Truth

The property a claim has when things are as it says they are - and the argument over what that amounts to.

Theme: Knowledge, truth and reason
Tradition: Cross-tradition
Term: ἀλήθεια · alētheia
Term fixed: c. 330 BC

In plain terms

Everyone can use the word without difficulty. A statement is true when things are as it says. The trouble starts when you ask what that relation is.

It cannot be a comparison, because you cannot step outside your beliefs to hold one up against reality and check. Every check you run is itself another belief. So the correspondence has to be real without ever being directly inspected, which is uncomfortable.

The alternatives all give something up. Say truth is coherence with everything else you hold, and a consistent fiction becomes true. Say it is what works, and useful falsehoods become true. Say the word adds nothing - that asserting a claim is true is just asserting it - and you have avoided the question rather than answered it, though a surprising number of philosophers think that is the right thing to do.

As a philosopher would put it

Correspondence theories make truth a relation between a truth-bearer and a fact, and their difficulty is saying what facts are without defining them as whatever makes true statements true. Tarski's semantic conception supplies a formal adequacy condition - the T-schema, that snow is white is true if and only if snow is white - and a recursive definition for formal languages, sidestepping metaphysics while satisfying the intuition.

Coherence theories make truth consistency within a system, which handles the inspection problem and struggles with the objection that more than one consistent system is possible. Pragmatist theories tie it to enquiry: Peirce's is the opinion fated to be agreed by all who investigate, which is realist in intent; James's looser version, that the true is what works, drew the charge that it confuses truth with utility.

Deflationism denies there is a substantive property at all - the truth predicate is a device for generalising and for endorsing, and once the T-schema is granted there is nothing further to explain. The liar paradox sits underneath all of these, since a sentence asserting its own falsity shows that the intuitive notion is not simply consistent.

Where you meet it

Scots law has three verdicts. A jury may return guilty, not guilty, or not proven - and the third acquits exactly as the second does, with no legal difference in effect.

What differs is what is said. Not proven is widely understood to mean that the jury suspected the accused and were not satisfied by the evidence, and it is used in about a fifth of acquittals in solemn cases. Reform has been debated for decades: campaigners argue it stigmatises the acquitted and is used disproportionately in sexual offence cases, and legislation to abolish it has been repeatedly proposed.

What the verdict makes visible is a distinction the rest of us usually leave implicit. A court can only ever establish what was shown, and the question of what happened is not the same question. Every legal system knows this and Scotland alone says it out loud, which is why the verdict is both useful and impossible to defend comfortably.

What is actually disputed

Whether truth is a substantive property.

Inflationists hold that it is - that being true consists in some relation to reality, and that an account of what that relation is remains owed. Deflationists hold that the truth predicate is a logical convenience for endorsement and generalisation, that the T-schema exhausts its content, and that centuries of searching for the relation found nothing because there is nothing to find.

Underneath is a dispute about realism. If truth is correspondence, then claims can be true beyond any possible verification, which some regard as the whole point and others as unintelligible. Pragmatist and epistemic accounts tie truth to what enquiry would deliver, which keeps it connected to us at the cost of making unknowable truths impossible - and it is not obvious that they are.

Defined against

Appearance and reality If appearances can be systematically misleading, truth is precisely what they can fail to deliver

Philosophers

Aristotle The formulation everything else is a response to
Thomas Aquinas Truth as the adequation of intellect and thing
Friedrich Nietzsche Truths as metaphors whose metaphorical character has been forgotten
Charles Sanders Peirce Truth as the settled opinion of enquiry
William James The looser pragmatist version, and the objections it attracted
Martin Heidegger Truth as unconcealment, before it is a property of statements

Works

Metaphysics The classical formulation
Summa Theologiae The scholastic account of adequation
Pragmatism Truth tied to what works
How to Make Our Ideas Clear Peirce's version, and the difference from James's
Tractatus Logico-Philosophicus Truth as picturing, in its most austere form
Being and Time Truth as disclosure rather than correspondence

Schools

Peripatetic school (Aristotelianism) The correspondence intuition in its original form
Thomism Its scholastic development
Classical Pragmatism Truth as the outcome of enquiry
British Idealism The coherence account at its strongest
Vienna Circle Verification as the criterion
Phenomenology Truth as disclosure

Related concepts

Knowledge c. 369 BC · Correspondence c. 1270 · Coherence 1807 · Belief c. 380 BC · Objectivity 1781

Aristotle's formulation - to say of what is that it is - is the ancestor of every correspondence account; Tarski gave it a formal treatment in 1933 that changed what the question could look like.