What Is Knowledge in Philosophy?

First specify what kind of knowing you mean

This page concerns propositional knowledge: knowing that something is the case. Examples include knowing that a meeting starts at noon, that a particular key opens a door, or that a mathematical conclusion follows from stated premises. A proposition is the content that can be true or false. Asking whether someone knows it requires a claim precise enough to evaluate.

Other uses of ‘know’ raise related but different questions. Knowing a neighbour is a form of acquaintance; knowing how to ride a bicycle concerns an ability. Someone might explain the mechanics of balance without being able to ride. Philosophers disagree about how these kinds of knowledge are related, so the analysis of knowing that should not silently become a theory of every use of the word.

Further reading: Internet Encyclopedia of Philosophy · Stanford Encyclopedia of Philosophy

Truth separates knowledge from a reasonable mistake

Suppose Leah reads a familiar clock that shows 3:00 p.m. She has no reason to suspect a fault, but the clock is ten minutes slow. Her belief that it is 3:00 is false. On the standard understanding of propositional knowledge, she does not know that it is 3:00, even if consulting the clock was reasonable. This is the factive feature of knowledge: if someone knows that p, p is true.

The example keeps two evaluations apart. We can ask whether Leah used her available information responsibly and whether she got the time right. A sympathetic judgement about the first does not settle the second. Nor does later discovering the error automatically show that she was careless at the time. Truth concerns the claim; justification concerns the basis on which the person accepts it.

Further reading: Stanford Encyclopedia of Philosophy · Internet Encyclopedia of Philosophy

Belief is not another word for unsupported opinion

In philosophical usage, a belief is an attitude of taking something to be true. A belief can be carefully supported, casually inherited, correct or mistaken. Calling a view a belief does not by itself criticize it. On most familiar accounts, knowing that the door is locked includes believing that it is locked. The two categories therefore overlap; knowledge and belief are not opposing teams.

Confidence is a further matter. A person can feel completely certain about a false claim, while a cautious person can know something without advertising certainty. Philosophers also distinguish outright belief from degrees of confidence and debate whether knowledge always entails belief. For this introductory discussion, the ordinary belief requirement is the working assumption, with that debate left visible rather than presented as closed.

Further reading: Stanford Encyclopedia of Philosophy · Stanford Encyclopedia of Philosophy

A good basis matters—but what counts as one?

Compare two correct answers to a timetable question. One person consults the current official timetable, checks the station and date, and sees that the train leaves at eight. Another picks eight at random. Both may arrive at a true belief, but they reach it differently. The first has a relevant basis; the second has only a coincidence. This is why truth plus belief looks too weak as an account of knowledge.

A justification requirement is meant to capture that difference. Its details are disputed. Some accounts emphasize reasons available from the believer’s perspective. Others emphasize features such as whether the process producing the belief is reliable. These approaches can disagree when a person has apparently good reasons but an unknown defect in the situation. They should not be compressed into a checklist that treats evidence, reliability and competence as interchangeable.

Further reading: Stanford Encyclopedia of Philosophy · Internet Encyclopedia of Philosophy

Why a stopped clock causes trouble

Sam consults a clock that has always been accurate and is normally well maintained. Unknown to Sam, it stopped at 3:00 earlier. Sam happens to look at exactly 3:00 p.m. and believes that it is 3:00. This time the belief is true. Sam also appears to have a good reason to trust the clock, judged from the information available. Yet many people hesitate to call this knowledge: the clock would have given the same answer at the wrong time.

The case challenges the sufficiency of justified true belief, often abbreviated JTB. It asks whether a person can meet those three conditions while still being right through the wrong sort of luck. The stopped-clock example predates Edmund Gettier’s 1963 paper; it is not one of his two original examples. It is useful for introducing the wider problem his paper made central to the analysis of knowledge.

A counterexample has a limited logical job. If Sam meets the proposed conditions but lacks knowledge, those conditions are not sufficient. That would not show that truth is irrelevant, that every belief is a guess, or that nobody knows anything. Also, an account that evaluates justification through reliability might reject the claim that Sam is justified. Explaining that disagreement is more informative than assuming every theory must classify the case in exactly the same way.

