I’m embarassed to bring this up again, because I seem to quote steven0461 too often—but, in something close to his words; “When you have eliminated the impossible, whatever remains is likely more improbable than an error in one of your impossibility proofs.”
I reject that entirely,” said Dirk, sharply. “The impossible often has a kind of integrity to it which the merely improbable lacks. How often have you been presented with an apparently rational explanation of something which works in all respects other than one, which is just that it is hopelessly improbable? Your instinct is to say, `Yes, but he or she simply wouldn’t do that.’”
There is no evidence that is so strong that it will justify a statement no matter how improbable you initially considered it. Thus, as Oscar points out, this quote is off.
In Bayes/Pearl terminology, knowledge of an effect destroys the causes’ independence (d-connects them), and ruling out a cause shifts probability onto the remaining causes.
Arthur Conan Doyle
I’m embarassed to bring this up again, because I seem to quote steven0461 too often—but, in something close to his words; “When you have eliminated the impossible, whatever remains is likely more improbable than an error in one of your impossibility proofs.”
...Or you’ve just missed something. If all you’re left with is improbable you notice that you are confused. I’ve always thought that quote was off.
Then again, Sherlock never did miss anything.
I also just noticed a Sherlock quote with exactly this meaning:
Sherlock’s a more rounded rationalist than he’s given credit for.
To the contrary, he was roundly defeated on at least one occasion.
I am sorry gentlemen, but this quote of Holmes is the very essence of rationalism as I see it.
Douglas Adams
There is no evidence that is so strong that it will justify a statement no matter how improbable you initially considered it. Thus, as Oscar points out, this quote is off.
In Bayes/Pearl terminology, knowledge of an effect destroys the causes’ independence (d-connects them), and ruling out a cause shifts probability onto the remaining causes.
How does a Bayesian rule out a cause?
As a rationality quote, ”… must contain the truth” would have been better.