Gardi-Lori: PhD physicist review

Reviewer role: PhD physicist, mainstream view. Inputs: facts.json, claims.md (Step 1). I re-read the key passages myself: Gardi 2017 "A Medium for the Propagation of Light Revisited" (full text), theOMparticle 2023 section 4, and rechecked the numbers below with NumPy. I did not re-read the Yee-Gardi equation pages as images; for those I rely on Step 1's page-image readings (p.7, 8, 28, 29 of the Spacetime paper) and say so where it matters.

Source gap (repeated from Step 1): the numbered equations of the Yee-Gardi papers (Electric Universe, Mass and Charge, Geometry of Particles, Geometry of Spacetime) are not in the text layer, and the companion papers on the mole, the fine-structure constant and Pi, plus Distinti's and Yee's wave-constant papers and Gardi's 2016 Physics Essays paper, were not available. Verdicts on the "dozen derived constants" therefore rest on the pages Step 1 rendered (the G derivation, p.28) plus a general argument that Planck-unit identities cannot yield independent predictions. Any statement about the companion works is [memory] or unknown, not read.

Web checks (budget about 6): the search tool was out of budget, so I used 3 page fetches. Results are tagged [web]. Two checks failed and stay [memory]; each is flagged.

Corrections to Step 1

1. facts.json says the Fermi 0.4 s transient was "not confirmed by INTEGRAL" at [memory] medium. Now [web] checked: Fermi GBM reported "a weak transient above 50 keV, 0.4 s after the GW event, false alarm probability 0.0022 (2.9 sigma)" (https://arxiv.org/abs/1602.03920). INTEGRAL found no significant emission, with E_gamma/E_GW < 1e-6 (https://arxiv.org/abs/1602.04180). So the "S-wave" counterpart is a 2.9-sigma candidate that another instrument did not see. It is not an established event.
2. facts.json (Medium paper) calls 1.022 MeV a "binding energy". Gardi does the same, but physically 1.022 MeV = 2 m_e c^2 is the annihilation (rest-energy) release, not a lattice binding energy. Her "derivation" is therefore E = mc^2 rearranged. Step 1's conclusion was right; the label needs this care. I rechecked: sqrt(2 m_e c^2 / 2 m_e) = 299792458 m/s identically, and her A = 1.6374e-13 J equals 2 m_e c^2 to 8 digits.
3. The Fermi GRB 090510 Lorentz-violation point stays [memory] medium. My fetch of the arXiv number returned an unrelated page, so the check failed. The claim that Planck-scale lattices are constrained by gamma-ray timing is not needed for any verdict below; the lattice verdict does not depend on it.
4. Mandelbrot paper: Step 1 cites the EHT ring only from memory. [web] the EHT M87* image is an asymmetric ring about 690 AU across, with asymmetry attributed to relativistic beaming (https://en.wikipedia.org/wiki/Messier_87). The same page reports that a 2010 suggestion that the M87 black hole is offset from the galaxy centre was not found significant in 2011, and a 2018 study attributed the apparent offset to jet brightness variability, not a physical displacement. This weakens the paper's "22 light-year offset" evidence. Note: a secondary source, not the primary papers.
5. Otherwise I confirmed Step 1's numbers: NaCl speed 3699 m/s; m c^2 / r_e = 29.05 N; Z0/(2 pi) = 59.958 ohm; e^2/Q_p(h)^2 = alpha/(2 pi) = 0.00116141 (Q_p(h)^2 = 4 pi eps0 h c, checked numerically); l_P c^2 / m_P = 6.6743e-11 = G. The 2023 paper's section 4 text is as Step 1 describes [read].

