Each work is a pair. The left panel is a photograph as it was made. The right panel is the same photograph carrying one intervention derived from its own measurements. Side by side, the difference often lives at the threshold of vision, and it is stated exactly, down to fractions of a single tonal step. The numbers are the work, and anyone may measure them again. You are offered almost nothing to see, and everything to check. · · · One paint, one color, brightness all but untouched, so the wall no longer reports which of the two lights reached each brick. Binned by their own brightness into five bands, 332,233 wall pixels reveal that this painted wall records two different colors for one coat of paint. The darkest band sits at chromaticity r = 0.6716 and the brightest at 0.5647. Split a second way, below its 20th brightness percentile the paint reads 0.1310 redder in r than above its 80th. The lit face reads as the lamp's and the shadowed face as mostly the red wall's own bounce, two lights of different color on one surface. It is in nearly every photograph of a painted wall, and no one sees it, because we expect shadows to be moodier. The wall has been given a single chromaticity, the luminance-weighted mean the wall itself reports, so lit and shadowed brick both had a say in what the paint is. It measures r 0.6032, g 0.2418, b 0.1550. The shadow-to-light r split falls from 0.1310 to 0.0145, 89% removed, measured from the finished file. The intervention is constrained to preserve luminance, and it nearly does: the median change is 0.00031 in linear light, 0.0017 at the 99th percentile, and eighty-four pixels of 590,000 exceed 0.01, every one already clipped in red before the intervention. The hanging plumes move by at most 1.91 levels. Presentation. The mount is RGB (86, 54, 43), the paint itself at the darkest luminance the wall reaches, its 2nd percentile. The intervention's rule was that chromaticity is given to the paint and luminance is left to the light, so the board is the same paint in deeper shadow than anything in the frame. It also measures. The corrected panel's shadowed wall sits 0.0105 from the board's hue, and the untouched panel's sits 0.0821, eight times farther. The color is what makes the mount an instrument here. There is no rule, because the board is the paint continued, and a line would say the paint stops. To check. 477,167 pixels move, none by more than 44 of 255. Download this work's original file, 07_one_coat.png, not a screenshot or a marketplace preview, attach it to an AI that can run code, and paste this: Run this exact line on the attached file, changing only the file path if needed. Show me its exact output, and say whether that output is exactly: peak 44 changed 477167 python3 -c "import numpy, PIL.Image as I; c=numpy.asarray(I.open('07_one_coat.png').convert('RGB')).astype(int); a=c[691:2995,691:2995][::3,::3]; b=c[691:2995,3341:5645][::3,::3]; d=abs(b-a); print('peak', d.max(), 'changed', (d.max(2)>0).sum())"