What if reality obeys a hidden structure?

Compute the field structure. Keep its rules fixed. Compare the constants and behavior that follow.

Abstract—Unified Maxwellian Field Theory is a proposed theory of the nature of reality. It takes energy and spacetime as mutually interacting dynamical fields, and seeks particles, forces, and matter as their organized behavior rather than as a fundamental inventory of subatomic constituents. Its Maxwell-inspired gauge structure draws on extended electromagnetic geometry, including Lindgren’s work, and on independently proposed return, winding, and boundary dynamics. The claim is that their composite equations can produce the numerical organization of reality, not merely redescribe it.

This paper begins with the computed consequences: a transverse eigenvalue near the inverse fine-structure constant, its unchanged reuse in the charged-lepton mass relation, atomic mode capacities, and fixed nuclear openings. It then follows the field construction into a solved charged core, its radiation channels, and the global meaning of half-winding. Each result retains its supplied inputs, measured comparison where one exists, and model-selection history. Agreements, independent predictions, and familiar identities are not interchangeable.

What comes out when the rules stay fixed?

The empirical starting point is a set of regularities: charged leptons share charge and spin despite a large mass hierarchy; atoms organize into repeatable mode families; and one dimensionless electromagnetic coupling reappears across otherwise different measurements. Established physics explains much of their behavior, but takes the charged-lepton masses and low-energy coupling as inputs rather than deriving their numerical values. UMFT asks whether interacting energetic and spacetime structure can make that numerical organization a consequence.

The transverse boundary calculation returns \(\alpha^{-1}=137.035999156447\). Reusing that coupling in the loaded lepton relation returns \(m_\tau=1776.938315\,\mathrm{MeV}/c^2\), from the supplied electron and muon masses. The tau comparison is \(1776.93\pm0.09\,\mathrm{MeV}/c^2\).²² No tau mass or measured alpha is read by this calculation.

These close numbers deserve to be visible. Their interpretation also depends on how the equations were selected: the \(3{:}10\) structure and transverse correction were developed with the measured organization already known. Reproducing their output with the target data removed demonstrates a fixed calculation, not retrospectively blind discovery. The stronger result sought is that the common field action selects those same coefficients and predicts additional observations without modification.

Three connected calculations carry the argument. Boundary geometry supplies a dimensionless coupling candidate; the return-amplitude rule uses that coupling in a mass relation; pressure–tension balance explains why coherent amplitudes would enter mass quadratically. They are evaluated separately so that a failure in one is not concealed by agreement in another.

The calculated numbers stay put.

The vertical line is the unchanged model output. Points and horizontal bars are separately loaded determinations and their reported one-standard-uncertainty intervals. The two panels have different units. The electron-moment point infers alpha using Standard Model theory; it is not a UMFT calculation of the magnetic moment.

Show comparison marks on the plot Also show the uncorrected 10:3 mass relation

calculated inverse alpha: 137.035999156447
calculated tau mass: 1776.938315 MeV/c²
transverse equation residual: loading calculation

Comparison data do not feed back into Equations (C1)–(C3).
Comparison	Measured / evaluated	Calculated minus comparison	Difference / reported uncertainty

Computing the model before loading its comparison data.

One coupling, reused without adjustment

The proposed composition has physical motivation: three transverse rest-space channels and ten components of a symmetric four-dimensional tensor organize primitive and composed returns. Turning those dimensions into coherent increments also requires equal channel normalization and an admissible boundary coupling. The following calculation tests the resulting fixed formulas; the component counts alone are not a derivation of their physical amplitudes.

The proposed primitive and composed return weights are \(3\) and \(10\). Their graph operator has nonzero eigenvalues \(13\pm\sqrt{79}\), so its compliance is \(C=13/45\). Matching \(4\pi^2/C\) to the odd spherical boundary spectrum selects \(b=137\):

\[ L=\begin{pmatrix}3&-3&0\\-3&13&-10\\0&-10&10\end{pmatrix},\qquad C=\frac1{13-\sqrt{79}}+\frac1{13+\sqrt{79}}=\frac{13}{45}, \] \[ b=\underset{2\ell+1,\;\ell\ge0}{\operatorname{argmin}} \left(2\ell+1-\frac{4\pi^2}{C}\right)^2=137. \tag{C1} \]

The half-winding junction and neighboring-sector correction define the retained scalar eigenproblem. With \(\rho=\sqrt{8/(\pi\Lambda)}\), its complete form is

\[ q=\frac{\rho}{1+\sqrt{1-\rho^2}},\qquad v=\frac{q}{4\pi^2},\qquad \Lambda=b+\frac{\pi^2/2-1/(4\Lambda)}{\Lambda} +\frac{v^2}{\Lambda-(b-2)},\qquad \alpha=\Lambda^{-1}. \tag{C2} \]

The solver starts from the selected integer branch, not a measured decimal value. Its output enters the mass relation unchanged:

\[ s=\sqrt{1-\frac{\alpha}{2\pi^2}},\qquad m_\tau=\left[\sqrt{m_\mu}+\frac{10}{3}s \left(\sqrt{m_\mu}-\sqrt{m_e}\right)\right]^2, \tag{C3} \] \[ m_e=0.51099895069\,\mathrm{MeV}/c^2,\qquad m_\mu=105.6583755\,\mathrm{MeV}/c^2. \]

The inputs in Equation (C3) are measured masses; neither is claimed as an output. The loading factor is fixed, not adjusted by the comparison controls. A small numerical residual does not by itself prove the physical origin of the weights, the junction normalization, or the selected boundary sector.

What would a sharper tau mass decide?

The loaded relation gives a fixed number that a future measurement can challenge. Exact Koide geometry gives a nearby but different number. These are alternatives, not simultaneous constraints: setting \(Q=2/3\), with \(a=\sqrt{m_e}\) and \(b=\sqrt{m_\mu}\), selects the heavy root

\[ m_K=\left[2(a+b)+\sqrt{3(a^2+4ab+b^2)}\right]^2 =1776.969027\,\mathrm{MeV}/c^2. \tag{K1} \]

It lies \(30.7123\,\mathrm{keV}/c^2\) above Equation (C3). Neither formula derives the electron or muon input mass. Their propagated input uncertainties are below \(0.041\,\mathrm{keV}/c^2\), even allowing every correlation between those inputs. A future measurement with \(5\,\mathrm{keV}/c^2\) total standard uncertainty could therefore resolve these fixed alternatives; the current reference average cannot.

Change the precision, not the prediction.

The curve shows the distance between the two fixed mass relations in units of a prospective measurement’s uncertainty, including the larger input-error bound. It assumes the future central value equals one relation. It is a planning calculation, not new data, a measured significance, or a guarantee that either relation is correct.

Total measurement uncertainty / keV/c²

Computing the two fixed mass relations and their shared-input uncertainty.

loaded relation / MeV/c²: 1776.938315
exact Koide / MeV/c²: 1776.969027
fixed separation / keV/c²: 30.7123

A testable number needs a controlled error

A published tau-threshold proposal projects approximately \(1\,\mathrm{keV}/c^2\) precision at a future high-luminosity facility. This is a sensitivity study, not an achieved measurement. Its fit includes tauonium; omitting that bound-state contribution shifts the fitted mass by about \(4\,\mathrm{keV}/c^2\). Beam calibration, energy spread, and the radiative and threshold model therefore matter at the precision needed here.²⁶

The displayed square root already includes every order of the fixed uniform-cycle kernel; truncating it at first order changes the mass by only \(0.046\,\mathrm{keV}/c^2\). A physical error bound must also address deformation of the junction profiles and other omitted interactions. No such full theory-error bound is assigned to the curve above. Standard QED in the experimental mass extraction is likewise a measurement assumption, not an additional UMFT derivation.

A result compatible with the loaded number would support that fixed conditional relation, not uniquely establish its field ontology. A result compatible with neither relation must remain a possible outcome. The equations and mass convention must be fixed before unblinding; adjusting a loading coefficient afterward would define a new prediction.

Agreement must survive the comparison set

The tau output is within \(0.10\) of the PDG average’s reported uncertainty. The alpha output differs from the CODATA 2022 recommendation by about \(-0.98\) reported uncertainties. The 2025 evaluations of cesium and rubidium recoil give discrepancies of approximately \(+4.13\) and \(-5.03\); the electron-moment inference using updated Standard Model theory gives \(-0.44\).²³²⁷ The calculation is numerically close, but does not agree with every precision determination at its quoted uncertainty.

The recoil points reevaluate the 2018 cesium and 2020 rubidium experiments with updated conversion constants; they are not new experiments.²⁴²⁵ The electron-moment point uses the 2023 measurement with revised QED and hadronic contributions. Agreement with that inferred coupling does not establish that UMFT predicts the magnetic moment. Recoil extraction also inherits external mass and spectroscopic inputs. These dependencies matter for a theory proposing to replace the underlying description.

CODATA is an adjusted recommendation, not a third independent experiment to count alongside its inputs. These discrepancies are descriptive comparisons, not discovery significances: the construction has no established theory uncertainty or quantified model-selection penalty. Changing a coefficient to follow a preferred measurement would be a new model and would leave this comparison intact.

Why square-root mass is the natural coordinate here

At fixed coherent return amplitude \(x\), the reduced closure energy balances Maxwell compression against quartic return tension. The same two coefficients apply to every amplitude:

\[ U(R,x)=\frac{\mathcal A}{R}+\mathcal T x^4R, \quad R_* =\sqrt{\frac{\mathcal A}{\mathcal T}}\,x^{-2}, \quad U_*=2\sqrt{\mathcal A\mathcal T}\,x^2, \quad U_*R_*=2\mathcal A. \tag{C4} \]

Thus \(\sqrt m\propto x\), heavier closures have smaller return radii, and the two energy sectors are equal at equilibrium. Those are consequences of one fixed energy law, not three separately assigned masses. The quartic tension and fixed-amplitude constraint are assumptions of this reduction; the calculation does not yet select the three physical amplitudes or an absolute electron mass.

Change the amplitude. The energy law does the rest.

Dimensionless units set \(\mathcal A=\mathcal T=c=1\). The curves are compression, return tension, and their sum, on fixed logarithmic axes. The minimum is solved from Equation (C4); it is not fitted to a particle mass. The slider changes a fixed constraint, not physical time.

Fixed return amplitude x 1.00

equilibrium radius: 1.0000
equilibrium rest energy: 2.0000
energy × radius: 2.0000

What is an atom?