Claim 1: light and all forces propagate in a physical medium (aether)

a. Mathematics. The 2017 argument is an analogy, c = sqrt(1/(eps0 mu0)) rewritten as sqrt(mu0^-1 / eps0) to look like sqrt(stiffness/density). That is algebra, not structure: eps0 is not a mass density and 1/mu0 is not a modulus (units differ). The "pair medium" calculation divides A by (B + C), where A = 2 m_e c^2 and B + C = 2 m_e, so it returns c by construction; it also "works" for proton-antiproton or quark pairs for the same reason, which shows it carries no information about electrons. A mass "per unit volume" is quoted but no volume appears (the units are kg, not kg/m^3), so even the NaCl analogy is dimensionally loose [read]. Verdict on (a): internally inconsistent in the derivation; the weaker idea (a medium exists) is not itself self-contradictory.

b. Experiments. Michelson-Morley and its modern successors (rotating cryogenic and optical-resonator tests, Kennedy-Thorndike, Ives-Stilwell) show no preferred frame to very high precision [memory, high]. Gardi's rebuttal, that using light to detect the medium of light is "circular", is wrong: the experiment compares light speed in two directions, which would differ if a medium frame existed. Her air analogy fails the same way: a sound-based MM experiment in air does detect wind relative to the apparatus; a null result in an apparatus co-moving with the air is expected, and no one concludes air does not exist. The ultrasonic test she cites is unofficial and unreferenced beyond a ref [8]. Also, a lattice of electron-positron pairs would have enormous dispersion, charge screening and gamma-ray scattering effects that are not observed [memory, medium; reasoning only]. A vacuum with an Fermi-timing-scale Lorentz-violation limit: [memory, medium, check failed].

c. Tested? Aether theories were tested directly from 1881 on (Michelson-Morley, 1887; later many repeats) and failed [memory, high]. Gardi's specific pair-lattice model has not been tested by anyone other than by the broad Lorentz-invariance tests above; it makes no stated quantitative prediction of its own, apart from claim 5.

d. If true: special relativity's postulate of no preferred frame would have to be replaced by a Lorentz-ether theory that mimics every SR result (as Lorentz and Poincare once did), and quantum electrodynamics's vacuum would need to be rebuilt without Lorentz-invariant virtual pairs.

Verdict: Historically superseded but instructive (the 1900-era aether); the specific pair-lattice derivation is Internally inconsistent (circular, E = mc^2 restated). The 2023 "Planck-granule medium" form: Not testable as stated, since no measurable difference from existing physics is offered.

Claim 2: charge is wave amplitude in metres; mass is standing-wave energy; Coulomb, electron mass and gravity follow from wave geometry

a. Mathematics. From Step 1 and my own algebra: replacing the coulomb by a length makes eps0 and mu0 carry odd units (mu0 becomes a linear density, k a force). Within those redefinitions, "deriving" Coulomb's law gives F = m c^2 r_e / r^2, which is k e^2 / r^2 because r_e is defined by m c^2 = k e^2 / r_e; checked: 8.187e-14 J / 2.818e-15 m = 29.05 N [read: CNPS 2018, my recheck]. "Deriving" m_e = mu0 e^2 / (4 pi r_e) is that same definition solved for m. Dimensional analysis in SI treats charge as a base quantity by convention; one can set it to a derived unit (as in Gaussian units, where charge has dimensions of sqrt(force) x length), so "charge is not a base dimension" is not itself wrong. What is wrong is the claim of a derivation. Equation-level checks of the Mass and Charge paper are limited by the text-layer gap, so this is checked through Step 1's quotations only.

b. Experiments. Coulomb's law to 1/r^2 holds to within about 1e-16 in the exponent (Williams-Faller-Hill 1971 and later) [memory, medium; not web-checked]. Quantised charge, electron-positron pair creation and charge conservation are tested to extreme precision [memory, high]. A model that only re-expresses these can match them, but equally it cannot be distinguished from them. Metres for charge also loses the sign and quantisation of charge unless added by hand, and the paper offers no wave mechanism for either.

c. Tested? Not tested by anyone; there is no numerical prediction separating it from standard electrodynamics. The authors' own check is the tautology above.

d. If true: charge would be a derived geometric amplitude; quantum electrodynamics's coupling constant would be a geometric property of a lattice; gravity and electromagnetism would be unified at the classical level.

