The P6 Relative-Sign Extension Residue Is Gauge Removable
Question
Entry 1765 finds intrinsic relative monodromy
on the split algebraic submodule, but no diagonal lift through the marked extension. Can a nonsplit triangular nearby-cycle extension intrinsically attach the marked quotient to the (-1)-monodromy line?
Frozen diagonal residues
The source-derived wall quotient connection of Entry 855 has denominators
all coprime to (P_6). Hence at generic (P_6=0),
Entry 867 gives
in the ordered split algebraic lines
Hom-indicial calculation
For a regular triangular gauge
the degree-zero residue operator is
Its linewise ranks and kernels are
Therefore (L_0) is invertible on every residue column landing in the (-1)-monodromy line. Such a residue can always be removed by a regular triangular gauge.
Result
The only resonant local extension slot is the rank-three space landing in the (+1)-monodromy line (mathcal L_{D_1}). It cannot implement the selective companion deletion of Entry 1761.
Thus both proposed lifts of the relative sign are closed:
- diagonal lift: obstructed by Entry 1765;
- triangular logarithmic lift: gauge-trivial on the relative-sign line by this entry.
This is local at generic (P_6=0). Deeper intersections may have different indicial spectra, but they must be tested as existing-carrier intersections, not as new carrier strata.
Durable artifacts
- checker:
research/benincasa/checkers/p6_hom_indicial_gate.rs; - result:
research/benincasa/results/p6-hom-indicial-gate.json; - convention note:
research/benincasa/p6-hom-indicial-gate.md; - allocator claim:
seqclaim-8de3b82fb0f9ff13d8cbe1ca.