Further reading: Stanford Encyclopedia of Philosophy · Internet Encyclopedia of Philosophy · OpenStax

Several responses, with different costs

One response asks whether a belief is safe: roughly, whether it could easily have been false when formed in the relevant way. Another asks whether the belief-forming process is reliable. Virtue approaches connect true belief with intellectual competence or ability. These ideas each attempt to explain why getting the answer right is insufficient when the success is badly disconnected from how the person arrived at it.

They are not simply alternative names for one agreed extra condition. A generally reliable method can face a misleading local environment. A person can exercise an ability in a setting where getting the truth still seems fortunate. Philosophers construct further cases to test which features matter and how the conditions should be stated. The lesson is to identify the proposed explanation and its difficult cases, rather than memorize a formula as the final answer.

Knowledge-first approaches reverse the usual direction of explanation. Instead of defining knowledge by combining independently understood belief, truth and justification, they treat knowledge as a basic starting point for explaining other epistemic notions. This does not remove the need to examine examples. It changes what a successful theory is trying to provide: knowledge may illuminate evidence or justification without first being reduced to a short list of simpler ingredients.

Further reading: Stanford Encyclopedia of Philosophy

Does knowledge require absolute certainty?

Many fallibilist views allow knowledge even though a person cannot rule out every conceivable possibility of error. In an ordinary setting, a properly functioning clock may let Maya know the time without proving that she is not dreaming. A skeptical argument presses the demand for stronger grounds. These are disputes about the standards for knowledge, not merely observations about how confident Maya feels.

Avoid two shortcuts. ‘I cannot prove error impossible’ does not automatically establish ‘I know nothing.’ Conversely, ‘this is what we normally accept’ does not answer every skeptical challenge. State the claim, the available grounds and the alternative possibility under discussion. A specific question about a faulty clock can then be distinguished from a much broader question about whether perception could ever justify belief.

Further reading: Stanford Encyclopedia of Philosophy · Internet Encyclopedia of Philosophy

Knowing through other people

We do not personally repeat every measurement or witness every event we know about. Much learning depends on testimony: another person tells us something, a document records it, or a specialist explains it. Dependence on others is therefore not automatically a defect. The philosophical question is when that dependence can put us in a position to know, and what happens when the source or the chain of transmission goes wrong.

For a concrete example, a museum guide gives the date of an object using its catalogue record. Ask what the exact claim is, whether the record supports it, and whether the visitor has a reason to suspect a mistake. A bare statement that the guide is ‘an authority’ leaves those details out. This is an example of how to make the inquiry more precise, not a universal procedure that guarantees knowledge whenever a source looks respectable.

Further reading: Internet Encyclopedia of Philosophy · Stanford Encyclopedia of Philosophy

Use the distinction on one claim at a time

Begin by writing a complete proposition. ‘The timetable is useful’ is different from ‘the train leaves this station at eight today.’ Identify who believes it and the basis they actually used. Then separate facts supplied by the example from information available to that person. Sam’s clock is stopped as a fact of the story, even though Sam does not know this. Losing that distinction makes the stopped-clock problem disappear by accident.

Next ask what explains the correct answer. Was the person guided by the relevant facts, or did a defective method happen to coincide with them? A helpful revision to Sam’s case would provide an independently functioning time source and an appropriate basis for trusting it. Merely making Sam more certain leaves the defective clock untouched. This test helps expose the issue without pretending to settle every theory of knowledge.

For classroom discussion, compare all four clock cases in the linked worksheet: a working clock, a lucky guess, a confident mistake and a stopped clock. Each changes a particular feature. Explain which condition changes and why your verdict follows. If two readers disagree, ask whether they disagree about the described facts, the meaning of justification, or the conditions a theory should impose. Those are different disagreements and can require different arguments.

Further reading: Stanford Encyclopedia of Philosophy · Internet Encyclopedia of Philosophy