(OTOH I think language should also depend on what you value: if your utility function is the number of inwardly-thrice-bent metal wires capable of nondestructively fastening several standard sheets of paper together at an edge in the universe, it’s handy to have a single word for ‘inwardly-thrice-bent metal wire capable of nondestructively fastening several standard sheets of paper together at an edge’, whether that’s a natural category or not. But you shouldn’t pretend it’s a natural category.)
A: No human thinks red shirts are better than blue shirts. B: Lots of psychopaths think red shirts are better than blue shirts. A: I meant true humans. Psychopaths aren’t really humans, so don’t count. B: What about my friend Billy? He is not a psychopath but thinks red shirts are better than blue shirts. A: True humans are non-psychopaths who are not your friend Billy.
Not “No true Scotsman”:
A: No human thinks red shirts are better than blue shirts. B: Lots of psychopaths think red shirts are better than blue shirts. A: I meant true humans. Psychopaths aren’t really humans, so don’t count. B: What about my friend Billy? He is not a psychopath but thinks red shirts are better than blue shirts. A: Oh, I guess I was wrong—some humans think red shirts are better than blue shirts.
The second is just using a nonstandard definition, not redefining the word to fit the line of argument, so does not fall under the No True Scotsman fallacy. Even if you’re gerrymandering reality ahead of time, it doesn’t count as No True Scotsman (At the very least, that isn’t even an argument yet, so can’t be a fallacious argument!)
Clearly not. As I noted upthread, True Scotsman requires that the redefinition is happening for rhetorical, rather than objective, reasons. The relevant reference class could have been picked out ahead of time, and I wouldn’t predict those two folks are going to continue that dispute.
Well, I’d say it depends on the complexity of those objective reasons. “The way to carve reality at its joints, is to draw simple boundaries around concentrations of unusually high probability density in Thingspace. Otherwise you would just gerrymander Thingspace.”
(OTOH I think language should also depend on what you value: if your utility function is the number of inwardly-thrice-bent metal wires capable of nondestructively fastening several standard sheets of paper together at an edge in the universe, it’s handy to have a single word for ‘inwardly-thrice-bent metal wire capable of nondestructively fastening several standard sheets of paper together at an edge’, whether that’s a natural category or not. But you shouldn’t pretend it’s a natural category.)
“No true Scotsman”:
Not “No true Scotsman”:
The second is just using a nonstandard definition, not redefining the word to fit the line of argument, so does not fall under the No True Scotsman fallacy. Even if you’re gerrymandering reality ahead of time, it doesn’t count as No True Scotsman (At the very least, that isn’t even an argument yet, so can’t be a fallacious argument!)
“Everybody likes to watch a beautiful sunset”
“Fred doens’t. Mind you, he’s blind”.
“Then he doesn’t count”
True Scotsman or not?
Clearly not. As I noted upthread, True Scotsman requires that the redefinition is happening for rhetorical, rather than objective, reasons. The relevant reference class could have been picked out ahead of time, and I wouldn’t predict those two folks are going to continue that dispute.