Experiment does not answer with a drawing. It answers with persistence, spatial extent, scattering, ionization, recoil, and a spectrum. An atom accepts only particular energies. It releases the same differences as light. When one electron is removed, a definite amount of work has been done.

These observations define the first construction problem. A model of the atom must possess a finite normalized state; it must remain stationary while its phase advances; it must admit discrete excited states; and its transitions must conserve the energy and momentum carried away by radiation.

Start with the smallest atom. Do not decide what its parts are yet.

Density remains. Phase moves.

The curve and field marks are generated from the selected numerical state. The phase clock is physical time. The convergence trace is solver time.

Physical phase0.00π 3D view azimuth0.21π Solver convergencestationary

configuration: 1s¹
outer mean radius: 1.500 a₀
calculated opening: 13.5984 eV
evaluated opening: 13.5984 eV
residual: 0.0000 eV

What must the simplest atom do?

Let \(\mathbf r_e\) and \(\mathbf r_p\) be the two observed charge coordinates. The minimal benchmark introduces no fixed nucleus and no orbiting bead. It evolves both termini under one Hamiltonian. This established two-body operator tests the effective atomic behavior that UMFT must recover; its agreement is not independently counted as a derivation of the termini from the UMFT field action:

\[ H= -\frac{\hbar^2}{2m_e}\nabla_e^2 -\frac{\hbar^2}{2m_p}\nabla_p^2 -\frac{e^2}{4\pi\epsilon_0|\mathbf r_e-\mathbf r_p|}, \qquad i\hbar\frac{\partial\Psi}{\partial t}=H\Psi . \tag{1} \]

The center and the relative field separate under

\[ \mathbf R=\frac{m_e\mathbf r_e+m_p\mathbf r_p}{m_e+m_p}, \qquad \mathbf r=\mathbf r_e-\mathbf r_p, \qquad \mu=\frac{m_em_p}{m_e+m_p}, \tag{2} \] \[ \left[ -\frac{\hbar^2}{2\mu}\frac{d^2}{dr^2} +\frac{\hbar^2\ell(\ell+1)}{2\mu r^2} -\frac{e^2}{4\pi\epsilon_0r} \right]u_{n\ell}=E_nu_{n\ell}, \tag{3} \] \[ u_{n\ell}(0)=0,\qquad \lim_{r\to\infty}u_{n\ell}(r)=0,\qquad \int_0^\infty|u_{n\ell}(r)|^2\,dr=1. \tag{4} \]

Measured masses and the electromagnetic coupling enter Equation (1). Finite radius, stationary density, moving phase, binding energy, and transition wavelength leave the solution. Hydrogen gives \(13.5984\ {\rm eV}\) and \(121.568\ {\rm nm}\) for the \(2\to1\) opening; the evaluated values are \(13.598433\ {\rm eV}\) and the resolved Lyman-\(\alpha\) doublet near \(121.5\ {\rm nm}\).¹

The pair model explains atomic relative motion. It has not yet explained what either terminus is.

The mode count repeats without a new fit

A spherical Maxwell boundary admits interior radial modes proportional to \(r^\ell\) and decaying exterior modes proportional to \(r^{-\ell-1}\). Their normal-derivative jump gives the boundary eigenvalue \((2\ell+1)/R\), with angular multiplicity \(2\ell+1\). For the admitted atomic sectors \(0\le\ell\lt n\), supplying the two-state half-winding multiplicity gives

\[ \mathsf S_RY_{\ell m}=\frac{2\ell+1}{R}Y_{\ell m},\qquad N_n=2\sum_{\ell=0}^{n-1}(2\ell+1)=2n^2. \tag{C5} \]

The computed capacities for \(n=1,2,3,4\) are 2, 8, 18, 32. One rule gives the sequence; no capacity table enters the sum. The two-state factor and the allowed angular sectors remain declared assumptions here. This recovers the familiar orbital count, not yet the exclusion principle, the actual energy ordering of a many-electron atom, or the different nuclear magic-number sequence.

What changes when the atom has two electrons?

Independent hydrogen copies would leave each electron responding to a field that the others do not alter. The next hypothesis is smaller: every occupied mode contributes to the Maxwell potential sampled by every other mode, while its own field is removed from its equation.

\[ \left[ -\frac12\frac{d^2}{dr^2} +\frac{\ell_a(\ell_a+1)}{2r^2} -\frac{Z}{r} +V_{\rm all}(r)-V_a(r) \right]u_a(r)=\epsilon_au_a(r), \tag{5} \] \[ \rho_{\rm all}(r)=\sum_a q_a|u_a(r)|^2,\qquad V_{\rm all}(r)= \frac1r\int_0^r\rho_{\rm all}(s)\,ds+ \int_r^\infty\frac{\rho_{\rm all}(s)}s\,ds, \tag{6} \] \[ E_{\rm atom}= \sum_aq_a\epsilon_a- \frac12\sum_aq_a \langle u_a|V_{\rm all}-V_a|u_a\rangle, \qquad \int_0^\infty|u_a|^2dr=1. \tag{7} \]

Equations (5)-(7) are iterated until every occupied mode reproduces the field that acts upon it. The solution contracts the inner modes as \(Z\) rises and places lithium's first \(2s\) mode far outside its \(1s^2\) pair. The shell reset falls out before the measurements are joined.

Does the shared field survive measurement?

The model gets the ordering right: helium closes tightly, lithium opens a broad outer mode, and beryllium tightens that mode again. It misses all three thresholds.

That residual is not patched. It asks what the spherical density discarded.

The uncorrected shared field is shown.

What if identical closures remember their orientation?

For occupied states \(i,j\), the direct Maxwell interaction and its exchanged ordering are

\[ J_{ij}=\iint \frac{|\psi_i(\mathbf r)|^2|\psi_j(\mathbf r')|^2} {|\mathbf r-\mathbf r'|}\,d^3r\,d^3r', \qquad K_{ij}=\iint \frac{\psi_i^*(\mathbf r)\psi_j(\mathbf r) \psi_j^*(\mathbf r')\psi_i(\mathbf r')} {|\mathbf r-\mathbf r'|}\,d^3r\,d^3r', \tag{8} \] \[ E= \sum_i\langle i|h|i\rangle+ \frac12\sum_{i\ne j} \left(J_{ij}-\delta_{s_is_j}K_{ij}\right). \tag{9} \]

Equation (9) moves lithium and beryllium toward their measured openings. Helium does not move: its opposite orientations admit no exchange. The mixed result says exactly one thing. Orientation is active, and orientation exchange is not the whole missing energy.

What physical motion remains possible when an occupied pair opens briefly into the unoccupied atomic modes?

Do occupied modes move independently?

The residual left by Equation (9) identifies a motion excluded by every stationary single-mode picture. Two occupied modes may couple through the Coulomb field to an unoccupied pair and return. No new force is introduced. The Hilbert space admitted by the same Hamiltonian is enlarged.

\[ H=\sum_{p=1}^{N}\left(-\frac12\nabla_p^2-\frac{Z}{r_p}\right) +\sum_{p\lt q}^{N}\frac1{r_{pq}}, \qquad H_0\Phi_0=E_0\Phi_0, \qquad \langle\Phi_I|\Phi_J\rangle=\delta_{IJ}, \tag{10} \] \[ \langle ij\Vert ab\rangle= \iint\psi_i^*(\mathbf r)\psi_j^*(\mathbf r') \frac{\psi_a(\mathbf r)\psi_b(\mathbf r')- \psi_b(\mathbf r)\psi_a(\mathbf r')} {|\mathbf r-\mathbf r'|}\,d^3r\,d^3r', \tag{11} \] \[ E_{\rm corr}^{(2)}= \frac14\sum_{ij\in{\rm occ}}\sum_{ab\in{\rm virt}} \frac{|\langle ij\Vert ab\rangle|^2} {\epsilon_i+\epsilon_j-\epsilon_a-\epsilon_b}, \qquad I_{\rm corr}=\bigl[E_0^{N-1}+E_{\rm corr}^{N-1}\bigr] -\bigl[E_0^N+E_{\rm corr}^{N}\bigr]. \tag{12} \]

The virtual set is not a fitted correction. It is generated from the same radial boundary, antisymmetrized by Equation (11), and enlarged by angular order, energy cutoff, and state count. The observable is the difference between two independently corrected total energies.

Open the omitted modes. Keep the miss.

The left trace is helium correlation energy as angular sectors enter. The right trace is carbon ionization as the virtual-energy boundary expands. Neither trace contains the evaluated measurement.

Maximum angular orderℓ ≤ 1 Virtual energy cutoff0.5 Ha

helium correlation: −0.03249 Ha
carbon model: 8.8286 eV
carbon evaluated: 11.2603 eV
remaining residual: −2.4317 eV

The same omitted-motion operator lowers helium and raises carbon’s opening toward observation. At the converged boundary it still leaves carbon about \(1.08\,\mathrm{eV}\) low. Correlation is required by this evaluation; this correlation operator is not sufficient.

What leaves an atom?

The preceding operators describe stationary atomic recurrence. Spectroscopy adds an event: prepare an excited recurrence and the atom later occupies a lower one while energy and momentum cross the laboratory. The interval is discrete at the atom and extended in space.

A detector does not receive an orbiting object extracted from the atom. It receives a finite electromagnetic event with a leading front, a trailing front, polarization, direction, bandwidth, and one total quantum of energy. The event is called a photon. The first question is therefore not what picture resembles it, but what complete field behavior arrives.

The atom earns the photon. The photon does not enter as decoration.

What is inside a wavepacket?

