Conductor Normal-Link Fold and the Occurrence-Loaded Trace Boundary
Record
Date: 2026-08-14
Status: proved integral carrier theorem at the first conductor normal-link grade; the strict common-rank-one PC lift is falsified, while the intended noninvertible Gysin lift remains open.
Scope: the factorization-marked six-point normalization–conductor geometry
of entries 93–94, with its positive normal cones and declared normal
orientations. This result is governed by docs/nima-research-objective.md
and docs/research-lifecycle.md.
The source differential previously described as absent is canonical at this specific grade. It is the dual cellular/Cousin differential of the two positive projectivized conductor normal cones. The first unresolved arrow is therefore no longer an unspecified “scalar differential”. It is the occurrence-loaded coefficient/Gysin trace which must lift this carrier fold to the full PC complexes.
Claim
On the two normalization branches of entry 93, the conductor ideals are
[ J_+=(x_1,x_3,x_5), \qquad J_-=(x_0,x_2,x_4). ]
Their positive projectivized normal cones are canonically marked triangles
[ L_+=\mathbb P_+(N_{Z/F_+})\simeq\Delta^2, \qquad L_-=\mathbb P_+(N_{Z/F_-})\simeq\Delta^2. ]
These are normal-link vertex figures, not faces of the ordinary hexagon associahedron. Let (f_+,f_-) be their oriented dual top cells and let (e_0,\ldots,e_5) be the six sheet-resolved edge generators. The geometric top differential is
[ d_2(f_+)=e_1+e_3+e_5, \qquad d_2(f_-)=e_0+e_2+e_4. ]
It is derived from the two coordinate normal cones before using (K_{\rm alt}). With the independently scalar-derived QTDS incidence
[ d_1=C_{\rm QTDS}
\begin{pmatrix} 1&1&0&-1&-1&0\ 0&-1&-1&0&1&1\ -1&0&1&1&0&-1 \end{pmatrix} ]
and road augmentation (d_0=\epsilon=(1,1,1)), this gives the augmented normal-link carrier
[ C_{\rm link}: \mathbb Z^2\xrightarrow{d_2}\mathbb Z^6 \xrightarrow{C_{\rm QTDS}}P_{\rm road} \xrightarrow{\epsilon}\mathbf1. ]
It satisfies (d^2=0) integrally. There is a degreewise-surjective chain map to the entry-94 augmented triangle resolution
[ C_\triangle: \mathbf1_{\rm or}\xrightarrow{\Delta}P_{\rm tag} \xrightarrow{\partial_\triangle}P_{\rm road} \xrightarrow{\epsilon}\mathbf1 ]
given by
[ G_2=(1,-1), \qquad G_1=K_{\rm alt}, \qquad G_0=\operatorname{id}, \qquad G_{-1}=\operatorname{id}. ]
The two nontrivial chain identities are
[ K_{\rm alt}d_2=\Delta(1,-1), \qquad \partial_\triangle K_{\rm alt}=C_{\rm QTDS}. ]
The integral kernel is
[ \ker G_2=\mathbb Z\langle f_++f_-\rangle, ]
[ \ker G_1
\mathbb Z\langle e_0+e_3, e_1+e_4, e_2+e_5 \rangle, ]
with differential (1\mapsto(1,1,1)). Hence
[ 0\longrightarrow [\mathbb Z\xrightarrow{\Delta}\mathbb Z^3] \longrightarrow C_{\rm link} \xrightarrow{G}C_\triangle \longrightarrow0 ]
is a short exact sequence of integral complexes, and
[ H_1(\ker G)
\mathbb Z^3/\mathbb Z(1,1,1) \simeq A_2 ]
canonically through (\partial_\triangle). There is no torsion and no division by two or three.
One-step cyclic transport acts positively on roads and tags, by (e_j\mapsto-e_{j+1}) on polarity-loaded source edges, and by the signed sheet swap
[ f_+\mapsto-f_-, \qquad f_-\mapsto-f_+. ]
The target relation generator is invariant. All chain and fold squares commute for all six powers. Assigning a minus sign to the target top generator is incompatible with both (\Delta) and (G_2).
Evidence
The exact certificate is
research/voevodsky/check_conductor_normal_link_fold.rs.
It checks:
- construction of (d_2) from the two normal-coordinate triples without reading (K_{\rm alt});
- both square-zero identities and all three fold squares;
- saturated integral kernels and degreewise right inverses;
- Smith factors and the torsion-free homology calculation;
- the canonical identification (H_1(\ker G)\simeq A_2);
- all six cyclic powers and the forced target-top sign.
Reproduce with:
$src = "research/voevodsky/check_conductor_normal_link_fold.rs"
$exe = Join-Path $env:TEMP "marici-conductor-normal-link-fold.exe"
rustfmt --edition 2021 --check $src
rustc --edition=2021 -D warnings -O $src -o $exe
& $exe
Certificate SHA-256:
61ebadf9eb8e106c69833c912ec6667dd929547f86550d17ae440906a11f8718
The sharp coefficient negative control is
research/voevodsky/check_occurrence_pc_trace_obstruction.rs.
