Paper 9: the flavour pipeline

Paper 9 is the framework’s attempt at the hardest numbers in physics: the particle masses. It is also the second paper in the series to be written twice, after its first version failed, and the difference between the two versions is the difference between two ways of doing this kind of theory. The second version is much better.

Why mass is the wall

The Standard Model’s masses and mixings are numbers, not laws. The electron is about 206 times heavier than the muon is a fact of the same stature as Newton’s constant, and no current theory derives it. Any framework that claims to complete physics must eventually face these numbers, and most frameworks die on this hill, because the numbers have structure, hierarchies within hierarchies, that resists every clean mechanism. The topological category has no scale in it at all, so deriving mass from it means first importing a scale and then a hierarchy, and both imports are opportunities to fit rather than derive.

The failed first version

The first Paper 9 posited a power-law ansatz for the Yukawa couplings, with generation exponents built from the category’s data. It failed by a factor of sixty at the top: predicted top-to-charm of about 5 against an observed 300. The audit’s verdict was that the ansatz was a fit with too few parameters, presented as a derivation. Archived, with the usual reasoning: the failure mode, hierarchy from a one-parameter law, is the default mistake and needs a documented corpse.

The surviving second version

The new Paper 9 changes the mechanism from a power law to modular forms. The boundary theory is placed on a torus whose complex modulus tau parametrises the flavour sector, with the modular symmetry restricted to the level-6 subgroup matching the category’s centre. Yukawa couplings become modular forms of specific weights, and their size is controlled by the parameter q, the exponential of 2 pi i tau. At the framework’s fixed point, tau about 0.10 plus 1.22 i, q is about 4.7 times 10 to the minus 4, and the three generations with weights 4, 2 and 0 give couplings of order one, of order root q, and of order q. The ratios land at about 300 and about 400, against observed values of 300 and 400, and that hierarchy, which the power law missed by sixty, falls out of weights and a fixed point with nothing tuned.

The paper also derives the Cabibbo angle: a background flat connection induces a holonomy phase of pi over 14, and sine of pi over 14 is 0.2225, against the observed Cabibbo sine of 0.2250. That is a match to one per cent of a mixing angle that has resisted derivation for fifty years.

The honest caveats

Three, and they are big. The fixed point tau is input, not derived: the framework states it as a conjecture with a falsification route, a functional renormalisation group computation of the torus modulus from the bulk, and if that computation lands elsewhere, the mass ratios go with it. The modular form weights are chosen, not forced. And one per cent agreement on one angle is one number, not a fit to the full mixing matrix, whose other elements the paper gets only hierarchically. The status is theorem for the modular-form structure, conjecture for the fixed point, and the paper’s own text keeps those separate. It is the best mass mechanism in the programme, and its honest summary is that it turns the wall into a doorway that is still locked, with the key possibly in the bulk.

The next paper takes one particle out of that pipeline and does the arithmetic to the end: Paper 9b, neutrinos.