Comments on: Sounds like a fallacy fallacy to me

By: Xrlq

Xrlq — Thu, 27 Apr 2006 01:01:28 +0000

Eliminating the imperfections in a theory is always desirable, but not always feasible. If no one ever drew a conclusion solely from premises that merely increase the probability of the conclusion, rather than absolutely proving it, no civil or even criminal trial could ever be decided. There’s always some doubt, however far-fetched.

By: tgirsch

tgirsch — Wed, 26 Apr 2006 21:06:37 +0000

But that logician would (or should) just as quickly encourage you to refine the argument to eliminate the fallacy, rather than to throw the baby out with the bath water. If the facts are on your side, you shouldn’t need to resort to fallacious reasoning to make your case.

Also, with most of the logical fallacies, the “fallacy” part comes not from introducing the premises, but from trying to draw a conclusion solely from those premises.

By: Xrlq

Xrlq — Tue, 25 Apr 2006 22:17:47 +0000

TGirsch, it sound like you misunderstood my point. I wasn’t arguing that one can make stupid, baseless arguments and randomly land on the correct answer anyway. Everyone understands that is possible. My point was that just because an argument is deemed “fallacious” under the rules of formal logic, that does not mean it’s a generally invalid argument that should be ignored in other contexts, as the connotations of the word “fallacy” suggest. The reason why it’s sometimes OK to be “fallacious” (in the formal logic sense of the word) is because the rules of formal logic have zero tolerance for errors. In the real world, show me an argument A that correctly predicts conclusion C 99.999% of the time, and I’ll show you a pretty strong argument. Show the same argument to a logician, and he’ll show you a “fallacy.”

By: tgirsch

tgirsch — Tue, 25 Apr 2006 20:31:50 +0000

Meh. Basically, a long-winded way of saying that just because someone uses a logical fallacy doesn’t necessarily make them wrong. He’s right, it doesn’t, but I suspect most of us already knew that. It does, however, make that portion of their argument suspect. And when someone uses them a lot, it’s a pretty good clue that they’re full of shit (although that itself is probably fallacious reasoning…).

By: Stormy Dragon

Stormy Dragon — Mon, 24 Apr 2006 22:57:11 +0000

Part of the problem is Xrlq is confusing formal fallacies (e.g. undistributed middle) and informal fallacies (e.g. causal slippery slope).

The former always represents an invalid argument; the later may or may not be invalid depending on the particulars of the specific case under discussion.