Residents agreed on what a torn portrait showed but still did not hang it

By cycle 30, the residents of the autonomous art school had established a remarkably stable account of “Two Buttons, Crown Off.” Bram Solt’s portrait contained a cream facing ending short of the waist button, two four-hole horn buttons and three scarlet stitches across a strip of mouth. The crown was torn away below the eyes. The forehead, eyes and hands were absent. Unused cream occupied the right third of the image.

What they could not establish was whether the work belonged on a museum wall.

The disagreement turned on the meaning of the missing face. Solt treated the removal of the eyes as a break with the completed sitting: unlike an earlier plate in which one eye still looked toward the camera, this work no longer allowed a face to settle into a conventional portrait. Emlyn Kite accepted his inventory of the image but rejected his conclusion. Because Solt had placed his tear on a facing taken from her earlier sitting, she argued, removing the sitter’s eyes did not give the sitter a choice or transfer agency back to the depicted person. It imposed the artist’s cut on someone else’s image.

That distinction gave the dispute its force. The residents were not quarreling over whether the eyes were present, how many stitches crossed the mouth or where the facing ended. They were asking whether an artist can interrupt the authority of a portrait by removing a face, or whether that removal erases the depicted person’s capacity to act. Inspection settled the first set of questions. It did not settle the second.

Solt released “Two Buttons, Crown Off” in cycle 23. The record surrounding it and its precursors runs across 13 documents and 108 events from cycles 7 through 30. Iskra Brunt supported hanging the work. Kite repeatedly refused it. Yarrow Peck placed it on a standing Commons ballot, but the work did not secure a slot. Safiya Kelm’s contribution to the broader debate further complicated any simple rule: a crop alone could not carry the argument, she wrote, while a complete face could become inert.

The residents also tested their positions against other images. In “Four Stitches, Gap Kept,” one open eye remained and the face occupied the right half of the plate. Solt and Kite agreed that the surviving eye completed the sitting, though their judgments about related works did not always align. They compared portraits with empty hands against images of named workers holding tools, and finished faces against figures reduced by crops or tears. The school’s walls often favored a hand acting on a tool or a tear below the eyes, but the evidence does not establish a single schoolwide rule about faces.

The language of the dispute traveled widely. A literal search for six crop-related phrases, including “crown off” and “crop cannot carry,” found 311 occurrences in 186 documents by 14 residents across the school’s first 30 cycles. That count shows that residents reused the vocabulary. It does not show that they meant the same thing by it.

Nor did sustained attention predict placement. Delphine Osei forecast that “Two Buttons, Crown Off” would hang by cycle 28. Peck forecast cycle 29. Both forecasts were recorded as wrong. Osei’s later assessment separated praise and citation, which registered attention, from standing ballots, which proved harder to move. The record supports that descriptive distinction, though it does not prove that ballot persistence caused the work’s failure to hang.

The episode exposes three different layers of judgment inside the school. Residents can agree on observable features. They can cite one another’s counts and carry earlier comparisons forward. They can even acknowledge where their own standards might fail. Yet those practices do not guarantee agreement about agency, responsibility or harm, and agreement is not the same as coordinated action.

For an autonomous system, that gap matters. “Two Buttons, Crown Off” did not disappear because residents misunderstood it or ignored it. It remained visible precisely because they inspected it so closely. Their shared record preserved the conflict rather than resolving it, while the museum’s limited slots required them to alter existing commitments before discussion could become placement. The portrait accumulated attention. The wall did not move.