McGrew (2013) argues humans systematically over-trust arguments from silence because the decisive Bayesian term is exactly the kind of quantity we misjudge toward overconfidence

Having decomposed the argument from silence into a product of three probabilities — notice, record, survive (claim-mcgrew-argument-from-silence-is-notice-record-survive-product) — McGrew's central diagnostic claim is that our intuitions about such arguments are biased in a predictable direction. "There is, then, a mechanism at hand to account for overconfidence with respect to inferences from silence: the critical term in a Bayesian reconstruction of the argument is just the sort of thing we are apt to misevaluate in the direction that would lead to overconfidence."

The mechanism is a conjunction effect. The decisive quantity is a product of probabilities (a conjunction of "would notice AND would record AND would survive"), and people routinely estimate conjunctions too high relative to their factors — the Kahneman–Tversky conjunction fallacy. So the very cases where a silence feels conclusive are exactly the cases where the underlying product has been silently over-counted. McGrew's stock illustrations are of confident silences that turn out worthless: Greek historians do not mention Rome; Pliny does not mention the destruction of Herculaneum; Bacon and Shakespeare never mention each other.

The lesson is not to abandon the argument from silence but to grade it, and to distrust the intuition that a silence is airtight. This bears directly on a vault that specializes in negative findings — "the primary source is silent, therefore the repeated attribution is false." The conjunction-fallacy framing connects to the vault's separate treatment of that bias in probabilistic judgment (claim-quantum-like-models-fail-conjunction-fallacy), and it sharpens the methodological warning already standing at reflection-recurring-tool-summary-is-not-the-source: the tool that keeps the vault honest is also the one most likely to make it overconfident. The two seed instances graded against this warning are at observation-bjork-dreyfus-same-argument-from-silence-graded-differently; the cluster hub is moc-argument-from-silence.