All 35 Asymmetric Disjoint-Cut Proper Pairs Fail the Four-Focus Landau Test
Sequence claim: seqclaim-c2fbfa1e853ab370e35f9af9.
Frozen systems
Entry 1261 leaves 35 source-compatible proper-wall pairs whose two labelled cut supports are disjoint. Each pair therefore has four distinct focus occurrences and two independent cut roots .
For every labelled pair, reconstruct from Entry 1257:
- the four ordered focus vectors;
- the full routing Gram matrix;
- its nonzero determinant;
- the realization equation for the common integration point;
- the three minors imposing collinearity of the two wall gradients.
No cyclic representative or inherited distance is used.
Exact staged elimination
First eliminate between the realization equation and each of the three collinearity minors. Then eliminate between pairs of the resulting necessary conditions.
For every one of the 35 systems, at least one second-stage resultant is
A common Landau solution would have to annihilate every staged necessary condition. The unit certificate therefore excludes such a solution.
Consequence for all proper--proper pairs
The 150 source-compatible proper--proper pairs now split as
Thus no proper--proper pair produces anomalous two-wall support on the asymmetric physical family.
Across the full 245-pair census, 175 pairs are now classified. The only remaining class is the 70 pairs containing one one-cut total wall.
Artifact
research/benincasa/marici-gm/src/bin/five_site_asymmetric_disjoint_cut_pairs.rsresearch/benincasa/results/five-site-asymmetric-disjoint-cut-pairs.json
Next falsifier
Derive the correctly typed endpoint/stationarity equations for the 70 one-cut-total pairs and eliminate them without treating the one-cut total wall as an ordinary two-focus proper-region wall.