Yes, people who say “all X are Y” usually do know at least one person who happens to be an X and whom they don’t actually alieve is Y—but I think that in certain cases what’s going on is that they don’t actually alieve that person is an X, i.e. they’re internally committing a no true Scotsman. Now, I can’t remember anyone ever explicitly saying “All X are Y [they notice that I’m looking at them in an offended way] -- well, you’re not, but you’re not a ‘real’ X so you don’t count” (and if they did, I’d be tremendously offended), but I have heard things that sound very much like a self-censored version of that.
The reasoning you described reaches valid (object level) conclusions in the different cases under consideration, but you still prefer to analyze it as full of fallacies for some reason.
Huh, no. If an argument has premises “all X are Y” and “John is an X” and conclusion “John is not Y”, it is broken. Whether the conclusion happens to be true because one of the premises is false is irrelevant.
The argument’s stated premises were “X are Y”, you decided to interpret the ambiguous statement as “all X are Y” and then complain that it makes the argument formally false.
[...] but I have heard things that sound very much like a self-censored version of that.
What exactly do you count as a self-censored version of that? Pointing out that you’re an exceptional X, that you have characteristic Z, which correlates negatively with Y, or some such thing? If so, the answer is: well, of course, what do you expect?
If people make a generic generalization along the lines of “(all) X are Y”, then naturally, you have to be an exceptional X in order to be Y. One could say that it’s enough that you are Y, because then you are an exceptional X in virtue of that. But that’s not how generic generalizations work. People make such generalization usually not purely on the basis of statistical data, but because in their model, something about X causes Y (or they have a common cause). So if you’re X, but not Y, chances are you have additional characteristic Z, which is rare among Xs, and which counteracts X’s influence on Y.
It’s just like saying “dogs have four legs—well, not Fido, obviously, but he’s had an accident and one of his legs had to be amputated”. This kind of thing might sound a bit like a self-censored version of “but Fido isn’t a true dog”, but what it really says is “but Fido isn’t an ordinary dog”, which is entirely correct!
Maybe you’re aware of all this anyway, but I just thought it’d be worth pointing out.
In the context of human groups and human sub-groups, I’m not sure “ordinary” member of the group is used differently than “true” member of the group. Witness those who claim the community organizer is not “really” black because he did not live the ordinary life experiences of a black male child (i.e. he didn’t live in a poverty stricken inner city while growing up).
I’m inclined to argue, as some linguists would, that tabooing “ordinary” is impossible in this context, because people are intuitive essentialists, and that generic statements make reference to such postulated essences, which define what makes for an “ordinary” X. (Hence a lot of Aristotelian nonsense.)
This does, indeed, fit very well with your observation—with which I agree—that sometimes, the borderline between “ordinary” and “true/real” becomes blurred. However, I think one should still be wary of suspecting mentions of “extraordinary” of being censored no-true-Scotsman-arguments without further evidence.
Yes, people who say “all X are Y” usually do know at least one person who happens to be an X and whom they don’t actually alieve is Y—but I think that in certain cases what’s going on is that they don’t actually alieve that person is an X, i.e. they’re internally committing a no true Scotsman. Now, I can’t remember anyone ever explicitly saying “All X are Y [they notice that I’m looking at them in an offended way] -- well, you’re not, but you’re not a ‘real’ X so you don’t count” (and if they did, I’d be tremendously offended), but I have heard things that sound very much like a self-censored version of that.
I generally avoid criticizing reasoning that reliably reaches correct conclusions.
I’m not sure what the relevance of that to my comment is.
The reasoning you described reaches valid (object level) conclusions in the different cases under consideration, but you still prefer to analyze it as full of fallacies for some reason.
Huh, no. If an argument has premises “all X are Y” and “John is an X” and conclusion “John is not Y”, it is broken. Whether the conclusion happens to be true because one of the premises is false is irrelevant.
The argument’s stated premises were “X are Y”, you decided to interpret the ambiguous statement as “all X are Y” and then complain that it makes the argument formally false.
Re-read the fifth word of this comment. (Or am I missing something?)
You may want to (re)read this comment to see/remember how this discussion started.
What exactly do you count as a self-censored version of that? Pointing out that you’re an exceptional X, that you have characteristic Z, which correlates negatively with Y, or some such thing? If so, the answer is: well, of course, what do you expect?
If people make a generic generalization along the lines of “(all) X are Y”, then naturally, you have to be an exceptional X in order to be Y. One could say that it’s enough that you are Y, because then you are an exceptional X in virtue of that. But that’s not how generic generalizations work. People make such generalization usually not purely on the basis of statistical data, but because in their model, something about X causes Y (or they have a common cause). So if you’re X, but not Y, chances are you have additional characteristic Z, which is rare among Xs, and which counteracts X’s influence on Y.
It’s just like saying “dogs have four legs—well, not Fido, obviously, but he’s had an accident and one of his legs had to be amputated”. This kind of thing might sound a bit like a self-censored version of “but Fido isn’t a true dog”, but what it really says is “but Fido isn’t an ordinary dog”, which is entirely correct!
Maybe you’re aware of all this anyway, but I just thought it’d be worth pointing out.
Perhaps taboo “ordinary” and “true”?
In the context of human groups and human sub-groups, I’m not sure “ordinary” member of the group is used differently than “true” member of the group. Witness those who claim the community organizer is not “really” black because he did not live the ordinary life experiences of a black male child (i.e. he didn’t live in a poverty stricken inner city while growing up).
I’m inclined to argue, as some linguists would, that tabooing “ordinary” is impossible in this context, because people are intuitive essentialists, and that generic statements make reference to such postulated essences, which define what makes for an “ordinary” X. (Hence a lot of Aristotelian nonsense.)
This does, indeed, fit very well with your observation—with which I agree—that sometimes, the borderline between “ordinary” and “true/real” becomes blurred. However, I think one should still be wary of suspecting mentions of “extraordinary” of being censored no-true-Scotsman-arguments without further evidence.