The reason an argument from authority is sometimes correct

Appeal to authority is taught as a fallacy, and it is one when the authority is irrelevant or when the matter is settled by available evidence. In most cases it is neither. The listener has no access to the primary material, could not evaluate it if they did, and must therefore decide whom to believe rather than what to believe.

What everyone sees

The standard advice is to examine the evidence yourself. Applied consistently, this would require expertise in every field one holds a belief about, which nobody has. The rule is therefore not being followed by anyone; what varies is only whether the deference is admitted or disguised as independent judgement.

What is actually happening

Hardwig argued that epistemic dependence is unavoidable and rational: much of what any person knows rests on testimony they cannot check, and refusing such testimony would leave them with less knowledge rather than more. Goldman then set out the practical problem — how a novice can assess competing experts — and identified usable indicators, including the quality of argumentation, agreement among independent experts, track record, and the presence of interests that might bias the claim. The fallacy is therefore narrower than the slogan suggests. It lies in deferring without applying those indicators, not in deferring at all.

Why it stays hidden

The confusion persists because deference is invisible when it succeeds. A belief absorbed from a reliable source feels like something one knows, not something one accepted on testimony. Only failures are recognised as appeals to authority, so the whole functioning apparatus of dependence goes unnoticed and the practice appears to be an error.