Forensic Examination of the Kinematic Term
in Special and General Relativity

D. J. Hallman — SCG@azfn.com
2026 (v9) — DOI: 10.5281/zenodo.21201773
CC BY 4.0 © 2026 D. J. Hallman

1. The Michelson-Morley Experiment

A difficulty was identified following Maxwell’s unification. Under the Galilean transformation of Newtonian mechanics, the form of Maxwell’s equations was argued to change between frames in relative motion. The Michelson-Morley experiment of 1887 produced a null result for aether drag — removing one proposed explanation for the constancy of \(c\), but not establishing that Maxwell’s equations fail under Galilean transformation. The difficulty was mathematical, not empirical.

The Michelson-Morley apparatus split a beam into two perpendicular arms and compared return times between them. The design was deliberate: this symmetric there-and-back geometry cancels all effects first-order in \(v/c\) by construction. Both arms see the same first-order contribution from Earth’s motion through the medium; it drops out of the comparison algebraically before any fringe is counted. The apparatus was purpose-built to isolate the second-order term \(v^2/c^2\) specifically because first-order effects were already well understood and accounted for elsewhere.

It is worth being precise about what this means. Doppler — the confirmed first-order medium effect, operating at \(v/c \approx 10^{-4}\) for Earth’s orbital velocity, uncontested in every moving source and receiver since 1842 — was invisible to Michelson-Morley not because it was absent, but because the instrument was engineered to cancel it. The medium was announcing itself at first order continuously, in every stellar spectrograph, every naval sonar, every laboratory that had ever pointed an instrument at a moving source. The experiment designed to detect the medium was looking at an order where the medium’s primary effect could not be seen.

The apparatus looked for a fringe shift of approximately \(0.37\) fringes from Earth’s orbital velocity; the expected aether-drag signal scaled as \(v^2/c^2 \approx 10^{-8}\). Small shifts were observed but were far smaller than the Newtonian drag prediction and were judged inconsistent with a stationary aether. The result was reported as effectively null at the second-order level. What followed from that null result is the subject of the next section.

2. Four Investigators and the Decision

In the years between the 1887 experiment and Einstein’s 1905 paper, four physicists engaged directly with the Michelson-Morley result and proposed frameworks to accommodate it. All four worked within a medium framework. None made the decisive interpretive step.

George FitzGerald proposed in 1889 that matter physically contracts in the direction of motion through the aether by the factor \(\sqrt{1-v^2/c^2}\), offering a mechanism that would cancel the expected fringe shift. FitzGerald retained the aether throughout.

Hendrik Lorentz developed the same contraction hypothesis independently in 1892, and by 1904 had incorporated the factor into transformation equations that preserved the form of Maxwell’s equations between moving frames. Lorentz treated the factor as a mathematical tool encoding the dynamical interaction between matter and aether. He explicitly declined to assign it physical significance as a statement about time itself — a restraint he maintained even after 1905.

Joseph Larmor argued in 1897 that charged particles traversing their paths do so in times shorter by \(\sqrt{1-v^2/c^2}\), without identifying the physical mechanism and without the nuclear atomic model that would not exist until Rutherford’s 1911 experiments.

Henri Poincaré, in his 1900 Paris Congress address, asked openly whether the aether exists, and by 1904 had articulated what he called the principle of relativity — that the laws of physics must be the same for observers in uniform relative motion and that absolute motion is undetectable in principle. Poincaré came closer than any of the four to the 1905 framework, yet his derivations in 1905 remained anchored in an aether. He did not take the step of eliminating the medium.

In 1904, Emil Cohn applied the factor directly to the rate of clocks in moving frames without identifying the mechanism.

In 1905, Albert Einstein derived the same factor from two postulates — the equivalence of physical laws in all inertial frames, and the invariance of \(c\) — and attached to it the interpretive claim all four predecessors had declined to make: that the slowing it describes is a property of time itself, not of the propagation geometry. That step was a choice, not a logical necessity forced by the mathematics or by any experimental result. The factor had existed for eighteen years. It had been handled by four physicists without that attribution. Einstein attached it. That decision is the origin of kinematic time dilation.

That decision carried two errors from the MM null result directly into the new framework.

Error 1: A second-order null used to deny a first-order medium. The apparatus found no second-order aether drag. Einstein declared that the first-order medium does not exist. This is not a valid inference. The orders are independent. A null at \(v^2/c^2\) says nothing about what exists at \(v/c\). The apparatus could not have detected first-order effects regardless of whether the medium existed — that cancellation was built into the design. The medium was not tested at first order. It was convicted on evidence that could not have seen it.

Error 2: No second-order effect was detected, yet SR added one anyway. Michelson-Morley found no second-order aether drag. Special Relativity then introduced kinematic time dilation — a second-order effect scaling as \(v^2/c^2\) — as a law of motion governing all moving objects. The experiment that found nothing at second order was used to justify adding a second-order term to physics. The experiment’s own null finding was violated by the theory it was used to found.

One boundary consequence went unremarked. At \(v = c\) the factor \(\sqrt{1-v^2/c^2}\) returns zero. A clock whose rate reaches zero has no worldline. It can neither propagate nor carry the oscillatory structure that defines its frequency. The factor does not degrade gracefully at its own boundary — it annihilates the object it is applied to. Lorentz’s refusal to assign the factor physical significance left this boundary behavior unexamined. It was inherited by everything built on the factor thereafter.

3. Consequences

3.1 What the 1905 Decision Required

Every entry below was a necessary consequence of a prior decision. The list is chronological. Each entry stands on the one before it. It is not complete.

1905 — The null worldline. Eliminating the aether and assigning the Doppler propagation geometry to the source clock left the photon without a medium to carry its extended oscillation. At \(v = c\) the factor \(\sqrt{1-v^2/c^2}\) returns zero. The photon’s proper time vanished. Its worldline went null. With no proper time, the photon required a replacement evolution parameter. An affine parameter \(\lambda\) was introduced: an arbitrary quantity that increases monotonically along the null geodesic but has no physical content. Wave-particle duality was required to account for its simultaneous wave-like propagation and point-like detection.

