Philosophy › Socrates, Plato, Aristotle & the Axial Age › Plato
What can we know?
asked since c. 369 BC· Big question· Knowledge, logic & science
Mostly agreedGood evidence; some details uncertain.
The central question of epistemology: what knowledge is, how we get it, and whether anything can be known for certain.
See it on the Philosophy map →The idea
In the Theaetetus (c. 369 BC) Plato tested, without accepting, the idea that knowledge is true belief backed by an account, later summed up as 'justified true belief'. Being right is not enough: his example is a jury talked into the correct verdict by clever speeches, which believes truly but does not know. Rival answers followed. Skeptics doubted certainty is possible; rationalists trusted reason, empiricists experience; Kant argued the mind shapes what we can know. The main objection to the classic definition: Gettier's 1963 counterexamples.
Why it matters
Science, law and due diligence all depend on it: each needs a standard for when evidence is good enough to say we know.
Think of it like
Like a title search: a clean title isn't just what you believe; it needs a documented chain that holds up if challenged.
Where it comes from
Tracked through key texts: Plato's Theaetetus, Sextus Empiricus, Descartes' Meditations (1641), Hume, Kant's Critique of Pure Reason (1781), and Gettier's three-page paper (1963).
Still debated
There is still no agreed definition of knowledge after Gettier, and no consensus reply to the skeptic who says we might be dreaming.
Connections (10)
▸The Socratic method and "the unexamined life" probed What can we know?Mostly agreed
The connection
By asking 'What is courage?' or 'What is piety?' and refuting each answer with counterexamples, Socrates showed that confident people often cannot explain what they claim to know. His own wisdom, he said, was knowing that he did not know. That exposed the question underneath: what would count as real knowledge?
When and how
Athens, c. 430–399 BC, in conversations with generals, politicians and poets, as recorded in Plato's early dialogues such as the Laches, Euthyphro and Meno.
How sure are we?
Plato's early dialogues show the method in action, and Aristotle credits Socrates with seeking general definitions; how faithful the portrayal is remains uncertain.
Debates and open questions
Whether Socrates thought we need definitions before knowing anything, and how a method that refutes answers can ever produce positive knowledge.
▸The Allegory of the Cave illustrates What can we know?Mostly agreed
The connection
The Cave dramatizes a skeptical worry and Plato's answer. The prisoners are certain about the shadows yet know nothing real, so certainty is not knowledge. Real knowledge, Plato suggests, comes only by turning the mind from appearances toward unchanging truths, a slow and painful education rather than piling up more information.
When and how
Plato's Republic, around 375 BC; echoed later by Descartes' 'evil demon' (1641) and modern thought experiments such as the brain in a vat.
How sure are we?
Plato presents it as an image of 'our nature in its education and want of education'; later skeptical scenarios explicitly cite or echo it.
Debates and open questions
Whether Plato's answer helps at all: how could a freed prisoner know he has left the cave, and is not just in a bigger one?
▸Zhuangzi and the butterfly dream raised playful doubts about What can we know?Mostly agreed
The connection
Zhuangzi asked how we can tell dreaming from waking, and whether our standards are just one creature's view: people sicken sleeping in damp places, eels don't, so who knows the proper home? On a bridge he claimed to know the fish were happy; his friend Huizi asked how he could, not being a fish.
When and how
China, 4th–3rd century BC, in the Zhuangzi's chapter 2 (the butterfly dream and Wang Ni's questions) and chapter 17 (the fish of the Hao River).
How sure are we?
The passages survive in the text; comparisons with Greek skepticism and Descartes' dream argument are modern parallels, not historical influence.
Debates and open questions
Whether Zhuangzi denied that knowledge is possible or only that any single viewpoint is final; his humorous style makes his position hard to pin down.
▸Ancient Skepticism (Pyrrho to Sextus Empiricus) suspended judgement on What can we know?Mostly agreed
The connection
The Skeptics gave the most radical answer: perhaps nothing. Arguing both sides of every question, they showed that appearances conflict and that any proof needs a further proof, without end. Rather than claim nothing can be known, itself a claim, Pyrrhonists suspended judgment and lived by appearances and custom.
When and how
Greece and Rome, c. 300 BC – AD 200: Pyrrho, Arcesilaus, Carneades, Aenesidemus and Sextus Empiricus, whose Outlines of Pyrrhonism survives.
How sure are we?
Sextus's works set out the arguments in detail, including Agrippa's five modes, whose regress of justification is still discussed today.
Debates and open questions
Whether anyone can truly live without beliefs, and whether modern replies to the regress, such as foundationalism or coherentism, actually defeat it.