Maxwell’s construction supplies no carrier beneath the electric and magnetic field. In covariant form the field tensor is generated by a potential, constrained by the homogeneous equation, and sourced by four-current:³

\[ F_{\mu\nu}=\partial_\mu A_\nu-\partial_\nu A_\mu, \qquad \partial_{[\lambda}F_{\mu\nu]}=0, \qquad \partial_\mu F^{\mu\nu}=\mu_0J^\nu . \tag{13} \] \[ T^{\mu\nu}=\frac1{\mu_0} \left(-F^{\mu\alpha}F^{\nu}{}_{\alpha} +\frac14\eta^{\mu\nu}F^{\alpha\beta}F_{\alpha\beta}\right), \qquad u=\frac{\epsilon_0}{2}|\mathbf E|^2+\frac1{2\mu_0}|\mathbf B|^2, \qquad \mathbf S=\frac1{\mu_0}\mathbf E\times\mathbf B, \tag{14} \] \[ \frac{\partial u}{\partial t}+\nabla\!\cdot\!\mathbf S=-\mathbf J\!\cdot\!\mathbf E, \qquad \lim_{|\mathbf x|\to\infty}F_{\mu\nu}^{\rm in}=0 \quad\hbox{outside the prepared incident mode}. \tag{15} \]

Here \(\eta=\operatorname{diag}(1,-1,-1,-1)\), so Equation (14) has positive \(T^{00}=u\) and a timelike total momentum has positive \(P_\mu P^\mu\). Equations (13)-(15) identify what propagates and how its energy moves. They do not yet determine whether one normalized packet transfers its energy to one atomic transition. That requires the atom and the selected Maxwell channel to be evolved together.

If there is no ball inside the wavepacket, can its leading and trailing fronts change whether the atom absorbs it?

\[ H=\hbar\omega_0\sigma_+\sigma_- +\int_{-\infty}^{\infty}\hbar\omega(k)a_k^\dagger a_k\,dk +i\hbar\sqrt{\frac{\Gamma_m}{2\pi}} \int_{-\infty}^{\infty} \left(\sigma_+a_k-\sigma_-a_k^\dagger\right)dk+H_\perp, \qquad \int_{-\infty}^{\infty}|\xi_{\rm in}(t)|^2dt=1, \tag{16} \] \[ \tau=\Gamma t, \qquad \frac{dc_e}{d\tau}=-\left(\frac12+i\frac{\Delta}{\Gamma}\right)c_e -\sqrt{\eta}\,\xi_{\rm in}(\tau), \qquad c_e(-8)=0, \tag{17} \] \[ \xi_{\uparrow}(\tau)=\frac{e^{\tau/2}}{\sqrt{1-e^{-8}}}\,\mathbf 1_{[-8,0]}(\tau), \qquad \xi_{\downarrow}(\tau)=\frac{e^{-(\tau+8)/2}}{\sqrt{1-e^{-8}}}\,\mathbf 1_{[-8,0]}(\tau), \tag{18} \] \[ \xi_g(\tau)=\mathcal N_g e^{-(\tau+4)^2/(4\sigma^2)}\mathbf 1_{[-8,0]}(\tau), \qquad \sigma=\frac43, \qquad \mathcal N_g^{-2}=\int_{-8}^{0}e^{-(\tau+4)^2/(2\sigma^2)}d\tau, \tag{19} \] \[ \xi_{\rm out}=\xi_{\rm in}+\sqrt{\eta}\,c_e, \qquad \xi_\perp=\sqrt{1-\eta}\,c_e, \tag{20} \] \[ \int_{-\infty}^{\tau}|\xi_{\rm out}|^2d\tau' +\int_{-\infty}^{\tau}|\xi_\perp|^2d\tau' +|c_e(\tau)|^2 +\int_{\tau}^{\infty}|\xi_{\rm in}|^2d\tau'=1. \tag{21} \]

The parameter \(\eta\) is the spatial and polarization overlap with the atomic dipole; \(\Delta\) is the detuning; \(H_\perp\) contains every radiative mode not drawn. Equations (18)-(19) are the actual finite pulses integrated below. Equation (21) forbids the animation from losing energy between the field, the atom, and the undisplayed channels.

Reverse the envelope. Do not change its energy.

The stage, excitation trace, outputs, and four-term ledger are integrated from Equations (16)-(21). Pulse shape, detuning, and overlap are model inputs.

Detuning Δ/Γ0.00 Mode overlap η1.00 Physical time Γt−8.00

peak excitation: 99.93%
current excitation: 0.00%
incident norm: 1.0000
energy closure: 0.0

Does the front of the packet matter?

The model receives unit incident norm in every case. On resonance with perfect overlap, a truncated rising exponential reaches \(99.93\%\) peak excitation; its time reverse reaches \(54.15\%\). Equal energy is not equal interaction.

Leong et al. prepared rising and decaying one-photon envelopes with equal Lorentzian power spectra. The rising pulse produced \(56\pm11\%\) greater peak excitation.⁵ The measured atom therefore distinguishes temporal ordering that an energy-only particle picture discards.

The ideal one-channel model yields an \(84.6\%\) relative increase. The experiment yields \(56\pm11\%\). The direction survives; the magnitude does not. The next evaluation must introduce the experiment’s measured mode overlap and loss channels rather than call the ideal amplitude a match.

What if the electron is actually the same energetic field prevented from completing open propagation—closed into a persistent, charged recurrence?

What did the light acquire?

The question is licensed by a reversible observation. Electromagnetic events can produce an electron and a positron; an electron and a positron can return their energy and momentum to electromagnetic events. The transformation preserves total four-momentum, charge, and angular momentum. It does not show that either lepton is a photon bent into a circle.

Begin with the part that follows without an ontology. For two photons of energies \(E_1,E_2\) meeting at angle \(\theta\), their summed four-momentum possesses an invariant energy even though each photon separately has none:⁶

\[ k_1^\mu=\left(\frac{E_1}{c},\frac{E_1}{c}\hat{\mathbf n}_1\right), \qquad k_2^\mu=\left(\frac{E_2}{c},\frac{E_2}{c}\hat{\mathbf n}_2\right), \qquad k_1^2=k_2^2=0, \qquad \hat{\mathbf n}_1\!\cdot\!\hat{\mathbf n}_2=\cos\theta, \tag{22} \] \[ E_{\rm cm}^2=c^2(k_1+k_2)^2 =2E_1E_2(1-\cos\theta), \qquad \gamma+\gamma\rightarrow e^-+e^+ \quad\Longleftrightarrow\quad E_{\rm cm}\ge 2m_ec^2, \tag{23} \] \[ p_*c=\sqrt{\left(\frac{E_{\rm cm}}2\right)^2-m_e^2c^4}, \qquad K_{\rm cm}=E_{\rm cm}-2m_ec^2, \qquad e^-+e^+\rightarrow\gamma+\gamma+\cdots . \tag{24} \]

At a head-on collision, two \(0.51099895069\,\mathrm{MeV}\) photons supply exactly the \(1.02199790138\,\mathrm{MeV}\) rest-energy threshold fixed by the measured electron mass.⁷ Co-propagating photons never reach it, however large their separate energies, because their total momentum remains null.

Energy is necessary. Geometry decides what is available.

The moving fronts and invariant-energy curve are evaluated from Equations (22)-(24). The figure computes permission for pair production, not its cross-section.

Energy of each photon0.750 MeV Collision angle180° Physical time to event−1.000

total incident energy: 1.5000 MeV
invariant energy: 1.5000 MeV
pair threshold: 1.0219979 MeV
available kinetic energy: 0.4780 MeV

The combined field event has enough invariant energy for an electron–positron pair.

Can light possess a rest frame collectively?

The stress-energy tensor already supplies the calculation. Integrate a complete field configuration over a spacelike slice. Its total energy and directional momentum define one four-vector and therefore one invariant mass:

\[ \partial_\mu T^{\mu\nu}=0, \qquad P^\nu=\frac1c\int_{\Sigma_t}T^{0\nu}(\mathbf x,t)\,d^3x, \qquad M^2c^2=P_\nu P^\nu, \tag{25} \] \[ E=\sum_aE_a, \qquad \mathbf P=\frac1c\sum_aE_a\hat{\mathbf n}_a, \qquad M^2c^4=\left(\sum_aE_a\right)^2- \left|\sum_aE_a\hat{\mathbf n}_a\right|^2, \tag{26} \] \[ \hat{\mathbf n}_a=\hat{\mathbf n}\ \forall a\Rightarrow M=0, \qquad E_1=E_2=E_0,\ \hat{\mathbf n}_2=-\hat{\mathbf n}_1 \Rightarrow \mathbf P=0,\ Mc^2=2E_0. \tag{27} \]

Equation (27) establishes a narrow result: locally null energy can form a timelike total system when its directional momentum closes. It does not establish an electron. Source-free linear Maxwell evolution supplies no mechanism that keeps the opposed flows localized, no net electric charge, and no half-integer spin.

The electron hypothesis must therefore be stated as a construction problem. Find a finite recurrent solution whose complete field action produces the measured invariants

\[ P_\mu P^\mu=m_e^2c^2, \qquad Q=\epsilon_0\lim_{R\to\infty}\oint_{S_R}\mathbf E\!\cdot d\mathbf A=-e, \qquad W_\mu W^\mu=-m_e^2c^2\,\frac34\hbar^2, \tag{28} \] \[ \lim_{|\mathbf x|\to\infty}T^{00}=0, \qquad \int_{\mathbb R^3}T^{00}d^3x=m_ec^2, \qquad \Phi(\tau+\mathcal T)=e^{i\alpha}\Phi(\tau), \tag{29} \] \[ W^\mu=\frac12\epsilon^{\mu\nu\rho\sigma}P_\nu J_{\rho\sigma}, \qquad J^{\rho\sigma}=\frac1c\int_{\Sigma_t} \left(x^\rho T^{0\sigma}-x^\sigma T^{0\rho}\right)d^3x. \tag{30} \]

Pair production motivates the closure question. Equations (28)-(30) decide whether any proposed closure deserves the electron’s name.

Does zero total momentum keep the energy together?

It need not. For a smooth source-free Maxwell field in flat, unbounded space, let \(u=T^{00}\) and \(\mathbf S\) be its energy density and Poynting vector. Assume finite energy and finite energy-weighted second moment, with sufficient decay to remove the surface terms. Local conservation and the traceless Maxwell stress tensor then give the following result by two integrations by parts:¹⁹

\[ I=\int r^2u\,d^3x,\qquad \dot I=2\int\mathbf x\cdot\mathbf S\,d^3x,\qquad \ddot I=2c^2\int\sum_iT^{ii}d^3x=2c^2U. \tag{30a} \] \[ \mathbf P=0,\quad\dot I(0)=0 \quad\Longrightarrow\quad R_{\rm rms}^2(t)=R_{\rm rms}^2(0)+c^2t^2, \qquad R_{\rm rms}^2=I/U. \tag{30b} \]

The origin here is the stationary energy centroid. A field may contract toward its minimum radius and then expand; it cannot repeat a finite-radius cycle under these assumptions. This is an obstruction to an isolated classical Maxwell packet, not a theorem excluding charged quantum states or every coupled spacetime theory. A charged Coulomb tail, for example, has infinite \(I\), so Equation (30b) cannot be applied to that tail.

