Argument from authority

An argument supports a claim by citing an authority who asserts it. It is fallacious when the authority lacks relevant expertise or the assertion lacks adequate support.

Connect

Appeal to authority: An argument that treats an authority’s judgment as evidence for a claim. This is the common name for the reasoning pattern, including both reasonable testimony and fallacies.
Ad verecundiam: A traditional name for the fallacy of appealing improperly to authority. The Latin label names the faulty form distinguished from legitimate expert testimony.
Credentialism: Excessive reliance on formal qualifications as a measure of a person’s knowledge or competence. Credentials can signal expertise but do not establish authority on every question.
Inductive reasoning: Reasoning that supports conclusions probabilistically from observations or premises. Testimony generally raises a claim’s probability rather than proving it deductively.
Expert testimony: Evidence provided by someone with specialized knowledge relevant to a disputed question. Relevant expertise can make an authority’s assertion rational evidence rather than a fallacy.
Argument from ignorance: An argument that treats a claim as true because it has not been disproved, or false because it has not been proved. It substitutes absence of contrary evidence for a relevant authority’s assertion.
Scientific consensus: The prevailing judgment among relevant scientists based on accumulated evidence and scrutiny. It shows how expert authority can summarize a substantial body of evidence rather than replace it.
Reliability: The tendency of a source or process to produce accurate results across relevant cases. An authority’s track record helps determine how much confidence its assertion warrants.
Testimony: A person’s report offered as evidence about an event, fact, or claim. Authority-based arguments are a special case of relying on another person’s report.
Bandwagon effect: A tendency to adopt a belief or behavior because many others do so. Popularity is not expertise, though both can be mistaken for evidence.