Episteme

The problem it answers

What does it take to know something, as opposed to merely believing it correctly? You might believe correctly that it is raining (by chance, or by guessing, or because someone told you). The Greek philosophical tradition insists that this kind of correct belief is not yet knowledge. Knowledge requires something more: a grasp of why the belief is true, with reasons that ground it, organized into a structure that could not easily be undone. Episteme (Greek epistēmē) is the term the tradition uses for this stronger cognitive state.

The distinction between knowledge and mere belief is foundational to Western epistemology. What is knowledge? is one of the few questions philosophers have continued to ask in essentially Platonic terms across 2,400 years.

The core claim

The core ancient claim about episteme has three parts.

Episteme is more than true belief. A correct belief that is not properly grounded does not constitute knowledge. The famous Meno problem (a slave boy guesses the correct geometric construction without understanding why it is correct) makes the point sharply: he has true belief but not episteme.

Episteme requires the grasp of reasons (logos). What converts true belief into episteme is the ability to give an account — to explain why what is known is so, to derive it from more basic principles, to defend it against objections. The logos requirement is the engine of the Socratic elenchus.

Episteme is stable. Mere true belief is unstable; it can be shaken by any clever objection or change in mood. Episteme, because grounded in logos, is stable across challenges. This is what makes it knowledge in the strongest sense rather than merely lucky cognitive contact with truth.

History in one paragraph

Socrates in the Meno (97a–98a) develops the foundational distinction between knowledge and true belief, arguing that what converts the latter into the former is being tied down by reasoning about the cause. Plato in the Republic and the Theaetetus develops the contrast between episteme (knowledge of the eternal Forms) and doxa (opinion about the changing visible world); the Theaetetus canvasses three definitions of knowledge (perception, true belief, true belief with an account) and rejects each, ending in characteristic Platonic aporia. Aristotle in the Posterior Analytics gives the canonical ancient treatment: episteme is demonstrative knowledge derived from first principles that are themselves known by nous (intuitive grasp). The Hellenistic Skeptics (Pyrrho, Sextus Empiricus) attacked the very possibility of episteme; the Stoics defended it through the doctrine of the cataleptic impression. The medieval Scholastic tradition (especially Aquinas in his commentary on the Posterior Analytics) integrated the Aristotelian account with Christian theology. Early modern epistemology shifted the central question: Descartes in the Meditations sought indubitable foundations; Locke and the empiricists asked how knowledge can be derived from sense experience; Hume produced a devastating skepticism about most putative knowledge. Kant reframed the question entirely in the Critique of Pure Reason: knowledge is not the mind's conformity to objects but objects' conformity to the structures of the mind; this allowed for a category of synthetic a priori knowledge that the empiricists had denied. Contemporary epistemology has been organized since Edmund Gettier's 1963 paper, which presented counter-examples to the traditional justified true belief analysis of knowledge and launched a half-century of efforts to refine the account.

The traditional analysis: justified true belief

The standard Western analysis, traceable to Plato's Theaetetus, is that knowledge is justified true belief — a belief that is in fact true and that the holder is justified in holding. The three components:

Belief. You cannot know what you do not believe. (One important caveat: some philosophers distinguish knowing-how from knowing-that, with the former not requiring belief in the standard sense.)

Truth. You cannot know what is false. A false belief, however well-justified, is not knowledge.

Justification. The belief must be properly held. The exact nature of this requirement is the most contested element. Justification might be evidentialist (you have appropriate evidence), reliabilist (your belief-forming process is reliable), foundationalist (the belief rests on basic beliefs), coherentist (the belief fits with the rest of your beliefs), or virtue-theoretic (the belief expresses appropriate intellectual virtues).

The Gettier problem

In 1963 Edmund Gettier published a three-page paper presenting counter-examples to the traditional analysis. The structure of the counter-examples: cases in which someone has a justified true belief that is nonetheless intuitively not knowledge, because the justification connects to the truth only by coincidence.

Classic example: Smith and Jones have applied for the same job. Smith has strong evidence that Jones will get the job and Jones has ten coins in his pocket. From this Smith infers that the person who will get the job has ten coins in his pocket. As it happens, Smith himself will get the job, and Smith happens to have ten coins in his pocket. So Smith's belief that the person who will get the job has ten coins in his pocket is true, and his belief was inferred from properly held premises (hence justified). But intuitively he does not know it — the truth of his belief is coincidental.

