Why Justified True Belief Is Not Enough: The Gettier Legacy

The classical definition of knowledge runs like a familiar chord: knowledge = justified true belief. But Gettier's 1963 paper struck a dissonant note that has haunted epistemology ever since. To understand why JTB fails, we have to see exactly which job each condition performs — and where the whole machinery grinds to a halt.

The truth condition ensures that you cannot know something false. The belief condition ensures that knowing involves a mental attitude of assent. The justification condition is meant to rule out sheer guesswork or reckless assertion. It is the justification condition that is supposed to bridge the gap between you and the truth, making your true belief more than a coincidence.

In Gettier's scenarios, however, the bridge is defective. Smith's justification is impeccable as far as it goes — testimony from the boss, actual counting of coins — but it happens to be justification for a proposition that is true for reasons entirely disconnected from Smith's evidence. The belief is true, but not because of the justification; the two are held together by mere happenstance.

This suggests that JTB is not a matter of missing a minor ingredient. The entire relationship between justification and truth is far looser than the classical definition assumed. Justification can be perfectly rational and still fail to be properly connected to the truth-maker. In short, epistemic luck can contaminate the JTB formula.

The obvious next step was to add a fourth condition: perhaps knowledge is justified true belief plus something that eliminates luck. But what exactly? The simplest candidate was the no false lemma condition: the belief must not be deduced from one or more false premises. In the job counterexample, Smith's inference is valid, but it is premised on a false belief — that Jones will get the job. Dropping that false premise seems to break the lucky chain.

Yet Gettier's second case contains no visible false premise. Smith believes a disjunction, "Jones owns a Ford or Brown is in Barcelona," justified by the true (for Smith, justified) belief that Jones owns a Ford. The disjunction is true because Brown is in Barcelona, but the justification is the false first disjunct. No-false-lemma fails to block this because the false premise isn't a separate lemma; it's embedded in the disjunction. Philosophers soon constructed "Gettierised" cases with no false beliefs at all, proving that the problem is deeper than a mere mistake in reasoning.

"The Gettier problem is not merely about false lemmas or sloppy inference. It is about the structure of luck. Once luck can sneak through even the most carefully constructed evidence chains, we are forced to ask what knowledge is really made of."

Subsequent repair attempts — causal theories, reliabilism, virtue epistemology, safety and sensitivity conditions — all tried to anchor knowledge in the origin of the belief. Reliabilism says the belief must be produced by a process that tends to produce true beliefs. Virtue epistemology says the belief must be credited to the intellectual character of the believer. Safety and sensitivity say the belief must track the truth across nearby possible worlds.

Each of those proposals is richer, but each has generated its own intricate counterexamples. The Gettier problem did not kill the definition of knowledge as much as it inaugurated a research program. It converted a cozy certainty into a vibrant, unsettled field where the boundaries of knowledge are constantly renegotiated. And perhaps that is precisely what epistemology needs — not a final answer, but a permanent humbling.

If you found this valuable, consider supporting our work.