Burden of Proof

Examples

A housemate makes an unusual claim about the apartment.

Leo: “There’s an invisible dragon living in the garage.” Mia: “That seems unlikely. Any evidence?” Leo: “Can you prove it isn’t there? Until you do, I’m right.”

Leo has it backwards. He made the claim, so demonstrating it is his job — Mia isn’t obligated to disprove something nobody has shown any reason to believe.

The same pattern appears constantly online:

Post: “This supplement cures insomnia. Prove me wrong.” Reply: “Do you have any studies showing it works?” Post: “You can’t prove it doesn’t. So it does.”

Nobody can prove a negative about every possible untested claim, and that’s exactly why the burden sits with whoever is asserting something, not with whoever is skeptical of it.

Why the reasoning fails

This is sometimes called Russell’s teapot, after the philosopher who imagined claiming a china teapot orbits the sun between Earth and Mars, too small for telescopes to see — and then insisting no one could disprove it. The point is that unfalsifiable or unsupported claims aren’t rendered credible just because they resist disproof. The person introducing a claim is the one who owes the evidence; shifting that job onto the listener inverts how reasoning is supposed to work. If simply failing to disprove something made it true, every unfounded claim anyone ever invented would be equally valid, which makes the standard useless.

How to respond

• Redirect the burden plainly: “You’re the one making the claim, so the evidence needs to come from you.”
• Name the shift when it happens: “I’m not obligated to disprove this — what’s the actual support for it?”
• Distinguish extraordinary claims from ordinary ones. Everyday claims (“I had coffee this morning”) need little evidence; unusual ones need more, and that’s not unfair, it’s proportionate.
• Don’t confuse this with genuine open questions — if a claim is plausible and simply untested, “let’s find out” is a reasonable response, not burden-shifting.