1908 — The spacetime manifold. Minkowski extended Einstein’s single propagation term to three spatial dimensions, producing the spacetime manifold. Each spatial term is a squared propagation distance carried forward into a new geometric dimension. The four-dimensional manifold became the substrate of all subsequent relativistic physics.

1913 — Quantization by postulate. Bohr’s atomic model imposed quantization of electron orbits by fiat — as a rule without a geometric origin. The allowed radii were declared rather than derived. The model worked numerically. The mechanism was unidentified.

1916 — The metric and its singularities. Schwarzschild inherited the Minkowski manifold through his boundary condition. The metric produced a coordinate singularity at the event horizon and a physical singularity at \(r = 0\).

1923 — Wave-matter duality. de Broglie extended the wave-particle duality of the photon to matter. The reduced wavelength \(\bar{\lambda} = \hbar/p\) was placed at the center of wave mechanics without a geometric origin.

1925–1926 — The wavefunction and probability. Heisenberg, Schrödinger, and Born formulated quantum mechanics on the point-particle framework. The Born rule declared \(|\psi|^2\) to be the probability of detection. The uncertainty principle was formulated as a fundamental limit rather than a resolution limit. Hilbert space — an infinite-dimensional complex vector space — was required to represent the quantum states of objects that had been compressed to points.

1927 — The Copenhagen interpretation. Bohr and Heisenberg formalized the measurement problem: the wavefunction exists in superposition until measured, at which point it collapses to a definite outcome. The collapse mechanism was declared outside the scope of physical description. The observer was introduced as a necessary element of the theory.

1928 — The Dirac equation. Dirac constructed a wave equation compatible with special relativity, required because Schrödinger’s equation was not Lorentz-covariant. The equation required four-component spinors and predicted antiparticles. The positron was confirmed in 1932.

1929 — The expanding universe. Hubble’s observations of galactic redshifts were interpreted within the FLRW metric as recession velocities — spacetime itself expanding. The metric, built on the Minkowski backbone, required a \(T = 0\) origin: the Big Bang singularity.

1930s — Virtual particles and QED divergences. Quantum electrodynamics, built on point-particle interactions, produced infinite results when interactions were computed at zero spatial separation. Virtual particles — unobservable intermediate states — were introduced to regulate the interaction geometry.

1935 — EPR. Einstein, Podolsky, and Rosen showed that quantum mechanics as formulated implied either incompleteness or nonlocality. The question arose from the measurement problem and the absence of a physical account of wavefunction collapse.

1948 — Renormalization. Feynman, Schwinger, and Tomonaga developed systematic procedures for removing the infinities QED produced. The procedure subtracted one infinity from another and extracted the finite remainder. The necessity of the procedure followed from the point-particle assumption.

1950s–1960s — The particle zoo. Accelerator experiments produced a proliferation of short-lived resonances. Each was classified as a new fundamental particle. By the early 1960s hundreds had been catalogued. The point-particle framework provided no organizing principle for the proliferation.

1964 — The quark model and Bell’s theorem. Gell-Mann and Zweig introduced quarks to organize the particle zoo. In the same year, Bell showed that any local hidden-variable theory must satisfy certain inequalities that quantum mechanics violates.

1964 — The Higgs mechanism. Mass had no natural origin in the point-particle framework. The Higgs field was introduced to give particles mass through spontaneous symmetry breaking. The mechanism required an additional scalar field permeating all of space.

1967–1973 — The Standard Model. The electromagnetic and weak interactions were unified. Combined with QCD the Standard Model assembled three of the four fundamental forces. Nineteen free parameters — masses, coupling constants, mixing angles — were required and inserted by hand. No geometric origin was identified for any of them. Gravity remained outside the framework.

1957/1973 — Many-worlds interpretation. Everett proposed that wavefunction collapse does not occur — instead all outcomes are realized in branching parallel universes. The interpretation was a response to the measurement problem.

1968–present — String theory. String theory replaced point particles with one-dimensional strings to regulate the infinities at zero spatial extent and to address the singularities the Schwarzschild metric requires. The theory requires six or seven additional spatial dimensions. It inherits the manifold and the singularities.

1970s — Hawking radiation and the information paradox. Hawking showed that black holes emit thermal radiation through quantum effects near the event horizon. The result implied that black holes eventually evaporate, raising the question of whether the information of infalling matter is preserved or destroyed. The paradox requires the singularity at \(r = 0\) to be physical.

1970s–present — Dark matter. Galactic rotation curves could not be accounted for by visible mass distributions. A non-luminous, non-baryonic mass component interacting only gravitationally was proposed. It has not been directly detected.

1981 — Inflation. The FLRW framework produced a horizon problem and a flatness problem. A brief epoch of exponential expansion in the early universe was proposed to resolve them. The inflationary parameter space became effectively unconstrained.

1984 — Quantum cryptography. The BB84 quantum key distribution protocol was developed. Security relied on the assumption that a photon’s polarization state is genuinely indeterminate until measured.

1990s — Supersymmetry. The hierarchy problem — why gravity is \(10^{36}\) times weaker than electromagnetism — motivated supersymmetry. No supersymmetric particles have been observed at the Large Hadron Collider.

1994–present — Quantum computing. Quantum algorithms demonstrated that superposition and entanglement could outperform classical computation for specific problems. The computational advantage relies on the ontological interpretation of the point-particle wavefunction.

1998 — Dark energy. Observations of distant supernovae indicated an accelerating expansion. The cosmological constant was reintroduced as dark energy: a repulsive energy density filling all of space. Its magnitude requires fine-tuning to one part in \(10^{120}\) relative to the value quantum field theory predicts.

2000s–present — The string landscape and the multiverse. The unconstrained parameter space of string theory produced \(10^{500}\) or more possible vacuum states. The multiverse was proposed as the explanation for the observed values of physical constants. The anthropic principle was invoked in place of a derivation.

Present — Unresolved. The black hole information paradox is unresolved. The hierarchy problem is unresolved. The measurement problem is unresolved. Quantum gravity does not exist. The unification of the four fundamental forces is incomplete. Loop quantum gravity and string theory are incompatible with each other. Each open problem traces to the structures built after 1905.

