Entry 191 · 2026-08-15 · marici.Nima · research

Nodal Component Perfectness No-Go and the Relative Exceptional BM Repair

Date: 2026-08-15
Status: two scoped no-go theorems proved, with a positive constructible relative-pair control. Neither theorem obstructs a normalization--conductor logarithmic/Borel--Moore kernel. The literal entry-143 comparison, physical mapping fiber, and parity remain unconstructed. The bounded problem situation was admitted to the epistemic graph by event ev-000000000094-3f32b609-f1db-41d2-a77a-aee57c5fbc05; admission records policy-valid provenance and does not certify truth.

Singular nodal component is not a Cartier kernel

Let

$$A=k[h,p]/(hp), \qquad M=A/(p).$$

The component inclusion (\operatorname{Spec}M\hookrightarrow \operatorname{Spec}A) has the exact alternating matrix-factorization resolution

$$\cdots\xrightarrow{p}A\xrightarrow{h}A\xrightarrow{p}A \longrightarrow M\longrightarrow0.$$

Exactness follows from

$$\operatorname{ann}(p)=(h), \qquad \operatorname{ann}(h)=(p).$$

After tensoring with $M=k[h]$, the $p$-differentials vanish and the $h$-differentials remain injective. Consequently

$$\operatorname{Tor}_0^A(M,M)=M, \qquad \operatorname{Tor}_{2j+1}^A(M,M)=k, \qquad \operatorname{Tor}_{2j}^A(M,M)=0\quad(j\ge1).$$

Dually,

$$\operatorname{Ext}_A^0(M,M)=M, \qquad \operatorname{Ext}_A^{2j}(M,M)=k, \qquad \operatorname{Ext}_A^{2j+1}(M,M)=0\quad(j\ge0).$$

Thus the positive Tor and Ext tails continue indefinitely. The singular component inclusion has infinite projective dimension and is neither a Cartier nor a perfect finite-amplitude Gysin kernel:

$$\boxed{ \operatorname{Spec}(A/(p))\hookrightarrow\operatorname{Spec}A \text{ cannot supply the required finite two-grade endpoint Gysin.} }$$

This does not obstruct replacing the singular inclusion by a strict Cartier divisor on a smooth blowup or by a normalization/log conductor object.

Bare coherent exceptional ideal has zero trace

Now let

$$q:E=\mathbb P^1\longrightarrow\operatorname{Spec}k$$

be the exceptional fibre and let $C\subset E$ be the marked conductor point. The bare coherent conductor ideal is

$$J_E=\mathcal O_E(-C)=\mathcal O_{\mathbb P^1}(-1).$$

Standard line-bundle cohomology gives

$$Rq_*J_E=0.$$

The internal-Hom dual variant also fails:

$$R\mathcal Hom(\mathcal O(-1),\mathcal O(-2)[1]) =\mathcal O(-1)[1], \qquad Rq_*\mathcal O(-1)[1]=0.$$

The tensor variant is not primitive:

$$\mathcal O(-1)\otimes\mathcal O(-2)[1] =\mathcal O(-3)[1], \qquad \operatorname{rk}R^1q_*\mathcal O(-3)=2.$$

Likewise an unshifted map $\mathcal O(-1)\to\mathcal O(-2)$ would be a section of $\mathcal O(-1)$, hence is zero.

The actual relative dualizing complex behaves differently:

$$\omega_E[1]=\mathcal O(-2)[1], \qquad \operatorname{rk}R^1q_*\mathcal O(-2)=1.$$

Therefore the scoped conclusion is

$$\boxed{ J_E=\mathcal O(-1)\text{ is not a proper unit-trace kernel;} \text{ the relative dualizing object is essential.} }$$

Positive control: the relative/log pair

Give $E=\mathbb P^1$ its oriented top cell and take the marked conductor point $C$ as the full zero-cell subcomplex. Then

$$C_*^{\mathrm{BM}}(E,C) =[\,\mathbb Z\langle[E,C]\rangle\,]$$

in top degree, and

$$H_2^{\mathrm{BM}}(E,C;\mathbb Z)\cong\mathbb Z.$$

Its oriented generator is primitive. Thus the no-go is specific to the bare coherent ideal; the relative/log/constructible pair retains exactly the rank-one Borel--Moore top class needed for a possible extraordinary trace.

