Questions: Justificatory Chains and Support Relations

Question 1

Three independent witnesses each report seeing the same crime. Under a Bayesian (probabilist) model of justification, their combined testimony:

• A. Provides no more justification than a single witness, since all three observed the same event
• B. Provides weaker justification than a single witness, due to transmission loss at each inferential step
• C. Provides stronger justification than any single witness alone, because independent convergence is unlikely if the crime didn't occur
• D. Provides exactly three times the justification of one witness, by simple addition

Answer: (C)

Under a Bayesian model, independent evidence compounds: if E₁ and E₂ independently support hypothesis H, their conjunction raises H's probability more than either alone, because having multiple independent corroborating witnesses is much less likely if H is false. Option A ignores the independence benefit. Option D treats evidence as linearly additive, which misses the multiplicative nature of probabilistic combination. The key is genuine independence — if the witnesses colluded or all observed the same misleading cue, the combination is far weaker.

Question 2

In classical foundationalism, when a valid deductive inference is drawn from a fully justified belief, the resulting belief is:

• A. Less justified than the premise, because each inferential step introduces transmission loss
• B. Fully justified, because valid deductive inference preserves justification without degradation
• C. Justified only if the conclusion is also self-evident or directly observable
• D. Only partially justified — the conclusion requires independent corroboration to achieve full justification

Answer: (B)

Classical foundationalism holds that valid deductive inference transmits justification fully — if the premises are justified and the inference is valid, the conclusion is justified with no loss. This contrasts with probabilistic models where inductive inference only partially transmits justification. The foundationalist model is demanding about starting points (basic beliefs must be certain or self-justifying) but generous about what valid inference produces from them.

Question 3

A coherentist holds that justification flows through chains from basic foundational beliefs to conclusions, just as foundationalists claim.

• T. True
• F. False

Answer: False

Coherentism explicitly rejects the chain metaphor. Rather than justification flowing directionally from foundations through inferential links to conclusions, coherentism holds that each belief is justified by its coherence with the whole web — mutual support, consistency, explanatory integration. There are no 'basic' beliefs from which chains begin. The 'chain' metaphor is foundationalist; coherentism replaces it with a web or network where support is holistic and non-directional.

Question 4

Multiple independent pieces of evidence that each support a hypothesis can, in total, provide no more justification than the single strongest individual piece.

• T. True
• F. False

Answer: False

Independent convergence amplifies support. If each piece of evidence is more likely given H than given ¬H, then multiple independent pieces make H substantially more probable — their conjunction is even less likely to occur by chance if H is false. This is the logic of cumulative cases in law, science, and everyday reasoning: multiple independent lines of evidence (motive, means, opportunity, physical evidence, testimony) are more convincing than any single element, even when each alone is inconclusive.

Question 5

What is 'transmission loss' in justificatory chains, and which epistemological theory most readily accepts it? Which theory denies it, and on what grounds?

Model answer: Transmission loss is the weakening of justification as it passes through inferential steps — each link degrades the support slightly, so a conclusion many steps from basic evidence is less justified than a direct inference. Probabilism (Bayesian epistemology) most naturally accommodates this: inductive inference only partially raises a conclusion's probability, and long chains of probabilistic inferences accumulate uncertainty. Classical foundationalism denies transmission loss for deductive chains: valid inference from a justified premise yields a fully justified conclusion. The foundationalist grounds this in the meaning of validity — a valid inference cannot have true premises and a false conclusion, so if premises are fully justified (certain), the conclusion cannot fail to be.

The debate matters practically: if transmission loss is real, then even valid argument chains from strongly justified beliefs might produce only weakly justified conclusions. If foundationalists are right, one certain basic belief can anchor an entire system through valid inference. The practical upshot: evaluate whether evidence sources are genuinely independent, and check whether 'convergence' traces back to a single shared origin.