A Non-Empty Parameter Region for a Foliation-Based Relativistic MOND Theory: Eleven Gates Passed Below the Galaxy Scale, and a Factor 1.49-1.99 Mass Deficit Above It

A search over 666,082,872 grid points per acceleration footing finds, for the first time in this programme, a non-empty admissible region for a covariant modified-gravity action built on a preferred foliation: general relativity's metric sector exactly, a khronometric clock scalar, and a MOND scalar sourced by the clock's acceleration and carrying a coherence (healing-length) operator. An explicit point in that region — KB = 0.2, c14 = 1.0 × 10−6, σ = 1.6793, c2 = 1.68 × 10−6, |K2| = 1.93 × 106, ξ = 0.10 pc (canonical) / 0.15 pc (alt), p = 1.7538, and no condensate — passes all eleven admissibility gates on both of the framework's acceleration footings (a0 = 9.3619 × 10−11 and 1.1279 × 10−10 m s−2), and no subset of those gates up to size four is empty. The theory reproduces Milgrom's law at the action's own a0 with coefficient exactly 1.000000, and with Φ = Ψ to better than 10−4 out to 1 Mpc, so lensing and dynamics share one potential with no slip and nothing fitted to make them do so. It passes the Solar-System gates (the Cassini quadrupole, the Pitjev-Pitjeva Saturn phantom-mass bound, the sunward-acceleration bound), the preferred-frame gates (α1 = −4.5 × 10−6, α2 = −2.0 × 10−7, α3 = 0, γ = 1), the tensor-speed gate (c13 = 0 identically, so cT = c exactly at every KB), Hadamard well-posedness, gravitational Cherenkov, and causality — the clock's leaves are spacelike for both cones, so no closed causal curve can be built even though the clock mode is 29.6% superluminal. It carries four propagating modes, all healthy, against the programme's own requirement of two; that requirement, read as a total count, fails. Above a galaxy the theory fails. Its single fatal instability was a clock tachyon sourced by the scalar condensate's background; curing it required setting the condensate amplitude to zero, and the condensate was the action's only dark component. The theory is consequently short in mass at R500 by a factor 1.618 ± 0.022 (canonical) / 1.493 ± 0.020 (alt) against hydrostatic masses and 1.987 ± 0.246 / 1.834 ± 0.227 against weak lensing, the two probes agreeing with each other at 1.55σ, so no lensing-sector repair can reach the deficit; its ΔΣ log-slope is +0.53 ± 0.06 shallower than measured, a 9σ error in the shape of the shear; the cluster-core acceleration excess of 3.37 a0 at 40 kpc exceeds the kernel's own parameter-free ceiling by 5.2×; and with no dark component the CMB acoustic peaks, σ8/S8 and the matter power spectrum are unaddressed. Significances taken from samples that carry coherent systematics are labelled statistics-only and their systematics-floor values are given: the cluster shortfall's 13-15σ becomes 4.4σ, the isolated binary-galaxy amplitude's 19.6σ becomes 7.2σ or 3.0σ, and the Coma ultra-diffuse-galaxy amplitude (a factor of 14) is 4.9σ, not the 19.4σ once quoted; in each case the direction survives and only the number falls. The relation a0 = κ c √(GρΛ) is fitted, not derived: κ = 0.4998 canonical and 0.6023 alt, so the tie does not even select a footing. Distinctive predictions are stated with numbers and falsifiers, among them a parameter-free ceiling on the acceleration excess and a deep-MOND baryonic Tully-Fisher zero point flat to better than 1% out to z = 5. Every number is produced by a committed script with checks that can fail (fable_independent_2026/L43_assemble_theory.py and its supporting lanes in the public repository). AI-assisted research draft; not peer reviewed.