The belief was minor, but the story is entertaining:
A while ago a guy walked into the bookstore and asked me for a copy of The Art of War—by Machiavelli.
I’ve developed the habit of being polite when customers are mistaken about details, taking (and often inventing) every possible opportunity to help them save face, so I handed him a copy of Sun Tzu without comment—though you can be sure that internally I was feeling all kinds of smug at the chance to display my superior knowledge of extremely common classic books. He glanced at it and left—mortified, I imagined.
A few months later, I looked it up and discovered that Machiavelli did, in fact, write a treatise called The Art of War.
But that isn’t the embarrassing part.
The embarrassing part is that, in the moment I went to check, what I was thinking was not “Hmm, I wonder if I could have been mistaken”; it was “Heh, I wonder if anyone else has made the same mistake as that idiot!” My error was corrected only incidentally—in the course of my efforts to reinforce it.
I wonder what the chances of the guy actually asking for the Machiavelli tract are relative to the chances of him being wrong about the author? When I run into a namespace collision like that, I try to be extremely clear about it precisely so that I don’t run into situations like you described—i.e. “Machiavelli’s Art of War, not the one by Sun Tzu”.
You’re absolutely right. Anyone who knew about the existence of both books would also be aware of the need to clarify which he meant (unless he was deliberately testing me so he could feel smug at his superior knowledge). The chances he was simply mistaken are still pretty good.
Had I considered that possibility, and rejected it on grounds of low prior, maybe I would have been entitled to a Rationality Cookie; but alas, what actually happened was that I didn’t think at all.
Wait. “Rationality Cookie” is that a real thing? I can’t find it but it sounds like a good idea to train rationality the classical way via quick rewards.
The belief was minor, but the story is entertaining:
A while ago a guy walked into the bookstore and asked me for a copy of The Art of War—by Machiavelli.
I’ve developed the habit of being polite when customers are mistaken about details, taking (and often inventing) every possible opportunity to help them save face, so I handed him a copy of Sun Tzu without comment—though you can be sure that internally I was feeling all kinds of smug at the chance to display my superior knowledge of extremely common classic books. He glanced at it and left—mortified, I imagined.
A few months later, I looked it up and discovered that Machiavelli did, in fact, write a treatise called The Art of War.
But that isn’t the embarrassing part.
The embarrassing part is that, in the moment I went to check, what I was thinking was not “Hmm, I wonder if I could have been mistaken”; it was “Heh, I wonder if anyone else has made the same mistake as that idiot!” My error was corrected only incidentally—in the course of my efforts to reinforce it.
I wonder what the chances of the guy actually asking for the Machiavelli tract are relative to the chances of him being wrong about the author? When I run into a namespace collision like that, I try to be extremely clear about it precisely so that I don’t run into situations like you described—i.e. “Machiavelli’s Art of War, not the one by Sun Tzu”.
You’re absolutely right. Anyone who knew about the existence of both books would also be aware of the need to clarify which he meant (unless he was deliberately testing me so he could feel smug at his superior knowledge). The chances he was simply mistaken are still pretty good.
Had I considered that possibility, and rejected it on grounds of low prior, maybe I would have been entitled to a Rationality Cookie; but alas, what actually happened was that I didn’t think at all.
Wait. “Rationality Cookie” is that a real thing? I can’t find it but it sounds like a good idea to train rationality the classical way via quick rewards.
Not that I know of—but it could still work. I hear Eliezer once had some success with M&Ms!