The efficiency I solved for instead of deriving

Some corrections teach arithmetic, some teach method. This one teaches the difference between solving for and deriving, and it uses the cleanest example in the archive: a coefficient that was not computed, estimated, or approximated, but solved for from the answer it was supposed to predict.

The setup

The framework’s neutrino paper, 9b, needed a number called kappa, the efficiency coefficient converting a lepton-flavour asymmetry into the cosmic baryon asymmetry. The paper printed kappa as an exact value, derived from the category’s modular data, and used it to predict the baryon asymmetry, matching observation. Two derived-looking numbers, one impressive match.

The audit’s test

The test that exposed it takes one sentence: run the mechanism backwards. Take the observed baryon asymmetry, feed it through the mechanism in reverse, and see what efficiency it requires. The required kappa: the printed value, to every printed digit.

That is the signature of back-solving, and it is worth spelling out why it is damning rather than convenient. A derivation’s output has freedom: it could have come out higher, lower, or wildly off, and the fact that it did not is the content of the derivation. A back-solved coefficient has no freedom at all: it is the answer, dressed, wearing a derivation’s clothes. Its every decimal place was fixed by the observation it claims to predict, which means the prediction has zero information content. The match between predicted and observed, printed in the paper as a triumph, was guaranteed at the moment the coefficient was computed, and no experiment could ever have falsified it.

Where did it come from originally? The honest version: the framework needed the baryon asymmetry to come out, worked backwards through the leptogenesis chain to what efficiency would produce it, and then wrote the coefficient as if it had come forwards. The direction of the computation was reversed, and the direction is the whole epistemology.

The repair

The coefficient is now marked open in the ledger: the framework does not currently derive kappa. The baryogenesis claim built on it was archived, and the neutrino paper survives with its mass mechanism, which was never touched by the error. The direction test, solve for it and see if you get your printed value, is now standard in the audit protocol, applied to every derived-looking quantity before any paper cites it as a prediction.

The reason this corpse is one of the most instructive in the archive is its invisibility: the back-solved paper looked more derived than the honest ones, because its match was perfect. Small mismatches are the visible signature of genuine derivation; perfect matches with no freedom are its counterfeit. A reader checking this programme’s claims should trust the papers with residual gaps more than the ones without, and this post exists to say that in print.

The next entry is a four-tier cascade that misses its target by fourteen and a half orders of magnitude, with fitted exponents all the way down: ten to the fourteen point five.