knowledge

← Encyclopedia

Knowledge is the condition of having grasped what is true — of holding something not merely as opinion or guess but as something secured. That much is widely shared. What divides the traditions that have thought hardest about it is what “secured” requires, and whether the highest kind of knowing is a matter of the mind at all. The English word carries all of this under one roof; the traditions that built their lives on the distinction did not, and the most interesting history of the concept is the history of the lines they drew that English erases.

Marble bust of Plato (Musei Capitolini, MC1377), a Roman copy after a Greek original attributed to Silanion; the Theaetetus and Meno set the terms of the Western inquiry into knowledge. — Marie-Lan Nguyen, via Wikimedia Commons (public domain)

The first inquiry: the Theaetetus

The Western philosophical inquiry into knowledge takes its terms from Greek. The Theaetetus of Plato, the first sustained attempt in the European record to ask what knowledge is, sets the pattern for everything after it by refusing to answer too quickly. Set in Megara and read aloud from a written record, the dialogue has Socrates press the young geometer Theaetetus through three candidate definitions and dismantle each. Knowledge is perception: rejected, because perception is private and shifting and cannot reach the stable relations — sameness, difference, number, being — that the mind grasps without any organ of sense. Knowledge is true belief: rejected, because a jury can be talked into a true verdict by a skilled speaker and believe truly without knowing, having been persuaded rather than brought to see. Knowledge is true belief with an account (logos): rejected after the account itself dissolves under examination, each sense of “account” either adding nothing or quietly presupposing the knowledge it was meant to supply. The dialogue ends without a settled definition. That is itself a kind of result — a demonstration that the obvious answers fail, which is the necessary clearing before any harder one. Socrates frames the whole exercise as midwifery: he claims to know nothing himself and only to deliver the thoughts of others, testing each newborn to see whether it is a true child or a wind-egg. The temper of the search is set early, when Socrates tells the boy that his perplexity is the mark of a philosophical nature — that wonder is the feeling of a philosopher, and philosophy begins in wonder. The first European treatise on knowledge thus opens onto something closer to a discipline of not-yet-knowing than to a doctrine.

From the Theaetetus discussion descended the long-standing formula that knowledge is justified true belief: a claim one holds, that happens to be true, and that one holds for good reason. For most of two millennia the formula looked secure enough to teach. Then, in a paper of three pages — Edmund Gettier’s Is Justified True Belief Knowledge?, published in Analysis in 1963 — it came apart. Gettier constructed cases in which a person holds a belief that is true, and holds it on good grounds, yet plainly does not know, because the belief is true by an accident that the justification fails to track: the reasoning is sound, the conclusion correct, and the link between them severed by luck. The half-century since has been spent patching the formula — adding a ban on false premises, a causal tie between fact and belief, a clause against defeating evidence, a demand that the belief-forming process be reliable — and finding that each patch admits a new counterexample. Timothy Williamson’s Knowledge and Its Limits (Oxford University Press, 2000) drew the radical conclusion: perhaps knowledge cannot be built out of belief plus extra conditions at all, because the project has the order of explanation backward. On his account knowledge is not a compound to be analyzed but a primitive — a mental state in its own right, the most general factive state, with belief the fainter thing that knowledge degrades into when its grip on the world fails. The questions Plato raised are, recognizably, the questions modern epistemology argues over, and the most influential recent move is to suspect that they resisted answer because they asked knowledge to be made of something simpler than itself.

Epistēmē and gnosis: the line the word blurs

Greek drew a line the English word blurs. Epistēmē — from the verb for standing over or being master of something — named demonstrable, teachable knowledge: what the geometer has of the triangle, what the physician has of the body, knowledge that can be set out in order, defended with reasons, and handed on intact to a student. Gnosis came to name a knowing by acquaintance, a direct contact with its object rather than a conclusion reached about it — the difference between knowing a proof and knowing a face. The opposition is partly a later hardening; in classical prose the two words overlap, and Aristotle’s catalog of intellectual virtues — craft, epistēmē, practical wisdom, philosophic wisdom, and nous, the intuitive grasp of first principles — gives gnosis no technical place at all. The reification of gnosis into a distinct, salvific mode of knowing is a later, Hellenistic-religious development. But the distinction the two words eventually carried is real and consequential, because the religious traditions reach almost always for the second. What the mystic claims to know is not a proposition that could be written down and checked, but a presence met. The contemplative term sophia and the faculty-term nous belong to the same Greek vocabulary, and the salvific career of gnosis — the long story of how a word for plain recognition became the name of a knowledge that saves — is a study in its own right, traced where the saving sense rather than the general concept is at issue.