▸Nyaya: Indian logic and theory of knowledge listed valid sources for What can we know?Mostly agreed
The connection
Nyaya answered 'what can we know?' with a list of valid sources of knowledge (pramanas): perception, inference, comparison and reliable testimony. Any claim must trace back to one of them. That turned the question into a checklist, much as a lender asks which document supports each number.
When and how
India, in the Nyaya Sutras (roughly 2nd century AD) and Vatsyayana's commentary (c. 5th century).
How sure are we?
The opening sutras list the four pramanas; rival schools accepted, rejected or added to that list, showing it set the terms.
Debates and open questions
Whether comparison and testimony are truly separate sources or reduce to inference was argued by Buddhist and Vaisheshika critics.
▸"I think, therefore I am" (the cogito) offered a doubt-proof foundation for What can we know?Mostly agreed
The connection
Plato asked what knowledge is and how it differs from belief. Descartes answered the skeptics with one certainty, the cogito, then tried to build all knowledge on it, like a building on bedrock. This 'foundationalism' set the agenda of epistemology, the theory of knowledge, for centuries.
When and how
Descartes' Discourse (1637) and Meditations (1641); Plato's Theaetetus (c. 369 BC) had framed the older question.
How sure are we?
Epistemology textbooks treat Cartesian foundationalism as a central answer, and the empiricists, Kant and 20th-century philosophers all responded to it explicitly.
Debates and open questions
Many now doubt that knowledge needs certain foundations at all; coherence and reliability theories offer rival pictures.
▸Critique of Pure Reason redrew the limits of What can we know?Mostly agreed
The connection
Kant asked, in his own words, 'What can I know?' His answer split knowledge in two: we can know the structure of all possible experience with certainty, because the mind supplies it, but never things in themselves, nor God, the soul or freedom by pure reason. 'I had to deny knowledge to make room for faith.'
When and how
Critique of Pure Reason, 1781 and 1787; the three questions 'What can I know? What ought I to do? What may I hope?' appear near its end.
How sure are we?
The text states the questions and limits explicitly; later thinkers, from Hegel to the logical positivists, framed their views for or against those limits.
Debates and open questions
Whether things in themselves are separate objects or the same objects viewed another way, and whether modern physics undermines his account of space and time.
▸The Gettier problem refuted the classic answer to What can we know?Well documented
The connection
Many philosophers had taken knowledge to be justified true belief, a definition often traced to Plato's Theaetetus. Gettier's cases showed all three conditions can be met while a belief is right only by luck, so the three are not enough. The question of what knowledge is reopened.
When and how
Wayne State University, Detroit, 1963; Gettier targeted versions of the definition in A. J. Ayer (1956) and Roderick Chisholm (1957).
How we know
Nearly all philosophers accept the counterexamples; a 2017 study found the intuition across many cultures, though an earlier 2001 study had claimed differences.
Debates and open questions
Historians question whether anyone, even Plato, actually held the 'classic' definition in the neat form Gettier attacked.
▸Postmodernism and post-structuralism (Foucault, Derrida) questioned objective answers to What can we know?Debated
The connection
Postmodern thinkers argued that what counts as knowledge depends on language, institutions and power, not a neutral view from nowhere. Foucault studied how expert fields like psychiatry decide what is true; Lyotard doubted any grand story could justify all knowledge. Critics say this slides into relativism, where no claim is better supported than another.
When and how
France, then American universities, from 1966; Lyotard's The Postmodern Condition: A Report on Knowledge (1979) was commissioned by Quebec's universities council.
How sure are we?
The texts make these claims, but how radical they are is the dispute: Foucault and Derrida both denied being relativists, so readings of this link vary widely.
Debates and open questions
Whether postmodernism undermines objective truth or only exposes bias in how it is pursued, and whether it fed 'post-truth' politics, are fiercely debated.
▸Nick Bostrom: superintelligence and the simulation argument gave a new skeptical twist to What can we know?Mostly agreed
The connection
Bostrom's simulation argument updated the old dream and demon worries with probability. If advanced civilizations run many detailed simulations of their ancestors, simulated minds would vastly outnumber real ones, so we should think it likely we are simulated, unless such civilizations rarely arise or rarely bother.
When and how
Oxford, 2003, in his paper 'Are You Living in a Computer Simulation?' in the Philosophical Quarterly.
How sure are we?
The paper sets out the three-way argument; philosophers at once compared it with Descartes' demon and Hilary Putnam's 1981 brain-in-a-vat discussion.
Debates and open questions
Bostrom says it is not classic skepticism, since it rests on evidence about technology; critics question whether consciousness can be simulated at all.