4 ms·
Right. The article calls these things "fallacies", but that only strictly means that they're not slam-dunk arguments. Logical fallacies aren't necessarily bad h
by repsilat 8y ago
Right. The article calls these things "fallacies", but that only strictly means that they're not slam-dunk arguments. Logical fallacies aren't necessarily bad heuristics, they just don't necessarily assure strict logical implication.
(Often they are actually decent Bayesian "evidence", though I don't want to make that argument for the cases in TFA.)
- GauntletWizard 8y agoThe Roman Trivium was Logic, Rhetoric, and grammar. Most of these arguments are logical fallacies - X does not guarantee Y - but are valid Rhetorical devices. Slippery slope is easiest to explain. If A, 90% chance of B, if B, 90% chance of C, if C, 90% chance of D. Therefore, if A, 73% of D - A implies D, if not perfectly. This is completely valid in Rhetoric, where you can't perfectly model anything, just approximate, but not logic, where truth or falsehood is digitized and binary. Most arguments that humans get into are rhetorical, but our focus on logic has left us vulnerable to misunderstanding. We are ignoring clear correlations because there's not a logical certainty.