This is not a complete list.

4. The Refutation

4.1 Definitions

The refutation rests on four definitions. Each is stated precisely before the contradictions are drawn.

c: The Recovery Rate of the Aether

Every wave-supporting medium has two properties that together determine how fast a disturbance propagates through it: its resistance to being disturbed, and its tendency to restore itself. For sound in air, these are the bulk modulus \(B\) and the density \(\rho\). The Euler fluid equation gives the speed of sound as:

The stiffer the medium, the faster it restores; the denser, the slower. This result predates Maxwell by two centuries. It describes a mechanism everyone understands: disturb the medium, it pushes back, the disturbance propagates forward at the rate the medium restores. The medium does not move. The disturbance does. The propagation speed is a property of the medium at a point — not a velocity of anything moving through it.

Maxwell’s 1865 equations revealed that the electromagnetic field has exactly the same structure. \(\varepsilon_0\) is the medium’s resistance to forming an electric field — its electromagnetic permittivity. \(\mu_0\) is its resistance to forming a magnetic field — its magnetic permeability. Their product governs the restoration rate:

This equation contains no velocity, no observer, no reference frame. \(c\) is the recovery rate of the aether. It is not a velocity in the Galilean sense. The recovery rate of a medium is a property of the medium at a point — how quickly that point returns to equilibrium after being disturbed. It does not transform with the observer’s motion.

The aether is not strange. It is the electromagnetic analogue of the compressible fluid Euler described. It predates Maxwell — Maxwell measured its properties for the first time in 1865. Every wave in nature propagates through a medium whose recovery rate is determined by two competing properties. The aether is that medium for electromagnetic waves.

This single recognition dissolves the problem the four investigators faced. Galilean velocity addition applies to objects moving through a medium. It does not apply to properties of the medium itself. Every observer measuring \(c\) locally gets the same number — not because of a special transformation, but because they are all measuring the same local property of the same field. The Michelson-Morley null result is then a confirmation, not a crisis. A field whose recovery rate is a local scalar would produce exactly that result regardless of Earth’s motion through it.

Time

Aristotle identified the nature of time in the fourth century BC: time is the number of motion with respect to before and after. Every clock ever built confirms this. A pendulum counts spatial traversals. A cesium atomic clock counts electromagnetic oscillations cycling through spatial field configurations at 9,192,631,770 Hz. In both cases what is measured is spatial change — things moving, fields oscillating, positions shifting, counted and compared.

Time is a relation. It requires no origin. A relation has no geometric depth. You cannot stretch a count. Kinematic time dilation requires a temporal geometric axis with genuine depth that velocity can act on and stretch. That substrate was never there.

Doppler

The Doppler effect is a relation between source, medium, and receiver. The frequency shift lives in the propagation path between them — specifically in the changing distance between successive signal emissions and the observer. It belongs to no single body. It requires a medium: a fixed propagation substrate relative to which source and receiver velocities are measured. There is no Doppler effect without a medium. It has been confirmed continuously since Doppler’s 1842 paper in every radar gun, sonar system, spectrograph, and fiber optic network ever operated.

The Doppler formula is structurally identical for sound and for light. The only difference is the propagation speed of the medium. If the SR interpretation is correct — that the frequency shift reflects a genuine slowing of the source clock — then applying the identical interpretation to a moving train requires that the train’s clock runs slow, causing its whistle to emit at a reduced rate \(f_0\sqrt{1-v^2/v_s^2}\) rather than \(f_0\). That prediction is wrong. Train whistles follow the classical Doppler formula exactly, with no clock-rate correction. The path geometry does the entire job in both media. A wrong interpretation in the sound case does not become a correct interpretation in the light case simply because the medium is different.

The Lorentz Transforms

The Lorentz transforms are the exact mathematical description of what a moving observer measures relative to a source in an electromagnetic medium with propagation speed \(c = 1/\sqrt{\varepsilon_0\mu_0}\). They are Doppler geometry — the relationship between source, medium, and receiver in relative motion — expressed as coordinate transformations. Lorentz derived them as descriptions of how electromagnetic signals propagate between relatively moving frames in a medium, and explicitly declined to assign them physical significance for the internal rate of clocks. They contain no clock internals. They identify no process by which velocity alters the rate of any oscillation. They are perspective transforms: what an observer measures given the changing propagation geometry of the medium between them.

4.2 The 1905 Derivation

A clock moves at velocity \(v\) relative to a stationary observer. Two successive ticks constitute two events. In the observer’s frame they are separated in both time \(dt\) and space \(dx = v\,dt\) — the distance the clock has traveled between ticks, which is the additional propagation distance each successive signal must cover to reach the observer. This is the classical Doppler effect. It is present in the geometry before the invariance condition is applied.

Einstein’s postulate that \(c\) is the same for all observers requires the quantity \(c^2dt^2 - dx^2\) to take the same value in both frames:

The term \(dx^2\) carries the classical Doppler propagation geometry into the invariance condition. Dividing by \(c^2\,dt^2\) and substituting \(dx = v\,dt\):

The quantity \(dx = v\,dt\) does not describe anything the clock does. It describes where the clock is when it emits each successive tick — the additional distance each signal must travel to reach the observer. That distance is a property of the propagation path, not the source. The invariance condition did not generate it. It absorbed it and reassigned it to the clock’s rate. A legitimate attribution would require identifying a physical mechanism internal to the clock by which velocity alone alters the rate of the oscillation constituting the tick. No such mechanism appears in the derivation.

