Epistemology

This is Unit Two of Philosophy: Epistemology, the study of knowledge. The unit opens on a definition that stood for roughly two thousand years and was destroyed in 1963 by a paper three pages long. What replaced it is not a better definition — it is decades of proposed repairs, each meeting a new counterexample, and a serious suggestion that the thing may not be definable. Then the other direction: not what knowledge is, but whether we have any. The skeptical arguments here are built deliberately to be as strong as possible, by a philosopher constructing the best case against himself. The rest of the unit is about justification — where it comes from, whether it can stop anywhere, and whether it has to be something you could actually produce if asked. And it ends on a problem that has no accepted solution, only a way of living with it. By the end you will be able to do five things. One: state the classical analysis of knowledge, and say why one lucky case defeats it. Two: tell the three theories of truth apart, and name what each is accused of. Three: say what the skeptical arguments claim — and the narrower thing they do not. Four: tell the two great traditions apart by what each says reason alone can deliver. Five: state the regress problem, and the two structures offered against it.

One entry of housekeeping, and it earns its place. The English word covers three different accomplishments, and the debates in this unit are all about one of them. Let the three run together and the arguments stop making sense.

The one this unit is about.

Philosophy 0045 Propositional Knowledge

Knowing a capital city, knowing how to ride a bicycle, and knowing a neighbour are three unrelated achievements. Everything ahead concerns the first — keep them apart or the debates blur.

Three conditions, jointly sufficient, tracing back to Plato. Then the counterexample that broke them — and what is instructive is not the break but what followed it: thirty years of repairs, each answered by a new case.

The three conditions.

Philosophy 0046 The Justified True Belief Analysis

A juror who thinks the defendant guilty, on strong evidence, about a man who is guilty — believed, so, and well grounded. On this analysis that settles it.

And the case that broke it.

Philosophy 0047 The Gettier Problem

Glance at a stopped clock at precisely the hour it shows. The belief is grounded, and it is correct, and the correctness arrived by accident — the grounds and the truth met by luck rather than connection.

The repairs.

Philosophy 0048 Repairing the Analysis

Rule out false assumptions, or rule out luck. A rock shaped like a sheep persuades you a sheep is present — and one is, out of sight behind a rise. Each repair met new counterexamples, and some concluded the thing resists definition.

One of the three conditions is agreed on by everyone and understood differently by everyone. Three accounts of what truth amounts to, each with a real objection — and the objections are more useful to remember than the theories.

The first.

Philosophy 0050 The Correspondence Theory of Truth

A claim about a cracked window is true when the window is cracked, and nothing more elaborate is needed. The difficulty is explaining the matching — what a belief about justice is supposed to match.

The second.

Philosophy 0051 The Coherence Theory of Truth

A detective accepts the account where every clue and timeline fits without contradiction. The objection is short: a novel can be perfectly consistent throughout, so what ties a consistent system to the world?

The third.

Philosophy 0052 The Pragmatic Theory of Truth

Navigators relied on Newtonian mechanics because ships arrived and eclipses came on schedule. The reply asks who it worked for and for how long — useful and true can part company, and there they did.

Now the other question: whether anyone knows anything at all. What follows is a method with a deliberate escalation built into it — each step constructing a stronger opponent than the last, by a philosopher who intended to defeat all of them.

The position.

Philosophy 0055 Philosophical Skepticism

The claim that knowledge is unavailable — in one area, or across the board. Note what it is not: a mood or a pose. It arrives as an argument and has to be answered with one.

The tool.

Philosophy 0056 Descartes' Method of Doubt

Treat anything that can be doubted as though it were false, and see what is left standing. He takes the building down to the ground — and the demolition exists so that something can be rebuilt.

The first escalation.

Philosophy 0057 The Dream Argument

He recalls dreaming he sat by a fire while lying asleep. Nothing within the experience marked it as a dream — so experience by itself cannot certify that you are awake. It never claims you are dreaming; only that you cannot rule it out from inside.

And the maximum.

Philosophy 0058 The Evil Demon

A deceiver strong enough to manufacture every experience, so that even two plus three becomes doubtful. This is doubt pushed as far as it will go, not a claim about anything supernatural — he is building the hardest opponent available.

One thing the strongest possible deceiver cannot reach. Then a modern restatement of the same problem, and a philosopher who declines to engage with it on grounds worth taking seriously.

The survivor.

Philosophy 0059 The Cogito

Doubt everything and the doubting is still going on. A deceiver requires somebody to deceive. But note how little this establishes: a thinking thing exists, not a body, a past, or a world.

The modern version.

Philosophy 0060 The Brain in a Vat

A brain fed a flawless simulation would have experiences matching yours in every detail, and no test run from inside could separate the two. The demon, rewired.

And the refusal.

Philosophy 0061 Moore's Two Hands

He holds up a hand, then the other, and takes that as more certain than any premise a skeptical argument could offer. Where a conclusion is less credible than what it denies, the argument is the thing to doubt.

A distinction about where a justification comes from, and then the two traditions that divide over how much the first source can deliver. Both traditions are routinely caricatured, so each entry says what the position does not claim.

The distinction.

