# The order-46 centralizer augmentation gate **Date:** 25 October 2026 **Status:** quotient structures identified; scalar trace route subsequently exhausted by the full order-23 centralizer census **Evidence:** [centralizer-augmentation ledger](../../../evidence/data/moonshine-order46-centralizer-augmentation-v1.json) **Certificate:** [quotient-order verifier](../../../verification/verify/verify_moonshine_order46_centralizer_augmentation.mjs) ## Result The cyclic orientation no-go does not return the programme to an unbounded search. The smallest data outside the cyclic subgroup are finite and tiny. For an order-46 root \(h\), every element of \(C_{\mathbb M}(h)\) commutes with \(h\), hence with \[ g=h^2,\qquad z=h^{23}. \] Insertions differing by a power of \(h\) remain inside the cyclic modular orbit already exhausted. The genuinely new commuting directions are therefore indexed by \[ Q_h=C_{\mathbb M}(h)/\langle h\rangle. \] ATLAS centralizer orders give \[ |Q_h|= \begin{cases} 184/46=4,&h\in46A,46B,\\[2mm] 92/46=2,&h\in46C,46D. \end{cases} \] Thus the 2B/Leech rail has three nonidentity auxiliary cosets and the 2A/Baby-Monster rail has one. These are the complete minimal noncyclic insertion spaces at the order-46 junction. ## 1. Resolution through the two involution centralizers The quotients have a precise relation to the centralizer lifts already found. On the 2A rail, \[ C_{\mathbb M}(z)=2.B. \] The image \(\bar h\) has order 23 in \(B\), and its Baby-Monster centralizer has order 46. The cyclic subgroup generated by \(\bar h\) has order 23, so the remaining quotient has order two: \[ Q_{46CD}\cong C_B(\bar h)/\langle\bar h\rangle, \qquad |Q_{46CD}|=2. \] On the 2B rail, \[ C_{\mathbb M}(z)=2^{1+24}.Co_1. \] Projection sends \(h\) to an order-23 Conway element whose centralizer is exactly its order-23 cyclic subgroup. The kernel of the restricted centralizer projection has order eight, because \(184/23=8\). Since its central involution \(z\) already lies in \(\langle h\rangle\), \[ Q_{46AB}\cong \ker\!\left(C_{\mathbb M}(h)\to C_{Co_1}(\bar h)\right)/\langle z\rangle, \qquad |Q_{46AB}|=8/2=4. \] This identifies where the earlier order-eight lift residue goes: after the cyclic parity generated by \(z\) is removed, it leaves the four-coset augmentation space. Moreover, the kernel lies in the extraspecial normal subgroup \(E=2^{1+24}\). Since \(E/\langle z\rangle\) is elementary abelian, every subgroup of it is elementary abelian. Therefore \[ Q_{46AB}\cong C_2^2. \] The order-four ambiguity is consequently resolved without having to decide whether the order-eight preimage is dihedral or quaternionic. On the other rail, the order-two quotient is necessarily \(Q_{46CD}\cong C_2\). ## 2. The next trace family Choose a representative \(k\in C_{\mathbb M}(h)\) of a coset in \(Q_h\). Then \(k\) commutes with \(g\) and \(z\), so the next admissible observables are centralizer-resolved traces such as \[ Z(g,zk;\tau) =\operatorname{Tr}_{V^\natural_g} \!\left(zk\,q^{L_0-1}\right). \] The identity coset recovers the cyclic trace already shown to be orientation-blind. Only the nonidentity cosets need new evaluation: - one on the 46CD/2A rail; - three on the 46AB/2B rail. This is an upper bound on new trace types, not a promise that every coset gives a distinct function. Conjugacy inside the centralizer, projective characters or further modular relations may identify them. ## 3. The exact orientation test Fifth power exchanges each oriented class pair. Precisely because \(46A\) and \(46B\), and likewise \(46C\) and \(46D\), are distinct Monster conjugacy classes, this exchange is **not** implemented by an element of the Monster normalizer of \(\langle h\rangle\). If such an \(n\) existed with \(nhn^{-1}=h^5\), the two elements would be conjugate and the ATLAS classes would coincide. The transport to compare is instead the cyclotomic Galois or power-map transport on character values and twisted modules. Write \(\sigma_5\) for that transport and \(k^{\sigma_5}\) for the corresponding insertion label