An explicit three-dimensional solution makes the distinction observable. In units \(c=\epsilon_0=\mu_0=1\), with arbitrary length unit \(\sigma\), choose a scalar wave and obtain divergence-free vector fields from its derivatives:

\[ (\partial_t^2-\nabla^2)\psi=0,\qquad \psi(r,0)=e^{-r^2/2},\quad\partial_t\psi(r,0)=0, \qquad\psi\to0\ (r\to\infty), \tag{30c} \] \[ g(s)=s e^{-s^2/2},\qquad \psi(r,t)=\frac{g(r-t)+g(r+t)}{2r},\qquad \psi(0,t)=g'(t), \tag{30d} \] \[ \mathbf A=\nabla\psi\times\hat{\mathbf z},\quad\phi=0, \qquad\mathbf E=-\partial_t\mathbf A,\quad \mathbf B=\nabla\times\mathbf A,\qquad u=\tfrac12(E^2+B^2). \tag{30e} \]

These are vector fields in three dimensions, not spherical electromagnetic monopoles. Only their plotted energy is integrated over angle. The initial electric field vanishes, the magnetic field does not, and symmetry keeps total momentum zero. Field integration gives \(U=5\pi^{3/2}/4\) and \(R_{\rm rms}^2(0)=19/10\) in these units. No electron mass or radius has been supplied.

The energy has a rest frame. It still spreads.

The green curve is the angular integral of the exact Maxwell energy density from Equations (30c)-(30e); grey is its initial profile. The dashed line marks the radius calculated from that field, not a prescribed moving envelope. Negative time shows incoming focusing, positive time outgoing propagation.

Physical time 0.00 σ/c

Field-integrated rms radius: 1.37840 σ
Energy / initial energy: 1.000000000
Difference from exact radius² law: 0 σ²

The quadrature extends nine initial widths beyond \(|ct|\); it is not a reflecting boundary. Time advances only when played or scrubbed. This classical solution tests persistence, not a single-photon detection probability.

Can the return field choose its own shape?

Confinement needs more than a circulating arrow. The field must pay for its gradients, carry its own transport curvature, and approach a specified vacuum. A useful candidate is a directional order matrix \(G\), distinct from the causal spacetime metric. Its inverse-matrix kinetic norm measures the cost of changing eigenvalues as well as rotating their axes. A rotational connection transports those axes; a Weyl dilation alone cannot undo their changing orientation.

The distinction changes the computation. Restricting the matrix too early gives a curved field-space metric with a conical core. Allowing the otherwise unsourced eigenvalue shape to relax removes that artificial core restriction. The remaining two-coordinate metric is smooth but still curved:

\[ d\ell_{\rm field}^{\,2}=dt^2+F(t)^2d\theta^2,\qquad t=\sqrt K\,u,\qquad F(t)=2\sqrt K\sinh\!\left(\frac{t}{2\sqrt K}\right), \qquad {\cal K}=-\frac1{4K}. \tag{R1} \]

No change of field coordinates makes that curvature zero. The familiar flat complex-field vortex is instead a controlled weak-anisotropy limit. With the compatible potential \(V=(\mu_v-\mu)^2/(2\kappa_d)\), where \(\mu=4K[\cosh(u/2)-1]\), the first-order, or Bogomolny, equations give equal scalar and connection vacuum masses and tension \(2\pi n\mu_v\). This fixes a relation within the candidate action, not its physical vacuum or absolute energy scale.

Solving the finite-vacuum equations produces a family rather than one universal profile. In units set separately by each member’s vacuum mass, the half-flux radius increases from 1.7962 in the flat limit to 5.5248 at split \(v=8\), and 14.3335 at \(v=12\). An independent bound, \(R_p\ge2\sqrt p\cosh(v/4)\), follows from the maximum field strength. These are dimensionless radii, not a prediction that a particle grows by that factor in metres.

The vacuum changes which radiation channels can open.

The points come from solved backgrounds and their action-derived scalar–connection fluctuation operator. The line at \(2\omega=1\) is the massive continuum threshold in vacuum-frequency units. A harmonic above it can propagate only if the full interaction supplies a nonzero source. A harmonic below it does not establish stability against other fields.

Computed vacuum split

Loading the source-verified vacuum family.

The fluctuation Hessian factors as \(H=D^\dagger D\). Its scalar partner is \(-\Delta+(F/F_v)^2\), not an assumed normalized-field-squared potential. The squared mode frequency is approximately 0.434280 at \(v=4\) and 0.217662 at \(v=6\): the second harmonic crosses the threshold between those sampled cases. No particle data select either vacuum.

Circular motion closes one loss channel, not every channel.

A separate relative-orientation mode of the flat candidate has frequency approximately 0.88174, below its own unit continuum threshold. The fundamental therefore has no outgoing flux in that linear channel, while its second harmonic can radiate. Varying the existing alignment energy \(U=h^2|\mathbf r|^2/2\) supplies the actual quadratic source \(h_0v_{\rm mode}^2\), rather than an assigned damping percentage.

For \(\mathbf r=\mathbf p\cos\omega t+\mathbf q\sin\omega t\), the scalar second harmonic is controlled by \((|\mathbf p|^2-|\mathbf q|^2)/2\) and \(\mathbf p\cdot\mathbf q\). Equal, orthogonal components cancel it exactly; linear motion maximizes it at fixed mean norm. Allowing the connection to respond increases the static susceptibility from 0.070048 to 0.079419. At twice the mode frequency, the outgoing response is approximately \(-0.042259+0.013221i\), with 48.68% of the outgoing flux in the connection even though only the scalar is directly driven.

\[ P_{\rm driven}=\frac{\omega\,\operatorname{Im}G_{\rm out}}{Z_{\rm bg}}\,M^2\chi^2 \simeq\frac{0.011658}{Z_{\rm bg}}M^2\chi^2,\qquad \chi^2+\left(\frac{\omega|\mathbf L|}{E}\right)^2=1. \tag{R2} \]

Here \(M\) is mean squared mode amplitude, \(\chi\) its scalar harmonic contrast, and \(Z_{\rm bg}\) the background action weight relative to the mode. All values use the stated unit-action normalization. Source work equals the independently calculated outgoing boundary flux. This is a derived, polarization-dependent classical loss channel; it is not a measured lepton lifetime. Circular cancellation leaves other tensor harmonics and other interactions to be tested.

The primitive’s static inverse response also depends on what is held fixed: approximately 0.10930 when unrestricted, 0.052755 with its amplitude fixed, and 0.12412 for its normalized shape Hessian. These are three physical questions, not three estimates of one stiffness. Its neutral source obeys \(\widetilde J(k)=k^2\widetilde v(k)\), so a zero source monopole can coexist with a finite static Green response. Neither a finite-box subtraction nor a static pole alone supplies a decay probability.

A positive order field is not a light cone.

A radial potential does not determine its extension to every matrix deformation. Explicit invariant potentials can agree on the same rank-one profile and disagree in the missing scalar-shape direction. The extension with no shape penalty has a compact negative-energy variation; adding a positive penalty is an extra choice, not a consequence of the radial fit. The charged calculation below states that choice openly.

There is a separate spacetime issue. A Lorentzian causal metric paired with a positive spacetime order tensor selects a unique timelike eigenline and hence a positive reference norm. This is a pointwise construction, not a preferred observer imported from outside; its dependence on the order field must also be varied in the action. It must not be identified with the internal order matrix used in the following diagnostic.

Directly replacing the Maxwell metric by the positive tensor gives the wrong hyperbolic structure: positivity of a matrix does not make it a causal propagation metric. Mixing with a causal Maxwell term can restore positive energy and frozen-background hyperbolicity, but anisotropic constitutive weights can produce birefringence and a different characteristic cone. A positive scalar multiplier of the ordinary causal Maxwell action preserves its cone. No optical parameter, preferred vacuum or fully coupled causal particle has been selected by these checks.

A charged core can be solved rather than drawn.

A definite internal \(SO(4)\) extension provides a constructive test. The group, potential and vacuum are stipulated; their identification with nature is not established. They are held fixed throughout the following calculation. The order field has logarithmic eigenvalues \((u/2,-u/2,f,-f)\). Its two-component charged shape has phase \(\theta\), transforms with weight two under the surviving internal rotation, and obtains its current from the same covariant kinetic term that governs its motion:

\[ {\cal L}_{G}=K\!\left[\frac12(\partial u)^2+2(\partial f)^2+ 2\sinh^2 f\,(\partial\theta+2a)^2\right], \qquad j^\mu_{\rm Gauss}=-8K\sinh^2 f\,(\partial^\mu\theta+2a^\mu). \tag{R3} \]

The source convention is \(\kappa\,\partial_\mu F^{\mu\nu}=j^\nu_{\rm Gauss}\); the coefficient obtained by differentiating the matter Lagrangian with respect to \(a_\nu\) has the opposite sign. Set \(\theta=-\omega t\), \(x=m_vr\), \(w=\omega/m_v\), and \(b=(\omega-2a_0)/m_v\). The neutral shape, charged amplitude and electric constraint are then solved together—not by prescribing a charge density. The dimensionless potential and all three radial equations are

\[ U(u,f)=\frac{2[\cosh(v/2)-\cosh(\sqrt{u^2+4f^2}/2)]^2}{\sinh^2(v/2)} +\gamma\left(\frac{u^4f^2}{8}+\frac{u^2f^4}{2}\right), \tag{R4} \] \[ u''+\frac{2u'}x=U_u,\quad f''+\frac{2f'}x=\frac14U_f-\sinh f\cosh f\,b^2,\quad b''+\frac{2b'}x=\operatorname{csch}^2(v/2)\sinh^2 f\,b . \tag{R5} \]

Regularity sets all three derivatives to zero at the centre. Far away, \(u\to v\), \(f\to0\), and \(b=w-c_Q/x+\cdots\). Because the charged amplitude vanishes in the vacuum, the surviving electric connection has a Coulomb exterior. This differs from neutral order coupled only through \(F_{\mu\nu}\): without a charged source or inner boundary, that smooth source-free system cannot produce a nonzero net displacement flux through a surrounding sphere.

