The tripartite (justified true belief) definition of knowledge, the distinction between propositional, ability and acquaintance knowledge, Gettier cases, and the main responses to Gettier including infallibilism, no false lemmas, reliabilism and virtue epistemology.

What this dot point is asking

AQA wants you to distinguish three kinds of knowledge, analyse the tripartite (justified true belief) definition of propositional knowledge, explain why Gettier cases appear to refute it, and evaluate the main responses: adding an infallibilist condition, a no-false-lemmas condition, replacing justification with reliabilism, or replacing it with virtue epistemology.

Three types of knowledge

The tripartite definition: justified true belief

The classic view, traced to Plato's Theaetetus, defines knowledge as justified true belief. S knows that p if and only if: (1) p is true, (2) S believes that p, and (3) S is justified in believing that p. Each condition is claimed to be individually necessary and the three jointly sufficient.

You can argue each is individually necessary. The truth condition: you cannot know a falsehood, since "I know that the earth is flat" is self-defeating once we accept the earth is round. The belief condition: knowledge is a mental state held by a subject, so you cannot know what you do not even believe; a candidate who writes the right answer while sincerely doubting it does not know it. The justification condition: a lucky guess that happens to be true is not knowledge, which is why we add that the belief must be held for good reasons or on adequate evidence. The justification condition is the one that does the most philosophical work, and it is also the condition Gettier exploits.

JTB is meant to rule out two failure modes: false beliefs (via truth) and accidentally true beliefs such as guesses (via justification). Gettier is powerful because it produces a third failure mode the definition did not anticipate, a belief that is justified and true yet still only accidentally connected to the truth.

Gettier cases

A second Gettier-style case is the stopped clock: you look at a clock that reads 2 o'clock and form the justified true belief that it is 2 o'clock, but the clock stopped exactly twelve hours ago. True, justified, believed, yet not knowledge.

Every Gettier case shares one structure: a subject reasons competently from good evidence, the belief turns out true, but it is true for a reason unconnected to the evidence that justified it. Two ideas explain the failure. First, justification is fallible: good evidence can justify a false belief (Smith's evidence about Jones was excellent but the proposition he inferred through was false). Second, truth can arrive by luck that bypasses the justification entirely. The lesson is that knowledge requires the truth to be non-accidentally related to how the belief was formed, and JTB does not capture that anti-luck condition.

Responses to Gettier

• Infallibilism (and the J+T+B+ no-luck idea). Strengthen justification so it guarantees truth: only certain, indubitable beliefs count. This avoids Gettier luck but is too strong, ruling out most ordinary empirical knowledge.
• No false lemmas (Clark). Add a fourth condition: the justified true belief must not be inferred from any false belief. Smith's belief rests on the false lemma that Jones will get the job, so it fails. Objection: some Gettier cases (for example fake barns) involve no false lemma.
• Reliabilism. Replace justification with: the belief is produced by a reliable cognitive process. This handles some cases but faces the fake barn county, where a reliable process still yields only lucky knowledge, and worries about clairvoyance counterexamples.
• Virtue epistemology (Zagzebski, Sosa). Knowledge is true belief that is the product of intellectual virtue (the belief is "apt", true because of skill). This ties truth to the believer's competence rather than luck. Sosa's archery analogy is the standard illustration: a shot can be accurate (it hits), adroit (it is skilful) and apt (it hits because it is skilful). A Gettier belief is accurate and adroit but not apt, because it is true by luck rather than competence; only apt belief is knowledge.

Across all four responses there is one dialectic the examiner rewards. Strengthen the conditions and you risk scepticism (infallibilism makes ordinary knowledge impossible); keep them realistic and you risk new Gettier cases (no false lemmas falls to fake barns). The fake barn case is the key test: in barn-facade country you look at the one real barn and form a reliable, no-false-lemma belief that it is a barn, yet you could so easily have faced a facade that the belief is too lucky to be knowledge. A strong evaluation tracks each response against this case.