4.3 The Algebraic Contradiction

All electromagnetic process rates are governed by \(\varepsilon_0\mu_0\) and geometry. For a canonical electromagnetic oscillator of length \(L\):

KTD asserts \(f(v) = f_0/\gamma\). Substituting and solving for what this requires of the medium:

This is not an interpretation. It is a necessary algebraic consequence: for KTD to physically reduce any electromagnetic process rate by \(\gamma^{-1}\), the local product \(\varepsilon_0\mu_0\) must increase by \(\gamma^2\). But SR’s second postulate, its requirement of homogeneity and isotropy, and the Lorentz transformation jointly establish that \(\varepsilon_0\) and \(\mu_0\) are invariant under velocity. The three statements cannot simultaneously hold:

• \(c = 1/\sqrt{\varepsilon_0\mu_0}\) — Maxwell (1865)
• \(f(v) = f_0\sqrt{1-v^2/c^2}\) — KTD
• \(\varepsilon_0,\,\mu_0 = \text{invariant}\) — SR Postulates

Any two may hold. All three cannot. The contradiction is internal to the orthodox framework, algebraic, and exact. No external theory is required to expose it.

4.4 Three Independent Lines of Impossibility

Three independent lines establish that KTD is not merely wrong but impossible, each sufficient on its own:

1. The misattribution is algebraically identifiable. The quantity \(dx = v\,dt\) in the 1905 invariance condition describes the additional propagation distance between successive signal emissions and the observer — a property of the signal path, not the source clock. The step that assigned it to the clock’s rate is locatable and contains no clock internals.
2. KTD contradicts SR’s own postulates. For KTD to physically reduce any electromagnetic process rate by \(\gamma^{-1}\), the local product \(\varepsilon_0\mu_0\) must increase by \(\gamma^2\) with velocity. SR’s postulates require \(\varepsilon_0\) and \(\mu_0\) to be invariant under velocity. The three statements — Maxwell, KTD, and the SR postulates — cannot simultaneously hold.
3. The required substrate was never there. KTD requires a temporal geometric axis with genuine depth that velocity can act on and stretch. Time is the count of spatial change — a relation, not a coordinate. A relation has no geometric depth. All curvature in the \(\varepsilon_0\mu_0\) framework is spatial. The temporal axis KTD would need to deform does not exist.

Each line is independent. Any one is sufficient. Together they establish that KTD is not a theory awaiting better experimental resolution — it is an algebraic impossibility within the framework that asserts it.

5. The Ground That Remains

5.1 What the Removal Preserves

Removing the misattribution is not a destruction. It is an excavation. What the kinematic term was riding on was always real. What it was obscuring was always sufficient.

The invariance of \(c\). Einstein’s foundational postulate — that the speed of light is the same for all observers regardless of the motion of the source — is preserved completely. The kinematic term was not the invariance of \(c\). It was the second-order reception geometry of the Doppler effect — \(dx^2 = v^2dt^2\) — absorbed into the invariance condition and misattributed to the clock’s rate.

Gravitational time dilation. The gravitational term in the Schwarzschild metric stands alone and sufficient. Clocks at different gravitational potentials run at different rates. That effect is real, confirmed, and uncontested, and it is gravitational Doppler: the field-ratio shift between two \(\varepsilon_0\mu_0\) environments at different gravitational potentials.

Maxwell’s equations. The complete description of electromagnetic phenomena is preserved. The null worldline that exiled the photon from its own evolution space dissolved with the manifold that required it.

Three spatial dimensions. The three spatial dimensions are preserved. Gravity is preserved — not as curvature but as field densification, derived in Section 5.2.

5.2 The Medium

The \(\varepsilon_0\mu_0\) field is the aether. It fills all of space. Its local value determines the local propagation speed \(c = 1/\sqrt{\varepsilon_0\mu_0}\) and the local impedance \(Z_0 = \sqrt{\mu_0/\varepsilon_0}\).

The field is position-dependent. Where it is elevated, \(c\) is lower and gravity is stronger. A photon climbing out of an elevated region loses frequency — its wavelength is set in a denser medium and read in a sparser one. A photon descending into an elevated region gains frequency. The frequency shift of light traversing a gradient is the direct observation. What has been called clock slowing is that same shift applied to the oscillation standard of the local detector: a clock deep in the field oscillates at the rate the local medium supports, which is slower than a clock in a sparser region. The comparison between them is gravitational Doppler.

The Acceleration Law

The Euler equation for a compressible fluid with pressure \(P\) and density \(\rho\) gives:

The \(\varepsilon_0\mu_0\) field has exactly this structure. \(\varepsilon_0\) is the field’s electromagnetic compressibility. \(\mu_0\) is its electromagnetic inertial density. Their product plays the role of density. Applying the Euler result:

This is not a postulate. It is the Euler fluid equation applied to the electromagnetic field. The mechanism is identical to sound. The result is gravity.

Mass and Gravity as Field Densification

A mass is a stable closed configuration of the \(\varepsilon_0\mu_0\) field: a region where the field is locally elevated and the closure geometry is satisfied in a standing mode. The unique spherically symmetric field profile consistent with the acceleration law and recovering the Newtonian limit at large \(r\) is:

Space is not curved by mass. It is densified by it. Objects follow the paths of least action through a medium whose density increases toward the mass. What appears as gravitational attraction is the acceleration law acting on all propagating structures in the field gradient.

5.3 Frequency Shift Taxonomy

All frequency shifts are Doppler. Doppler is a relation between source, medium, and receiver. There is no frequency shift that is not a consequence of one of three geometric configurations in the \(\varepsilon_0\mu_0\) medium.

Emission Doppler. A source moving through the medium during the electron transition physically sets the wavefront spacing deposited into the field. The photon’s physical length in the medium is \((c \pm v_s)\Delta t\) depending on whether the source recedes or approaches. The resulting frequency \(f = c/\ell\) is genuinely different from the rest-frame frequency:

Gravitational Doppler. A photon emitted in a \(\varepsilon_0\mu_0\) environment at density \((\varepsilon_0\mu_0)_1\) and received in an environment at density \((\varepsilon_0\mu_0)_2\) is measured at a different frequency. The photon itself does not change — \(Z_0\) is invariant, the energy flux of the propagating disturbance is conserved. What changes is the rate at which the detector reads the arriving wave against its own oscillation standard:

What has been called gravitational time dilation is gravitational Doppler.

Reception Doppler. A receiver moving relative to incoming wavefronts already in the field encounters them at a different rate. The photon’s wavelength — fixed at emission — is unchanged:

The clock-tick scenario in the 1905 derivation is reception geometry. Kinematic time dilation is reception Doppler misattributed to the source.