The fixed diagnostic choice \(v=4,\gamma=0.1,w=1.2\) yields a nonzero stationary solution. In the conventions \(q=Q/(32\pi K/m_v^2)\) and \(e=E/(4\pi K/m_v)\), the refined finite-volume result is \(q\simeq32.9327\), \(e\simeq148.3701\). Domain extension, source–flux balance and an independent virial identity check the solution. No elementary charge, mass in MeV, or radius in metres has been supplied or inferred.

The shape and its electric field close together.

These are freshly recomputed profiles of Equation (R5), not an illustration fitted to an electron. The plot uses the higher-order background solution; the electric field outside the displayed core continues to infinity. Changing the displayed radius does not rerun or refit the model.

Displayed radius \(x\) 16

Coulomb coefficient \(c_Q\): 2.503601
discrete background residual: Loading
physical mass / length scale: Not selected

Reconstructing the solved background and its Gauss flux.

Stationary is not the same as protected.

The core’s energy per charge is below the passive scalar threshold but above a lighter charged-vector threshold from the same action. In common units the three values are 1.126313, 1.264911 and 0.648054 respectively. A scalar-only binding test would pass while missing the cheaper way to carry the charge.

The electric field must readjust during a deformation. Solving Gauss and preserving total charge removes a spurious frozen-electric negative direction. The tested scalar–electric response is restoring; the translational mode approaches zero under refinement. Yet moving a small amount of charge into a sufficiently broad, distant mixed-vector packet lowers the leading constrained energy. This is an energetic escape direction, not a calculated breakup time.

Real-frequency scattering makes the channel tangible. A signal at \(\Omega_s=0.7\) can emerge with greater flux and a partner at \(\Omega_p=0.5\), in the same vacuum-frequency units. The refined conversion occupation is \(N\simeq0.01313975\). The signed wave-flux identity is \(G-N=1\), so the signal-number gain is about 1.01313975 and the total outgoing energy is about 1.022525 times the incident energy. The excess energy and charge have the same ratio \(\omega/2\) as the core’s independently derived chemical relation.

\[ G-N=1,\qquad \frac{E_{\rm out}}{E_{\rm in}}=G+\frac{\Omega_p}{\Omega_s}N,\qquad \frac{\Delta E}{\Delta Q}=\frac{\omega}{2}. \tag{R6} \]

Amplification needs an incident classical disturbance. It does not by itself prove that the isolated core has an exponentially growing mode. The outgoing-wave searches below retain Coulomb transport, electric constraints and independent refinement. Their null counts describe only their stated windows.

Numerically searched frequency windows, \(\sigma=\operatorname{Re}\sigma+i\,\operatorname{Im}\sigma\), with time dependence \(e^{-i\sigma\tau}\).
Sector	Window in vacuum-frequency units	Result and remaining boundary
Magnetic-parity vectors, \(\ell=1,2,3\)	\(\|\operatorname{Re}\sigma\|\le0.61\); \(10^{-4}\le\operatorname{Im}\sigma\le0.61\)	No poles counted under the recorded refinements. Weaker growth, higher multipoles and other parity sectors remain unexcluded.
Spherical longitudinal sector	\(\|\operatorname{Re}\sigma\|\le0.85\); \(0.02\le\operatorname{Im}\sigma\le1\)	Zero winding on the refined adaptive contour. Coarse apparent counts disappeared when rapid phase changes were resolved. The strip below 0.02 remains open.

These are converged numerical screens, not interval-certified stability theorems. The transverse analytic bound \(|\sigma|\le w/2\) restricts that sector’s possible growing modes; it does not supply a count at the real axis. Higher-order background reconstruction reduces the longitudinal Gauss defect approximately sixteenfold when the grid spacing halves. Neither a small residual nor a zero count gives the core an observed particle lifetime.

What changes if the outgoing fields are quantized?

Adding ordinary canonical bosonic quantization to this same scattering sector changes its empty-input behavior. The partner enters through a creation operator, and its commutator supplies a \(+1\) term. This is an additional quantum assumption, not a derivation of quantum mechanics from UMFT.³²

\[ a_{s,\rm out}=A\,a_{s,\rm in}+B\,a^\dagger_{p,\rm in},\qquad |A|^2-|B|^2=1,\qquad \langle n_s\rangle=\langle n_p\rangle=|B|^2=N \quad\hbox{for incoming vacuum}. \tag{R7} \]

At the displayed \(\sigma=0.1,\ell=1\) probe, \(N\simeq0.01313975\) is a mean pair occupation, not a decay probability for the core. Summing both mixed pairs and \(\ell=1,2,3\) over the interior band \(0.02\le\sigma\le0.24\) gives a partial pair flux of approximately 0.002363084 per dimensionless time. The physical conversion needs the frequency scale; the fractional backreaction also needs \(\hbar/\kappa\). Neither is selected. The calculation omits the remaining frequency band, higher multipoles, other sectors and depletion.

The two emitted waves carry the same sign of the internal charge, drawn from the rotating core. They are not automatically an electron–positron pair. This conditional quantum channel can exist without a detected classical growing pole, so classical null searches alone cannot establish quantum persistence.

Quantizing the core’s actual collective phase adds another restriction. Its half internal turn leaves the pure core unchanged, giving ordinary core-only charge labels \(Q/\hbar=2n\) in weight-one carrier units. An odd intrinsic carrier can change that to \(2n+1\). These are allowed representation labels, not solved low-charge bound states. The fixed-shape rotor also gives a finite pair-energy step \((2n-1)\hbar^2/C_e\), rather than treating a small core as an infinite classical reservoir. Internal charge is not spatial spin.

Which half-turn belongs to the particle?

A director can return as its negative while its dyad returns unchanged: \(n(\chi+2\pi)=-n(\chi)\), but \(nn^T\) is periodic. A torus has two independent paths: a meridian around the tube and a longitude along the major ring. A physical rotation of the whole configuration is a third operation. They cannot be used interchangeably to infer spin.

There is an exact exterior constraint. The preferred, untwisted longitude of an isolated unknotted ring bounds a disk outside its tube. A nondegenerate real eigenline defined throughout that disk must have even sign return. An odd longitudinal return must therefore live in a field that vanishes before reaching that exterior, encounter another defect or boundary, or use different global bundle data. This concerns an eigenline’s sign topology, not an arbitrary connection’s Wilson phase: curvature can give a nonzero transport phase around a contractible loop.

Linking supplies a different possibility. The longitude parity of ring \(i\) must equal \(\sum_{j\ne i}\mathrm{Lk}_{ij}m_j\pmod2\), where \(m_j\) is the other ring’s meridian parity. A once-linked pair can support odd longitudinal signs. A framing twist instead changes the measured path to \(\lambda_f=\lambda+f\mu\), changing its sign by \(fm\). That may count the meridional winding along a new curve rather than create another independent longitudinal winding.

The arrow reverses. The axis returns.

The arrow is a chosen lift of a director; the unoriented axis represents its dyad. Move through one or two circuits. This is exact sign geometry, not a simulated particle trajectory or a demonstration of quantum spin.

Circuit angle / \(\pi\)

One circuit changes the director sign; the dyad is unchanged.

A periodic geometry must also carry a well-defined charge.

If the sign-changing quantity is the Maxwell potential, its nonzero curvature \(F=dA\) changes sign too. A periodic stress tensor does not by itself make that curvature a single-valued ordinary Maxwell field. Treating the cover as physical, treating \(A\) only as a director lift, and gluing through charge conjugation are different theories of the global observable. The latter resembles the charge-reversing transport possible in gauge theories with disconnected stabilizers.³³ A central minus sign in an Abelian rotation group alone does not reverse its charge generator.

The chiral ring’s half-cover circulation magnitude \(4A_0R\) is reproduced exactly. But its \(2\pi\) segment joins opposite sheet points: the complete \(4\pi\) integral cancels. Under ordinary cover gauge freedom, the half-arc integral changes by an endpoint term. A physical exposure or junction can make it part of a boundary observable, but that boundary current and measurement coupling must be included. The potential-dependent geometric action is not silently assumed to have ordinary Maxwell gauge freedom.

Finally, the quantum sign under a spatial rotation belongs to the state over the full configuration space, including all fields and boundaries. A compact toroidal dyad with isotropic exterior admits an explicit contraction through vanishing order; its chosen director’s minus sign does not prevent it. A fixed anisotropic vacuum or additional topology could change that conclusion. Establishing half-integer spin requires the actual rotation and exchange loops, not the chosen eigenvector alone.³⁴

What changes up the lepton chain?

The electron, muon, and tau carry the same charge and spin. They do not carry the same rest energy or persistence. The muon is \(206.77\) electron masses and decays after \(2.1969811\,\mu\mathrm{s}\) on average; the tau is \(3477.23\) electron masses and decays after about \(2.903\times10^{-13}\,\mathrm{s}\).⁹

The sameness isolates a charged exterior. The decay isolates what changes. A negative muon does not emit only energy and become an electron. Its ordinary decay contains an electron, an electron antineutrino, and a muon neutrino:

\[ \mu^-\rightarrow e^-+\bar\nu_e+\nu_\mu, \qquad \mathcal L_{\rm eff}=-\frac{4G_F}{\sqrt2} \bigl(\bar e\gamma^\alpha P_L\nu_e\bigr) \bigl(\bar\nu_\mu\gamma_\alpha P_L\mu\bigr)+{\rm h.c.}, \qquad P_L=\frac{1-\gamma^5}{2}. \tag{34} \] \[ \Gamma_0=\frac{G_F^2m_\mu^5}{192\pi^3}, \qquad \frac{d\Gamma}{dx}=2\Gamma_0x^2(3-2x), \qquad x=\frac{2E_e}{m_\mu c^2}, \qquad 0\le x\le1, \tag{35} \] \[ \int_0^1\frac1{\Gamma_0}\frac{d\Gamma}{dx}\,dx=1, \qquad \tau_\mu^{(0)}=\frac{\hbar}{\Gamma_0}, \qquad \frac{dN}{d\tau}=-\frac{N}{\tau_\mu}, \qquad \frac{N(t)}{N_0}=e^{-t/(\gamma\tau_\mu)}. \tag{36} \]

Equations (34)-(36) are the unpolarized, tree-level weak model with the final-state masses neglected in the plotted spectrum.¹⁰ It predicts the electron-energy distribution and advances the decay clock in proper time. It also provides a numerical loss: its \(2.18735\,\mu\mathrm{s}\) lifetime is \(0.438\%\) shorter than the measured value because the displayed operator omits the known finite-mass and radiative terms.

