# Moonshine junction to the frontier-physics target set **Date:** 25 October 2026 **Status:** corpus crosswalk and ranked theoretical work programme; no physical promotion **Authority:** [frontier target set](../frontier-physics-targets/frontier_physics_target_set_v1.md) **Evidence:** [machine-readable crosswalk](../../../evidence/data/moonshine-frontier-crosswalk-v1.json) **Certificate:** [crosswalk verifier](../../../verification/verify/verify_moonshine_frontier_crosswalk.mjs) ## Result The Moonshine programme now touches seven maintained frontier targets, but it does so at three different distances. FPT-002 and FPT-019 can move immediately by computation: the scalar/genus-one quotient is already known, and the cost of each proposed refinement can be audited. FPT-007, FPT-009 and FPT-022 have exact finite carriers but still require either a physical source or an intertwiner. FPT-010 and FPT-017 remain comparisons only: the current chain does not produce a new positivity bound or a renormalization-group flow. This prevents two opposite errors. Moonshine is no longer isolated from the frontier lane, but neither is its mathematical closure counted as experimental or four-dimensional physical closure. ## Target map | target | exact asset contributed by the Moonshine branch | distance to target | next gate | stop rule | |---|---|---|---|---| | FPT-002 inverse detector design | explicit losses: \(23A/B\to T_{23}\), 22 charges to one charged character, \(\mathrm{Dic}_{11}\to D_{11}\), \(\chi_5\to\chi_{10}\) | immediate computational | find the least additional trace/insertion that separates the desired orientation after nuisance quotients | stop if every candidate is constant on the same fibres | | FPT-007 defects and boundaries | order-23 Tambara--Yamagami defect channel; order-46 2A/2B junction; ten-sector genus-two seam carrier | mathematical near, physical far | construct the twisted--twining/genus-two junction and state what it transports beyond quantum dimension or sector count | stop if equal defect indices admit scrambled orientation and seam data | | FPT-009 quantum-code and synthetic-gauge emulators | 529-label finite quadratic module, Fourier quarter-turn and order-44 \(\mathrm{Dic}_{11}\) instruction set | emulator-near | realize \(D_5\) and \(S\) from source-native controls and predict forbidden/reachable words | stop if only an arbitrary universal compiler realizes the group | | FPT-010 scattering, causality and positivity | explicit admissibility quotients and no-goes constrain what an observable may retain | comparison only | derive an inequality excluding an otherwise unitary, causal and passive response | stop if it is standard positivity in new notation | | FPT-017 scale-stable admissibility | a composable chain from characteristic-two code through number field, lattice and VOA | translation-near, RG-far | define pushforward measures/errors at three actual physical scales and test direct versus staged descent | stop if arrows are changes of mathematical category without coarse-graining | | FPT-019 computational admissibility | finite certificates for Rule 60, quadratic forms, modular label orbits, transition groups and carrier counts | immediate audit | publish input size, arithmetic cost, source dependence and missing analytic tables for every next calculation | stop any claim hiding full VOA construction or unrestricted character data as an oracle | | FPT-022 quotient scars | exact central-sign loss \(\mathrm{Dic}_{11}\to D_{11}\), orientation loss \(23A/B\to T_{23}\), and order-eight 2B-lift kernel | strong comparative asset | test whether an oriented record resolves the central/lift residue while scalar records remain identical | do not identify this \(C_2\) quotient with \(D_4\to G_2\) triality | ## The closest no-new-experiment tranche The next work should stay entirely on the mathematical side until it either produces a discriminator or fails: 1. enumerate the four oriented order-46 cases and their power-map labels; 2. modularly transport the known order-46 Hauptmoduln to \(Z(h^2,h^{23};\tau)\) and normalize them compatibly; 3. group the 46A/B and 46C/D results by their Conway and Baby-Monster centralizer lifts; 4. record the resulting no-go: fifth power fixes the involution and leaves each orientation pair scalar-degenerate; 5. identify the complete \(C_{23}\times S_4\) centralizer and place the \(Q_{46AB}\cong C_2^2\), \(Q_{46CD}\cong C_2\) quotient labels inside its five insertion classes; 6. close every single-insertion scalar route: the product orders \(23,46,46,69,92\) are cyclic and all A/B scalar series pairs agree; 7. construct the smallest matrix kernel from the surviving order-23 branches to the ten even genus-two characteristics; 8. reject the join if the modular representations or anomaly lines cannot intertwine. Steps 1--6 close the scalar part of FPT-002. The projective-centralizer audit then closes the remaining genus-one \(S_4\) representation route: the transgressed cocycle is trivial and the five class traces determine the module. Step 7 is therefore the FPT-007 higher-genus junction. Step 8 is useful even when negative: it converts a speculative bridge into a typed no-go. ## The finite emulator tranche The defect plane already supplies a compact synthetic-gauge target: \[ A=\mathbb F_{23}^2, \qquad q(a,b)=ab/23, \qquad S(a,b)=(b,-a), \qquad D_5(a,b)=(5a,14b). \] An emulator need not realize a Monster VOA. It must realize these two operations with independently calibrated controls, preserve the declared quadratic data, and demonstrate the predicted 44-element closure before projective readout and the 22-element closure after it. The decisive record is the central sign: coherent label readout must see it; projective readout must not. That is a source-restricted FPT-009/FPT-022 experiment. A generic qudit compiler reproducing the matrices on demand would verify software, not the proposed physical provenance. ## Where the physical path stops No present arrow maps the VOA grading variable to laboratory time, the Monster defect to a material defect, or the genus-two sewing modulus to a spacetime metric component. The Spin adjacency also remains finite and arithmetic: \[ \mathrm{Dic}_{11}