Did medieval judicial torture exist specifically to convert a 'half-proof' into the 'full proof' of a confession?
claim-roman-canon-law-rated-one-witness-equal-to-a-private-document carries the capture's [unverified-mechanism -- needs primary] flag on its darkest and most load-bearing detail: that judicial torture in the Roman-canon law of proofs functioned as the licensed instrument for completing a half-proof — turning strong-but-insufficient suspicion into the confession that alone counted as full proof. The claim currently rests only on the Tier-3 Wikipedia "Half-proof" entry.
What would answer it
- John H. Langbein, Torture and the Law of Proof (University of Chicago Press, 1977) — the standard monograph on exactly this mechanism; check whether it states that torture required a half-proof threshold before it could be applied and that its purpose was to produce a confession completing the proof.
- Mathias Schmoeckel or other historians of the ordo iudiciarius and the medieval law of proofs, for the doctrinal rule linking semiplena probatio to the preconditions for torture.
- A primary or near-primary statement of the two-witness / confession = "full proof" rule and the fractional-proof system attributed to the 1190s Glossators (Azo).
Why it matters
If confirmed, the note becomes a clean, well-sourced instance of a legal system that legislated evidence into fractions and then built torture into the arithmetic — the load-bearing point that makes the half-proof rule more than a curiosity. If the causal link is looser than "torture existed to complete half-proofs," the mechanism claim needs softening. Either way it should not leave seedling on an encyclopedia paraphrase.
Update 2026-07-11 (still open)
Promotion of 10-inbox/raw/2026-07-11-dup-risk-amari-gates-half-proof-bridge.md added claim-medieval-judicial-torture-required-a-half-proof-and-produced-the-completing-confession, which raises the mechanism's support from a lone Tier-3 encyclopedia entry to: James Franklin's Tier-1 chapter ("torture … could be applied only when there was half-proof of guilt (to produce the other half)"), Whitman p. 115 ("could not be ordered unless there was 'semiplena probatio'"), and Langbein (the system "was unworkable standing alone," torture "extended" it). This is materially stronger, but the reads were via WebSearch/a legal-blogger quotation and a tls:unverified PDF, not a direct read of Langbein's monograph or a medieval primary. Kept open: the "needs primary" gap the question names is not yet closed. The new claim-note is held at seedling for the same reason.