Most people are taught what to think and never how to think. That leaves their beliefs unguarded, so anyone with a confident story can walk off with one. The defence is six rules for weighing evidence: Falsifiability, Logic, Comprehensiveness, Honesty, Replicability, Sufficiency. Drop the vowels and you get FiLCHeRS. Apply all six to every claim and nobody can steal a belief from you.
A claim that can’t lose says nothing
It must be possible to imagine evidence that would prove the claim false. That sounds backwards, but every true claim passes this test. “Humans live less than 200 years” would fall the moment someone reached 200. “Water freezes at 32°F” would fall if it froze at 34. We don’t expect either to fall, but either could. A claim that survives every possible outcome isn’t true. It’s empty. It tells you how the speaker feels, not how the world is.
There are two ways to dodge the test. The first is vagueness. “Crystals restore balance to your spiritual energy.” What is unbalanced energy, how would you detect it, and what would show it hadn’t been fixed? A prediction that someone’s year will be “filled with changes” can’t miss either. It just means “whatever will be, will be.”
The second is the endless excuse. Starlight billions of years old refutes a 10,000-year-old universe, so God must have created the light already in transit. The evidence for UFOs is missing because of a cover-up. The healing failed because your faith was weak. The spoon didn’t bend because a skeptic was watching. The claim is heads I win, tails you lose.
Valid is not the same as sound
An argument is valid if the conclusion follows from the premises, and sound if it is valid and the premises are also true. You don’t need formal logic to check validity. Look for one counterexample. “All dogs have fleas; Xavier has fleas; so Xavier is a dog” fails the moment you picture a flea-ridden cat named Xavier. “All dogs have fleas; Xavier is a dog; so Xavier has fleas” is valid, but it’s unsound, because not all dogs have fleas.
Pseudoscience usually fails both tests at once. “This old map is so accurate it could only have come from satellite photos” is invalid, since plenty of other methods could produce an accurate map. Its premise is false too, because the map is riddled with errors.
All the evidence, faced honestly
You can’t count the hits and throw away the misses. Plane crashes on days when the pilots’ “biorhythms” were low prove nothing if you ignore the larger number of crashes on high days. A psychic who called an election everyone else also called is no prophet once you count the thousands of predictions that failed. Choose your evidence carefully enough and the earth is flat.
Then be honest about what the whole record shows. If it goes against the belief, give the belief up. A century of psychic research that keeps failing to replicate doesn’t prove psi is “elusive”. It shows the early results were luck. When you catch a psychic bending the spoon with his hands, the honest conclusion isn’t that he only cheats sometimes. And when the evidence is overwhelming, staying agnostic is no better than staying credulous.
Once is coincidence
A single result never settles anything. Every experiment can carry hidden bias, error, or fraud. Independent repetition corrects all three, even when nobody ever finds out what went wrong. If I call the next roll of the dice, make me do it again. A “psychic detective” who makes thousands of predictions will land a few hits by chance. Pointing to those few and ignoring the rest is not evidence.
The burden is on the claimant
The evidence has to be enough to establish the claim, and three rules follow. First, the person making the claim has to prove it. Not being disproved isn’t proof: you can’t disprove that Hitler is living in Argentina, and that doesn’t make it so. Second, extraordinary claims need extraordinary evidence. My word is enough to tell you it rained on my way to work. It isn’t enough to tell you aliens took me to the far side of the moon. Third, testimony never establishes a paranormal claim. People lie, and sincere people are still wrong. Perception is selective and memory rewrites itself, and credentials guard against neither.
Belief at a fair price
The first three rules are logical necessities. A claim has to mean something, the argument for it has to hold, and the evidence has to be complete. The last three are practical ones. People deceive themselves, deceive others, make mistakes, and misremember, so the evidence has to be faced honestly, checked by repetition, and strong enough for what it claims.
If a claim fails any one of the six, reject it. If it passes all six, you’re entitled to real confidence in it. That isn’t certainty, because tomorrow’s evidence may overturn it, and you should be ready to change your mind cheerfully when it does. That is all skepticism means: believe exactly as far as the evidence takes you, and no further.
Pay full price for your beliefs, and nobody can filch them.