The Gettier problem launched a half-century of efforts to repair the analysis. The major responses include adding a fourth condition (no false premises; no defeaters; tracking of truth; safety; sensitivity), abandoning the analytic project altogether (Williamson's Knowledge and Its Limits, 2000, treats knowledge as unanalyzable and primitive), and shifting to virtue epistemology (Sosa, Greco, Zagzebski).

Knowledge of the Forms

In the Platonic tradition, episteme is knowledge specifically of the eternal Forms, as distinct from opinion (doxa) about the changing visible world. The famous Divided Line in Republic VI puts this hierarchically: at the lowest level, eikasia (imagination, the cognitive state of seeing shadows and reflections); above it, pistis (belief about visible particulars); above it, dianoia (mathematical reasoning, which depends on hypotheses); at the top, noesis (the direct intuitive grasp of the Forms). Episteme in the strict sense names the upper two stages.

This Platonic restriction of knowledge to the eternal has been contested throughout the tradition. Aristotle disagreed: there is genuine episteme of the natural world, where natural kinds and causes provide the necessary stability. The modern scientific tradition, in deepening the empirical study of nature, has implicitly sided with Aristotle against Plato on this point.

Common confusions

Episteme is not the same as the modern science. The English word science derives from Latin scientia (the Latin translation of episteme) but has narrowed in modern usage to mean the natural and social sciences. Ancient episteme covers any systematic, demonstrable knowledge, including ethics, mathematics, and theology.

Episteme is not the same as certainty. A claim can be known even if it could in principle be wrong; what episteme requires is appropriate grounding, not infallibility. (Descartes's program of seeking indubitable foundations is one specific epistemological strategy, not the meaning of knowledge as such.)

Episteme is not the same as expertise. Modern English expertise implies practical mastery of a domain; ancient episteme primarily names theoretical understanding. The Aristotelian framework distinguishes episteme (theoretical knowledge) from technē (productive expertise) and phronesis (practical wisdom).

Live debates

Knowledge-first epistemology. Timothy Williamson's Knowledge and Its Limits (2000) argues that knowledge is conceptually primitive and cannot be analyzed in terms of belief and justification; rather, belief should be understood in terms of knowledge. The position has reshaped a substantial part of contemporary epistemology.

Virtue epistemology. A family of positions (Linda Zagzebski, John Greco, Ernest Sosa) that ground epistemic evaluation in the character traits of knowers rather than in properties of beliefs. The intellectual virtues parallel the moral virtues; knowledge is appropriately understood as cognitive success arising from such virtues.

Social epistemology. The recognition that most of what we know depends on testimony, division of cognitive labor, and the credibility-attribution practices of communities. The work of Alvin Goldman, Miranda Fricker (Epistemic Injustice, 2007), and Sanford Goldberg has made social epistemology a major subfield.

The relation between knowledge and understanding. Recent work (especially Catherine Elgin, True Enough, 2017) argues that understanding is a distinct cognitive achievement irreducible to knowledge, and that scientific theories often provide understanding without strictly amounting to knowledge.

Contemporary engagement

Epistemology is one of the largest subfields of contemporary analytic philosophy. The standard reference works include the Stanford Encyclopedia of Philosophy (which has dozens of relevant entries), the Routledge Companion to Epistemology, and the Oxford Handbook of Epistemology. The major journals (especially Philosophical Studies, Philosophy and Phenomenological Research, Mind, Synthese, Episteme) publish extensive ongoing work. Recent influential monographs include Williamson's Knowledge and Its Limits, Duncan Pritchard's Epistemic Luck (2005), and Miranda Fricker's Epistemic Injustice. The Gettier problem and its successors remain among the most-discussed problems in the analytic tradition.

Further reading

• Plato — the Theaetetus and the Meno on the analysis of knowledge
• Aristotle — the Posterior Analytics on demonstrative knowledge
• Form — the proper object of episteme in the Platonic tradition
• Hume — the problem of induction and the limits of empirical knowledge
• Kant — the Critical reframing of the conditions of knowledge
• Dialectic — the method by which episteme is acquired in Platonic philosophy

This is a Pillar concept. Satellite concepts should link here as their parent.