Standard Model structure from the bundle of Lorentzian metrics

Overview This paper asks how much of the Standard Model is fixed once one takes seriously a single geometric idea: that the fundamental configuration space is not spacetime itself but the bundle of all pointwise Lorentzian metrics over a four-dimensional spacetime. At each point of spacetime the set of possible metrics is a ten-dimensional space; assembling these fibres over the four-manifold gives a fourteen-dimensional total space Y¹⁴ → X⁴. The claim of the paper is that the internal structure of this object — its natural geometry, symmetries, spinors and topology — already encodes the gauge group, the fermion content, the number of generations, a tower of mass scales, and a list of testable predictions, with comparatively little put in by hand. The construction proceeds in a fixed chain. The natural ("DeWitt") metric on the space of metrics has signature (6, 4), forced by the requirement that the graviton carry positive energy. Its symmetry group has maximal compact part SO (6) ×SO (4), whose spin cover is exactly the Pati–Salam group SU (4) × SU (2) L × SU (2) R. The fibre spinors furnish precisely one chiral generation — a single 16 of Spin (10), including a right-handed neutrino. The reduction from Pati–Salam to the Standard Model gauge group is achieved geometrically by a discrete (ℤ₃) Wilson line supported on the spatial topology, with no Higgs phase transition and hence no monopole problem. Electroweak breaking is supplied by the metric section itself, which plays the role of the order parameter. The number of generations is then not a free choice but a group-theoretic invariant of the relevant exceptional structure (E₆), equal to three. A central methodological feature is stratification discipline: every claim is labelled by how much it earns — unconditional (pure representation- and lattice-theory, given a short list of structural inputs), branch-local (proved within a specified spectral-action class), forward (a parameter-free consequence), anchored (a quantity calibrated to data), or open / no-go. Each computational claim is backed by a committed verification script. The paper is careful to state what it does not derive as clearly as what it does. What is derived (the unconditional core) The construction rests on just two foundational postulates — that the configuration space is the bundle of Lorentzian metrics, and the noncommutative-geometry / spectral-action principle (the spectrum of the Dirac operator is the fundamental observable) — together with one structural identification: colour realised as a primitive A₂ inside an E₈ lattice summand (the E₈-singularity locus). From these, the following form a connected chain of machine-checked theorems, with no free discrete, topological, or generation parameter: the (6, 4) signature and the Pati–Salam gauge algebra; one chiral 16 per generation, with correct hypercharges; the Standard Model gauge group via the canonical ℤ₃ Wilson line; the generation count = 3, as the discriminant invariant of E₆. Two features that might look like extra assumptions are in fact consequences of the above, not independent inputs: the (6, 4) signature follows because the DeWitt fibre metric is the unique natural one (up to scale) at the general-relativity value, and the order-three (ℤ₃) structure is simply the centre of the observed colour group SU (3) — a canonical Wilson line, not a choice. The single remaining structural identification — colour realised at the A₂⊂E₈ locus — is the one place the construction reaches beyond its two postulates, and its status is now pinned down precisely. It is certified as the unique primitive lattice placement compatible with the observed colour SU (3) and three chiral generations: an exhaustive enumeration of the alternatives shows every other configuration fails one of these two requirements, so the identification carries no free geometric content beyond the observed spectrum — it is the geometric encoding of "colour SU (3) plus three generations, " in the same way the order-three structure reduces to the centre of SU (3). It remains an honest input rather than a dynamical output: the spectral-action vacuum energy, in both its static and full dynamical form, actually disprefers this locus (a clean no-go), with the only candidate selector being a non-static cosmological trapping effect outside the variational principle. The identification is therefore provably minimal — the unique placement the observed gauge group and generation count allow — but not energetically forced. A dedicated three-step analysis pins this single identification down to its sharpest form and certifies it irreducible within the framework. The colour A₂ is realised as the exceptional cycles of the gauge-enhancement region, which are simultaneously the massless gauge modes and the cohomology class — one object, not a coincidence. Demanding that the fibre matter and the cohomological matter be the same then forces the