after a compatible choice of representatives. Orientation is recoverable at genus one only if at least one normalized auxiliary trace fails the Galois covariance equality \[ Z(h^2,h^{23}k;\tau) = \sigma_5^{-1}Z(h^{10},h^{23}k^{\sigma_5};\tau). \] The notation deliberately does not assert a canonical group isomorphism between the two centralizers. Such an isomorphism, or a common rational form of their character data, is part of the computation rather than input. The finite task is consequently: 1. enumerate the character orbits of \(Q_{46AB}\cong C_2^2\) and \(Q_{46CD}\cong C_2\) as they occur in the two actual centralizers; 2. compute the fifth-power Galois transport between the 46A/B and 46C/D character data, without replacing it by nonexistent Monster conjugation; 3. evaluate one plus three nonidentity-coset trace families; 4. quotient by all forced conjugacy and modular identifications; 5. declare orientation recovered only if a normalized function actually splits a power-five pair. If no split occurs, genus-one orientation recovery is exhausted for the full order-46 centralizers, not merely for their cyclic subgroups. ## 4. Relation to the seam The cardinalities two and four are tempting because the corpus contains binary seam and metaplectic data. No identification follows from the counts. The quotient \(Q_h\) is a finite group-theoretic insertion carrier. The genus-two seam is a characteristic orbit with analytic theta multipliers. A bridge requires an explicit character/intertwiner whose degeneration and modular action agree on both sides. The centralizer quotient is nevertheless the right source to test. It is the smallest extension that: - commutes with the established order-23 twist; - is not generated by the already exhausted cyclic root; - remembers the different 2A and 2B lift geometries; and - has a finite complete enumeration. ## 5. Subsequent scalar exhaustion The [full order-23 centralizer census](order-23-centralizer-exhaustion.md) closes the scalar question left open here. For \(g=h^2\), \[ C_{\mathbb M}(g)/\langle g\rangle\cong S_4. \] The quotients \(C_2^2\) and \(C_2\) found above sit inside centralizers of the two involution types of that \(S_4\). Every individual auxiliary insertion has order \(1,2,3\), or \(4\), coprime to 23, so together with \(g\) it generates a cyclic subgroup. Its generalized trace is therefore a modular branch of an ordinary series from one of \[ 23AB,\quad46AB,\quad46CD,\quad69AB,\quad92AB. \] All five A/B series pairs coincide. Thus the four nonidentity quotient cosets do not recover orientation as normalized scalar genus-one traces. The projective-centralizer audit further proves that the transgressed cocycle is trivial and the five traces determine the full ordinary \(S_4\) module, so no additional genus-one representation calculation is warranted. ## 6. Exhaustion boundary The quotient types close the label search space. The scalar failure is now established by the larger \(S_4\) census rather than left as a future conditional. No claim is made that \(Q_{46AB}\) is the Rule-60 deck carrier, that \(Q_{46CD}\) is fermion parity, or that either quotient acts on the ten even-characteristic genus-two family. Those are exactly the next intertwiner tests. The programme must move to cross-grade VOA composition or genus-two sewing. ## Sources and corpus links - [ATLAS Monster conjugacy classes and centralizer orders](https://brauer.maths.qmul.ac.uk/Atlas/v3/spor/M) - Breuer, Magaard and Wilson, [*Verification of the conjugacy classes and ordinary character table of the Monster*](https://arxiv.org/abs/2412.12182) - [Order-46 observable ladder](order-46-observable-ladder.md) - [Order-46 Galois orientation gate](order-46-galois-orientation-gate.md) - [Order-23 centralizer exhaustion](order-23-centralizer-exhaustion.md) - [Order-23 projective-centralizer no-go](order-23-projective-centralizer-no-go.md) - [Joined order-46 bridge](joined-bridge-map.md) - [Monster fermionization bridge](../lattice-subroutine-gravity/monster-fermionization-bridge.md)