I agree, but I don’t think that people are putting it in such logical terms. In practice, we’re likely talking about clusters of related beliefs. If a cluster contains a reprehensible belief, then all beliefs in that cluster end up with a taint, even the reasonable ones. C may even acknowledge that belief Y is reasonable, but it’s unreasonable for B to express it, because A has done the same and also has reprehensible belief X (see “Blue lives matter”, etc.).
I disagree and I think “Blue lives matter” goes right to the heart of it. People may not express themselves in terms of formal logic but that doesn’t mean they aren’t thinking logically.
It is considered unreasonable to express “Blue lives matter” is because expression of that phrase has a history that implies opposition to “Black Lives Matter” (belief X). So what you really have is two beliefs, Y and Z. Y, belief that the lives of police matter, C admits is reasonable. Z, belief that it is appropriate to express Y as “Blue Lives Matter”, C would say is unreasonable. Not solely because of A, but because it is associated with X.
~Max
So, how many inductive leaps is reasonable? You can get from anywhere to anywhere with a sufficient number of them, because none of them are perfect in a logical sense. The more leaps you consider acceptable, the more polarization you are guaranteed to generate, because each step has some error and that error represents a person you shoved into the wrong category: someone that (perhaps) you called a racist because they expressed a belief that many racists express. Perhaps you say that a small amount of error is acceptable. But how much is too much?
Only one inductive leap is ever made - generalizing Z to X. Going from general to specific is logically sound deduction. The error (if there is one) is thinking the belief itself is racist.
~Max
Well, it’s absolutely false that everyone that states “Blue lives matter” is racist. And yet that’s a commonly made inference, because a whole lot of those people are racist.
And the same is true for any statement about people that isn’t basically tautological. So if induction is to have any role here, it has to have some fuzziness to it. Which means it will have some error. We accept a certain amount of error when classifying people–to do otherwise is impossible–but the more steps are present, the greater the error and the more polarization you get.
Well it obviously depends on context - there’s use-mention distinction for example in this discussion. But there’s still only one necessary generalization, only one inductive step, and that’s whether it is ever appropriate (not racist) to state “Blue lives matter” in a given context.
~Max
Now I’d like you to consider my point of view.
- Person B holds belief Y.
- Person C identifies belief Y with X. (Because of the history of Y being applied to bring about X)
- Person C calls B an X because he holds belief Y.
C might be willing to explain why he or she believes the history of Y implies X, or maybe not (esp. if it’s common knowledge). This just seems, to me, altogether more straightforward and realistic. Suppose this is the real state of things, but
- A holds belief X
- B (ETA: maybe C as well) mistakenly believes: Person C calls B an X because he agreed with A
So B challenges C for comparing him to A. B makes great efforts to distinguish himself from A.
B is wasting his efforts. B eventually writes C off as irredeemable, or vice versa.
~Max
We’re at risk of getting sidetracked. We can go with transphobes, QAnon, flat Earthers, whatever.
I’m having a slightly hard time following your example, because it seems you’re using X and Y in slightly different ways. In my example, A/B/C were people, and X/Y were stated beliefs. I didn’t give letters to conditions or goals.
If A says “all members of race R are evil”, then we can directly infer that they’re racist. They meet the basic definition. You can’t directly conclude that if they say “F lives matter”. Maybe A1 through A999999 all say both things, and so there’s a certain association. But it’s at best probabilistic unless you can prove there’s no B that also says “F lives matter”.
Well, yes. C has concluded the wrong thing about B and there’s nothing B can say to convince them otherwise. That sounds like a bad outcome to me. B and C may have much in common, but C has rejected all that by making a false inference. If C had focused less on dogwhistle detection and more on what is actually the case, we’d have less polarization.
If a person expresses racism, is that person racist? The conclusion arguably follows directly from the definition of racist, adjective, i.e. a person who expresses racism is racist. No need to prove there’s no non-racists who express racism. That’s tautological.
