Fallacy Library

Attacking the person advancing an argument rather than the argument itself. Their character has no bearing on whether their reasoning holds.

Treating a personal story or an isolated case as though it outweighed systematic evidence. Individual cases are real, but they do not establish general patterns.

Citing an authority where the authority is outside their field, is not representative of expert opinion, or where the question calls for evidence rather than endorsement. Citing relevant, genuine expertise is not a fallacy: this is about misuse.

Substituting an emotional response for a reason. Emotions are often relevant to ethical questions; the fallacy lies in using feeling in place of an argument rather than alongside one.

Arguing that an action is not worth taking because it will not, by itself, solve the whole problem. It mistakes a contribution for a complete solution.

Treating the absence of proof against a claim as proof of it, or the absence of proof for a claim as proof against it.

Treating 'natural' as evidence of 'good', or 'unnatural' as evidence of 'bad'. Whether something occurs in nature tells you nothing about whether it is beneficial or right: hemlock, malaria and infant mortality are all perfectly natural.

Treating widespread belief or practice as evidence that something is true or right. Also called the bandwagon fallacy, or argumentum ad populum. Majorities have been wrong about a great many things.

Treating the length of time a practice has continued as a reason to continue it. Age is not a justification; it only tells you the practice is old.

Demanding that others disprove a claim rather than supporting it. The burden sits with whoever makes the assertion.

Presenting only the evidence that supports a conclusion while leaving out evidence that tells against it.

Assuming the conclusion somewhere among the premises, so the argument never supplies independent support. The formal name is begging the question, or petitio principii.

Composition: assuming what is true of the parts must be true of the whole. Division: assuming what is true of the whole must be true of each part.

Defining a term so that the conclusion is built into the definition, winning the argument by stipulation rather than evidence.

Using a single word in two different senses within one argument, so that a conclusion appears to follow when it does not.

Inferring that one thing caused another from the fact that they occurred together or one after the other. Correlation and sequence are not causation.

Presenting two options as though they were the only ones available when further options exist. Also called a false dichotomy, or black-and-white thinking.

Treating two things as comparable when the differences that matter to the argument are large.

Judging a claim by where it came from rather than by whether it is true. The origin of an idea is not evidence about its content.

Drawing a general conclusion from a sample too small or too unrepresentative to support it.

Moving from a description of how things are to a claim about how they ought to be, without supplying the moral premise that would connect the two. Identified by David Hume.

Assuming the world is fundamentally fair, and therefore that those who suffer must in some way have deserved or invited it.

Asking a question that builds in an unproven assumption, so that any direct answer concedes it.

Treating a previous good act as permission for a later harmful one, as though ethics ran on a balance sheet.

Raising the standard of proof once the original standard has been met, so that no amount of evidence is ever sufficient.

In G. E. Moore's original sense, defining 'good' as identical to some natural property (pleasant, natural, evolved) so that the moral question is settled by definition. Frequently confused with the is–ought fallacy; they are related but distinct. This one is an error about definition, is–ought is an error about inference.

Rejecting a workable option because it falls short of a perfect one. Also called the perfect solution fallacy. The relevant comparison is with the realistic alternative, not with perfection.

Defending a generalisation against counterexamples by redefining the category so the counterexamples no longer count.

Introducing a point that is not relevant to the question under discussion, drawing attention away from it rather than answering it.

Dismissing a problem by pointing to a worse one. That something worse exists does not make the thing in question acceptable.

Claiming that one step must lead to an extreme outcome without showing why each step in the chain follows from the one before it.

Applying a principle generally while claiming an exemption for one's own case, without giving a relevant reason for the exception.

Preferring existing arrangements simply because they are the existing arrangements, and treating any change as carrying a burden of proof that the present state does not.

Replacing someone's actual position with a weaker, distorted version, then refuting that instead. The original argument is left untouched.

Continuing with something because of what has already been invested in it, rather than judging it on its merits from this point forward. Past costs are gone either way.

Rejecting a claim by pointing to inconsistency in the person making it. Hypocrisy may be worth noting about the person, but it has no bearing on whether what they said is true.

Meeting a criticism by raising a different issue instead of answering it. A form of tu quoque used to change the subject rather than to rebut.