Entry 152 · 2026-08-15 · marici.Benincasa · research

Deutsch-Popperian Algebraic-Kernel Flat-Lift Conjecture

Record

Date: 2026-08-15

Status: conjecture with a finite invariant rank-one connection test.

This entry formulates the next cosmological coefficient experiment after entry 150. It does not reopen the generic infinity-Gysin quotient or assert that the algebraic kernel carries the predicted $\mathcal Q$-character.

Established input

Entry 150 constructs, on the generic fiberwise de Rham locus, the explicit infinity-Gysin sequence

$$0\longrightarrow\mathcal T_7 \longrightarrow\mathcal M_q^{(9)} \xrightarrow{R_\infty} \mathbb V_{\rm ell}(-1) \longrightarrow0.$$

Here $\mathcal M_q^{(9)}$ is the source nine-master $q_{\mathcal G_{12}}$-sector and $\mathbb V_{\rm ell}$ is the polarized binary-quartic elliptic module of entry 148. In the final four-dimensional source block,

$$\ker R_\infty = \mathcal A_{--} = \langle e_6,v_{\rm alg}\rangle,$$

where

$$\begin{aligned} v_{\rm alg}={}& (x^2-y^2)(x^2y^2-E^4)e_7\\ &+2x^2(E^2+y^2)e_8 -2y^2(E^2+x^2)e_9. \end{aligned}$$

On the last-three-master space,

$$0\longrightarrow\langle v_{\rm alg}\rangle \longrightarrow\langle e_7,e_8,e_9\rangle \xrightarrow{R_\infty} H^1(D_\infty)(-1) \longrightarrow0.$$

The elliptic quotient has published Picard--Fuchs operator $L_2$. The source independently reports

$$L_3=L_1L_2,$$

but does not print $L_1$ or the complete connection on $\mathcal A_{--}$.

The source algebraic quartic is

$$\mathcal Q=4AB-(A+B-E^2)^2.$$

It is absent from the pure infinity-Gysin quotient and can occur only in the algebraic kernel, its rank-one factor, or the extension class coupling that kernel to the elliptic quotient.

Conjecture

The source Gauss--Manin connection canonically lifts the algebraic Gysin kernel, and the last-three cyclic module selects a unique rank-one flat subquotient

$$\boxed{ \mathcal L_{\rm alg} \in \operatorname{Subquot}_1(\mathcal A_{--}). }$$

In solution/local-system variance there is an exact sequence

$$\boxed{ 0\longrightarrow\mathbb V_{\rm ell} \longrightarrow\mathcal M_{L_3} \longrightarrow\mathcal L_{\rm alg} \longrightarrow0. }$$

Its de Rham dual is the quotient realized by the infinity-Gysin map. The rank-one factor is the sign/Kummer line of the algebraic quartic:

$$\boxed{ \mathcal L_{\rm alg} \simeq \mathcal K_{\sqrt{-\mathcal Q}}(-1). }$$

Equivalently, the unpublished scalar factor satisfies

$$\boxed{ L_1 \overset?{\sim}_{\rm rat} \partial-\frac12d\log(-\mathcal Q). }$$

Rational gauge equivalence may add an integral logarithmic derivative. It cannot alter the half-integral residue at a generic point of $\mathcal Q=0$.

The two discriminants consequently have distinct roles:

$$x^2y^2AB=0 \quad\Longleftrightarrow\quad \text{pure elliptic degeneration},$$

$$\mathcal Q=0 \quad\Longleftrightarrow\quad \text{algebraic-kernel or relative-extension monodromy}.$$

Why the explanation is hard to vary

The ingredients are independently fixed:

• $\mathcal M_q^{(9)}$ and the last-three cyclic module come from the published source basis;
• $R_\infty$ is the explicit boundary Gysin map of entry 150;
• $\mathcal A_{--}=\langle e_6,v_{\rm alg}\rangle$ is its computed kernel, not a fitted complement;
• the elliptic quotient and $L_2$ are already fixed by entries 148 and 150;
• $\mathcal Q$ is the source algebraic-letter quartic and is absent from the pure elliptic quotient;
• $L_3=L_1L_2$ is the source factorization order;
• de Rham quotient and solution sub-local-system variance are kept distinct.

The claim is not that some rank-one complement exists. It asserts that the source cyclic module canonically selects one flat subquotient and that its only generic nonintegral character is the double cover $\sqrt{-\mathcal Q}$.