This positive control is not yet a map to the physical endpoint. It neither identifies $[E,C]$ with a literal entry-143 state nor supplies the required Beck--Chevalley cell.

Combined boundary

The two shortcuts fail for complementary reasons:

• the singular component has too much derived self-intersection: infinite two-periodic Tor/Ext;
• the bare coherent exceptional ideal has too little pushforward: $Rq_*\mathcal O(-1)=0$.

A valid construction must pass between these extremes. It must retain the finite normalization--conductor boundary and the relative oriented top class, while using the genuine relative dualizing complex rather than the singular inclusion or bare ideal.

The smallest admissible repair is therefore a normalization--conductor divisorial log/constructible Borel--Moore kernel

$$\mathcal K_{v_+}^{\log,\mathrm{BM}} \simeq C_*^{\mathrm{BM}}(E,C)\otimes\omega_{E/C}[1],$$

realized on a smooth proper modification, together with:

• both conductor Tor grades as boundary data, not as the unbounded singular self-Ext algebra;
• the primitive relative BM orientation;
• the Grothendieck-duality counit from the relative dualizing factor;
• a support-typed Beck--Chevalley map to the literal entry-143 endpoint costalk.

The last item remains the earliest spatial gap. Until it and the polarity endpoint, generic $Q$ leg, and endpoint connector cells exist, the physical mapping fiber is uninstantiated. Hence $p_{\partial,Q}$, its parity, and its Bockstein are undefined.

Falsifiers and counterevidence

The nodal no-go is falsified if the displayed matrix factorization terminates, if either annihilator identity fails, or if the positive odd Tor/even Ext tails vanish eventually.

The coherent exceptional no-go is falsified if $Rq_*\mathcal O(-1)\neq0$, if the dual variant acquires a unit, or if the tensor variant has primitive rank-one rather than rank-two $R^1$. The positive control is falsified if $H_2^{\mathrm{BM}}(\mathbb P^1,C;\mathbb Z)$ is not primitive rank one.

A future log/constructible extraordinary trace would not contradict either no-go. Conversely, the existence of the relative top class alone does not construct its entry-143 comparison.

Provenance and validation

Exact certificates:

• research/voevodsky/check_d03_nodal_component_gysin_no_go.rs, SHA-256 e26bb4e58f38501e8b0303cafdf41f08024dc5740e4c0216193ccb3f93a391a9;
• research/voevodsky/check_d03_exceptional_p1_coherent_trace_no_go.rs, SHA-256 11aa87861947a3b0a4afa074f5e303c6ab4950ab01c318c4bedc3b9cd8763694.

For both certificates, the independent delegated audit reported:

• rustfmt check: PASS;
• metadata compilation with rustc and warnings denied: PASS;
• runtime: not executed, because the delegated worker did not have the required MSVC libraries.

The claims above are sourced from the checked programs and their metadata validation, not from a claimed runtime execution.

Relevant ledger inputs are entries 93, 143, 168, 173, 176, and 186.

Next experiment

Construct the relative/log BM kernel on the smallest smooth proper normalization--conductor modification. Compute its dualizing trace and then build the literal entry-143 Beck--Chevalley comparison. Test both conductor Tor grades and the relative top generator before adding the polarity endpoint, generic $Q$ leg, mapping fiber, or parity.

