The Classic Account: Justified True Belief

You wake up and glance at your phone: it is 7:42 AM. You believe it. The phone is working correctly, so the belief is true. You checked a reliable device, so the belief is justified. That seems like knowledge. For most of the history of Western philosophy, this triangulation of truth, belief, and justification was considered the complete story.

The project of epistemology (the branch of philosophy concerned with knowledge, justified belief, and the nature of understanding) begins with a deceptively simple question: what distinguishes knowledge from mere opinion? You might believe truly that the stock market will rise tomorrow while having no good reasons; a lucky guess is not knowledge. You might have all the justification in the world and still be wrong; a justified false belief is not knowledge either. The classical analysis, traceable to Plato's dialogue Theaetetus and systematized by modern philosophers, says knowledge is Justified True Belief (JTB): a belief that (1) is true, (2) is held by the subject, and (3) is adequately justified.

Each condition does real work. Truth rules out lucky guesses about the future that happen to come true. Belief rules out knowledge of facts the subject has never considered. Justification, the most contested condition, rules out accidentally correct beliefs formed by unreliable methods. Justification is what connects the believer epistemically to the truth, what makes the true belief non-accidental.

For much of the twentieth century, justification was understood in internalist terms: what justifies a belief is the evidence and reasoning that the believer has access to from the inside, their experiences, memories, and logical relations between their beliefs. On this view, two subjects in exactly the same psychological state are justified or unjustified in exactly the same way, regardless of what the outside world is actually like.

The internalist picture is intuitive. When your friend says "I know it will rain today," you naturally ask what their evidence is, you want to assess the justification they have access to. If they say "I just feel it," you withhold the label knowledge. If they say "the radar shows 90% chance, dark clouds are building, and the barometric pressure has dropped," you grant it. Justification, on this view, is a matter of the believer's epistemic position, not of hidden causal mechanisms.

A Justified True Belief counts as knowledge if and only if the belief is true, the subject holds it as a belief, and the subject is epistemically responsible in forming it, with the grounds of the belief establishing its truth, not merely accidentally correlating with it.

— Standard formulation of the JTB account, as articulated in Edmund Gettier's discussion and Oxford epistemology literature

This account dominated philosophy for millennia. Then in 1963 Edmund Gettier published a three-page paper that detonated the consensus.