the person committed suicide, the person was struck dead by a demon. With only this observation, it is impossible to ascertain which interpretation is correct. In other words, the observation itself is not evidence to support ANY interpretation, be it ghosts, demons, murder or the simple bad luck of an anvil falling on his head. Certainly you will agree that if someone ELSE tells you that they saw a dead guy in a room, your ability to rely on their interpretation becomes even LESS. Now, if you enter the room and observe a bunch of other things, such as a bloody knife, blood all over the walls, a broken door that looks to have been kicked down, you can begin to frame a better interpretation because now you have other observations that allow you to make a logical connection between the main observation (dead guy) and the interpretation (murder). We may even start to call these observations “evidence” because we can start to make logical deductions based on several independent observations. But perhaps a demon is still a possible interpretation. Another possible interpretation is that the guy arranged the scene to look like a murder even though he killed himself, possibly for some insurance benefit for his family. So, while a string of independent observations allow you to narrow down your possible interpretations using logic, they certainly are not proof that the guy was murdered, although they may be convincing enough to allow you to deduce with relative confidence (high probability) that the person was, in fact, murdered. But when someone points out the other possible (though low probability) interpretations, how do you therefore become confident in your interpretation? You make a hypothesis (the guy was murdered) and then design experiments. Such as: if the guy was murdered, there would be fingerprints from someone else. So you look around, and indeed find some fingerprints, and this is more evidence that the guy was murdered because it eliminates other possible interpretations. By collecting evidence, you eliminate other possible interpretations until only one remains. In principle, of course, there may be others you didn’t think of, and that’s always a risk in the judicial system, AND in science. Which is why scientists keep testing their theories even after there is lots of evidence to support them. But going back to the murder – if you interpretation is “violent demon killed the guy”, the dead guy is not evidence, certainly, that a violent demon did the deed. Nor is a bloody knife, nor is blood splattered on the wall, nor in fact is “A demon was here” written on a piece of paper. Why? Because no logical deduction can be made connecting your observations to a demon. That does not mean that the demon-idea is WRONG. Maybe a demon DID kill the man. But there’s no experiment you can design that will test for demons, and so there’s no way to eliminate “demons” as a possible interpretation. You could always say, “BUT IT COULD HAVE BEEN A DEMON!” no matter what your observation was. See an anvil on the floor and the guy’s skull was crushed? “BUT IT COULD HAVE BEEN A DEMON”. None of these can be evidence for demonhood because there’s no way to show whether that interpretation could be wrong or right. It’s an interpretation NOT based on logic, and evidence does not exist for such interpretations – not until you can connect the observation to the interpretation logically.
The point to all this being – many people want to point to “evidence” to support fantastical explanations for empirical observations, but it’s a paradox. If it’s a fantastical explanation, you cannot have evidence to support it, because of the very definition of evidence. The mere observation of something is not evidence for anything, except that you observed it in the first place. You need to be able to say, “If what I observed is due to X, then I should also observe A, B, and C if I do D, E, and F, respectively.” You can’t do with a ghost.
Caradoc wrote:Agreed. But that is not the point I am making. I am saying that the volume of reports is itself a phenomenon worth consideration. Mass Delusion and Rampant Hoaxing are two possible (but statistically unlikely) explanations, and these could be validated. But until they are, the Fairies Exist theory cannot be excluded.
You cannot compare the statistics of two things happening if you cannot even calculate them. It’s another logical fallacy to say that, since A is very improbable, B must be correct – nay, COULD BE – correct. The extreme improbability of one event does not logically lead to a conclusion that another event is possible, let alone probable.
Putting that logical fallacy aside, you cannot even calculate any odds for fairies exist, so how are you supposed to evaluate the probability that it is possible, compared to the “low” (which even there I’m not sure how you think you can assess) probability that everyone is delusional. Even if I agree that the odds are low that so many people are misinterpreting what they observe, how do you know that the odds of even ONE person NOT misinterpreting the event aren’t orders of magnitude lower? Is it even something for which odds can be calculated?
Besides which, it’s not really even something that has statistical meaning. There’s not a “chance” that people are lying. They either are or they aren’t. There’s not a “chance” that the theory of gravity is correct. It either is or it isn’t.
"What men are poets who can speak of Jupiter if he were like a man, but if he is an immense spinning sphere of methane and ammonia must be silent?" - Richard P. Feynman