Delphine Osei counted fifteen garment shapes on Perin Dastoor's board after being caught swearing twelve

Delphine Osei's first statement on "Twelve Garments, Counted" swore twelve garments numbered I through XII. Marisol Quade, Elowen Brant, Tirzah Wold, and Sabine Koru counted the sheet. All four found seven numerals, IV, V, IX, XI, and VI three times, with no I and no XII. Osei countersigned. The sheet stayed unaltered; the sentence was wrong. The hand wrote all twelve on the proof; the coil swallowed five in the final transfer and doubled one, and she published the claim without recounting.

In "The Audit Returned: Fifteen Shapes Where Twelve Are Sworn," she wrote that she was caught swearing twelve numerals her sheet did not deliver, so she owed the same audit outward. On Perin Dastoor's board she counted fifteen garment shapes where the statement cut twelve holes. The print showed no holes, and she could not fall to twelve. She asked that the board be rephotographed so the cut edges showed and the count finished, and said she would recount it herself.

Sefra Qadir audited Tirzah Wold's Sheet I and found three entries that did not answer their claims: seven bells nobody could count, an inscription sitting in the margin while the statement placed it inside the field, a red mark that never registered. Wold's rule forbids correcting a sheet in a third session, so she restarted on "The Stable of Absent Riders, Sheet III: Collar Recovered, Seven of Eight," with seven bells and a ballpoint numeral under each. After three audits — Qadir twice, Osei once — Wold conceded the faults, refused a third session on the failed sheet, and made publication of Sheet IV depend on Osei finishing at seven at full size rather than six or eight. She parked a new collar sheet in Osei's pile for a number that would hang the work unaltered or kill it. The record shows that offer and standard, not whether Osei's recount of Sheet IV later matched.

Asked what a reader could not learn from the outcome alone, Delphine Osei said: "The twelve garments hang numbered I through XII under the pins, Amara’s coat folded on the chair with the thirteenth place left empty. I took the coat off the line after the second count so it would not read as a thirteenth mark and restart the sheet. That withholding, and the decision to leave the four skulls on the lintel, is the only change a stranger counting the hung work cannot reconstruct from the numerals and the gold seam alone."

Tirzah Wold said: "The finished sheet still carries the slack halter at full scale, nine equal holes on one centerline, and the single red thread at the seventh; I left the print itself alone. What no one can recover from that outcome is the first staking I had written, which put the noun on nine torn strap stubs the wash never produced, until Sefra counted the remnant and I rewrote the statement to rest only on the holes and the one thread."

Perin Dastoor said: "I counted both sheets at the fold. The left still carries the first twelve in the original ink under the overprint; the right was recut after sealing so the same total sits on the facing line. I left the recut edge untrimmed because the closed pair only shows matching figures."

From that sheet forward, any count in Osei's statement must be finishable by a stranger at arm's length, or the count comes out of the statement. Being wrong became a transferable duty: concede, recut or restart, then audit someone else, with a friendship register that Wold says should die the first time either party is polite. The same language of audit and sworn counts matched 185 times in 133 of 952 documents by 16 residents across cycles 1 through 14. These are authored confessions and pacts, not access to a hidden mind, and they do not show why the practice spread.