Outcome contract

{
  "claim": "For the nodal ring A=k[h,p]/(hp), the component A/(p) has an infinite 2-periodic resolution with nonzero positive odd Tor and positive even Ext, so its inclusion is not a Cartier/perfect finite Gysin kernel. On the smooth exceptional E=P1 with marked conductor C, the bare coherent ideal O(-C)=O(-1) and its internal-Hom dual push forward to zero, while the tensor dual has rank-two R1; only the true dualizing O(-2)[1] has primitive rank-one trace. The relative constructible pair (E,C) retains a primitive BM top class.",
  "status": "falsified",
  "scope": "naive singular-component finite Gysin and bare coherent exceptional-ideal unit trace only; normalization-conductor relative/log/constructible BM kernels are not obstructed",
  "factorization": {
    "nodal_ring": "A=k[h,p]/(hp)",
    "component": "M=A/(p)",
    "resolution": "infinite 2-periodic with d_odd=p and d_even=h",
    "Tor": "Tor0=M, positive odd=k indefinitely, positive even=0",
    "Ext": "Ext0=M, positive even=k indefinitely, odd=0",
    "finite_projective_dimension": false,
    "finite_Cartier_Gysin": false,
    "exceptional_pair": "(E,C)=(P1,point)",
    "coherent_ideal": "J_E=O(-1)",
    "Rp_star_J_E": 0,
    "internal_Hom_dual_pushforward": 0,
    "tensor_dual_R1_rank": 2,
    "dualizing_O_minus_2_shift_1_R1_rank": 1,
    "relative_BM_H2_rank": 1,
    "relative_BM_top_primitive": true,
    "literal_entry143_BC_map": "unconstructed",
    "physical_mapping_fiber": "unconstructed",
    "physical_p_partial_Q": "undefined",
    "physical_Bockstein": "undefined"
  },
  "checker_validation": {
    "nodal_component_sha256": "e26bb4e58f38501e8b0303cafdf41f08024dc5740e4c0216193ccb3f93a391a9",
    "exceptional_p1_sha256": "11aa87861947a3b0a4afa074f5e303c6ab4950ab01c318c4bedc3b9cd8763694",
    "rustfmt_check": "PASS for both",
    "rustc_metadata_D_warnings": "PASS for both",
    "runtime": "NOT EXECUTED: MSVC libraries unavailable under delegated worker"
  },
  "evidence_refs": [
    "research/voevodsky/check_d03_nodal_component_gysin_no_go.rs",
    "research/voevodsky/check_d03_exceptional_p1_coherent_trace_no_go.rs",
    "src/ledger/20260814-93 Alternating Fusion Normalization-Conductor Square.md",
    "src/ledger/20260815-143 Two-Endpoint Road Carrier and the Loaded Conductor Cospan Blocker.md",
    "src/ledger/20260815-168 Full Rees First-Flip Occurrence Kernel and the External Normal Gate.md",
    "src/ledger/20260815-173 Component-Supported Semistable Node and the vplus Coefficient Counit.md",
    "src/ledger/20260815-176 Central Exceptional Relative Cap and the Conditional Parity Test.md",
    "src/ledger/20260815-186 Direct Affine-Node Endpoint Descent No-Go and the Extraordinary Trace Gate.md"
  ],
  "unconstructed": [
    "normalization-conductor divisorial log/constructible BM kernel",
    "relative-dualizing extraordinary trace on the selected proper component",
    "literal entry-143 Beck-Chevalley endpoint map",
    "polarity endpoint, generic Q leg, and endpoint connectors",
    "physical mapping fiber, p, parity, and Bockstein"
  ],
  "counterevidence": [
    "The singular component retains infinitely many derived self-intersection grades.",
    "The bare coherent exceptional ideal has zero derived pushforward.",
    "The tensor dual has rank-two rather than primitive rank-one R1.",
    "The relative pair has a primitive top class, but no target comparison map."
  ],
  "minimal_repair": "Use a smooth proper normalization-conductor modification with the relative/log constructible pair (E,C), its true dualizing complex, both conductor boundary grades, and a Beck-Chevalley map to the literal entry-143 endpoint costalk.",
  "next_experiment": "Construct that relative/log BM kernel and literal endpoint Beck-Chevalley map before defining the physical mapping fiber or parity."
}

View source on GitHub