It might be the case that everything you think you’re seeing is an illusion, a trick being played by some evil demon who has both malicious intent and perfect control over your experience. You could try to test things around you very carefully to see if you can find evidence of the demon’s trickery, but if they’re very clever they’ll reproduce every outcome with perfect precision and leave no dangling threads.
This, of course, sounds absurd, and (almost) nobody believes that this is true. But how do we know that it’s not going on? (Do we?)
It’s an old question, but it came back to mind because of a recent post on SlateStarCodex. Scott Alexander writes about two hypotheses for how dinosaur bones got put into their place inside Earth’s crust: (1) dinosaur bones originated in actual dinosaurs millions of years ago, who died and now their bones are where they left them; and (2) Satan put them there to fool us, carefully making it look precisely as they would if there had been actual dinosaurs (but there weren’t). He asks, how do we distinguish these hypotheses scientifically?
I think the correct response is to say that both theories explain the data, and one cannot empirically test which theory is true, but the paleontology theory is more elegant (I am tempted to say “simpler”, but that might imply I have a rigorous mathematical definition of the form of simplicity involved, which I don’t). It requires fewer other weird things to be true. It involves fewer other hidden variables. It transforms our worldview less. It gets a cleaner shave with Occam’s Razor. This elegance is so important to us that it explains our vast preference for the first theory over the second.
I like the example and the explanation. By construction, Satan doesn’t leave dangling threads so the predictions of the two hypotheses are precisely the same (with regard to dino bones). Someone named Faza in the comments notes that Satan is a “ZEVO”: a Zero Epistemic Value Object (good term!). If you subtract Satan you don’t lose any explanatory power. In fact, you can add more evil demons on top (“No, really what happened was Satan 2 played a trick to make it seem like Satan played a trick to make it seem like…”) and the outcome is the same. So we subtract the ZEVOs and, by Occam’s Razor, we decide to go with the simpler explanation.
(I’m going with “simpler”, mathematical rigor be damned. You know what I mean.)
I was careful to say “we decide to go with the simpler explanation”, a vague way to state the conclusion. Consider a few more precise alternatives:
By Occam’s Razor, we conclude that the simpler explanation is true.
By Occam’s Razor, we conclude that the simpler explanation is more likely to be testable.
By Occam’s Razor, we conclude that the simpler explanation is more rational to believe.
I think we should carefully distinguish these three different conclusions, and I want everyone to understand that (3) is the right one.
(To be clear, I don’t think Scott is misunderstanding Occam’s Razor in the linked post.)