The same charge returns. The rest does not vanish.

The survival curve and electron spectrum are calculated from Equations (35)-(36). Boost changes coordinate time, not the proper-time decay law.

Lorentz factor γ1.0 Coordinate time0.00 μs

proper time elapsed: 0.00 μs
surviving population: 100.00%
tree-level lifetime: 2.18735 μs
measured lifetime: 2.19698 μs
residual: −0.00963 μs

Did the heavier lepton contain only more twist?

No observation licenses that reduction. The heavier state has acquired whatever structure permits its larger rest energy while preserving the same exterior charge and half-spin. When the state opens, the difference is not emitted as one anonymous energy packet: the final state carries distinct neutral lepton channels.

The restrained hypothesis is structural rather than constituent. The neutral outputs may mark the opened boundaries of the transformation rather than intact objects formerly orbiting inside the muon. Tau decay strengthens the constraint: its electron and muon channels each carry a tau neutrino plus the antineutrino associated with the destination lepton, while its hadronic channels show that a simple three-rung nesting rule is insufficient.

\[ \tau^-\rightarrow e^-+\bar\nu_e+\nu_\tau, \qquad \tau^-\rightarrow\mu^-+\bar\nu_\mu+\nu_\tau, \qquad \tau^-\rightarrow\pi^-+\nu_\tau\quad\hbox{and other hadronic channels}, \tag{37} \] \[ Q_{\rm in}=Q_{\rm out}=-e, \qquad P_{\rm in}^\mu=\sum_fP_f^\mu, \qquad J_{\rm in}^{\mu\nu}=\sum_fJ_f^{\mu\nu}+J_{\rm orbital}^{\mu\nu}. \tag{38} \]

Any UMFT rung operator must generate the three charged masses without inserting them, make the electron stable, reproduce the muon and tau opening rates, and produce the observed source- and destination-specific neutral channels. Equations (34)-(38) describe the opening once it occurs; they do not generate the rungs or explain why only the lowest one persists.

What if leptons twist?

Then the twist is not the answer. The model must also generate what makes each twist dynamically possible—and what comes out when it fails.

But then what is the proton?

The working hypothesis treats proton and electron as complementary exposed termini of an atomic field closure, not mirror-image interiors. Hydrogen supplies equal and opposite exterior charge; the rest must come from the field solution. The proton is \(1836.15\) electron masses, has a finite charge radius, and carries a magnetic moment \(2.792847\) times the nuclear magneton.⁷ Those observations require internal distributed current or an equivalent form-factor structure. The proposed complementarity must reproduce them, not replace them.

\[ P_\mu P^\mu=m_p^2c^2, \qquad \epsilon_0\lim_{R\to\infty}\oint_{S_R}\mathbf E\!\cdot d\mathbf A=+e, \qquad W_\mu W^\mu=-m_p^2c^2\,\frac34\hbar^2, \tag{31} \] \[ G_E(0)=1, \qquad r_{E,p}^2=-6\hbar^2\left.\frac{dG_E}{dQ^2}\right|_{Q^2=0}, \qquad r_{E,p}=0.84075(64)\,\mathrm{fm}, \tag{32} \] \[ \boldsymbol\mu_p=G_M(0)\frac{e\hbar}{2m_p}\frac{\mathbf S}{\hbar/2}, \qquad G_M(0)=2.79284734463. \tag{33} \]

Equations (31)-(33) are not a proton theory. They are the measurement contract that prevents the next model from being named by resemblance. A candidate proton must solve them while also reproducing the hydrogen interaction already tested above.

What field or spacetime structure can hold null energy in a charged, half-spin recurrence—and can one construction admit both the electron and the very different proton without declaring them mirrors?

What does a neutron become?

A free neutron opens through \(n\rightarrow p+e^-+\bar\nu_e\). Inside a nucleus the same change alters the element, but whether it can occur depends on the total initial and final nuclear energies. The reverse charged-current grammar also exists: electron capture and muon capture convert a proton to a neutron while a neutrino leaves.

\[ n\rightarrow p+e^-+\bar\nu_e, \qquad p+e^-\rightarrow n+\nu_e, \qquad p+\mu^-\rightarrow n+\nu_\mu, \tag{39} \] \[ Q_{\beta^-}=\bigl[M(A,Z)-M(A,Z+1)\bigr]c^2, \qquad Q_n=(m_n-m_p-m_e)c^2=0.78233356\,\mathrm{MeV}, \tag{40} \] \[ P_i^\mu=P_f^\mu+p_e^\mu+p_\nu^\mu, \qquad Q_i=Q_f-e, \qquad A_i=A_f. \tag{41} \]

These transformations do not establish that a neutron is a proton and electron mechanically compressed together. They establish a shared transition whose open boundary carries a charged lepton and a neutral lepton channel. A structural model must reproduce when the channel is energetically open and how strongly the initial and final nuclear states overlap.

Why does the open channel wait 5,700 years?

Carbon-14 has more beta-decay energy than tritium: \(156.476\,\mathrm{keV}\) rather than \(18.5906\,\mathrm{keV}\). Yet carbon-14 persists for \(5700(30)\) years while tritium persists for \(12.32(2)\) years.¹¹¹² Energy availability points in the wrong direction.

The allowed beta model separates phase space from the nuclear transition strength. In electron-mass units, with the point-nucleus Coulomb factor used by the simulation,

\[ W=\frac{E_e}{m_ec^2}, \qquad p=\sqrt{W^2-1}, \qquad W_0=1+\frac{Q}{m_ec^2}, \qquad F(Z,W)=\frac{2\pi\eta}{1-e^{-2\pi\eta}}, \qquad \eta=\alpha Z\frac{W}{p}, \tag{42} \] \[ f(Z,W_0)=\int_1^{W_0}pW(W_0-W)^2F(Z,W)\,dW, \tag{43} \] \[ \lambda=\frac{\ln2}{t_{1/2}}=C\,f(Z,W_0) \left[g_V^2B_F+g_A^2B_{\rm GT}\right], \qquad B_{\rm GT}=\frac1{2J_i+1} \left|\left\langle f\left\|\sum_k\boldsymbol\sigma_k\tau_k^+\right\|i\right\rangle\right|^2. \tag{44} \]

Equation (43) gives carbon-14 about \(2081\) times the phase space of tritium. If their reduced transition strengths were equal, carbon-14 would last \(0.00592\) years—about \(2.16\) days. The measured \(5700\) years instead requires a strength \(1.04\times10^{-6}\) of the tritium comparator, or a transition amplitude about \(0.00102\) as large.

Carbon has more room to decay—and almost does not.

The spectrum is integrated from Equations (42)-(43). The lifetime comparison holds the reduced matrix strength fixed, then measures how far that counterfactual lies from the evaluated half-life.

Electron kinetic energy0.00 keV

endpoint energy: 156.476 keV
phase-space integral: 0.0059930
measured half-life: 5700 y
equal-strength half-life: 0.00592 y
required amplitude ratio: 0.001019

What would a half-cycle have to explain?

The evaluated ground states are \(^{14}\mathrm C:0^+\) and \(^{14}\mathrm N:1^+\). Their beta transition is allowed by the usual Gamow–Teller spin-parity rule; its extraordinary lifetime resides in the small matrix element, not in a closed phase-space channel.

A half-cycle or half-twist hypothesis becomes physical only when it generates the parent and daughter fields, inserts them into Equation (44), and produces the observed suppression without borrowing the half-life. Doubling an orbital period or a leakage opportunity by itself increases exposure; it does not explain why the overlap amplitude is nearly a thousand times smaller than the tritium comparator.

Formation adds a second boundary condition. Atmospheric carbon-14 is produced primarily through \(^{14}\mathrm N+n\rightarrow{}^{14}\mathrm C+p\): a neutron enters the nitrogen arrangement while a proton leaves. Any winding account must evolve that reaction into the carbon state and then recover the suppressed return to nitrogen under the same operator.

What atomic pressure and nuclear geometry make the same neutron–proton transition fast in one arrangement, forbidden in another, and almost closed in carbon-14?

The computed field model, not its historical precursors

The working construction couples energetic propagation to a spacetime return structure, with conserved flux, orientation transport, and an equilibrium between compression and tension. Its quantitative claim is that one set of field rules should produce mass relationships, atomic structure, and transition behavior without readjustment for each measured result.

Geometric Maxwell and proper-time equations are possible mathematical ingredients, not authorities that define this theory. Detailed checks of those borrowed operators and their interactive examples are retained in the mathematical comparisons. The evidence here belongs to the operators actually used in each calculation.

What does the current compact operator put where we have not looked?

A fixed forward output of the present model is a multipolarity ordering for particle–hole openings in doubly closed nuclei. This is an energy-level calculation, not yet a prediction of detectable resonance strengths. The construction begins from three spatial transport generators, two null orientations per generator, and two paired surface sectors. These choices fix the compact coefficients; they are hypotheses, not fitted resonance centroids:

\[ d=3\times2=6, \qquad N_{\rm phase}=2^d=64, \qquad \kappa=\frac3{64}, \qquad w=\frac12, \tag{57} \] \[ \varepsilon_{N\ell j}=N+\frac32 -2\kappa\,\mathbf L\!\cdot\!\mathbf S -\kappa w\ell(\ell+1), \qquad \mathbf L\!\cdot\!\mathbf S= \frac{j(j+1)-\ell(\ell+1)-3/4}{2}, \tag{58} \] \[ r_c=\frac{\hbar}{m_pc}, \qquad L_c=2\pi r_c, \qquad E_0=\frac{(\hbar c)^2}{2m_pL_c^2}=\frac{m_pc^2}{8\pi^2}, \qquad \hbar\omega_*=2E_0=23.76670964\,\mathrm{MeV}, \tag{59} \] \[ E_{a\to b}=(\varepsilon_b-\varepsilon_a)\hbar\omega_*, \qquad |j_a-j_b|\le L\le j_a+j_b, \tag{60} \] \[ \pi_a\pi_b=(-1)^L\quad(E L), \qquad \pi_a\pi_b=(-1)^{L+1}\quad(M L), \qquad L\in\{1,2\}. \tag{61} \]

The proton mass sets the only dimensional scale in Equation (59). Filling Equation (58) to the observed closed counts \(2,8,20,28,50,82,126\), then applying the angular-momentum and parity rules, produces ninety-eight frozen seeds below \(25\,\mathrm{MeV}\) in the selected window: the last three occupied levels, first eight vacant levels, and oscillator shells \(N\le14\). No measured resonance energy enters that execution path.