Philosophy 0065 A Priori and A Posteriori

Seven plus five needing no observation, against a kettle that does. And the distinction is about what justifies the belief, not how you came by the concepts — you needed experience to learn the numerals, and arithmetic is not thereby experimental.

The first camp.

Philosophy 0066 Rationalism

Some substantive truths about the world are available to reason without the senses, resting on ideas the mind brings rather than receives. This is not a rejection of the senses — it denies they are the whole story.

The second.

Philosophy 0067 Empiricism

Everything substantive about the world traces to experience, with reason unpacking what experience supplied. Pushed hardest, even cause and effect turn out to be something we learned by watching. Logic is still used — the quarrel is about knowledge of the world.

The image that carried it.

Philosophy 0068 Locke's Blank Slate

No ideas at the start; the materials all arrive through encounter and reflection. A child who has never seen red has no idea of it. But capacities are allowed — remembering, comparing, abstracting — so the paper is blank and the machinery is not.

One classic argument against relying on perception, and then two sources of knowledge that get overlooked precisely because they are so ordinary — and between them they account for nearly everything anyone knows.

The argument.

Philosophy 0069 The Argument from Illusion

A straight oar in water, a mirage, rails converging at the horizon — familiar sights, dependably wrong. Note the limit: occasional failure does not establish general unreliability. It opens the question rather than settling it.

The first overlooked source.

Philosophy 0070 Testimony

Your own date of birth, the boiling point of iron, the existence of Antarctica — all arrived on someone's say-so. Relying on this is not a weakness; the real question is which sources have earned it.

The second.

Philosophy 0071 Memory as a Source

You know the door is locked because you remember locking it, and nothing further is required. But it rebuilds rather than replays, and how sure you feel tracks accuracy less closely than anyone expects.

Now the structural problem sitting underneath all of it. Ask why you believe something, then ask the same of the answer, and keep going. Three endings are available and none is comfortable — the two theories in this topic each pick one and defend it.

The problem.

Philosophy 0075 The Regress Problem

The bridge is safe because of the inspection, and the inspection because of the inspector, and so on. The chain runs forever, doubles back on itself, or stops somewhere with nothing under it.

The first answer.

Philosophy 0076 Foundationalism

Certain beliefs are justified without resting on any others, and everything else is built on top — a pyramid. The pressure falls on which beliefs qualify, and whether they can carry the weight of the building.

The second.

Philosophy 0077 Coherentism

No base at all: beliefs hold each other up, the way planks interlock, with nothing touching bottom. Repairs happen while afloat — and the standing worry is that a well-joined raft can drift entirely free of the water.

One divide that reaches back into everything already covered, because it changes what the regress was even demanding. The question is whether justification has to be something you could produce if asked.

The divide.

Philosophy 0078 Internalism and Externalism

One side requires that what justifies you be something reflection can reach. The other allows facts outside your awareness to do the work — someone sorting chicks accurately without being able to say how may still count.

The developed version.

Philosophy 0079 Reliabilism

A belief qualifies when the process producing it dependably yields truths — daylight vision, sound arithmetic, a calibrated instrument. The difficulty is describing the process: vision, or vision in fog, or vision on Tuesdays?

The unit's last idea, and it is genuinely unresolved. Every inference from what has been observed to what has not rests on an assumption that cannot be established without using the very inference in question.

The problem.

Philosophy 0082 The Problem of Induction

Bread nourished yesterday and nourishes today. That it will tomorrow requires the future to run like the past — and that requirement is the thing wanting support.

The obvious defence, and why it fails.

Philosophy 0083 Defending Induction with Induction

Saying the method has a good record and will therefore continue uses the method to certify itself. A lender vouching for his own application — the record is genuine and the vouching is still circular.

And how to live with it.

Philosophy 0085 Fallibilism

Knowing something does not require being unable to be wrong; a belief can qualify while staying open to revision. Newtonian mechanics counted as knowledge for two centuries and was then revised. This lowers the bar rather than removing it.

Two procedures to close. One tests whether a claim to know actually holds up, with the decisive step coming last. The other takes the unit's most famous skeptical argument and rebuilds it into numbered lines, so it can be attacked where attacking is legitimate.

The audit.

Philosophy 0090 Auditing a Knowledge Claim

Run the claim past belief, truth, and grounds, then check whether the grounds and the truth are connected or merely coincided. Most claims clear the first three — the fourth is where luck hides.

And the reconstruction.

Philosophy 0091 Reconstructing the Dream Argument

Put the skeptical passage into numbered premises with its conclusion set apart, then judge it by the standard its own form invites. Attack a premise rather than the man — while every premise stands, a valid argument concedes nothing.

Five things you can now do. You can state the classical analysis — believed, true, justified — and say why a stopped clock read at the right moment defeats it: the grounds and the truth met by accident rather than by connection. You can tell the three theories of truth apart and name what each is accused of: explaining the match, floating free of the world, and confusing useful with true. You can say what the skeptical arguments claim and what they do not — the dream argument never asserts you are dreaming, only that nothing inside the experience would tell you. You can tell the two traditions apart by what each grants reason alone, and you know neither is the caricature: one does not reject the senses, and the other still uses logic. And you can state the regress problem — endless, circular, or stopping on nothing — with the two structures offered against it: a pyramid with a base, and a raft with none.