5.4 Sagnac

The Sagnac effect is the closed-path integral of emission Doppler plus reception Doppler. At every segment of a moving optical path, the travel-time difference per segment \(\Delta l\) moving at velocity \(v\) parallel to the propagation direction is:

Integrating around a circular loop of radius \(r\) with \(v = \omega r\):

This is the Sagnac formula, derived from first-order Doppler and geometry alone. No rotation physics, no preferred-frame postulate, no General Relativity. Wang et al. (2003, 2004) confirmed this directly by demonstrating the identical travel-time difference in a straight linearly-moving fiber with no rotation whatsoever — pure Doppler in an inertial frame. The rotation is the delivery mechanism that closes the path. It is not the physics.

SR Did Not Subsume Sagnac

Sagnac designed his 1913 experiment explicitly to demonstrate the aether and falsify Special Relativity. Paul Langevin provided SR’s response in 1921: the Sagnac effect is first-order in \(v/c\), and therefore presents no contradiction with SR.

This argument contains a fatal internal contradiction. First-order in \(v/c\) is the Doppler effect. The Doppler effect is a medium effect. It requires a fixed propagation substrate. There is no first-order \(v/c\) travel-time difference without something that \(v\) is measured relative to. That something is the \(\varepsilon_0\mu_0\) medium. Einstein eliminated the medium in 1905. Langevin claimed Sagnac for SR in 1921 on the grounds that it is first-order. Those two positions cannot both be held. A first-order result is a medium result. Langevin’s defence was a confession.

Wang closed the argument permanently. SR’s rotating-frame machinery requires a rotating non-inertial frame. Wang demonstrated the identical travel-time difference in a straight linearly-moving fiber in an inertial frame — no rotation, no closed path, no non-inertial coordinates. SR’s apparatus has no purchase on that result. The inertial-frame result is pure first-order Doppler in the \(\varepsilon_0\mu_0\) medium: exactly what Einstein declared did not exist in 1905.

5.5 The Closure Constant γ_cause

Any oscillation propagating at a single speed \(c\) traces a sinusoidal path whose total arc length per wavelength exceeds the forward propagation distance. Two independent arguments converge on the same condition:

The causal argument. Causality does not distinguish between frequencies: two oscillations of different frequencies traversing the same medium must cover the same arc-to-closure ratio, or the medium would impose a preferred frequency. For a sinusoidal oscillation \(y(x) = A\sin(kx)\), the unique value that introduces no external length scale is:

The least-work argument. Maupertuis’s principle of least action applied to a closure-constrained oscillation demands the same condition \(\beta = 1\).

With \(\beta = 1\) fixed, the arc length per wavelength is:

where \(E(\cdot)\) is the complete elliptic integral of the second kind. The arc-to-closure ratio is:

Any propagating oscillation in any medium finds this ratio geometrically enforced. It is as substrate-independent as \(\pi\). The closure condition also fixes the transverse extent directly: \(\bar{\lambda} = \lambda/2\pi\). The reduced wavelength has been present in every quantum mechanical calculation ever performed — in \(\hbar = h/2\pi\), in the Bohr radius, in the de Broglie relation. We have been using it for a century without knowing where it came from.

5.6 Quantization

The closure condition \(\beta = 1\) is a geometric requirement on any stable physical structure in any scalar field. A structure whose geometry does not close on itself without discontinuity is not stable — it disperses. What persists are the structures whose geometry satisfies \(\beta = 1\). Those structures are discrete by necessity.

At atomic scales the closure condition selects the orbital radii where the field gradient closes without discontinuity — the observed electron shells. At nuclear scales it governs which nucleon configurations form stable structures — the magic numbers. Discreteness is not a mystery imposed on top of classical physics. It is what a scalar field does.

5.7 The Sagnac Mass

The Equivalence Principle Is an Identity

Einstein declared that acceleration and gravity are indistinguishable. They are not merely indistinguishable. They are the same phenomenon. The acceleration law \(\mathbf{a} = c^2\nabla\ln(\varepsilon_0\mu_0)\) is both. There is one equation, one field, one mechanism.

A bicycle wheel’s rim is always accelerating toward the axle. That centripetal acceleration is, by the equivalence principle, a gravitational-equivalent field pointing inward. Sagnac’s experiment measures the phase shift produced by a rotating loop. It is therefore measuring the effect of that inward acceleration on light propagating around the loop. Sagnac measured gravity in a laboratory in 1913 without calling it that.

Inverting the Sagnac Formula

Applied to a stable field closure satisfying \(\beta = 1\) with \(\gamma_{\rm cause}\) as the least-work arc-to-closure ratio, the closure radius is:

Inverting for mass:

No free parameters. The closure radius is set by the geometry. Applying this to the electron and proton:

The mass ratio is a ratio of closure radii. It was not fitted. It fell out.

The Photon’s Sagnac Mass

At the apex of each half-cycle, the radius of curvature is at its minimum: \(R_{\rm apex} = \bar\lambda\). The Sagnac mass is maximum. The apex is at rest in the medium — constitutively stationary. \(E = mc^2\) applies exactly, literally:

At the zero crossing, the radius of curvature is \(R_{\rm zero} = \bar\lambda/2\sqrt{2}\). The zero crossing carries \(2\sqrt{2} \approx 2.83\) times more Sagnac mass than the apex. Here the contained Sagnac mass releases and drives the next half-cycle. The zero crossing is the propagation engine.

Momentum Without Paradox

The apex mass is real. It is at rest. It never closes into a persistent structure. A mass at rest without closure contributes no Newtonian momentum. Momentum comes from the transfer — the release of Sagnac mass at the zero crossing driving forward into the next apex. This delivers \(p = mc\) where \(m\) is the apex mass:

This is \(p = mc\) — the ordinary equation, applied without modification. \(E = pc\) follows from \(E = mc^2\) and \(p = mc\) combined. The apparent paradox of a massless momentum-carrier dissolves. The Sagnac mass scales as \(1/\bar\lambda \propto \nu\): this is why ultraviolet light breaks chemical bonds and infrared does not.