Enumerating every occupied-to-vacant pair within the same finite basis produces 117 seeds without changing an energy or coefficient. Nickel-78 grows from 17 to 19 seeds while its first E1 opening stays at 17.082 MeV. Lead-208 grows from 18 to 29 and gains a lower E1 opening at 14.854 MeV. The figure preserves the original restricted list; the wider enumeration is a sensitivity check, not a replacement chosen for closer agreement.

Every seed stays where the operator put it.

The left field shows the occupied-to-vacant transition graph at the probe energy. The right field retains every seed in the frozen window and joins measured bands only afterward. Ticks indicate energy and compatible multipolarity, not calculated intensity; the scan is not a physical excitation simulation.

Probe energy5.00 MeV

frozen source hash: loading
seeds in nucleus: 17
first primitive E1: 17.082 MeV
measured comparison: no full response

Where does the operator already fail?

It finds the gross E1 scale unevenly. A helium-4 seed at \(24.881\,\mathrm{MeV}\) lies below the measured peak near \(26\,\mathrm{MeV}\). Oxygen-16 places a seed at \(22.653\,\mathrm{MeV}\) near its prominent giant response. Tin-132 places one at \(15.968\,\mathrm{MeV}\) beside the \(16.1(7)\,\mathrm{MeV}\) giant resonance.¹⁵¹⁶ Lead-208 places the corresponding seed at \(15.968\,\mathrm{MeV}\), about \(2.72\,\mathrm{MeV}\) above the \(13.25\,\mathrm{MeV}\) centroid.¹⁷

The more important failure is structural. Tin-132 has measured low-energy E1 strength near \(9.8(7)\,\mathrm{MeV}\); the compact list contains no primitive E1 seed there, even after widening the occupied–vacant window. It generates single particle–hole openings, not the collective response that can move strength below them. The energy of a nearby seed is not a calculated giant-resonance centroid: that comparison still needs transition strengths.

Nickel-78 remains a forward research target. No full high-energy multipolarity response was identified in the sources checked on 9 September 2026; existing spectroscopy establishes low states including the first \(2^+\) near \(2.6\,\mathrm{MeV}\).¹⁸ The frozen operator puts M1/E2-compatible openings at \(7.798\) and \(10.027\,\mathrm{MeV}\), while primitive E1 begins at \(17.082\,\mathrm{MeV}\). Those locations survive the wider enumeration.

Parity and angular momentum permit a transition; they do not establish a nonzero matrix element. The seed calculation supplies no individual strengths or widths, so it cannot yet predict M1/E2 dominance at 8–10 MeV or a detectable E1 onset near 17 MeV. Low-energy E1 would expose the incompleteness of the bare seed description, but an absent peak cannot reject an unspecified strength.

One aggregate constraint can already be calculated. Minimally coupling the compact operator to an electromagnetic potential changes its current because both angular terms depend on momentum. Its coordinate double commutator is \([r_i,[h,r_j]]=\delta_{ij}-2\kappa w(r^2\delta_{ij}-r_ir_j)\). With independently filled oscillator subshells, equal constituent masses and a dipole coordinate relative to the centre of mass, nickel-78’s total E1 energy-weighted sum is 0.885325 times the momentum-independent value. This conditional sum covers all excitation energies, not just the plotted window. Explicit oscillator radial integrals and spin–angular recoupling now reproduce that sum line by line for all seven nuclei. Neither the aggregate check nor these matrix elements establish that the common field action selects this electromagnetic coupling.

Can the same field action supply the electromagnetic current and normalized nuclear states, so that these frozen energies acquire predicted intensities and an uncertainty budget before a new measurement is judged?

What changes when the openings acquire strength?

This separate conditional calculation uses the unchanged energies, oscillator wavefunctions and the specified intrinsic E1 coordinate. Heights are calculated B(E1), summed for coincident energies. There is no fitted broadening, collective mixing or detector response. The frozen-seed figure above remains unchanged.

Closed nucleus

Nickel-78 has 23 conditional E1 lines and a full strength-weighted centroid of 21.041 MeV. The frozen window contains 69.21% of its energy-weighted strength. The interactive plot is loading.

A nearby seed is not a strength-weighted centroid. This completion places the full E1 centroids at 20.527 MeV for tin-132 and 20.000 MeV for lead-208. A full-spectrum mean is not identical to a fitted giant-resonance centroid, but these results prevent the closest individual seeds from being presented as calculated mean responses. Collective shifts, continuum strength and a controlled uncertainty budget remain unresolved.

Does the nucleus yield enough to an electric field?

Electric dipole polarizability measures the linear response to a weak field. It weights every E1 strength by the inverse of its excitation energy, so low-energy strength matters more. With B(E1) expressed numerically in e² fm², the conversion is \(\alpha_D=(8\pi/9)(\alpha\hbar c)\sum_f B_f/E_f\), in fm³. The existing oscillator completion fixes every term; no measured polarizability enters its calculation.

Polarizability over matched excitation-energy intervals, in fm³
Nucleus / interval	Conditional model	Published extraction
⁴⁰Ca / 10–25 MeV	1.089	1.50 ± 0.02
⁴⁸Ca / 10–25 MeV	1.258	1.73 ± 0.18
²⁰⁸Pb / 5–20 MeV	4.733	18.9 ± 1.3

The calcium intervals come from Birkhan et al.; the lead interval is reported by Tamii et al.^{28, 29} These are extracted E1 responses with the papers’ quoted uncertainties, not raw detector counts. The conditional calculation falls below all three central values. Its full lead-208 result, 6.539 fm³, is smaller even than the measured low-energy contribution. The wider-coverage lead value, after subtracting estimated quasideuteron absorption, is 19.6 ± 0.6 fm³.³⁰ This is a discrepancy of the specified nuclear completion; without a controlled theory uncertainty, no significance level is assigned to UMFT as a whole.

Positivity makes the problem sharper. For strengths above a lowest energy \(E_{\min}\), \(\alpha_D\le(8\pi/9)(\alpha\hbar c)m_1/E_{\min}^2\), where \(m_1=\sum_fE_fB_f\). Even arbitrary redistribution preserving lead’s calculated energy-weighted sum and 14.854 MeV E1 floor cannot exceed 11.321 fm³. Reaching the corrected central value while retaining that sum requires some E1 strength at or below 11.289 MeV. That is a necessary condition, not a predicted new line: the spectrum, ground-state response or coupling must change. Existing measurements already impose this constraint; another experiment is not needed to reveal it.

Can a stable field supply the missing response?

An interacting completion must obtain the stationary nucleus and its response from the same energy. A separate nuclear density functional already supplies attraction, saturation and proton–neutral imbalance. Displacing its converged proton and neutral densities in opposite directions gives a positive local restoring contribution for calcium and lead, with saturation offsetting part of the attraction. These are not the analytic oscillator states plotted above, and this partial restoring force is not a complete polarizability.

Transporting the angular-orientation operators adds a small softening term, but their straightforward full-space extension fails a stability test. With the unchanged coefficients, its tangential kinetic factor is \(1-3r^2/(64R^2)\), where \(R\) is the centred rms radius. Beyond \(r/R=\sqrt{64/3}\approx4.619\), phase variations can lower the energy without changing the densities. A converged spherical calculation or a finite numerical box can conceal this direction. This rules out that unrestricted extension, not every possible UMFT construction.

Current terms must also distinguish common motion from opposing internal flows.³¹ A scalar mean-current correction can restore common-phase invariance while leaving an orthogonal pair’s opposite flows untouched: their total current is zero, but their kinetic energy is not. That correction alone does not cure the negative-energy direction. A stable internal kinetic/current structure, or a physical boundary with derived matching conditions, must precede a new nuclear response claim. These are theoretical consistency requirements, not missing experiments.

A positive-square inverse construction shows how the angular instability can disappear when higher-order terms are retained. In dimensionless oscillator radius \(x\), its tangential kinetic factor is \(1/(1+3x^2/64)\), positive everywhere. Its first-order expansion matches the compact angular terms, but the completed operator changes the states and gaps: an additional gap at count 70 becomes larger than the gap at 28. This construction is not selected by a common UMFT action; matching the old coefficients does not establish that physical identification.

Solving the static response with this operator’s own normalized radial states and conserved current gives a full lead-208 polarizability of 5.893 fm³, below the earlier 6.539 fm³. Its entire response is still smaller than the measured contribution below 20 MeV, 18.9 ± 1.3 fm³.²⁹ No old oscillator overlaps, fitted enhancement or line width enter the new solve. Positivity repairs this kinetic problem, but does not supply the missing nuclear response. The interacting ground state and collective response still have to follow from the selected action.

A transverse overlap is not yet a nuclear equation of state.

The primitive’s curved-target curvature moments can now be calculated under refinement. They do not, by themselves, choose the three-dimensional density functional, orientation ensemble, or electromagnetic current. The scalar overlap integral previously used as a density scale also needs a thermodynamic distinction. For the local polynomial energy density \(\varepsilon=-g\rho^2/2+h\rho^3/3\), the mean field vanishes at \(\rho=g/h\), but the pressure \(P=\rho\varepsilon'-\varepsilon\) vanishes at \(\rho=3g/(4h)\). With \(g/h=C\), the latter is \(3C/4\simeq0.311515\) in the model’s density units. It is not the experimentally selected saturation density.

The actual circulating vector core imposes another exact constraint. Its spatial vector integral is zero, so its vector autocorrelation also integrates to zero. Its leading low-wavenumber overlap is tensorial and proportional to \(k^2\), not a positive scalar bulk overlap. A smooth scalar smoothing kernel cannot restore the missing constant mode. A singular inverse field operator can change that conclusion, but then the complete propagated interaction—not the bare overlap—must be derived.

Kinetic mixing can induce a local field in a nominally massless connection while its conserved displacement flux remains zero. With no passive charge or nontrivial boundary, the massive active source does not become a Coulomb monopole merely by changing the field basis. These results sharpen the missing nuclear construction without retuning the frozen spectra or their measured discrepancies.

Which uncertainty actually needs a new measurement?

The fixed tau-mass comparison above remains the admitted precision target: the two stated relations differ by 30.7123 keV, while a 5 keV total measurement uncertainty would resolve that separation under the stated input and theory assumptions. The experimental method is an existing tau-threshold proposal, not a new invention of this calculation.