Plato himself gestured past argument toward such a knowing. In the Meno, confronted with the paradox that one cannot inquire into what one does not know — for one would not recognize the answer if one stumbled on it — Socrates answers that learning is really recollection. Drawing on a teaching he ascribes to priests and inspired poets, that the soul is deathless and has been born many times and has beheld all things, he has Socrates lead an untutored slave-boy, by questions alone, to construct the diagonal of a square he was never taught to find — eliciting the geometry rather than imparting it, on the argument that all inquiry and all learning is but recollection. Knowledge here is not poured into an empty vessel but recovered: something the soul once possessed, lost in the descent into a body, and uncovers again under the right questioning. The model is decisive for the later tradition. It makes knowing fundamentally an act of return rather than acquisition, and it ties the possibility of knowledge to the prior dignity of the knower — a structure that Neoplatonism inherits and that resurfaces, transposed, wherever a tradition holds that the deepest truth is not learned from outside but awakened from within.

The religious turn: knowing by acquaintance

The religious traditions sharpen the same move in their own vocabularies, and each sharpening is exact, fixed by a metaphysics that is not interchangeable with the others. They are best read as a family of distinctions between a knowledge that can be transmitted and a knowledge that can only be undergone — not as variants of one experience.

In classical Sufism the line runs between ʿilm, the knowledge that can be taught, transmitted, memorized, and argued — the possession of the jurist and the theologian — and maʿrifa, the intimate recognition of God that descends as a gift and cannot be manufactured by study. The earliest Sufi treatise in Persian, the Kashf al-Mahjub of al-Hujwiri (eleventh century, in Reynold Nicholson’s 1911 translation), opens its first chapter on the affirmation of knowledge by distinguishing knowledge that is divine from knowledge that is human, the latter worthless beside the former because God’s knowing is an attribute of an infinite Himself while ours is finite. The treatise is careful: it insists that knowledge severed from action is not knowledge, and that the theologians make no distinction between ʿilm and maʿrifa where the Sufi masters reserve the second word for a knowing bound up with religious practice and inward state. The placement of maʿrifa in the map of the path — whether it is a station the wayfarer climbs to or a state that descends unbidden — was itself disputed among the masters, which is doctrinally telling: the term sits exactly on the seam between earned discipline and given grace. Crucially, this is knowledge of an irreducibly personal Real who discloses himself; the cosmos is not a prison to escape but a theater of self-disclosure.

Advaita Vedānta set jñāna, the liberating knowledge of the self’s identity with the absolute, above all information about the world. But its structure inverts the Sufi one at the root. What is realized is not another to be known but the very Self of the knower; the formula tat tvam asi — that thou art — names not a relation between two parties but the collapse of the distance between them. And the decisive realization is, strictly, not an additional cognition but an unveiling: the removal of a constitutive misperception that had concealed what was always already the case. Liberation cannot be produced, the Advaitin reasons, because any product is impermanent; knowledge here does not add content but dissolves the ignorance that masked an eternal fact. The structural distinction Advaita draws — between mediate knowing, that the absolute is the self, taught by scripture or inferred, and immediate knowing, the direct non-objectifying realization — is its own version of the epistēmē/gnosis cut, and only the second liberates.

The Hermetic and Gnostic writings of late antiquity press the move to its limit: there knowing becomes a thing that saves, and ignorance the deepest affliction a person can suffer. The Corpus Hermeticum stages knowledge as a revelation granted rather than a conclusion reached — in its first treatise the mind of sovereignty unveils to Hermes the descent of light into darkness and the soul’s path of return, and the goal of those who attain gnosis theou is named as becoming divine. The Christian Gnostic systems made the salvific charge sharper still: in the Valentinian summary preserved by their opponents, what saves is the knowledge of who we were and what we have become, whence we came and whither we hasten — an awakening from forgetfulness rather than the learning of a fact. Here the metaphysics again diverges: the Hermetic cosmos is beautiful and providentially ordered, the Gnostic cosmos a flawed work to be seen through, and the knowing each prizes is colored accordingly. To call all of these “knowledge of God” is to let one English phrase cover a personal Real disclosing himself, a non-different ground unveiled, and a divine origin remembered — four structural divergences, at least, hiding under three words.