5.8 The New Photon

The null worldline manufactured a dimensionless point-particle by stripping the evolution space from Maxwell’s extended oscillation. The closure geometry restores it.

The Two Phases

The apex. The arc is at maximum curvature. The transverse velocity is zero. The Sagnac mass is maximum. The apex is a mass at rest in the \(\varepsilon_0\mu_0\) medium — stationary relative to the medium, not relative to the source or detector. \(E = mc^2\) is satisfied exactly. The photon achieves this twice per wavelength, by geometry alone, without permanent closure. Both \(\varepsilon_0\) and \(\mu_0\) are displaced in the same sense at the apex — a product perturbation. This is the gravitational face of the photon.

The zero crossing. The arc straightens. The Sagnac mass releases. The released energy drives the next apex. The photon is self-threading: the straights release what the curves built. No carrier particle. No virtual intermediary. Recovery cascade at \(c\).

Emission and Absorption

A photon is not ejected from an atom. The electron falls, vacating the field geometry it was sustaining. The \(\varepsilon_0\mu_0\) medium heals the abandoned geometry at \(c\) — the medium’s relaxation rate, concurrent with the fall. The photon is the medium restoring itself to equilibrium impedance \(Z_0\). Absorption is the exact time-reversal.

Wave-Particle Duality Dissolves

Every photon is a sine wave sustained by \(\beta = 1\) closure. It is wave-like because it is a physical oscillation in the \(\varepsilon_0\mu_0\) medium. It is particle-like because it is topologically stable, discrete, and non-dissipative over cosmic distances. Wave-particle duality dissolves because the sine wave IS the particle IS the wave. One geometric object that orthodox physics described from two incomplete angles simultaneously.

The three-polarizer experiment confirms that the point-particle photon does not exist in nature. Two crossed polarizers transmit no light. Inserting a third at 45° restores partial transmission. This result is impossible if photons carry pre-fixed binary polarization states. The only consistent account is that each polarizer redefines the polarization geometry of an extended field structure. The experiment is reproducible with three polarizing filters from a camera shop.

5.9 The ΛCDM Zoo and What Created It

The spacetime manifold absorbed a share of the total curvature budget into the temporal dimension. At galactic scales the curvature remaining in three spatial dimensions was systematically insufficient to account for observed rotation curves. The deficit was declared missing mass. Five distinct observational signatures were attributed to it. Fifty years later, no dark matter candidate has been detected in direct detection experiments, at the Large Hadron Collider, or in indirect astrophysical searches.

Each of the five deficits has a distinct geometric resolution in the \(\varepsilon_0\mu_0\) framework, from the same field, with the same invariant \(\gamma_{\rm cause}\), at zero additional parameters.

1. Rotation curves. The galaxy’s field naturally segments into causal domains with spacing \(\Delta r_i = \gamma_{\rm cause}\sqrt{r_i}\). Applied to 175 SPARC galaxies, the domain spacing law predicts kinematic transition locations with median RMSD 1.06 km/s, zero free parameters, no halos.

2. Gravitational lensing. \(\gamma_{\rm cause}\) scales the Einstein radius by the full spatial causal arc overhead, recovering the field depth GR’s temporal dimension absorbed. Systematic 18–21% enhancement, zero free parameters, across 186 lenses.

3. Cluster collisions. The field carries no electromagnetic cross-section and continues on the original trajectory while the plasma decelerates. Lensing follows the field. No particle required. The lensing centroid offset should decrease over time as the field relaxes — a testable prediction.

4. CMB acoustic peaks. The peak sequence arises from interference of \(\gamma_{\rm cause}\)-spaced causal shells. The odd/even asymmetry follows from the parity of the interference function \(F(k\tau_{\rm CMB})\) without a dark matter potential well.

5. Large-scale structure. The matter power spectrum — including the turnover near \(k_c \approx 0.02\,h\,\text{Mpc}^{-1}\), slope transitions, and filament spacing near 150 Mpc — all emerge from \(\nabla^2\ln(\varepsilon_0\mu_0)\) without dark matter seeding.

From one error came five problems. From five problems came a zoo. WIMPs, axions, sterile neutrinos, primordial black holes, self-interacting dark matter, fuzzy dark matter, warm dark matter, and MOND were each introduced as responses. No candidate has been detected. The five deficits share one resolution: \(\gamma_{\rm cause}\) applied to a scalar field that was always there.

5.10 G and M

\(G\) and \(M\) do not exist independently in the \(\varepsilon_0\mu_0\) framework. \(GM\) is a single field quantity: the integrated \(\varepsilon_0\mu_0\) field elevation over the closure volume of a mass configuration.

G as a Units Bridge

\(G\) appears constant because in every environment where it has been measured, \(\sqrt{\varepsilon_0\mu_0}\) is approximately uniform. It is not constant. It is locally stable. The persistent scatter in precision laboratory measurements of \(G\) — unresolved within the standard framework for decades — is a parameter-free prediction of this \(\varepsilon_0\mu_0\) dependence. The scatter is not experimental error. It is real.

M as Field Elevation

\(GM\) is what the field directly yields. \(G\) and \(M\) separately are two ways of reading the same field quantity in a unit system that was not designed for it.

Inertia

Inertia is the same mechanism as gravity. An object in a uniform \(\varepsilon_0\mu_0\) field has no preferred direction. An applied force locally compresses the field in the direction of acceleration, creating a gradient. The acceleration law then acts on that gradient. The resistance is the object’s inertia. There is one field, one gradient law, one mechanism.

5.11 The Fine-Structure Constant α

The fine-structure constant \(\alpha \approx 1/137.036\) is not a free parameter of nature. It is the coupling efficiency between the photon’s complete arc geometry and the electron’s circular closure geometry, expressed entirely in \(\gamma_{\rm cause}\) and \(\pi\), with no empirical input.

