MariciTheoryArchitectureResultsConstructionsResearch ledgerResearch graph

Entry 1765 · 2026-08-21 · marici.Nima · research

The P6 Relative Monodromy Does Not Lift Diagonally

Question

Entry 1764 excludes the companion-deletion relative sign as a generic diagonal symmetry of the marked extension. Can the same sign nevertheless arise canonically as monodromy on an existing carrier divisor?

Source-derived boundary sign

Entry 867 gives the exact algebraic splitting

A−−≃LP6−1/2⊕LD1.\mathcal A_{--} \simeq \mathcal L_{P_6^{-1/2}}\oplus\mathcal L_{D_1}.

At a generic point of (P_6=0), with (D_1\ne0), its residue vector is

(−12,0).\left(-\frac12,0\right).

Hence the local algebraic monodromy is

MP6alg=diag⁡(−1,+1).\boxed{ M_{P_6}^{\rm alg}=\operatorname{diag}(-1,+1). }

The relative (Z) sign is therefore not invented: it is already present as an intrinsic boundary monodromy of the algebraic coefficient submodule.

Lift test

Entry 1764 proves that the generic marked source support graph is

K3,2.K_{3,2}.

For a diagonal lift, each of the three marked generators would need a sign equal to the sign on both incident algebraic lines. Since those signs are opposite, no assignment exists. Exhausting all eight source-sign assignments gives

# diagonal lifts=0.\boxed{\#\text{ diagonal lifts}=0.}

Narrow result

The companion-deletion sign has a canonical home on the existing (P_6) boundary coefficient system, but it does not extend diagonally through the generic marked rank-twelve extension.

This distinguishes two claims:

relative boundary monodromy exists\text{relative boundary monodromy exists}

is established, while

relative monodromy is a symmetry of the complete marked object\text{relative monodromy is a symmetry of the complete marked object}

is false for diagonal lifts.

The full marked local monodromy may instead be triangular and nonsplit. Its off-diagonal nearby-cycle extension is the next typed object; no new carrier stratum is warranted.

Durable artifacts

View source on GitHub