surrounding E₆ × SU (3) family structure, with colour living inside E₆, from the one-generation spinor and the index alone — no colour input. What remains is a single ℤ₃ identification — the colour-centre with the geometric boundary group — which is shown not to be forced (the two ℤ₃'s are distinct subgroups). So the entire discrete sector reduces to this one irreducible identification, equivalent to the observed spectrum; deriving it would require a fundamentally different (string/F-theory-type) framework in which the gauge group arises from the internal geometry. This is the honest floor: a single, sharply-characterised structural input, certified irreducible, with everything around it — topology, Wilson line, generation count, the index completion's existence — a theorem or a derivation. Conceptual perspective Beyond specific numbers, the framework reorganises several familiar puzzles by changing what is treated as fundamental: not fields on a fixed spacetime, but the bundle of all pointwise Lorentzian metrics. The gravitational field is the configuration space — a finite-dimensional, ultralocal cousin of Wheeler–DeWitt superspace. Gauge symmetry has a gravitational origin. The Standard-Model gauge group is not put in by hand; it is the symmetry group of the space of metrics at a point (the DeWitt group SO (6, 4), maximal compact SU (4) ×SU (2) ×SU (2) ). Internal gauge structure and spacetime geometry are two faces of one object. (The full gauge connection is not purely gravitational — a stated reach limitation — but the gauge group is. ) Singularities become signature-change surfaces, not endpoints. Where the metric degenerates — where a curvature singularity would live — lies at infinite geodesic distance in the natural DeWitt metric. A metric history does not terminate there: it crosses the degenerate locus, and the signature turns from Lorentzian (1, 3) to Euclidean (4, 0). The Big-Bang and black-hole singularities are reinterpreted as the surface where time becomes spacelike, rather than as a breakdown of the theory. The Wick rotation is physical, not a trick. That same signature change is the Euclidean continuation — the path integral becomes well-defined exactly where a metric section crosses into Euclidean signature. This unifies three things usually treated separately: the definition of the quantum theory, the resolution of singularities, and the Hartle–Hawking no-boundary state that fixes the universe's initial condition. Three generations is a topological invariant (disc (E₆) = 3; it would be 2 for E₇, 1 for E₈), not a fitted parameter. Symmetry breaking without a phase transition. The reduction to the Standard Model is a topological Wilson line rather than a Higgs transition — hence no monopole problem, with the metric section itself acting as the electroweak order parameter. These are reinterpretations of the framework's structure. Where they extend into predictions — the singularity-sector outcomes (B−L violation, no black-hole remnants) — those are flagged below and in the paper as conditional on the metric section crossing the degenerate locus. Retrodictions (known physics the construction reproduces) The complete ledger is in the paper's consolidated prediction tables; the entries below cover it, grouped by how much each earns. Parameter-free (no continuous input beyond the single cutoff scale): The Pati–Salam gauge group SU (4) ×SU (2) L×SU (2) R and its reduction to the Standard-Model group SU (3) ×SU (2) L×U (1) Y. One complete generation — a 16 of Spin (10), including a right-handed neutrino — and three generations, the fixed invariant disc (E₆) = 3. A Higgs mass ≈ 125 GeV given the electroweak VEV (a conformal-weight-four quartic threshold, λ (Λ) ≈ 0. 0176, run to low energy; observed 125. 20 ± 0. 11 GeV). The de Sitter sign of the cosmological constant (RF = −36, a positive vacuum energy). A geometric strong-CP solution: θ̄ = 0 at tree level by a left–right parity, radiatively safe — hence no axion is required. The Starobinsky inflationary spectrum nₛ ≈ 0. 964 (with the R² coefficient c = 125/4). The electroweak coupling ratio ρ = sin²θW / (α₂/α₃) ≈ 0. 717, used as a consistency audit at the derived cutoff. Forward in structure, with data-anchored magnitudes (the flavour/neutrino ledger consumes a few measured anchors — the ratio mₛ/md, |Vᵤs|, and the neutrino mass-squared splittings — counted honestly as anchors, not as parameter-free predictions): CKM matrix: Cabibbo |Vᵤs| ≈ 0. 222, |Vcb| ≈ 0. 041, |Vᵤb| ≈ 0. 0038, and the CP phase δCKM ≈ 1. 23 rad (observed ≈ 1. 20). Charged-fermion ratios: the Georgi–Jarlskog relation m_μ/mₑ = 9 mₛ/md (to ~15%), mc/mᵤ ≈ 588, and mb/mₜ ∼ 10⁻². PMNS matrix: a maximal atmospheric angle sin²θ₂₃ ≈ 0. 475, a reactor angle sin²θ₁₃ ≈ 0. 022 tied to the Cabibbo angle, and the solar angle sin²θ₁₂ ≈ 0. 307. The lightest active-neutrino scale m⏜䃓 ≈ 0. 05 eV, and dark-matter relic concordance ΩDM h² ≈ 0. 12 (for a reheating temperature 10⁹–10¹¹ GeV). Predictions (falsifiable, not yet me