The photon’s arc has three independent geometric components contributing to the total arc-length ratio \(\gamma_{\rm total}\) in quadrature. Component 1 is the primary transverse oscillation with ratio \(\gamma_{\rm cause} \approx 1.2160\). Component 2 is the forward hemisphere correction:

Component 3 is the Sagnac depth oscillation with three-dimensional coupling factor \(13/4\). The three components add in quadrature:

The electron’s circular closure contributes two powers of \(\gamma_{\rm cause}\). Together:

The CODATA value is \(1/\alpha = 137.035999084\). The gap of 0.0015% is identified as contamination from kinematic time dilation assumptions in the empirical extraction procedure — not a missing geometric term. This derivation contains no empirical input and no adjustable parameters. The number 137 is not a mystery. It is the geometry of light coupling to charge.

5.12 ℏ and h

The closure condition \(\beta = 1\) forces the photon’s transverse radius to \(\bar\lambda = \lambda/2\pi\) for all frequencies. The photon’s momentum is \(p = E/c\). Their product:

Every step follows from the photon’s geometry. \(\hbar\) is momentum times the geometric radius of the oscillation that carries that momentum. It is the closure condition expressed in SI units. \(E = \hbar\omega\) is not a postulate of quantum mechanics. It is what the type-2 elliptic closure geometry of a propagating oscillation in the \(\varepsilon_0\mu_0\) field requires. Planck measured a geometric cycling cost in 1900 without knowing that was what he was doing. \(h\) is Maxwell’s constant, identified 35 years late.

5.13 Mercury

The \(\varepsilon_0\mu_0\) field profile surrounding the Sun expands to second order as:

The first term is the Newtonian inverse-square acceleration. The second is the leading relativistic correction — a \(1/r^3\) perturbation from the nonlinearity of the exponential field profile. Integrating over one complete orbit with semi-latus rectum \(\ell = a(1-e^2)\):

The factor \(6\pi\) is not a free parameter. It is a direct consequence of the exponential field profile and the orbital averaging. Substituting Mercury’s orbital parameters (\(a = 5.791\times10^{10}\) m, \(e = 0.2056\)):

The observed residual is \(43.6 \pm 1.0\) arcsec/century. Agreement is within observational uncertainty. Three ingredients: the exponential \(\varepsilon_0\mu_0\) profile, the orbital geometry, and the acceleration law. No kinematic term. No free parameters.

6. Claimed Confirmations and Corrections

The case against kinematic time dilation is complete. Three independent algebraic lines establish that KTD is impossible within SR’s own framework. The positive mechanisms — Doppler, Sagnac, gravitational field coupling — are derived from first principles in the preceding sections. What follows is not required to establish the argument. It is offered for completeness: each claimed experimental confirmation of KTD examined against the mechanisms the reader now has in hand.

Before examining individual experiments, two prior matters govern the logical status of any claimed confirmation. Section 4.4 establishes that KTD is algebraically impossible within SR’s own framework. A mechanism shown to be impossible cannot be confirmed by measurement. What measurement can do is reveal what is actually being measured.

The first prior matter is the Michelson-Morley double error identified in Section 1. The apparatus found no second-order aether drag. That null was used to deny the first-order medium — an inference across orders that is not valid. SR then added a second-order effect to motion anyway, as a law of nature, citing the same null result as justification.

The second is the Doppler misattribution identified in Section 4.2. The propagation geometry \(dx = v\,dt\) was assigned to the rate of the source clock rather than the propagation path. Every experiment cited as KTD confirmation measures either a frequency shift or a clock comparison. Frequency shifts belong to Doppler — including gravitational Doppler. Clock comparisons involve acceleration and gravitational potential.

Every major claimed confirmation shares a common feature: the measurement was performed inside a rotating reference frame, or in a regime where orbital velocity and centripetal acceleration are geometrically coupled. The Sagnac effect — derived in Section 5.4 as the closed-path integral of first-order Doppler, confirmed in a straight linearly-moving fiber by Wang et al. (2003, 2004) without any rotation — produces numerically identical results to the KTD prediction in every rotating-frame experiment examined here.

6.1 The First Kinematic Prediction

SR’s first kinematic prediction — the transverse Doppler shift at \(v^2/2c^2\) — is frame-dependent. The prediction changes depending on which frame the calculation is performed in. A physical effect cannot depend on the choice of calculation frame. The frame-dependence is a direct consequence of assigning the propagation geometry to the source clock: different frames disagree on how much of the Doppler shift to attribute to the clock rate and how much to the path geometry.

6.2 The Frame-Dependence of Photon Energy

If KTD is real, a photon emitted by a moving source carries less energy than a photon emitted by a stationary source at the same frequency, because the moving source clock runs slower. But the photon’s energy is fixed at emission by the field geometry — it is independent of what the source was doing before or after emission. The KTD interpretation requires the photon to carry information about the source’s velocity in its energy, which is not a property that photons carry. Doppler handles this correctly: the frequency shift is in the wavefront spacing, not in the photon energy per quantum.

6.3 The Geoid

The geoid is defined as the equipotential surface of Earth’s gravitational field at mean sea level. GPS clocks are corrected to run at the rate they would run on the geoid. This correction is purely gravitational — it uses the \(\varepsilon_0\mu_0\) field-ratio shift between orbital altitude and the geoid. KTD is applied separately, and then removed, leaving the gravitational correction as the net effect. A term that is applied and then removed performs no physical work. The GPS system operates correctly without KTD.

6.4 The GPS Clock Correction

The GPS clock correction of \(+38.2\,\mu\text{s\,day}^{-1}\) has three contributions.

Gravitational. The satellite operates at altitude \(r = 26{,}571\) km in a sparser \(\varepsilon_0\mu_0\) field than the geoid. The field-ratio shift gives \(+45.9\,\mu\text{s\,day}^{-1}\).

Sagnac. The satellite traverses a rotating path around Earth. The Sagnac integral for the GPS orbital geometry gives \(-7.2\,\mu\text{s\,day}^{-1}\). Note that the KTD prediction for this velocity is also \(-7.2\,\mu\text{s\,day}^{-1}\) — numerically degenerate at GPS altitude to four significant figures.

Total. \(+45.9 - 7.2 = +38.2\,\mu\text{s\,day}^{-1}\). Three physical ingredients. No kinematic term. No free parameters.

