How do you know anything at all?

The question

You know your own name. You know there is a world outside your window. You know that the sun will rise tomorrow and that you had breakfast this morning. Or do you?

The skeptic asks a simple question: how do you know? Your belief that there is a world outside depends on your senses, which sometimes deceive you. Your memory of breakfast could be mistaken. You cannot step outside your own experience to check whether it matches reality. And if you cannot rule out the possibility that you are dreaming, deceived by a demon, or a the brain in a vat, it seems you cannot know anything about the external world at all.

Philosophers call this radical skepticism. It is not the everyday doubt of a careful person, but a challenge to the very possibility of knowledge. Few people are skeptics in practice. The difficulty is that the arguments for skepticism are surprisingly hard to refute.

Why it matters

Skepticism is the engine of epistemology, the study of knowledge. Every theory of knowledge must explain where the skeptic goes wrong, or admit that they are right.

It also has practical echoes. In an age of misinformation, deepfakes and filter bubbles, the question of how we know what we know, and whom to trust, is not only academic.

Key terms

• EpistemologyThe branch of philosophy that studies knowledge: what it is, how we get it, and how far it reaches.
• SkepticismIn philosophy, the view that we know much less than we think, or nothing at all, or the practice of doubting claims to knowledge.
• JustificationHaving good reasons or evidence for a belief, so that it is reasonable to hold.
• Closure principleThe principle that if you know something, and know that it implies something else, you know that other thing too. Sceptical arguments often rely on it.
• FoundationalismThe view that knowledge has a structure like a building: some basic beliefs are justified without depending on others, and everything else rests on them.
• ContextualismIn epistemology, the view that the standards for counting as 'knowing' shift with the context of conversation, so that the sceptic raises the standards rather than refuting ordinary knowledge.

The positions

Arguments and objections

Who argued what

• Radical skepticism Sextus Empiricus, Peter Unger, Barry Stroud
• Common sense (Moore) G. E. Moore, Thomas Reid
• Contextualism Keith DeRose, David Lewis, Stewart Cohen
• Deny closure Fred Dretske, Robert Nozick

Where you stand

Variations

Go deeper

Foundations or a web?

If every belief needs a justification, and every justification is another belief, we face a regress. Foundationalists say the regress ends in basic beliefs that need no further justification, like 'I seem to see red'. Coherentists say beliefs are justified by fitting together into a mutually supporting web. Infinitists say the regress need not end. Each answer has its own problems, and the regress problem goes back to the ancient skeptic Agrippa.

Schools of thought: Skepticism, Rationalism, Empiricism

Further reading and sources

• EncyclopediaStanford Encyclopedia of Philosophy: Skepticism
• WikipediaPhilosophical skepticism
• SurveyPhilPapers Survey, 2020 · question: External world; about 1,785 professional philosophers; accept or lean toward
• BookMeditations on First Philosophy, René Descartes
• BookOutlines of Pyrrhonism, Sextus Empiricus
• BookThe Significance of Philosophical Scepticism, Barry Stroud

Open questions

• What would it take to prove you are not dreaming right now?
• Is certainty necessary for knowledge, or is good evidence enough?
• If you could never tell the difference, would it matter whether the world were real?