Verdict: split. Weak form (charge may be expressed in derived units; field energy lives in space): Established as a convention, Plausible but empty of new content. Strong form (Coulomb's law, electron mass and gravity derived from wave geometry): Internally inconsistent in the sense of circular (the "outputs" are inputs); and as physics, Not testable as stated, because no divergence from standard values is predicted.

Claim 3: the "derived" constants (c, k, alpha, a0, G, g-2) show the model reproduces measured values

a. Mathematics. Each case reduces to a Planck-unit or definitional identity. G: l_P c^2 / m_P = sqrt(hbar G / c^3) c^2 / sqrt(hbar c / G) = G, checked numerically 6.6743e-11 [read: p.28 as read by Step 1; my recheck]. Planck charge q_P = sqrt(4 pi eps0 hbar c), and e = sqrt(alpha) q_P, so every electric "derivation" from them recovers alpha by definition. The inputs include r_e, which already contains alpha, m_e and e, so electron quantities are not independent outputs. Five input constants, of which three are already related by hbar, G and c, give at most two independent numbers' worth of freedom, so "a dozen outputs" is expected to be rearrangements. Anomalous magnetic moment (2023 paper, eq. 8): e^2/Q_p^2 with Q_p built from h gives alpha/(2 pi) = 0.00116141 [read; recomputed 0.0011614097]. That is the first-order Schwinger term, not a model output, and the measured a_e = 0.00115965218 [memory, high; check below] differs by 1.76e-6, i.e. 0.15 percent, which is huge relative to the measurement's precision. Matching a leading-order term to 0.15 percent is not agreement with data that are known to 1e-13.

b. Experiments. The electron g-2 measured to roughly 1 part in 1e12 and agreeing with QED including terms through fifth order [memory, high; I could not run the web check]. The paper's value is outside experiment by many thousands of sigma, so as a claim about the measured moment it fails. Its further claim that alpha "is not required" contradicts the use of alpha throughout atomic spectroscopy.

c. Tested? The "tests" are comparisons with CODATA values that are inputs or identities; no independent party has tested a prediction because none is made.

d. If true: constants like alpha would be geometry, not measured; QED's higher-order corrections would need a different origin.

Verdict: Refuted by calculation for the "derivation" and g-2 sub-claims (the G chain is an identity, not a derivation; g-2 differs from the experimental value by 1.76e-6, see above). The wider "model reproduces constants" claim is Internally inconsistent in the sense of circular.

Claim 4: h is the energy of one cycle; photon energy is a one-second accumulation; MUA removes quantum uncertainty

a. Mathematics. The MUA example reproduces F = P/c because A F_s = (P / hf)(hf / c) = P/c algebraically [read: Step 1 and recheck], so it agrees with standard physics by construction and discriminates nothing. Treating "cycle" as a base unit is allowed (radians and cycles are dimensionless angle-like units; units of angular frequency vs frequency are a known source of 2 pi confusion), but a per-cycle energy that is the same at every frequency is a different physical statement from E = hf, not a re-labelling: E_cycle = hf/f = h would be frequency-independent and the 1 s "measure time" is arbitrary. The "perfect reflector" example uses P/c where the correct reflector force is 2P/c [memory, high], an internal inconsistency in the paper's own setup. The claim that uncertainty vanishes because the smallest energy step is 6.626e-34 J is a non sequitur: the uncertainty principle is about non-commuting observables, not about a minimum detectable step.

b. Experiments. The photoelectric effect has a frequency threshold independent of intensity (Einstein 1905, Millikan 1916) [memory, high]. A constant energy per cycle makes the energy of a short high-frequency pulse enormously different from what is observed: a single gamma-ray photon of 1 MeV has about 1e20 cycles but is detected as one quantum with energy 1 MeV, and single-photon counting, Compton scattering and photon antibunching [memory, high] show energy is delivered in quanta E = hf per photon. Bell-test violations (loophole-free ones in 2015) exclude local classical-wave explanations of the correlations [memory, high].

c. Tested? Yes, effectively: E = hf per quantum has been tested for over a century. Gardi's reinterpretation has not been experimentally tested by her or others; she proposes none.

d. If true: the photoelectric effect, Compton scattering, blackbody spectrum and Bell correlations would need classical wave explanations.