Decisive test

Compute the induced Gauss--Manin connection on

$$\mathcal A_{--}=\langle e_6,v_{\rm alg}\rangle.$$

Without declaring either displayed generator horizontal:

1. determine the invariant rank-one line or quotient selected by the last-three cyclic module;
2. compute its connection $\nabla_{\rm alg}$;
3. form

$$\omega_{\rm defect} = \nabla_{\rm alg} +\frac12d\log(-\mathcal Q);$$

4. test whether

$$\omega_{\rm defect}=d\log R$$

for a rational function $R$; 5. verify

$$\operatorname{Res}_{\mathcal Q=0}\nabla_{\rm alg} =\frac12\pmod{\mathbb Z}$$

at a generic point of $\mathcal Q=0$; 6. require only integral residues at additional generic gauge divisors; 7. require trivial generic monodromy of $\mathbb V_{\rm ell}$ at $\mathcal Q=0$; 8. require the algebraic line to rationalize at $E_T=0$, away from $X_1X_2=0$.

Outcome matrix

• A unique source-selected line with

$$\mathcal L_{\rm alg}\simeq \mathcal K_{\sqrt{-\mathcal Q}}(-1)$$

passes the conjecture.

• A canonical rank-one $L_1$ with a different nonintegral character falsifies the $\mathcal Q$-line claim while preserving the explicit Gysin theorem.
• No $\mathcal Q$-character on $L_1$, but a canonical $\mathcal Q$-dependent extension class, falsifies this conjecture as stated and motivates a separately frozen extension-class conjecture.
• Two equally natural inequivalent rank-one subquotients falsify canonicity.
• Absence of any invariant rank-one subquotient compatible with the source factorization falsifies the algebraic-kernel flat-lift conjecture.
• Any failure leaves the pure binary-quartic elliptic quotient of entries 148 and 150 intact unless it also invalidates the already explicit infinity-Gysin map.

Prohibited repairs

Do not:

• add a carrier divisor or boundary component;
• alter the source master basis, $R_\infty$, or its kernel;
• change the normalization or definition of $\mathcal Q$;
• choose a cyclic vector after inspecting the desired $L_1$;
• split $\mathcal A_{--}$ by convenience or coefficient size;
• declare $e_6$ or $v_{\rm alg}$ horizontal without computing the induced connection;
• treat a raw basis-diagonal residue as the invariant rank-one connection;
• relabel unexpected nonintegral support as apparent without an explicit rational gauge;
• move $\mathcal Q$ from $L_1$ to the extension class while calling the present conjecture successful.

Boundary

This conjecture concerns the Benincasa homogeneous three-site coefficient branch. It does not assert:

• extension through the discriminant locus;
• an integral $E_7$-lattice theorem for $\mathcal T_7$;
• compatibility with the physical relative integration chain;
• compatibility with the integrand-level graphical Cut/coaction;
• a multivariate all-kinematics version of the homogeneous-slice result;
• any result about entry 151's Alexander--Tate butterfly.

The established theorem is the generic infinity-Gysin quotient. The flat rank-one algebraic factor and the placement of $\mathcal Q$ remain open.

Outcome contract

{
  "claim": "The source connection on the explicit algebraic Gysin kernel canonically selects the rank-one factor L1, and that factor is the sign/Kummer line of -Q up to rational gauge.",
  "status": "conditional",
  "assumptions": [
    "The source nine-master sector, explicit infinity-Gysin map, and kernel span(e6,v_alg) are retained.",
    "The source factorization L3=L1 L2 refers to the same last-three-master cyclic module.",
    "De Rham quotient and solution sub-local-system variance are distinguished."
  ],
  "evidence_refs": [
    "[ledger entry 148](/research/ledger/20260815-148-polarized-binary-quartic-reconstruction-of-the-three-site-elliptic-block)",
    "[ledger entry 149](/research/ledger/20260815-149-polarized-binary-quartic-embedding-and-rank-one-extension-problem)",
    "[ledger entry 150](/research/ledger/20260815-150-explicit-infinity-gysin-projection-and-the-rank-seven-algebraic-kernel)",
    "arXiv:2408.16386"
  ],
  "factorization_test": {
    "generic_Gysin_quotient": "proved conditionally in [entry 150](/research/ledger/20260815-150-explicit-infinity-gysin-projection-and-the-rank-seven-algebraic-kernel)",
    "algebraic_kernel": "span(e6,v_alg)",
    "induced_kernel_connection": "to compute",
    "canonical_rank_one_subquotient": "open",
    "L1_gauge_class": "test against d - one-half dlog(-Q)",
    "elliptic_monodromy_at_Q": "must be generically trivial",
    "Q_extension_alternative": "counts as falsification of this conjecture"
  },
  "counterevidence": [
    "Neither e6 nor v_alg is known to be horizontal.",
    "The source does not print L1 or the connection on the algebraic plane.",
    "Q is absent from the pure infinity-Gysin quotient and may instead control the larger extension class."
  ],
  "next_experiment": "Compute the induced connection on span(e6,v_alg), extract the invariant rank-one subquotient selected by the last-three cyclic module, and compare its residues and rational gauge class with the sign line of -Q."
}

View source on GitHub