LogicalFallacies

Quora Answer : What is the statistical correlation between a person commiting logical fallacies and their conclusion being wrong?

Mar 8, 2016

None.

A logical fallacy is simply an inference that's invalid.

I assert "A implies B" but I'm wrong about that.

You can't infer anything about A or B themselves from the fact that I erroneously asserted A implies B.

In particular it doesn't make B less likely to be right.

Now that's the logic dealt with. What about the more general question.

If agent A is wrong about X is he / she more likely to be wrong about Y

For values of X and Y in general.

I don't know. I'm not sure we've done large enough experiments to be able to make serious claims one way or the other. We all know that knowledge is complex and compartmentalized. We've all been right about some things. And wrong about others. We all know people who have good intuitions in one domain and lousy ones in another.

You MIGHT want to say that a person who misunderstands a domain so much that they assume false implications may also misunderstand it so badly that they are ignorant of, or misinterpret crude facts.

For example, a person who subscribes to the gamblers fallacy may not have much background in or understanding of other statistical ideas.

But then, many of what are called "logical fallacies" are meant to be generic and independent of a domain. So why would applying an incorrect "no true Scotsman" argument in the domain of politics give us any reason to think that the person is ignorant of politics in general?

I'd suggest that the specific content of the claims is far more important than the logical structure when it comes to us making an assessment.

For example :

Q : why didn't Moonbeam just fly over the fence and escape the paddock?

A : dude, contrary to what you see in My Little Pony, real horses can't fly.

And

Q : how come conservatives are such hypocrites, going on about family values but having affairs on the side?

A : dude, no real conservative betrays his wife.

These are both "no true Scotsman" arguments. But it's obvious that you can't judge the depth of the speaker's knowledge or the strength of the conclusion by the fact that they used that particular structure of argument.

This actually gives us an interesting hint : the most damning thing you might find is that a person who uses what are called "logical fallacies" is really guilty of "over-generalization". In other words, there's nothing wrong with the "logic" in a "logical fallacy". It's that you can't apply it to the category because the category is too broad, and you are ignoring details.

Most people accept a blanket claim that horses can't fly. But most people probably assume that you can't generalize over conservatives or even "true conservatives" sufficiently well to make such a bald, all encompassing statement about their fidelity.

Now it may be that some people are inclined towards making stark generalizations. Perhaps they are too quick to embrace stereotypes or lack the care and attention to detail needed to understand the subtlety of an issue. These people might look like they're making a lot of "logical fallacies" because they are trying construct logical implications with categories that are too broad to generalize over such as "true conservatives".

I'd suggest that this is a better "theory" of mistake-making than "logical fallacies". It would make predictions that people who are wrong (ie. over-generalize in some cases) might be more likely to be wrong (ie. over-generalize) in others. And it would explain why they may NOT be so wrong in a third context (ie. a very familiar context or a specialization they studied) where they are more aware of and alert to subtleties.