The junction’s missing error bound is now more specific. In an explicitly defined periodic-gradient extension, a dimensionless profile stiffness of at least 0.12208745 bounds the extra relaxation shift by 5 keV for every common nonnegative finite-width contact in the tested class. Relaxation moves that mass downward, not toward the higher Koide value. The common action has not supplied this stiffness, so the bound remains conditional and does not alter either fixed prediction.

Nickel-78’s incomplete response, unselected optical couplings and the conditional core-radiation calculation remain research questions rather than new discriminating experiment proposals. They still lack an action-selected observable, its physical normalization or a controlled uncertainty budget. Existing calcium and lead data already constrain the nuclear completion. The remaining charge, spin and global-field questions require mathematical closure before another measurement can adjudicate them.

What entered, and what came out?

evaluationsupplied before solutionemitted by the modelmeasurement joined afterward

transverse coupling3:10 graph, odd-sector matching, fixed junction terms137.035999156447CODATA, reevaluated recoil, electron moment + Standard Model; historical selection not blind

loaded lepton relationelectron and muon masses; unchanged calculated coupling1776.938315 MeV/c²PDG 2025 tau average; relation motivated by known hierarchy

fixed-flux closurecompression, quartic return tension, fixed amplitudesquare-root mass coordinate, inverse mass–radius relationno independently measured return radius assigned

hydrogenmasses, \(e,\hbar,\epsilon_0\)normalized states, binding, transitionsNIST spectrum

multi-electron atom\(Z,N\), shared Maxwell operatorshell reset, radii, ionization openingsNIST ionization energies

photon absorptionpulse norm, detuning, mode overlapexcitation history and conserved ledgerLeong pulse-shape effect

pair productionphoton energies and collision angleinvariant energy and threshold permissionCODATA electron mass

muon decay\(G_F,m_\mu\), tree-level weak operatorMichel spectrum, \(2.18735\,\mu\mathrm{s}\)\(2.1969811\,\mu\mathrm{s}\)

carbon-14Q values, Coulomb phase space, tritium comparator2.16-day equal-strength counterfactual\(5700(30)\) years

nuclear seedsproton mass, closure counts, Equations (57)-(61)frozen energies and multipolaritiesknown response bands; nickel-78 lane empty

The new return-field figures load a source-hashed numerical packet: solved finite-vacuum profiles, their fluctuation eigenvalues, and a freshly recomputed nonzero charged core. The research directory records the equations, refinement studies, executable checks and limitations. The fixed internal group, shape penalty and vacuum enter the charged example; its profile, current balance and response follow. Canonical bosonic quantization enters the radiation calculation separately. No physical particle mass or lifetime is an output of that example.

Keeping measurement out of the calculation is necessary, but not sufficient for an independent prediction: the operator’s selection history must also remain visible.

What exists in this account?

A wavefront is the local advance of field phase and stress-energy. Maxwell’s equations govern its open propagation. A wavepacket is a finite normalized organization of fronts; the absorption calculation shows that its complete temporal and spatial mode is physically consequential. A waveknot is the name given to a recurrent organization whose directional momentum closes and whose exterior observables persist.

front

local propagation

\(F_{\mu\nu}(x)\)

→

packet

finite ordered event

\(\int|\xi|^2dt=1\)

→

knot

mass plus persistent recurrence

\(\mathcal O(t+\mathcal T)=\mathcal O(t)\)

Matter demonstrates persistent charged, massive, half-spin states. Interpreting those states as recurrent field knots is the hypothesis being tested, not an additional observation. A positive invariant mass alone does not demonstrate recurrence: the Maxwell packet above retains its invariant mass while spreading. Equations (28)-(33) state what the electron and proton constructions must calculate.

Neutrinos enter the working hypothesis as opened neutral return boundaries carrying source- and destination-specific transition information. The observed neutral channels constrain that interpretation. A complete test must generate their propagation, mixing, masses, and production amplitudes from the same boundary operator; neither naming a boundary nor importing an oscillatory equation completes that calculation.

The theory’s central wager is that open energetic propagation and timelike spacetime recurrence are not separate substances. The calculations now go beyond a schematic knot: they supply conditional matrix-field actions, finite profiles, a self-consistent charged core, driven and quantum radiation channels, and exact restrictions on its global interpretation. They do not yet identify that core with an electron or derive one uniquely selected action across all scales. The unselected vacuum and normalization, lighter charged escape channel, global field gluing, quantum rotation state, and nuclear response discrepancies are concrete parts of the remaining construction.

Can the same complete geometric action turn a front into a packet, a packet into a stable knot, and a knot back into measured radiation without changing its rules between scales?

References

NIST, Atomic Data for Hydrogen.
NIST Atomic Spectra Database, Ground States and Ionization Energies.
J. C. Maxwell, “A Dynamical Theory of the Electromagnetic Field”, Philosophical Transactions of the Royal Society 155 (1865).
M. Stobińska, G. Alber, and G. Leuchs, “Perfect excitation of a matter qubit by a single photon in free space”, Europhysics Letters 86 (2009).
V. Leong et al., “Time-resolved scattering of a single photon by a single atom”, Nature Communications 7, 13716 (2016).
G. Breit and J. A. Wheeler, “Collision of Two Light Quanta”, Physical Review 46, 1087 (1934).
E. Tiesinga et al., 2022 CODATA Recommended Values of the Fundamental Physical Constants, NIST (2024).
R. Pohl et al., “The size of the proton”, Nature 466, 213–216 (2010).
S. Navas et al. (Particle Data Group), “Review of Particle Physics: Summary Tables”, Physical Review D 110, 030001 (2024).
Particle Data Group, “Muon Decay Parameters”, in the 2024 Review of Particle Physics.
National Nuclear Data Center, Evaluated Nuclear Structure Data File, beta-decay evaluations for \(^{3}\mathrm H\) and \(^{14}\mathrm C\).
International Atomic Energy Agency, IAEA-C-3 Carbon-14 Reference Material, recommended \(5700(30)\)-year half-life.
J. Lindgren, A. Kovacs, and J. Liukkonen, “Electromagnetism as a purely geometric theory”, Journal of Physics: Conference Series 2987, 012001 (2025).
J. Lindgren, “On the Unitarity of the Stueckelberg Wave Equation and Measurement as Bayesian Update from Maximum Entropy Prior Distribution”, Quantum Reports 8, 18 (2026).
M. Murata et al., “Photodisintegration cross section of \(^{4}\mathrm{He}\) in the giant dipole resonance energy region”, Physical Review C 107, 064317 (2023).
P. Adrich et al., “Evidence for Pygmy and Giant Dipole Resonances in \(^{130}\mathrm{Sn}\) and \(^{132}\mathrm{Sn}\)”, Physical Review Letters 95, 132501 (2005).
A. Poltoratska et al., “Fine structure of the isovector giant dipole resonance in \(^{208}\mathrm{Pb}\)”, Physical Review C 89, 054322 (2014).
R. Taniuchi, “Competition of the Shell Closure and Deformations around the Doubly Magic \(^{78}\mathrm{Ni}\)”, Progress of Theoretical and Experimental Physics 2026, 04A103.
D. Tong, “Electromagnetism and Relativity”, Cambridge lecture notes: conserved, traceless Maxwell stress-energy tensor. Equations (30a)-(30e) give this publication’s moment derivation and explicit packet construction.
A. Delhom, “Minimal coupling in presence of non-metricity and torsion”, The European Physical Journal C 80, 728 (2020). Metric-volume and affine divergences differ in nonmetric geometry. Equations (47a)-(47b) are evaluated here in the convention of Equation (48).
M. J. Borchert et al., “A 16-parts-per-trillion measurement of the antiproton-to-proton charge–mass ratio”, Nature 601, 53–57 (2022).
Particle Data Group, Tau particle listing, 2025 update: mass average \(1776.93\pm0.09\,\mathrm{MeV}/c^2\).
NIST, 2022 CODATA complete constants table: inverse fine-structure constant and the electron/muon input masses.
R. H. Parker et al., “Measurement of the fine-structure constant as a test of the Standard Model”, Science 360, 191–195 (2018): cesium recoil.
L. Morel, Z. Yao, P. Cladé, and S. Guellati-Khélifa, “Determination of the fine-structure constant with an accuracy of 81 parts per trillion”, Nature 588, 61–65 (2020): rubidium recoil.
J.-H. Fu et al., “Novel method for identifying the heaviest QED atom”, Science Bulletin 69, 1386–1391 (2024). Projected tauonium and tau-mass sensitivity; not an achieved mass measurement.
R. Aliberti et al., “The anomalous magnetic moment of the muon in the Standard Model: an update”, Physics Reports 1143, 1–158 (2025), Equations (7.6), (7.7), and (7.10): updated alpha determinations.
J. Birkhan et al., “Electric Dipole Polarizability of ⁴⁸Ca and Implications for the Neutron Skin”, Physical Review Letters 118, 252501 (2017), pp. 3–4 of the manuscript: calcium partial integrals and high-energy completion.
A. Tamii, P. von Neumann-Cosel, and I. Poltoratska, “Electric dipole response of ²⁰⁸Pb from proton inelastic scattering: constraints on neutron skin thickness and symmetry energy”, The European Physical Journal A 50, 28 (2014), Section 5: partial and wider-coverage polarizabilities.
X. Roca-Maza et al., “Neutron skin thickness from the measured electric dipole polarizability in ⁶⁸Ni, ¹²⁰Sn, and ²⁰⁸Pb”, Physical Review C 92, 064304 (2015), Section II: quasideuteron subtraction.
N. Hinohara, “Energy-weighted sum rule for nuclear density functional theory”, Physical Review C 100, 024310 (2019), Sections III.3–III.4: common and isovector phase transformations; methodological context, not a validation of the UMFT completion.
C. M. Caves, “Quantum limits on noise in linear amplifiers”, Physical Review D 26, 1817 (1982). Canonical amplifier noise is an additional quantum framework, not a derivation from the classical core.
K. M. Benson and T. Imbo, “Topologically Alice strings and monopoles” (2004). Charge-conjugating transport requires the relevant global gauge structure.
S. Krusch and J. M. Speight, “Fermionic quantization of Hopf solitons”, Communications in Mathematical Physics 264, 391–410 (2006). Quantum spin restrictions depend on the full configuration space and its rotation loops.