← Back to Philosophy

What is knowledge? Belief, truth, and justification

We know many things, or think we do. Epistemology asks what knowledge is, what separates it from lucky guesses, and whether we have as much of it as we suppose.

Kinds of knowledge

Philosophers distinguish knowing that (propositional knowledge, such as knowing that Paris is in France), knowing how (practical skill, such as knowing how to ride a bicycle), and knowledge by acquaintance (being familiar with a person, place, or experience). Epistemology has focused mainly on propositional knowledge, though know-how has received growing attention.

The classical analysis

Since Plato's Theaetetus, knowledge has often been analysed as justified true belief. To know that P, you must believe P, P must be true, and you must have good reasons for believing it. Belief alone is not enough, since beliefs can be false; true belief alone is not enough, since a lucky guess can be true.

Gettier cases

In a famous three-page paper in 1963, Edmund Gettier offered counterexamples showing that justified true belief is not sufficient for knowledge. In one kind of case, someone has good evidence for a belief that happens to be true for reasons unrelated to their evidence. The belief is justified, true, and believed, yet it seems to be true only by luck, so not knowledge.

A Gettier-style example

You look at a clock that says 3 o'clock and form the belief that it is 3 o'clock. It is in fact 3 o'clock, but the clock stopped exactly twelve hours earlier. Your belief is true and justified, since clocks are usually reliable, yet most people judge you do not know the time, because you got it right by coincidence. This 'stopped clock' case predates Gettier but makes the same point.

Responses to Gettier

Philosophers proposed adding a fourth condition: no false beliefs in the reasoning, no defeating evidence that would undermine the justification, or a proper causal connection between the fact and the belief. Others proposed replacing justification with reliability or safety: a belief counts as knowledge if it could not easily have been false. No proposed analysis has won universal agreement.

Knowledge first

Timothy Williamson and others have suggested that knowledge cannot be broken down into simpler parts and should instead be treated as basic, with belief and evidence understood in terms of it. On this 'knowledge first' view, the long search for necessary and sufficient conditions was mistaken from the start.

A priori and a posteriori

A priori knowledge is justified independently of experience, such as knowing that 7 + 5 = 12 or that all bachelors are unmarried. A posteriori knowledge depends on experience, such as knowing that water boils at 100 degrees Celsius at sea level. Rationalists such as Descartes emphasised a priori reason; empiricists such as Locke and Hume emphasised experience.

Worked example

Ana sees her colleague Ben driving a red car every day and forms the belief 'someone in my office owns a red car'. In fact Ben borrows the car, but another colleague, unknown to Ana, does own a red car. Ana's belief is justified (good evidence), believed, and true, yet it is true only because of a fact she knows nothing about. This is a Gettier case: justified true belief without knowledge.

Key takeaways

Knowledge includes knowing that, knowing how, and knowledge by acquaintance; epistemology focuses on propositional knowledge. The classical analysis, from Plato's Theaetetus, is justified true belief. Gettier's 1963 counterexamples show justified true beliefs that are true by luck and so not knowledge, like the stopped clock. Responses add no-false-lemmas, no-defeater, causal, reliability, or safety conditions, none universally accepted. Knowledge-first approaches treat knowledge as basic. A priori knowledge is independent of experience; a posteriori knowledge depends on it, a divide between rationalists and empiricists.

Flashcard

Propositional knowledge

Knowing that something is so

Knowing that.

Check your understanding

What did Edmund Gettier's 1963 paper show?

Take the unit quiz

← Previous lesson Next lesson →