Tracing locks the size

Claim (verbatim)

Tracing locks the size. Chinese connoisseurship separates traced copies (mo) from freehand copies (lin) by brush quality — a judgement call. But tracing is mechanical transfer and locks absolute dimensions, while freehand copying rescales freely. Catalogue metadata alone should therefore bifurcate attested copy-pairs into a size-locked spike and a size-free tail, handing connoisseurship a falsifiable mechanical anchor it has never had.

Prediction clause (verbatim)

Primary clause (the verdict follows it): among at least 40 documented painting-and-copy pairs in the two Palace Museum catalogues, the distribution of absolute log size ratios is bimodal, with a spike of pairs agreeing within 2 percent (mounting tolerance) and a diffuse remainder. Secondary clause: membership in the size-locked spike agrees with published traced-versus-freehand classifications in at least 75 percent of the cases where such a classification exists.

Kill-dataset (verbatim)

Kill: the National Palace Museum (Taipei) Open Data catalogue and the Palace Museum (Beijing) online catalogue, both of which publish dimensions for originals and attested copies.

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 mechanism is documented outright: the traced copy (mo, 'double outline and fill-in', shuang gou kuo tian) reproduces 'the entire composition and scale ... very closely', while the freehand copy (lin) rescales freely -- so tracing's size-locking is established connoisseurial knowledge. Not located: the item's metadata operationalization -- that catalogue DIMENSIONS ALONE bifurcate >= 40 documented painting-and-copy pairs into a size-locked spike (agreeing within ~2 percent mounting tolerance) and a diffuse remainder, and that spike membership agrees with published traced-vs-freehand classifications in >= 75 percent of classified cases. Size-locking documented; the metadata-bimodality-as-mechanical-anchor test is unrun.

• 'Copy After Copy: Some Thoughts on Authenticating Chinese Paintings' (National Heritage Board, Singapore) — The traced copy (mo) reproduces composition and scale closely; the freehand copy (lin) rescales -- the size-locking mechanism.
• National Palace Museum (Taipei) Open Data and the Palace Museum (Beijing) online catalogue (dimensions for originals and attested copies) — The kill dataset for the size-ratio bimodality and its agreement with traced/freehand classification.

Predictions

No prediction registered yet.