What does it take to know something, and why are truth, belief and justification each thought to be necessary?
Explain the three conditions commonly held to be necessary for propositional knowledge - truth, belief and justification - and assess whether each is genuinely required
A focused answer on the three classic conditions for knowing that something is the case: truth, belief and justification. What each condition adds, why theorists treat each as necessary, and how to argue about them in an essay.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
Jump to a section
What this dot point is asking
SEAB wants you to be able to explain, and then critically assess, the three conditions that have traditionally been treated as necessary for knowing that something is the case: that the proposition is true, that the knower believes it, and that the belief is justified. The deeper task is to ask of each condition, "Could you have knowledge without this?" and to argue your answer. This is the foundation for the whole nature-of-knowledge area, including the Gettier problem.
The answer
Propositional knowledge
The subject focuses on propositional knowledge, or knowing that something is the case (knowing that water boils at 100 degrees Celsius at sea level), as opposed to knowing how to do something (riding a bicycle) or knowing a person or place by acquaintance. Propositional knowledge takes the form "S knows that p," where S is a subject and p is a proposition that can be true or false.
The truth condition
Knowledge is factive: you cannot know something false. If Maya knows that the meeting is on Tuesday, then the meeting is on Tuesday. Should it turn out to be on Wednesday, we withdraw the claim and say she only thought she knew. Truth is therefore a condition on the world, not on the knower. It is what makes knowledge worth having: it puts us in contact with how things actually are.
The belief condition
To know that p you must also believe that p. Belief is the mental attitude of taking a proposition to be true. The point of the condition is that knowledge is a state of a mind; a fact sitting in a textbook that nobody accepts is not anybody's knowledge. A standard test case: a nervous student who has revised thoroughly but is convinced she will fail, yet answers every question correctly. We hesitate to say she knew the answers, because she did not believe them. This is contested, but most accounts keep belief as a condition.
The justification condition
True belief is still not enough, because a true belief can be a matter of luck. Suppose a gambler bets the house on a horse purely on a hunch, and it happens to win. The belief was true, and it was believed, but it was not knowledge, because there was no good reason behind it. Justification is the condition that connects the believer to the truth in the right way: evidence, sound reasoning, reliable perception, or trustworthy testimony. It is what turns a fortunate guess into knowledge.
Internalism and externalism about justification
Theorists divide over what justification is. Internalists hold that the justifying factors must be accessible to the knower from the inside, reasons she could in principle cite. Externalists, especially reliabilists, hold that what matters is that the belief was produced by a reliable, truth-conducive process, whether or not the knower can articulate it. The disagreement matters because it changes which beliefs count as knowledge: a young child or an expert who "just knows" may pass the externalist test while failing the internalist one.
Examples in context
Example 1. The stopped clock. You glance at a wall clock that reads three o'clock and form the belief that it is three o'clock, and it happens to be exactly three. But the clock stopped twelve hours ago. Your belief is true and you have what looks like a reason (you read a clock), yet most people deny this is knowledge, because the truth and your reason came apart by luck. The case shows how much weight the justification condition carries, and previews the Gettier problem.
Example 2. The confident exam candidate. A student has thoroughly revised and answers correctly, but feels certain she has failed and so does not believe her answers are right. She produces true, well-supported responses without the corresponding belief. The case is used to test the belief condition: many conclude she does not yet count as knowing, which supports keeping belief in the analysis.
Try this
Q1. State the three conditions of the standard analysis of knowledge and explain what each one rules out. [6 marks]
- Cue. Truth rules out knowing falsehoods (factivity); belief rules out unaccepted facts; justification rules out lucky true beliefs (the hunch, the stopped clock).
Q2. Explain, with an example, why true belief is not sufficient for knowledge. [8 marks]
- Cue. Use a lucky-guess case (a hunch bet that wins). The belief is true and held, but with no good reason, so it is not knowledge; justification is the missing third condition.
Q3. Briefly distinguish internalist and externalist accounts of justification and say why the difference matters. [6 marks]
- Cue. Internalism requires reasons the knower can access; externalism (reliabilism) requires a reliable process. It matters because experts and children who cannot state reasons may count as knowing on the externalist view but not the internalist one.
Exam-style practice questions
Practice questions written in the style of SEAB exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
Original20 marksIs justification necessary for knowledge? Discuss with reference to the standard analysis of knowledge.Show worked answer →
A strong answer sets out the standard analysis first: S knows that p if and only if p is true, S believes that p, and S is justified in believing that p. The question targets the third condition.
Argue for necessity: a true belief that is a lucky guess (the gambler who simply feels a horse will win, and it does) is not knowledge, which shows truth and belief alone are not enough. Justification is what rules out luck and connects the believer to the truth for the right reasons.
Then test it. Reliabilists object that what matters is not a reason the knower can state but that the belief was produced by a reliable process; on this view a chicken-sexer or a child may know without being able to justify. So perhaps justification, understood as accessible reasons, is not strictly necessary, even if some truth-conducive condition is.
Judgement: defend that some third condition beyond true belief is necessary to exclude luck, while conceding the dispute is about its form (internalist reasons versus externalist reliability). Markers reward a precise statement of the analysis, a clear lucky-guess case, engagement with reliabilism, and a decided, defended conclusion rather than a survey.
Original20 marksCan a false belief ever count as knowledge? Defend your view.Show worked answer →
The expected answer defends the truth condition: knowledge is factive, so one cannot know what is false. Give the test: we readily say "she thought she knew, but she was wrong," which shows ordinary usage withdraws the word "know" once the belief turns out false.
Develop the point that this is a feature of the concept, not a high standard. To claim to know p is to commit to p being the case; if p is false there is no fact to have grasped, only a mistaken belief.
Consider the strongest opposition: in everyday speech people say "everyone knew the earth was flat." Reply that this is a loose, sociological use meaning "everyone was confident," not the strict sense the analysis targets; we do not really credit those people with knowledge once we judge the belief false.
Judgement: false belief cannot be knowledge in the strict propositional sense, while acknowledging looser uses of the word. Markers reward the factivity point, a clear example, fair handling of the ordinary-language objection, and a firm conclusion.
Related dot points
- 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
A focused answer on Gettier's challenge to justified true belief. How Gettier cases work, why they show the three conditions are not jointly sufficient, and the leading repairs - no false lemmas, defeasibility, causal and reliabilist accounts.
- Assess perception as a source of knowledge, contrasting direct realism, indirect realism and idealism, and evaluating the arguments from illusion and theory-ladenness
A focused answer on whether perception delivers knowledge of the external world. Direct and indirect realism and idealism, the argument from illusion, the veil-of-perception problem, and the theory-ladenness of observation.
- Distinguish a priori from a posteriori knowledge and analytic from synthetic truths, and evaluate the rationalist and empiricist accounts of the sources of knowledge
A focused answer on reason as a source of knowledge. The a priori versus a posteriori distinction, analytic versus synthetic truths, rationalism versus empiricism, and the contested status of synthetic a priori knowledge.
- Assess testimony as a source of knowledge, contrasting reductionist and anti-reductionist accounts and considering memory as a further source
A focused answer on testimony as a source of knowledge. Why so much of what we know rests on the word of others, the reductionist versus anti-reductionist debate over its justification, and memory as a preservative source.