Learned ignorance and the apophatic limit

A distinct strand runs counter to all of these without leaving the question of knowledge behind: the claim that the highest knowing is a disciplined recognition of what cannot be known. This is the territory of apophatic theology, where every name applied to the divine is in the same breath denied as inadequate, and it produces the sharpest paradox in the whole field. Nicholas of Cusa, the fifteenth-century cardinal and mathematician, gave it a name in his De Docta Ignorantia of 1440 — docta ignorantia, learned ignorance — and an argument: that between the finite and the infinite there is no proportion, so that the more truly the mind grasps God the more exactly it grasps that God exceeds its grasp. Reason reaches its summit not in a captured definition but in a precise understanding of why no definition can capture. Cusanus pressed the point with mathematics, watching the angles of a polygon inscribed in a circle multiply toward a curve they approach forever and never reach, a figure for finite reason nearing the infinite. The state he describes is not the absence of knowledge but its most demanding form — knowledge turned back on its own limit and made luminous there. It marks a fourth thing the word can mean: not a proposition secured, not an acquaintance granted, not an identity unveiled, but a measured seeing of the boundary itself.

These currents touch the much later, broader argument over whether religious knowing is even the kind of thing reason can adjudicate — the territory of revelation as a source set beside or above discursive reason, and of the negative claim that the divine lies past the reach of evidence altogether. The relevant point for the concept of knowledge is narrower: in the apophatic line, ignorance ceases to be the mere privation of knowledge and becomes one of its modes.

Scholarship and texts

The primary record for the philosophical inquiry is itself the literature. Plato’s Theaetetus and Meno, read in Benjamin Jowett’s nineteenth-century English, set the terms of the European discussion and remain the indispensable starting points; both are hosted here in full. The modern reopening of the question turns on two short texts that anyone can read against each other: Edmund Gettier’s Is Justified True Belief Knowledge? (Analysis 23.6, 1963), three pages that unsettled a two-thousand-year formula, and the program that grew from the failure of the repairs, set out at length in Timothy Williamson’s Knowledge and Its Limits (Oxford University Press, 2000), which argues that knowledge resists analysis because it is more basic than the belief it was supposed to be built from. The standard map of the analysis and its breakdown — no-false-lemmas, causal, defeasibility, and reliabilist repairs, each defeated in turn — is the connective tissue between them.

For the religious vocabularies, the public-domain apparatus is unusually rich. Nicholson’s 1911 translation of al-Hujwiri’s Kashf al-Mahjub gives the earliest Persian Sufi treatise in English, with its opening chapter on the affirmation of knowledge and its later chapter on the gnosis of God — maʿrifat Allah — laying out the ʿilm/maʿrifa distinction at first hand; the relevant chapter is hosted here. The comparative question of whether gnosis, maʿrifa, and jñāna name one reality or three — a question the modern study of mysticism has often answered too fast — is treated where the salvific sense of knowing is the subject, in the entry on gnosis; the disciplined conclusion of recent scholarship is that the terms form a family of resemblances under conditions of partial historical contact, neither a single hidden doctrine nor wholly unrelated, and that what convergence exists is itself a fact to be explained rather than a premise to argue from.

From possessing to undergoing

Across these settings one structural shift recurs, and it is the deepest thing the word holds. In its plainest sense knowledge is something the knower has — a possession carried, demonstrable, transferable, the geometer’s theorem or the jurist’s ruling, intact whether or not it changes the one who holds it. In its demanding sense it is something the knower is — a state undergone, in which the knowing and the change it works are not separable, where to come to know is to have become someone who could not have known before. The geometer can prove a theorem on a morning when he is distracted, faithless, asleep to everything but the diagram, and lose nothing of the proof. The ʿarif cannot be granted maʿrifa and remain unaltered, the Advaitin cannot realize the non-difference of self and ground and still stand at a distance from it, the one to whom the Hermetic vision is given does not file it among other facts but is, by the tradition’s own account, remade. The line does not separate East from West or faith from reason; it runs through Greek itself, between epistēmē that one masters and the nous that masters the one who exercises it, and it is the same line Cusanus crosses when he turns reason’s defeat into reason’s highest act. To ask what knowledge is, in the end, is to ask which of these one means: the thing carried, or the thing that carries the one who knows it away.

→ In the library: Plato — Theaetetus (Jowett) · Plato — Meno (Jowett) · Al-Hujwiri — Kashf al-Mahjub, ch. I: On the Affirmation of Knowledge (Nicholson, 1911)

→ Related: Gnosis · Nous · Soul · Neoplatonism · Plato · Socrates · Gnosticism · Corpus Hermeticum · Mysticism · Sufism · Al Hujwiri · Reason · Revelation · Apophatic Theology · Nicholas Of Cusa · Logos · Sophia