A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
If every justified belief requires a justifying reason, and every justifying reason is itself a belief that requires justification, then justification appears to generate an infinite chain. The Agrippa trilemma identifies three ways this regress can terminate: in an infinite chain (infinitism), in a circle (coherentism), or in beliefs that are self-justifying or justified without further reasons (foundationalism). Each option faces objections — infinite regresses seem psychologically unrealizable, circular reasoning seems viciously circular, and foundational stopping points seem dogmatically assumed. The regress problem is the central structural puzzle driving debates about the architecture of justification.
Draw the regress explicitly as a diagram with arrows representing 'is justified by'. Ask: where does the chain stop, and why there? Then evaluate each horn of the trilemma in turn before reading how foundationalists, coherentists, and infinitists respond.
The regress problem starts from a deceptively simple observation: if you claim to know something, you should be able to justify it. But any justification you offer is itself a belief — and can be challenged: "Why do you believe that?" You offer a further reason. "Why do you believe that?" And so on. The question is whether this chain can ever legitimately stop, and if so, how. This is not a merely academic puzzle — it is a direct challenge to the coherence of the concept of justification itself. If you can't stop the regress, then no belief is ever truly justified, which would mean no one knows anything.
Your prerequisite on justified true belief gave you the standard tripartite account of knowledge: knowledge is justified, true belief. The regress problem attacks the "justified" component by asking what it means for a belief to be justified by another belief. The picture is simple: belief B1 is justified by belief B2, which is justified by B3, which is justified by B4... The Agrippa trilemma names the three ways this chain can terminate (or fail to terminate). The trilemma is not solved by pointing to one horn — each horn is a position that serious philosophers defend, and each faces genuine objections that the others press. Understanding the trilemma means understanding why the problem is genuinely hard, not just naming the three options.
Foundationalism says the regress terminates in basic beliefs — beliefs that are justified without being justified by other beliefs. Candidates include beliefs about your current perceptual experience ("I seem to see red now"), beliefs that are self-evident to reason, or beliefs that are incorrigible. The challenge: why are these beliefs justified if not by further reasons? If "I seem to see red" is basic, what makes it justified rather than arbitrary? Strong foundationalists say such beliefs are infallible or self-justifying; more modest foundationalists say they have a kind of prima facie justification that doesn't require further support but can be defeated. Coherentism rejects the linear model entirely: beliefs are justified not by chains of support but by their coherence with the whole system of beliefs one holds. No belief is foundational; every belief is justified by fitting with the network. The objection your Core Idea notes is critical — coherence seems to permit internally consistent but globally false belief systems. Two people with radically different but internally coherent worldviews would each count as equally justified, which seems wrong.
Infinitism accepts the infinite regress and argues it is not vicious. On this view, a belief is justified if there exists an infinite chain of distinct supporting reasons, even if you have not actually traversed all of them. The objection: finite human minds cannot possess infinitely many beliefs, so the chain is never actually available. Infinitists typically respond that you need the chain to be *available* or *accessible*, not fully consciously held. The debate turns on whether potential or dispositional reasons can justify.
The regress problem's significance extends beyond epistemology. It is structurally analogous to problems in ethics (what justifies moral principles?), mathematics (what justifies axioms?), and law (what justifies the constitution?). In each domain, justification either bottoms out in something unjustified, goes circular, or regresses infinitely — and each option seems problematic. The regress problem reveals that the concept of "justified by" is harder than it first appears, and that the architecture of our knowledge — the structural relationships among beliefs — is a genuine philosophical problem, not a given.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.