Papers · William Alexander Aldred

A Number for the Angle

by William Alexander Aldred — Independent Researcher

Publicly deposited November 1, 2026 · Zenodo · 6 pp · DOI: 10.5281/zenodo.23867664

Abstract

The program I run has a standing debt, and it is one number: Θ, the rotation angle of the sky at recombination. Measurements by other people put it in a known band — 0.342 degrees, give or take 0.09 — and years of internal work have produced candidates, near misses, and refusals, but never a number of our own, derived from our own machinery, that we could seal before looking at the answer. This paper runs the deciding step one more time, in the open, and scores it. The construction-space census that closed this question in October was re-executed in this packet, line for line: it returns the same closure — twenty-eight candidate rows, exactly one inside the measured band, and that one is the fine-structure constant read as an angle, an external number with no mechanism, refused on three separate grounds. The closest owned exhibit, δ/(2π), recomputes to 0.250770 degrees — one thousandth of a degree below the band's low edge — and it is barred besides: δ is an adopted number, never derived, and an adopted toll does not become a sky amplitude by being divided by 2π after the fact. Everything the record gained in the day since the last attempt was audited for a new ingredient. Nothing arrived. So the honest result is a priced failure: there is no production, the cell ordered by the bench does not stand, and the missing object is one named thing — a slice-orientation law — with its three missing parts listed at full size. A forced number would have been easy to print. It would also have been a lie with a degree sign on it.

The result, in plain language

The program I run has a standing debt, and it is one number: Θ, the rotation angle of the sky at recombination. Measurements by other people put it in a known band — 0.342 degrees, give or take 0.09 — and years of internal work have produced candidates, near misses, and refusals, but never a number of our own, derived from our own machinery, that we could seal before looking at the answer. This paper runs the deciding step one more time, in the open, and scores it. The construction-space census that closed this question in October was re-executed in this packet, line for line: it returns the same closure — twenty-eight candidate rows, exactly one inside the measured band, and that one is the fine-structure constant read as an angle, an external number with no mechanism, refused on three separate grounds.

The closest owned exhibit, δ/(2π), recomputes to 0.250770 degrees — one thousandth of a degree below the band's low edge — and it is barred besides: δ is an adopted number, never derived, and an adopted toll does not become a sky amplitude by being divided by 2π after the fact. Everything the record gained in the day since the last attempt was audited for a new ingredient. Nothing arrived. So the honest result is a priced failure: there is no production, the cell ordered by the bench does not stand, and the missing object is one named thing — a slice-orientation law — with its three missing parts listed at full size. A forced number would have been easy to print. It would also have been a lie with a degree sign on it.

Program note

Paper 151 of the November 1, 2026 day charge — batch 2 of the writing bench, deposited the day it was written. Its verdict, as printed on its face: NO PRODUCTION — bench cell does not stand (KILL at the bench bar); board row H3b UNDECIDED

Read the paper on ZenodoAll papers

The program