Verdict: split. As a unit convention: Not testable as stated (it changes no measurement). As physics (energy per cycle constant, photon not fundamental, no uncertainty): Tested, not supported (by photoelectric, Compton, photon counting and Bell experiments, which were not run against her proposal but contradict its consequences) and Internally inconsistent where the reflector force and uncertainty arguments are concerned.

Claim 5: Michelson-Morley does not exclude an aether; GW150914 plus the Fermi transient are an aether P-wave and S-wave; every LIGO event should show the gamma follow-up

a. Mathematics. The seismic analogy is by visual resemblance of traces. P-waves in a solid are faster than S-waves and arrive first; here the supposed P-wave (the LIGO signal) arrives first and the 0.4 s later signal is a different instrument's weak transient. No speeds, distances or elastic constants are given to compute the 0.4 s delay from a medium model [read]. The claim cannot be checked numerically because there is no calculation.

b. Experiments. GW150914: Fermi GBM reported a 2.9-sigma transient 0.4 s later [web https://arxiv.org/abs/1602.03920]; INTEGRAL saw nothing, E_gamma/E_GW < 1e-6 [web https://arxiv.org/abs/1602.04180]. Later events: 2017's GW170817 was followed by GRB 170817A about 1.7 s after the merger, the neutron-star merger interpretation; this set a limit on the difference between gravitational-wave and light speed of order 1e-15 [memory, medium-high; not web-checked]. Most binary black-hole mergers have no gamma-ray counterpart [memory, medium]. So "all future LIGO events should show gamma follow-up" is contradicted by the many events without it, and the one robust case (GW170817) has the opposite order of significance (light and gravity at the same speed). I have no web-checked list for the later detections, so say this as [memory].

c. Tested? The prediction is clear and was tested by the LIGO-Virgo-Fermi follow-up programme run by others; it fails for most events [memory, medium]. Note this is a test of the specific prediction, not of the whole aether idea.

d. If true: gravitational waves would be longitudinal medium waves, not tensor spacetime waves; the polarisation measurements (LIGO-Virgo-KAGRA tensor-mode preference) would be wrong.

Verdict: split. Michelson-Morley "does not exclude an aether": Tested, not supported (the reasoning is wrong, see claim 1). The GW150914 P/S interpretation: Never properly tested as an interpretation (no alternative analysis exists), but Tested, not supported as a prediction (the follow-up prediction fails for most events, [memory, medium]). It was not dismissed without testing: the Fermi/INTEGRAL follow-up data exist and are public.

Claim 6: particles are standing-wave clusters of Planck wave centres; neutrino the fundamental unit; Mandelbrot set models black holes

a. Mathematics. Particles: the magic-number coincidence (K = 8, 20, 28, 50 for the four particles) is numerology unless the counts follow from a calculation. Step 1 notes no decay rates or lifetimes are calculated. Planck-mass oscillator energy matching a neutrino energy range upper estimate is a coincidence of a one-parameter choice. A neutrino as "fundamental unit" of an electron: neutrino mass < about 1 eV against m_e = 0.511 MeV [memory, medium (KATRIN limit around 0.8 eV is what I recall; not web-checked)], so the mass hierarchy is a factor 5e5 and the paper gives no mechanism. Mandelbrot: the mapping of z -> z^2 + c features to a Schwarzschild black hole's photon sphere and entropy is analogy; the author calls it "circumstantial and highly speculative" [read, Step 1].

