Justification (Philosophy HL)

The big idea: The whole topic keeps coming back to one word: justification — having good reasons.

But push on any reason and a problem opens up. Why believe it? Because of another reason. And why believe that? Another reason again. It never seems to stop — and that threatens to make justification impossible.

The regress problem: You believe the bus comes at 9. Why? The timetable says so. Why trust the timetable? The company published it. Why trust the company? … Each answer just needs a new reason underneath it. This endless chain is the regress problem. If it never ends, no belief is ever fully justified — and the sceptic wins.

Two big theories answer the regress in opposite ways — one stops the chain, the other loops it.

Foundationalism: some beliefs are basic: Foundationalism says the chain stops at bedrock. A few beliefs are basic — justified all by themselves, needing no further reason (like 'I'm in pain right now' or a simple truth of maths). Everything else is a building resting on those foundations. The regress ends because you hit the ground floor.

Coherentism: beliefs support each other: Coherentism rejects the whole idea of a bedrock. There are no 'basic' beliefs; instead, a belief is justified if it fits with all your others in one connected web — like threads holding a spider's web taut. Nothing is the foundation; everything supports everything else. The regress isn't stopped — it's turned into a supporting loop.

Checkpoint — the two shapes: In one line: foundationalism stops the regress at basic beliefs; coherentism loops it into a web where beliefs support each other. Now a different move entirely.

Both views so far assume justification is inside you — reasons you could point to. A third view breaks that assumption.

Reliabilism: justified by a reliable process: Reliabilism is the leading external view. What justifies your belief that there's a bird outside isn't a chain of reasons you could recite — it's that your eyesight is a reliable process that usually gets things right. This neatly dodges the regress (no chain of reasons needed) and fits how a child or an expert can know without arguing for it. It also links straight back to Gettier — the stopped clock fails because it wasn't a reliable process.

Go further — higher-level insight: See how reliabilism reframes the whole topic. It ties justification (this micro) to Gettier (3.2.2) — the stopped clock fails because the process wasn't reliable — and it lowers the sceptic's bar from 3.2.1: you don't need certainty or a complete chain of reasons, just a process that reliably tracks the truth. Linking all three across the topic is exactly the synthesis a §B essay rewards.

Checkpoint — the split: In one line: internal views want reasons you can state; reliabilism says a reliable process is enough, even if you can't spell out why. Three theories now sit on the table.

How Section B works: Section B gives you a choice of essay questions on the optional themes (Epistemology is one) [25]. No stimulus — just a claim to evaluate, using your own knowledge. 'Evaluate the claim that…' is the classic wording, and the whole of 3.2 arms you for it.

Common mistakes: 1. Describing the three theories instead of arguing between them. 2. Forgetting the regress — it's the reason the whole debate exists. 3. Only one theory — top bands need tension. 4. No conclusion — decide, with a reason. 5. Name-dropping — a theory earns nothing without its argument.