The Galileo constellation at orbital altitude 23,222 km breaks the degeneracy definitively. At GPS altitude the Sagnac and KTD velocity contributions agree to within \(0.004\,\mu\text{s\,day}^{-1}\). At Galileo altitude the same calculation gives Sagnac \(+7.60\,\mu\text{s\,day}^{-1}\) and KTD \(-6.46\,\mu\text{s\,day}^{-1}\): a difference of \(1.14\,\mu\text{s\,day}^{-1}\), well within the detection threshold of Galileo’s passive hydrogen maser clocks. GPS was placed at the one altitude where the two mechanisms are numerically degenerate. That coincidence of geometry postponed the experimental discrimination by decades. Galileo is operational. The discrimination is now available.

6.5 The Hafele-Keating Experiment

Hafele and Keating flew atomic clocks around the world in 1971 and compared them to ground clocks. The east-flying and west-flying clocks showed different time differences, attributed to the combination of gravitational time dilation and KTD. The Sagnac effect of Earth’s rotating frame accounts for the directional asymmetry exactly. The east-flying clock moves faster relative to the Earth’s center (ECI frame) and accumulates more Sagnac phase; the west-flying clock moves slower. The gravitational contribution is the \(\varepsilon_0\mu_0\) field-ratio shift at flight altitude. The result is fully accounted for without KTD.

6.6 The Seasonal Stellar Redshift

Stars observed at different times of year show a periodic frequency variation correlated with Earth’s orbital velocity. This has been cited as confirmation of the transverse Doppler effect and therefore KTD. The Sagnac effect of Earth’s orbit around the Sun — a closed path at orbital velocity \(v \approx 30\) km/s — produces a frequency variation of identical period and amplitude. The seasonal variation is the Sagnac effect of the orbital loop, observable in stellar spectra.

6.7 The Ives-Stilwell Experiment

Ives and Stilwell (1938) measured the second-order Doppler shift of fast hydrogen ions and confirmed the frequency shift \(\Delta f/f = v^2/2c^2\). This has been cited as the direct measurement of KTD. The second-order term \(v^2/2c^2\) follows from the classical Doppler expansion: the average of the forward and backward first-order shifts leaves a residual of exactly \(v^2/2c^2\). Ives himself stated that his result confirmed a classical wave-medium theory, not Special Relativity. The formula predates SR. It requires the medium. It is classical second-order Doppler.

6.8 The Muon

Cosmic ray muons reach sea level despite their short proper lifetime. This is cited as KTD evidence: the muon’s clock runs slow in the lab frame. The \(\varepsilon_0\mu_0\) field gradient through the atmosphere provides an alternative account. The muon propagates through a field that varies from the top of the atmosphere (sparse \(\varepsilon_0\mu_0\)) to sea level (denser). The field gradient acts on the muon’s internal oscillation rate in the same way it acts on a clock — gravitational Doppler applied to the decay rate. The effect is equivalent to the gravitational time dilation that the \(\varepsilon_0\mu_0\) framework provides without invoking velocity as the cause.

6.9 The Point-Particle Photon

The null worldline required the photon to be a dimensionless point particle. The three-polarizer experiment falsifies this directly. Two crossed polarizers transmit no light. Inserting a third at 45° between them restores partial transmission. This is impossible for a point particle with a fixed binary polarization state: a particle blocked by the first crossed pair cannot be unblocked by adding a third filter. The only consistent account requires an extended field structure whose polarization geometry is continuously redefined at each filter. Malus’s Law governs every step. The experiment is reproducible with three polarizing filters from a camera shop. KTD did not discover point-particle photons. It manufactured them as a requirement of the null worldline. The experiment shows what was manufactured does not exist.

Conclusion

One step. One paper. One year.

In 1905 the propagation geometry between a moving clock and its observer — the changing distance \(dx = v\,dt\) each successive signal must travel to reach the observer — was absorbed into an invariance condition and assigned to the rate of the clock itself. The assignment had no physical basis. The derivation contained no clock internals. No mechanism was identified by which velocity alone alters the rate of an oscillation. The geometry belonged to the propagation path. It was placed on the source.

That step was not forced by the mathematics. The factor \(\sqrt{1-v^2/c^2}\) had existed for eighteen years. Four investigators had handled it without that attribution. The Michelson-Morley null result cited as justification committed two errors simultaneously: using a second-order null to deny a first-order medium, and using the same null to license a second-order kinematic replacement of identical magnitude. The medium had been announcing itself at first order in every Doppler measurement since 1842. The apparatus used to declare it absent was designed to be blind to the primary evidence for it.

Three independent algebraic lines establish that kinematic time dilation is not merely wrong but impossible within the framework that asserts it. No experimental result, however precise, can confirm an algebraically impossible mechanism. What experiment can do is reveal what is actually being measured. Every claimed confirmation examined here resolves to Doppler, Sagnac, gravitational field coupling, or numerical degeneracy at a specific orbital geometry.

Remove the misattribution. What remains is Maxwell’s field, read directly. The \(\varepsilon_0\mu_0\) medium is the aether. Its recovery rate is \(c\). Its gradient is gravity. Its closure geometry is mass. From those three facts, without postulates and without free parameters: the acceleration law, the exponential field profile, the Sagnac effect derived from Doppler alone, \(\hbar\) from the closure condition, \(\alpha\) from three-component arc geometry, \(G\) as a units bridge, and Mercury’s perihelion precession of 42.9 arcseconds per century.

The Sagnac formula inverted, with \(\gamma_{\rm cause}\) as the least-work closure ratio, yields the electron mass, the proton mass, and their ratio of 1836.15 — from geometry, with no fitted parameters, matching the empirical values exactly. The photon’s apex satisfies \(E = mc^2\) exactly, twice per wavelength. The mass is real. The momentum comes from the transfer. The ordinary equations apply without modification or exception.

The geometry was always in Maxwell’s equations. The medium was always there. The passenger that obscured them rode for a century on a propagation distance that was never the clock’s.

All works are published open-access through Zenodo under a CC BY 4.0 license. Full publication list: Zenodo · ORCID 0009-0000-1710-3549 · Contact: SCG@azfn.com