This paper develops an ab-initio electroweak normalization in which the SU(2)L and U(1)Z gauge couplings—and therefore the electromagnetic coupling e and fine-structure constant α—are fixed by a coherent pre-geometric background rather than calibrated to Standard Model electroweak data. A single background stiffness τ(ρ0), together with fixed projection weights selecting the weak and Cartan directions, defines the neutral kinetic structure. The photon arises as the unique massless neutral mode, and its coupling is extracted without introducing electroweak fit inputs. Using these UV “boxed” boundary conditions at the coherence/matching scale Λq, we evolve the couplings to MZ (and to the Thomson limit) with a two-loop EFT renormalization group including threshold matching and quantified internal theory bands. For the canonical product-UV (isotropic) ledger (ωL,ωZ)=(1,1) with τ(ρ0)≈0.178, the baseline prediction is α−1(MZ)≈140.8 and sin2θW(MZ)≈0.258, which motivates a refinement: the realized coherent vacuum cannot be isotropic. A diagnostic inversion identifies the required deformation of the neutral stiffness kernel, and Appendix J exhibits a concrete microscopic mechanism—coherent coupled dynamics of the morton x-pair—that naturally generates the needed anisotropy and yields UV handles consistent with the observed electroweak normalization α−1(MZ)≈127, sin2θW(MZ)≈0.231 in the Scenario-B closure. The paper also includes a first-principles lattice computation of vacuum polarization in an internal S[θ,Z] current theory (a controlled analogue used for internal consistency/closure, not a claim of full QCD reproduction). Code, run manifests, and data products used to generate figures and tables are provided in the accompanying repository. Revision note: This update resolves the coherent (x)-pair coordinate convention used throughout Appendices J and K, standardizes the kernel/Hessian normalization, and clarifies the role of the microscopic displacement scale in the Axis framework. The electroweak normalization and associated RG analysis are unchanged. The remaining displacement-scale problem is now explicitly separated from the electroweak closure and reformulated as an independent scalar-coherence and twist-fluctuation scale-selection problem. A finite variational closure equation for the displacement scale is introduced and audited, establishing the formal structure required for internal scale determination without electroweak matching or external normalization inputs. No changes were made to the derived electroweak observables, RG evolution, benchmark closure results, uncertainty estimates, or phenomenological conclusions.