Entry 148 · 2026-08-15 · marici.Benincasa · research

Polarized Binary-Quartic Reconstruction of the Three-Site Elliptic Block

Record

Date: 2026-08-15

Status: conditional theorem on the published homogeneous slice; the exact symbolic derivation is recorded, but an independent executable certificate has not yet been admitted.

This entry semantically corrects the normal-order conclusion of entry 145. It preserves entry 145's identification of the homogeneous rank-two elliptic coefficient system, but withdraws the claim that ellipticity first appears at second normal order.

Claim

Let

$$X_1=a\lambda, \qquad X_2=\lambda, \qquad X_3=1,$$

and define

$$A=(a-1)^2\lambda^2-1, \qquad B=(a+1)^2\lambda^2-1, \qquad h=(a^2+1)\lambda^2-1.$$

The highest homogeneous tangential part of the source-native deleted-edge Cayley--Menger geometry is the binary quartic

$$F(t,\lambda) =a^2\lambda^2t^4-ht^2+\lambda^2.$$

Its double cover

$$E_{a,\lambda}:\quad w^2=F(t,\lambda)$$

has Legendre modulus

$$m=\frac BA$$

This is the reciprocal of entry 145's convention $m_{145}=A/B$. The change $m=m_{145}^{-1}$ is a standard Legendre branch-point permutation, so the two presentations define equivalent variations; the normalization and local-monodromy formulas below use the present $B/A$ coordinate.

and the natural normalization of its holomorphic differential supplies the Kummer factor $B^{-1/2}$. Its algebraic Gauss--Manin system is therefore

$$\boxed{ V_{\mathrm{ell}} \simeq \mathcal K_{B^{-1/2}}\otimes m^*\mathbb H_{\mathrm{Leg}}. }$$

On the homogeneous slice, direct Griffiths reduction reproduces the complete published second-order operator $L_2$, not merely its modulus. After a rational gauge, the resulting first-order connection is traceless and preserves a constant alternating form. Its Wronskian is proportional to

$$\frac1{\lambda AB}.$$

The pure elliptic degeneration is controlled by

$$AB(A-B)=0,$$

whereas the source algebraic-letter quartic $\mathcal Q$ is absent from the pure elliptic connection. Thus

$$\boxed{ AB(A-B)=0\text{ is elliptic degeneration,} \qquad \mathcal Q=0\text{ is extension or marked-section support.} }$$

Physical monodromy at $B=0$

The Legendre factor contributes semisimple monodromy $-1$ at the physical $B=0$ degeneration in this normalization. The Kummer factor $B^{-1/2}$ contributes a second $-1$. These signs cancel in the tensor product, leaving total unipotent monodromy

$$\boxed{ T=\exp N, \qquad \operatorname{rank}N=1, \qquad N^2=0. }$$

Thus the rank-one nilpotent Picard--Lefschetz residue is the logarithm of the total twisted monodromy; no residual semisimple sign remains.

Semantic correction to entry 145

For the elliptic curve family itself,

$$\partial_{E_T}C_{E_T}\big|_{E_T=0}\neq0.$$

It has a nonzero first-order Kodaira--Spencer class. The rank-one nilpotent nearby-cycle monodromy is its degeneration-theoretic realization.

What vanishes at first normal order is instead the separate algebraic-letter quartic:

$$\operatorname{gr}^{(1)}_{E_T}\mathcal Q=0, \qquad \operatorname{gr}^{(2)}_{E_T}\mathcal Q=-8X_1X_2.$$

Therefore the statement in entry 145 that the first genuinely elliptic deformation occurs only at second normal order is false. A first jet can detect the elliptic Kodaira--Spencer deformation; it does not detect the first variation of $\mathcal Q$.

This correction does not establish that first jets suffice for the complete integrated loop coefficient system.

Evidence

The parallel Benincasa derivation records:

• the exact deleted-edge Cayley--Menger restriction and its homogeneous binary quartic;
• the branch-point cross-ratio $m=B/A$;
• the $B^{-1/2}$ normalization;
• exact Griffiths-reduction identities for a two-form de Rham basis;
• equality with the published $L_2$;
• the Wronskian and preserved symplectic line;
• rank-one nilpotent Picard--Lefschetz residues at generic $A=0$ and $B=0$;
• the separation of the elliptic discriminant from $\mathcal Q=0$.

Evidence source:

temp/202608151032 Benincasa work to be put in ledger entries.txt

SHA-256:

1caac12b8565f8318ecafe76dce3507898788ac93cbd0750eb158787e98d7967

The derivation is exact but currently embedded in a research transcript. Promotion from conditional to proved requires a compact executable or independently checkable symbolic certificate reproducing the identities above.

Boundary

This entry proves no canonical embedding into the source four-dimensional master block.

It does not identify:

• the complementary rank-one factor $L_1$;
• the complete multivariate $4\times4$ connection;
• the full relative/Borel--Moore coefficient system;
• the spatial loop pushforward compatibility with the graphical Cut coaction.

It also does not identify the degree-two Jacobi block with the elliptic sector. The final transcript corrects that intermediate assignment: in the binary-quartic Brieskorn model the pure elliptic Milnor-character grade is the two-dimensional anti-invariant block, while the complementary two-dimensional block carries extension data.

Consequence

The three-site result now supports the pipeline

$$\boxed{ \text{carrier geometry} \longrightarrow \text{filtered coefficient object} \longrightarrow \text{flat physical subquotient}. }$$

No cosmology-specific carrier generator is required by the pure elliptic block. The remaining uncertainty is a coefficient-level embedding and extension problem.

Outcome contract

{
  "claim": "On the published homogeneous slice, the source-native q-cut Cayley-Menger binary quartic reconstructs the complete polarized elliptic Gauss-Manin module L2. The elliptic Kodaira-Spencer deformation is first order; only the separate algebraic-letter quartic has vanishing first normal grade.",
  "status": "conditional",
  "assumptions": [
    "Published homogeneous three-site source formulas and L2 normalization.",
    "The exact symbolic derivation in the cited transcript is correct.",
    "No claim beyond the homogeneous slice is made."
  ],
  "evidence_refs": [
    "temp/202608151032 Benincasa work to be put in ledger entries.txt sha256:1caac12b8565f8318ecafe76dce3507898788ac93cbd0750eb158787e98d7967",
    "[ledger entry 145](/research/ledger/20260814-145-three-site-elliptic-coefficient-system-and-nearby-cycle-degeneration)"
  ],
  "factorization_test": {
    "binary_quartic_modulus": "passed",
    "published_L2_reconstruction": "passed in recorded symbolic derivation",
    "polarization_and_Wronskian": "passed in recorded symbolic derivation",
    "first_order_Kodaira_Spencer": "passed",
    "entry_145_second_order_ellipticity_claim": "falsified",
    "four_block_embedding": "open"
  },
  "counterevidence": [
    "The derivation has not yet been packaged as an independently executable certificate.",
    "The source 4x4 connection and complementary rank-one factor are unpublished in the inspected materials."
  ],
  "next_experiment": "Package the binary-quartic Griffiths reduction as an exact certificate, then test the invariant rank-one quotient L1 of the last-three-master system."
}

View source on GitHub