Character Match and the Vanishing Star-Costalk Map
Record
Date: 2026-08-14
Status: exact integral (n=8) carrier theorem. The global orientation line of entry 91, tensored with the independently established polarity deck line, has the abstract character required by entry 89’s (\chi_N). Ordinary graph restriction nevertheless produces neither the primitive road counit nor the six-term alternating-conductor lift. The missing arrow is genuinely loaded Cousin/Gysin–Verdier data.
Primitive global quotient
Retain the notation
[ S\subset H_1(\Gamma_8;\mathbb Z), \qquad L_{\rm or}=H_1(\Gamma_8;\mathbb Z)/S\cong\mathbb Z ]
from entry 91. In the checker’s spanning-tree chord basis, the primitive quotient functional is
[ \ell=(0,0,1,1,1). ]
It annihilates all four local square cycles, has content one, and satisfies
[ \ell(o)=2 ]
on the outer octagon. An explicit integral cycle (g) with (\ell(g)=1) exists, so the quotient is primitive; the outer octagon is not.
The complete (D_8) transport is
[ \chi_{\rm or}(r^ks^e)=(-1)^e. ]
Let (L_{\rm pol}) be the independent alternating-polarity line, with one-step deck transport
[ \chi_{\rm pol}(r^ks^e)=(-1)^k. ]
Then the candidate loaded line
[ L_{\rm load}=L_{\rm or}\otimes L_{\rm pol} ]
has the abstract local character
[ \boxed{ \chi_{\rm load} =(\text{road rotation},\text{road reflection},\text{core exchange}) =(+1,-1,-1) =\chi_N. }
This is a character identity, not yet an identification of coefficient complexes.
What the physical stabilizer can see
The actual (D_8) stabilizer of the channel (D=03) is
[ {1,f_3}. ]
The nontrivial element reverses the three roads and exchanges polarity at the same time. Therefore
[ \chi_{\rm or}(f_3)=-1, \qquad \chi_{\rm pol}(f_3)=-1, \qquad \chi_{\rm load}(f_3)=+1. ]
The stabilizer sees only the product. It contains neither a three-cycle of the roads, nor a pure road reflection, nor a pure polarity exchange. Thus its agreement cannot separately certify the two factors of (\chi_N); the polarity deck involution must remain independently typed.
Ordinary star restriction lands in the wrong sector
Let
[ P_D=\mathbb Z\langle q_3,q_4,q_5\rangle, \qquad A_{2,D}=\ker(\varepsilon_D:P_D\to\mathbb Z). ]
Restrict an oriented graph cycle to the three edges incident at the vertex (D=03). The graph boundary equation at that vertex gives
[ \boxed{ \operatorname{res}D H_1(\Gamma_8) \subseteq A{2,D}. }
The four square boundaries restrict as
[ (1,-1,0),\quad(0,0,0),\quad(0,0,0),\quad(0,-1,1), ]
and span the full saturated (A_{2,D}) lattice. Hence raw restriction does not annihilate (S) and does not define
[ L_{\rm or}\longrightarrow P_D. ]
Passing both sides to primitive quotients does make a formal quotient square, but every cycle already has augmentation zero. Therefore the induced map is
[ \boxed{ H_1(\Gamma_8)/S \longrightarrow P_D/A_{2,D} \quad=\quad0. }
It has image rank zero and cokernel rank one. Ordinary star restriction is not the primitive road counit.
This is the expected variance warning: graph cycles naturally restrict to the local circuit/difference sector. A road quotient can arise only after the appropriate duality/Gysin operation, not by forgetting the signs of a restriction map.
Representative dependence and the six supports
Two representatives of the same primitive global class, differing by one local square, have road restrictions
[ (0,1,-1) \qquad\text{and}\qquad (1,0,-1). ]
Under entry 66’s supported conductor embedding, these give distinct six-vectors
[ (-1,-1,0,1,1,0) ]
and
[ (0,-1,-1,0,1,1). ]
Thus neither the global quotient class nor its character selects the six supported coefficients.
Even at the stabilizer level,
[ \operatorname{Hom}{\operatorname{Stab}(D)} (L{\rm load},A_{2,D})\cong\mathbb Z ]
is generated by
[ (1,-2,1). ]
No scale or sign is selected. Moreover the polarity-forgetting map
[ [I_3\ I_3]:\mathbb Z^3_+\oplus\mathbb Z^3_-\to P_D ]
has a rank-three kernel and no nonzero integral polarity-equivariant section. Character matching therefore cannot manufacture (\boldsymbol\sigma_{\rm alt}).
Correct formula objective
The exact negative result rules out
[ \pi_D^{\rm PC} \stackrel{\rm false}{=} \text{ordinary restriction of the global orientation cycle}.
The still viable construction has two derived stages:
[ \boxed{ \text{contact relative complex} \xrightarrow{;G_D^{\rm Cousin};} \mathcal R_{D}^{\rm circ,PC} \xrightarrow{;\mathbb D\Delta_D^{\rm circ};} \mathsf J_D^{\rm road,PC}. }
The first arrow must retain occurrence weights, the polarity conductor, ordered normal/tangential lines, and the actual scalar kinetic differential. Its associated symbol is the candidate (\boldsymbol\sigma_{\rm alt}). The second arrow is the twisted road/circuit duality already typed at associated grade in entry 89.
The next falsifier is therefore not another character calculation. Construct one coefficient-loaded Cousin/Gysin map (G_{03}^{\rm Cousin}) and test:
- its six supported coefficients against entry 66;
- its chain identity against the scalar kinetic/BRST differential;
- its endpoint Cousin terms against entry 86;
- its factorization through the twisted dual circuit resolution of entry 89.
Exact certificate
Run:
rustfmt --check research/voevodsky/check_n8_orientation_costalk.rs
rustc --edition=2021 -D warnings -O research/voevodsky/check_n8_orientation_costalk.rs -o "$env:TEMP\\marici-n8-orientation-costalk.exe"
& "$env:TEMP\\marici-n8-orientation-costalk.exe"
Certificate SHA-256:
e451d7dc14502b1608d1ad121419d848142d7e5aeeb27b5886a1f9ac8290b26e
Internal dependencies
- Entry 59: the integral circuit resolution.
- Entry 66: the alternating-conductor coefficient symbol.
- Entry 86: occurrence-resolved endpoint counit.
- Entry 89: associated-grade twisted road/circuit pairing.
- Entry 91: the global orientation quotient and relative contact carrier.
research/voevodsky/check_n8_orientation_costalk.rs.