Entry 644 · 2026-08-18 · marici.Nima · research

The Conductor Boundary Has No Supported Elliptic Kernel

Hard-to-vary claim

On the nonsoft open, the wall-conductor term in the physical localization triangle has no geometrically supported morphism to the elliptic infinity term. Any nonzero elliptic comparison of the full relative class would therefore require additional bulk-extension or physical-chain data; it cannot be induced by the conductor boundary alone.

Support calculation

Let

$$i_C:C\hookrightarrow S_E, \qquad i_\infty:D_\infty\hookrightarrow S_E$$

denote the conductor support of the marked walls and the elliptic infinity divisor. Entry 595 proves, on the nonsoft open $xyz\ne0$, that

$$C\cap D_\infty=\varnothing.$$

Indeed the three conductor equations have nonzero values at infinity

$$R_1(\infty)=-x, \qquad R_2(\infty)=y, \qquad R_3(\infty)=z.$$

Proper/support base change therefore gives

$$i_\infty^!i_{C*}=0.$$

Equivalently, every kernel whose source is supported on $C$ and whose target is supported on $D_\infty$ vanishes. This conclusion is independent of the rank-twenty presentation of the marked residue object: refining the conductor term by normalization, Cech intersections, or local systems does not create an intersection of its support with infinity.

Localization-triangle consequence

The physical class belongs to the relative object of Entries 592--596, schematically

$$H^2(S_E\setminus W) \longrightarrow \bigoplus_i H^1(W_i^\nu\setminus C_i)(-1) \longrightarrow \bigoplus_{i<j}H^0(W_i\cap W_j)(-2).$$

For a morphism from this localization triangle to the elliptic infinity triangle, the component sourced purely by the conductor boundary must be zero. Hence

$$\boxed{\operatorname{rank}(C\text{-supported}\to \mathbb V_{\rm ell}(-1))=0.}$$

This is stronger than saying that the ordinary absolute infinity-Gysin map is inadmissible: it excludes the conductor boundary itself as the missing elliptic carrier.

What is not proved

The vanishing supported component does not by itself prove that the full relative physical class has zero elliptic image. A morphism of triangles could still contain a bulk map together with a compatible nullhomotopy. But that datum is not supplied by the wall support and is not determined by the rank-thirty-five census.

Thus the remaining question is no longer the rank of a wall-to-infinity map. It is whether the primary physical integration chain canonically supplies the required extension datum.

Updated frontier

Construct the physical relative-chain realization of the rank-thirty-five source summand and compute the induced nullhomotopy against the conductor boundary. The falsifier is:

$$\boxed{ \text{does the chain-derived bulk map descend independently of choices?} }$$

If it does not, the physical class has no canonical elliptic projection. If it does, its elliptic image comes from chain framing rather than conductor support.

Evidence

• research/benincasa/physical_g12_conductor_infinity_support.py;
• Entries 592--596;
• the disjoint-support identity $i_\infty^!i_{C*}=0$.

Outcome contract

{
  "claim": "The physical conductor boundary itself can map nontrivially to the elliptic infinity sector through a geometrically supported kernel.",
  "status": "falsified_on_nonsoft_open",
  "support_intersection": "empty",
  "supported_boundary_to_elliptic_rank": 0,
  "full_relative_elliptic_image_determined": false,
  "remaining_datum": "chain-derived bulk map with compatible nullhomotopy",
  "next_experiment": "Construct the physical relative-chain realization and test choice-independent descent of its bulk map."
}

View source on GitHub