Explain the Gettier problem as a challenge to the sufficiency of the tripartite analysis and assess the main attempts to repair the definition of knowledge

What this dot point is asking

SEAB wants you to explain why the apparently watertight definition of knowledge as justified true belief was overturned, and to assess the attempts to fix it. The challenge comes from Edmund Gettier, who in a short 1963 paper produced cases in which all three conditions are satisfied yet we deny that the person knows. Your task is to show how such cases work, diagnose what has gone wrong, and weigh the proposed repairs.

The answer

The tripartite analysis recalled

The standard analysis says S knows that p if and only if p is true, S believes that p, and S is justified in believing that p. Each condition is meant to be necessary, and the three together sufficient. Gettier accepts that the conditions are necessary; what he attacks is the claim that they are jointly sufficient.

The structure of a Gettier case

A Gettier case is a recipe. Take a justified belief that happens to be false, draw from it a logical consequence that is also justified, and arrange the world so that the consequence is true, but true for a reason unconnected to the believer's evidence. The belief then satisfies all three conditions, yet the truth was a matter of luck, so it is not knowledge. The lesson is that justification can attach to a belief while failing to connect it to the fact that makes it true.

Diagnosis: epistemic luck

The common diagnosis is that Gettier beliefs are true by luck. Justification was supposed to rule out luck (that was its job in defeating the lucky-guess cases), but Gettier shows it can fail to do so when justification and truth come apart. So knowledge needs more than justified true belief: it needs the right kind of non-lucky connection between belief and fact.

Repair 1: no false lemmas

One repair adds a fourth condition: the justified true belief must not be inferred from any false premise. In the classic office case the believer reasoned through a false assumption (that a particular colleague owned the car), so the belief is disqualified. This handles inference-based cases neatly. Its weakness is that some Gettier cases involve no inference from a falsehood at all, such as perceptual cases.

Repair 2: defeasibility

A second repair says a justified true belief is knowledge only if there is no true proposition which, were it added to the believer's evidence, would defeat the justification. In Gettier cases there is always such a hidden truth (that the colleague had sold the car). The difficulty is that some genuine knowledge has misleading defeaters, true facts that would mislead if known, so deciding which defeaters disqualify knowledge is itself a problem.

Repair 3: reliabilist and causal accounts

A more radical move abandons internal justification. Causal theories require that the fact make the belief true through an appropriate causal chain. Reliabilism requires that the belief be produced by a reliable process. Both explain the anti-luck intuition: in Gettier cases the process or causal link is defective. But they face the fake-barn case, where the process (ordinary vision) is normally reliable yet fails locally, and the generality problem of specifying how broadly to describe the process.

Examples in context

Example 1. The fake-barn county. Henry drives through a region dotted with convincing barn facades and stops in front of the one real barn, forming the belief "that is a barn." The belief is true and produced by normal vision, yet because he could so easily have been looking at a fake, we hesitate to call it knowledge. This case pressures even reliabilist repairs, since the visual process is normally reliable but locally unsafe, and it involves no inference from a false premise.

Example 2. The reliable but unstated source. A seasoned birdwatcher identifies a rare bird at a glance and is right, but cannot articulate the cues she used. An internalist may say she lacks accessible justification, while a reliabilist credits her with knowledge because her perceptual process is reliable. The contrast shows how Gettier-driven repairs push the definition of knowledge toward externalism.

Try this

Q1. Explain the recipe for constructing a Gettier case. [6 marks]

• Cue. Start with a justified false belief, infer a justified consequence, and arrange the world so the consequence is true but for an unrelated reason, so all three conditions hold while the truth is lucky.

Q2. Why does the no-false-lemmas repair fail to handle every Gettier case? [8 marks]

• Cue. Some cases, like the fake-barn case, involve no inference from a false premise, so a condition banning false lemmas leaves them untouched.

Q3. Briefly explain how reliabilism responds to the Gettier problem and one objection to it. [6 marks]

• Cue. Reliabilism requires a reliable belief-forming process, which the defective Gettier process lacks; objection: fake-barn cases (locally unsafe but normally reliable) and the generality problem of how to describe the process.