# Justified true belief is not always knowledge For about 2500 years, knowledge was defined as [[Justified True Belief (JTB)|justified true belief (JTB)]] — a claim counts as knowledge when you believe it, it is true, and you have good reasons to believe it. In 1963, [[Edmund Gettier]] published two short thought experiments showing this definition is not enough. In a Gettier case, all three conditions are satisfied, yet the justification connects to the truth only by coincidence. The belief is true *by accident*, not because the reasons track reality. Intuitively, that is not knowledge. The general pattern: a justified belief passes through a false intermediate step (or an unrelated lucky fact), and the conclusion happens to be true anyway. The knower has no real grip on why the belief is true — they just got lucky. This matters beyond philosophy. In personal knowledge work, a note can be sourced, defensible, and factually correct while resting on a justification chain that does not actually support it: a miscited statistic that happens to be right, a chain of reasoning with a subtle error whose conclusion still holds, an AI summary that captured the answer while garbling the reasoning. Each is a Gettier case in miniature, and treating them as solid knowledge compounds hidden fragility. The defensive response — relevant to any [[Personal Knowledge Management (PKM)]] practice — is to track *why* a claim is held, not just *that* a source exists. Prefer primary sources for load-bearing claims, record the full chain of citations (not just the immediate one), build redundant justification for high-stakes claims, and accept that even impeccable justification can be lucky. This is the [[All knowledge is conjectural|fallibilist]] stance applied at the level of individual notes. ## References - Gettier, E. (1963). "Is Justified True Belief Knowledge?" *Analysis* 23(6), 121–123 - Stanford Encyclopedia of Philosophy — "The Analysis of Knowledge" - Plato, *Theaetetus* — the original JTB definition