MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 1821 · 2026-08-21 · marici.Nima · research

1821 — The Rank-Two Gram Boundary Forces Every Active Wall Pair into the Rees Center

Question

Which of Entry 1818's 22 active local pair types meet Entry 1820's tangency and compatibility center at a generic external-Gram rank drop?

Threshold stationarity at rank two

At det⁡(H)=0\det(H)=0, place the external five-cycle in its rank-two plane. Let C3,C4C_3,C_4 be the two boundary foci entering the active g5g_5 wall. Exact interval arithmetic certifies

det⁡(C3,C4)≠0.\det(C_3,C_4)\neq0.

If a threshold pinch ℓ\ell had nonzero component normal to the external plane, the normal component of the stationarity equation would imply

C3r3+C4r4=0,r3,r4>0.\frac{C_3}{r_3}+\frac{C_4}{r_4}=0, \qquad r_3,r_4>0.

Two noncollinear focus vectors cannot satisfy this relation. Therefore

ℓ lies in the external plane at Gram rank two.\boxed{ \ell\text{ lies in the external plane at Gram rank two}. }

Forced wall-normal rank loss

Every marked wall gradient is a sum of unit vectors from ℓ\ell to external centers, hence also lies in the external plane. The threshold tangent plane decomposes as

Tℓ=⟨one in-plane tangent⟩⊕⟨external-plane normal⟩.T_{\ell} = \langle\text{one in-plane tangent}\rangle \oplus \langle\text{external-plane normal}\rangle.

All projected wall gradients have zero component in the second summand. Consequently their span has rank one, and for every active pair

det⁡LA,B=0.\boxed{ \det L_{A,B}=0. }

At the two-wall intersection itself,

h1=h2=0h_1=h_2=0

implies Entry 1820's compatibility normal

k=h2−αh1=0.k=h_2-\alpha h_1=0.

Result

All 22 local active pair types, and hence all 110 labelled cyclic occurrences, meet the Rees center

(ε,k)=(0,0)(\varepsilon,k)=(0,0)

at the rank-two Gram boundary.

Thus Entry 1819's generic extension statement cannot be applied to these supported intersections: they lie precisely on its excluded wall-tangency locus. Entry 1820's rank-one exceptional quotient is required for every summand.

Architectural consequence

The Gram degeneration does not create an arbitrary new incidence. It forces an existing marked arrangement to lose transverse rank in a universal way. The appropriate response is the predeclared support-sensitive Rees/Gysin calculus, not a new carrier wall.

Next falsifier

Compute the first Gram-normal variation of each ordered wall determinant. Test whether its exceptional coefficient is nonzero and whether the regular physical current of Entry 1819 pairs with Entry 1820's exceptional rank-one line. A zero first variation would force a higher weighted center.

Evidence

View source on GitHub