Topic: "Usually X means Y" proposals

[Dallas]

Question... could this be a logical fallacy in how its used? If so, what would it be and why? Now, I think this can be used harmlessly in some cases like "Usually dark cloudy skies mean rain or storms." (Uses of it that propose a conclusion that is not necessarily 'wrong'). But it could also be used very poorly in creating a bad generalization (Example: "Usually when girlfriends say "I'm fine" to their boyfriends, they are really not fine.").

I am having talks with a relative about this and their reliance on these kinds of proposals just irk me. Even though I sometimes catch myself doing it too, in some situations. It just feels like a dangerous habit, though, to depend on reasoning too many things in this platform... like possibly becoming a "Common Knowledge =/= Personal Experience is Universal Fact" sort of mode of thinking.

Advice or thoughts on this?

Note: I'm not really looking for what I should say, really. This is just something I'm trying to identify in my head, more than anything.

[Blythe]

Not necessarily an outright fallacy. More like a case of a potentially false assumption, and in 'usually X means Y' arguments, that sort of argument easily lends itself to more subjective rather than objective perspectives.

It's not necessarily a bad thing, although like most methods of making a point, it can be used for bad things.

Generally if someone uses this sort of wording, try to apply a principle of charity to what they're saying and interpret whatever it is in the most positive manner it can be interpreted, and then ask if that's what they mean.

Edit: It should be noted that with 'usually x means y,' it's very easily possible to fall into making some fallacies, depending on how you justify 'usually x means y.'

[Dallas]

I see. Yeah, on one hand I felt weird calling it as if it were a straight up fallacy because of this could be used as a device. Kinda sorta (maybe?) like "If; Then" logic? It seems very human and natural to do, and it might not necessarily be bad... but could potentially be fallacious depending on the nature of what is  being proposed.

*laughs* I probably either read Veks' logical fallacies thread too much... or not enough. I am not quite sure, yet. Thanks for the perspective and patience, Blythe.

Sometimes its a bit dicey in trying to help me figure out what i am thinking and why I am thinking it.

[Vekseid]

Bayesian logic is a valid and well-studied form of logic. This sort of thing often runs simple 'AI' for your games and things like your thermostat and fridge.

For example, a manufacturer knows their temperature sensor has a certain degree of reliability. So rather than trusting it, this reliability informs how likely it should turn on or off at a given reading and temperature setting. This creates an overall more reliable experience.

[Dallas]

Ah, I think I understand this whole thing a bit better now. Thanks for your time, Veks.

Heh, the big challenge with me is precisely articulating what I am thinking (and/or feeling). So I am glad people (at least) see the wheels turning in my head. Anyways, I appreciate this a lot. I think this helped both identify this for what it is (on a wider scale) and the moments that bother me when this is abused/misused.

I would normally say that this has given me a lot to think about but... maybe I am thinking a bit too much tonight.

Thanks a bunch, you two

-Dall

[CopperLily]

I am having talks with a relative about this and their reliance on these kinds of proposals just irk me. Even though I sometimes catch myself doing it too, in some situations. It just feels like a dangerous habit, though, to depend on reasoning too many things in this platform... like possibly becoming a "Common Knowledge =/= Personal Experience is Universal Fact" sort of mode of thinking.

Advice or thoughts on this?

Note: I'm not really looking for what I should say, really. This is just something I'm trying to identify in my head, more than anything.

These often fall under the idea of 'Heuristics' as well, and well...humans are very, very good at developing them. Because many times, an approximately correct answer that arrives quickly is ideal.

But like all approximate algorithms, they can fail, and fail badly. That's part of what Veks is talking about with Bayesian logic. That gives value to "What is already known" when considering a new piece of information (the technical term is a 'prior').

The issue is when that information *doesn't* change given new information - or when contradictory information actually makes one dig in their heels. That way lies badly flawed reasoning.