Reproduce with:
$src = "research/voevodsky/check_occurrence_pc_trace_obstruction.rs"
$exe = Join-Path $env:TEMP "marici-occurrence-pc-trace-obstruction.exe"
rustfmt --edition 2021 --check $src
rustc --edition=2021 -D warnings -O $src -o $exe
& $exe
Certificate SHA-256:
3e4dee2b54dcaeb6147d3d7cf9c431fd676ec71d0dd2902b1fb5065db157e6da
Inherited inputs are entries 20, 38, 66, 86, 93, and 94 and their cited certificates. In particular, entry 20 supplies (C_{\rm QTDS}), entry 93 supplies the two regular conductor embeddings, entry 38 supplies the nonresonant normal-cone/Cousin framework, and entry 86 supplies the marked physical endpoint counit.
Boundary
This theorem is a carrier and first-associated-grade statement. It does not yet construct the full scalar total-specialization differential (d_{\rm sp,sc}), nor the filtered map (G_{03}^{\rm Cousin}) required by the current formula objective.
In particular:
- (L_\pm) are positive projectivized normal cones/vertex figures. No literal triangle is asserted to be an associahedral face, and entry 84’s warning about a global barycentric representative and its factor (1/2) remains intact.
- The certificate retains the integral carrier but not the actual (y_i)-weights, occurrence modules, Koba–Nielsen monodromies, or forced lower Cousin terms.
- Entry 86 fixes the boundary values on the marked road occurrences. It does not by itself define the image of the target relation generator or a trace between the two branchwise loaded edge systems.
- No “scalar BRST” differential is introduced. Gauge BRST remains downstream in Yang–Mills descent.
- The fold is a quotient of complexes, not an equivariant splitting or a transition automorphism.
The first sharp missing map is the occurrence-loaded trace
[ \operatorname{Tr}{\rm occ}^{\rm PC}: \operatorname{PC}(L+;\mathcal L_+) \underset{P_{\rm road}}{\sqcup^{\mathbb L}} \operatorname{PC}(L_-;\mathcal L_-) \longrightarrow \mathcal R_{03}^{\rm circ,PC} ]
whose carrier grade is (G). It must simultaneously:
- include every normal and lower Cousin term;
- send the relation-level source to the (\Delta) generator;
- restrict to entry 86’s four unit (D=03) road occurrences;
- intertwine the PC differentials; and
- obey the physical-Cut/Beck–Chevalley square.
The established data prove neither existence nor nonexistence of this trace. They do, however, falsify its strongest strict replacement. Put
[ R_u=\mathbb Z[u_0,\ldots,u_5], \qquad K(u_j)=[R_u\xrightarrow{u_j}R_u], \qquad u_j=q_j-1. ]
The three tag pairs selected by (K_{\rm alt}) are
[ (u_2,u_5), \qquad (u_0,u_3), \qquad (u_1,u_4). ]
Suppose a pair folded strictly over the identity universal-monodromy base to one supported rank-one target (K(v_i)), with the unit coefficients forced by the carrier fold. The two chain equations require
[ v_i\mid u_j, \qquad v_i\mid u_{j+3}. ]
The paired universal monodromy variables are independent, so their greatest common divisor is one. Hence (v_i) must be a unit and the target loses its boundary support. Therefore
[ \boxed{ \text{no strict supported common rank-one target exists over }R_u. } ]
Nonresonant localization can manufacture maps by ratios, but it erases this support and does not canonically choose a target character. This negative control does not rule out the desired Gysin span: its two legs may pull one coefficient object back to distinct source characters.
The occurrence calculation locates the missing coherence. In the normalized (D=03) road square, write the four corners as
[ (v_{00},v_{10},v_{01},v_{11}). ]
Entry 86 gives the two selected-edge supports
[ p_+=v_{00}+v_{10}, \qquad p_-=v_{00}+v_{01}. ]
Their difference (v_{10}-v_{01}) has two exact lower-Cousin primitives, one through (v_{00}) and one through (v_{11}). Their difference is exactly the boundary of the road-square top cell. Thus the endpoint periods fix the derived null class, but they do not select the strict lower-Cousin primitive or the top coherence realizing (\Delta).
A nonzero occurrence-relative class of the forced (\Delta)-boundary would be a stronger canonical falsifier. Defining the target coefficient system so that the trace exists tautologically would not be a solution.
Consequence
The blocker in entries 93–94 is narrowed by one full categorical level:
[ \text{normal-link carrier differential} \quad\text{is now canonical,} ]
while
[ \text{occurrence-loaded PC trace and its Beck–Chevalley naturality} \quad\text{remain open.} ]
Thus the integral augmented triangle is not merely an algebraic pattern. It is a quotient of an independently constructed scalar conductor normal-link complex, and its kernel is exactly the QTDS contact (A_2) sector.
The next discriminating experiment is the single paired correspondence
[ Z_0\longleftarrow W_{03}\longrightarrow Z_3. ]
Construct it in the factorization-marked scalar geometry, specify both pullbacks on universal normal tori and occurrence cosheaves, and compute its PC trace on the two road-square Cousin primitives. Geometry must select the lower-Cousin representative and send their top-cell difference to the (\Delta) relation. Only after this (D=03) square commutes should it be rotated or the primitive Cut square be promoted to a full chain-level theorem.