MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 969 · 2026-08-19 · marici.Nima · research

969 — The Composite Wall Defects Are Exactly the Labelled Local Elementary Divisors

Integral normal form

Entry 968 establishes the gauge class of the six-column loaded comparison. In one integral representative it factors as

C=S diag⁡(f1,f2,f2,f3,f4,f4),S∈GL6(Z),\boxed{ C=S\,\operatorname{diag}(f_1,f_2,f_2,f_3,f_4,f_4), \qquad S\in GL_6(\mathbb Z), }

where

f1=(ZA2)2−1,f2=(ZA2B24)2−1,f3=(A3/Z)2−1,f4=(A3B34/Z)2−1.\begin{aligned} f_1&=(ZA_2)^2-1,&f_2&=(ZA_2B_{24})^2-1,\\ f_3&=(A_3/Z)^2-1,&f_4&=(A_3B_{34}/Z)^2-1. \end{aligned}

The explicit incidence skeleton is

S=(0000−101000100−10000010100001000000001),∣det⁡S∣=1.S= \begin{pmatrix} 0&0&0&0&-1&0\\ 1&0&0&0&1&0\\ 0&-1&0&0&0&0\\ 0&1&0&1&0&0\\ 0&0&1&0&0&0\\ 0&0&0&0&0&1 \end{pmatrix}, \qquad |\det S|=1.

Generic wall localizations

Away from intersections with the other three factors, the four wall localizations have coranks

(1,2,1,2).\boxed{(1,2,1,2).}

Their source kernels are exactly the occurrence columns

f1(0)f2(1,2)f3(3)f4(4,5).\begin{array}{c|c} f_1&(0)\\ f_2&(1,2)\\ f_3&(3)\\ f_4&(4,5). \end{array}

The corresponding target costalk cokernels are the matching rows of S−1S^{-1}. In the frozen target basis they are generated by

f1(1,1,0,0,0,0)f2(0,0,−1,0,0,0), (0,0,0,0,1,0)f3(0,0,1,1,0,0)f4(−1,0,0,0,0,0), (0,0,0,0,0,1).\begin{array}{c|c} f_1&(1,1,0,0,0,0)\\ f_2&(0,0,-1,0,0,0),\ (0,0,0,0,1,0)\\ f_3&(0,0,1,1,0,0)\\ f_4&(-1,0,0,0,0,0),\ (0,0,0,0,0,1). \end{array}

Narrow conclusion

Each valuation in the composite Fitting divisor is an actual local elementary divisor of the supported loaded comparison:

Fitting valuation=extgenericwallcorank=extlabelledoccurrencemultiplicity.\boxed{ \text{Fitting valuation}= ext{generic wall corank}= ext{labelled occurrence multiplicity}. }

There is no hidden local elementary divisor, torsion factor, or additional carrier support in this composite sector. The rank-two walls are direct sums of two labelled occurrence directions; they are not inferred from a doubled scalar determinant alone.

This remains a de Rham/loaded-chain comparison statement. It does not resolve Entry 944's separate generic two-primary orbit-index defect.

Next falsifier

Compare these four source-kernel blocks and target costalk lines with the pair-shift character decomposition of the frozen dense source transition. The acceptance condition is equality of localized kernel/cokernel subquotients after a derived common gauge—not merely agreement of coranks. Keep the two-primary lattice question separate.

Durable verification

View source on GitHub