~Max
In my example, A, B, and C are people. X and Y are held beliefs but they could just as well be stated beliefs. I didn’t give any letters to conditions or goals, either.
~Max
This doesn’t follow from the premises. In my example, unlike yours, C uses a valid argument form, so one of the premises must be wrong for the conclusion to be wrong. I deliberately wrote the example so neither B nor C are objectively right or wrong.
~Max
Sure, if they believe it. For the sake of argument, I’m assuming that all these statements are made honestly.
“F lives matter” is obviously not a racist statement. You could conclude that someone is racist based on that only if you can show that no non-racist would ever say that. Which is highly unlikely to be the case.
Well–it might actually be the case in C’s particular circle. C and everyone like them wouldn’t be caught dead saying “F lives matter”, even if they believe it, because they know the inference everyone would make. So the belief about its validity turns into actual validity–but it only works for that community. C applying that logic to anyone else is invalid. B may simply have not gotten the message. Or, they may not care if C thinks they’re racist.
I’ll take it, and add that C may or may not care whether B got the message or is part of the community. And then it becomes a matter of whether B should voluntarily act differently out of deference to C, a matter of manners which B has the power to resolve with his own judgement, even in the face of fundamental differences of opinion and a defiant C.
~Max
You said “the history of Y being applied to bring about X”. You can’t bring about a belief, but you can bring about a goal. A belief could be “all members of race R should be oppressed”, while a goal might be “all members of race R are oppressed.”
Do they? “the history of Y being applied to bring about X” is never a valid argument, so there’s nothing for C to explain. It’s just the black swan problem. It doesn’t matter how many white swans you observe; you can never be certain that no black swans exist.
A minor semantic disagreement - you understand what I meant?
That’s also not the main argument form. It’s the inductive rationale for an argument, the conclusion of which is the second premises to the main argument. The main argument itself is valid deduction.
Person B holds belief Y.
Person C identifies belief Y with X.
Person C calls B an X because he holds belief Y.
i.e.
all B are Y
all Y are X
all B are X
~Max
I did; it just made it a little harder to follow.
Fair enough. That’s valid, for some reasonable meaning of “identifies”, etc. So yes, I dispute that the second premise can ever be true except in a probabilistic fashion.
It isn’t IME usually “all Y are X, therefore all B are X”. It’s more like “most Y are X; therefore if B is Y, there’s a pretty good chance that B is X.”
Agreed - as you said, every generalization has room for error! I just would like to point back to posts #157 and #158, which I think present the practical causes and solutions to this particular problem. It takes both parties to produce the polarized environment and it only takes one to depolarize - but that doesn’t mean both parties are equally in the wrong. One side might be definitely justified and the other unjustified.
@thorny_locust, in that case there would be two generalizations, first in making the general rule, second in applying it. It’s much weaker in my opinion, open to the fallacy of association, comparable to B naming “black friends” etc. Not necessarily wrong, though.
~Max
My point is that you can’t go with “whatever”. Ideas are different. Some of them can be dismissed with prejudice because we’ve already had ample information to judge them. We don’t have to re-hear every controversy.
QAnon is actually more dismissible than Nazis in this case. Nazis at least used some pseudo-scientific arguments that can be scientifically refuted (not that we need to, because it’s been done, but you could). Same goes for flat-Earthers. They have come up with zero new information since Copernicus’s time.
Transphobes yet again are a different case for various reasons, but by and large, we don’t need to listen to them. They’re bringing no new information to the conversation and don’t intend to be persuaded. The only thing most of them are bringing to the conversation is their personal “red line”, i.e. “I will accept everything except transwomen in women’s sports.” And a little investigation will show that those people follow no women’s sports, that they’re simply wielding a red line for the comfort of knowing that they still can.
You can’t reason people out of positions they didn’t reason themselves into. There’s no point trying. We don’t have to give every dumb controversy a fresh hearing just because someone rolled up late to the conversation. We can dismiss things with prejudice.