Gilded texts are corrupt

Claim (verbatim)

Gilded texts are corrupt. The deluxe illuminated copy and the textually best copy of a work are systematically different books: illumination routes production through commercial ateliers optimizing appearance, schedule, and a lay patron's ear — smoothing hard readings as they go — while plain copies made for professional readers preserve difficulty. Decoration grade should therefore predict textual corruption positively, making beauty a quantifiable warning sign for editors.

Prediction clause (verbatim)

Primary clause (the verdict follows it, adjudicated on the Canterbury Tales witnesses as the named arena): pre-1500 witnesses in the top decoration tier — miniatures or historiated initials — show at least 1.5 times the rate of smoothing variants plus omissions relative to plain witnesses of the same generation and region, with decoration grade taken from catalogue records and variant rates from published collations. Secondary clause: the same sign holds in at least one further collated tradition, such as a psalter recension.

Kill-dataset (verbatim)

Kill: the Canterbury Tales Project's published transcriptions and collations of the manuscript witnesses, with decoration grades from the British Library and Digital Bodleian catalogues.

Novelty / leakage triage

provisional — model-triaged, shepherd review pending

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. Provisional reading: Adjacent (closely related prior work exists).

The item's arena and its direction are documented at the flagship: for the Canterbury Tales the deluxe, richly illuminated Ellesmere shows 'a later attempt to regularise' the text, while the plainer Hengwrt (by the same scribe, Adam Pinkhurst) is now judged textually superior and preferred by most modern editors -- deluxe production routing toward a smoothed, regularized text exactly as the item claims. Not located: the item's systematic law -- that pre-1500 top-decoration-tier witnesses (miniatures/historiated initials) carry >= 1.5x the rate of smoothing variants plus omissions of plain contemporaries, holding in the Canterbury Tales collations AND in at least one further collated tradition. Flagship exemplar documented; the corpus-wide decoration-grade-predicts-corruption rate is unrun.

• Wikipedia, 'Hengwrt Chaucer' (Hengwrt textually superior and editor-preferred; Ellesmere regularized) — The deluxe Ellesmere regularized vs the plainer, better-text Hengwrt; the item's direction at the flagship.
• The Canterbury Tales Project transcriptions and collations, with decoration grades from BL and Digital Bodleian catalogues — The kill dataset for the decoration-grade vs variant-rate correlation across witnesses.

Predictions

No prediction registered yet.