Translation favors the self-contained

Claim (verbatim)

Translation favors the self-contained: across the great translation movements, commentaries crossed the language frontier at far lower rates than base texts — except lemmatized commentaries that embed their base text, which crossed at near-base-text rates. Patrons commissioned usable, self-sufficient objects; a commentary without its base text bundled is unusable without a second commission, so translation selects on textual self-containment. This one procurement rule shaped which scholastic apparatus each receiving culture ever saw, and it predicts countable differences between canons formed by different institutional regimes.

Prediction clause (verbatim)

Compare canon-level rates. Primary clause: in the Chinese Taisho canon, at least 80 percent of translated Indian commentarial works embed their complete root text (verses translated in situ), whereas in the Tibetan Tanjur — produced by a later, systematizing regime commissioning in bulk — under 50 percent do; the verdict follows this clause. Secondary clause: the Greek commentaries recorded as translated into Arabic in Ibn al-Nadim's Fihrist are predominantly of the lemmatized type.

Kill-dataset (verbatim)

Kill: the Taisho catalogue via CBETA, the Tohoku catalogue of the Derge Tanjur (published), and Dodge's English edition of Ibn al-Nadim's Fihrist (public).

Novelty / leakage triage

no prior located yet (provisional)

A provisional first pass authored by the model (Opus), not yet confirmed by the shepherd. It carries the same dated-search requirements as an authoritative verdict but is excluded from every headline figure and cannot underwrite a prediction until a shepherd confirms it.

[Independent blind re-audit 2026-07-08 (2nd pass, generator-independent), confidence low. Double-confirmed: both the provisional pass and this re-audit located no prior formulation of the specific operationalized claim. NOTE: low-confidence negative — an exhaustive search of specialist print scholarship was not achievable, so 'no prior located' is weaker here.] Studies of the great translation movements discuss what crossed and what did not, and the lemmatized-vs-detached commentary distinction is a familiar philological category in Buddhist (root-verses embedded in commentary in the Tanjur) and Greek->Arabic traditions. But I located no cross-canon, quantitative test of the specific procurement rule proposed: that commentaries crossed at far lower rates than base texts EXCEPT lemmatized ones, operationalized as Taisho >=80% of translated Indian commentaries embedding their complete root text vs Tibetan Tanjur <50%, with Ibn al-Nadim's Fihrist showing Greek-into-Arabic commentaries predominantly lemmatized. Confidence is low because the negative spans three separate philological literatures I could not exhaustively search; the closest prior work catalogues canon contents (Taisho/CBETA, Tohoku/Derge) without framing self-containment as the translation-selection variable.

• Survey work on the Tanjur commentaries embedding root text (e.g. Vasubandhu's Abhidharmakosa and its Tanjur commentaries)
• Bayard Dodge (trans.), 'The Fihrist of al-Nadim' — record of Greek works translated into Arabic (base texts vs commentaries)

Predictions

No prediction registered yet.