b. Experiments. Standard Model: deep inelastic scattering shows the proton contains quarks and gluons [memory, high]; lepton numbers are conserved separately; muon and tau decays have measured branching ratios the wave-centre model does not reproduce. Black-hole shadow: the EHT shows an asymmetric ring of about 690 AU with brightness asymmetry from relativistic beaming [web https://en.wikipedia.org/wiki/Messier_87]; a pear shape is not what is reported, and the M87 offset claim appears to have been explained away by jet variability (2018 study, per the same page). Pear-shaped Ra-224 nuclei are real [memory, high, Gaffney et al. 2013], but they are nuclear octupole deformations, unrelated to black holes.

c. Tested? The shadow shape could be tested with EHT images; the author's overlay is a hand-aligned image, not a test. No independent test of the particle model exists because no numeric prediction is given.

d. If true: the Standard Model's quark structure, lepton families and gauge bosons would be emergent from a lattice of identical Planck-scale units; and atoms would behave as black holes, which conflicts with their sizes far above Schwarzschild radii.

Verdict: Particle-cluster model: Not testable as stated (no computed lifetimes, masses or branching ratios), with the neutrino-fundamental part Refuted by calculation only in the weak sense that a sub-eV particle cannot trivially make up a 0.511 MeV electron without a mechanism the paper does not supply. Mandelbrot black-hole model: Not testable as stated as an analogy; the specific pear-shaped-ring prediction is Never properly tested in the strict sense, but the EHT ring images show the asymmetry attributed to beaming, so it is not supported by the data we have.

Claim 7: charge is obsolete, current is an electrigen diffusion wave (Dinu and Gardi 2020); dielectric and magnetic fields are exact duals (Steinmetz 1914); free space is an LC circuit with impedance Z0/(2 pi)

a. Mathematics. Steinmetz's duality of phi = L i and Psi = C e, energies L i^2/2 and C e^2/2, is standard circuit theory [read, Step 1]. The Maxwell equations are indeed nearly dual if magnetic monopoles are allowed; that duality is mainstream (the lack of monopoles breaks it). "Z0 / (2 pi) = 59.958 ohm is the correct impedance" is a convention: 376.73/(2 pi) = 59.958 [recomputed], and moving a 2 pi between quantities changes no observable. Electrigen: no equations in the text read; Step 1 notes no quantitative predictions.

b. Experiments. Electron charge, mass and charge-to-mass ratio, thermionic emission, the Millikan oil drop experiment, and the charge carriers in metals (Tolman-Stewart 1916: electron inertia in accelerated wires gives the right charge-to-mass sign and magnitude) [memory, high for Millikan, medium for Tolman-Stewart] contradict "electrigen is what flows". There is no neutral Group 0 element lighter than hydrogen in the periodic table [memory, high]. A liquid carrying only longitudinal waves cannot carry the transverse polarisation of light [memory, high].

c. Tested? Steinmetz's analogy is textbook. The electrigen is untested and has no stated experiment; the periodic table and electron-based conduction have been tested extensively.

d. If true: the electron as the carrier of current would be wrong, and Maxwell's equations would need a mechanical fluid origin; chemistry would have a missing element.

Verdict: split. Steinmetz field duality and energy in the field: Established. Z0/(2 pi) as "the" impedance: Not testable as stated (convention). Electrigen and "charge does not exist": Tested, not supported (electron-based conduction and quantised charge are tested extensively; no controlled test of electrigen exists because none is proposed, so strictly Never properly tested for the new element, but accepted physics already contradicts its consequences).

Summary table

What is worth keeping (fair to the author)

• The complex-number-as-2x2-matrix paper is standard, correct mathematics presented as pedagogy [read, Step 1].
• Steinmetz's field duality and the "energy lives in the field" point are mainstream.
• Asking whether the "cycle" is a hidden unit in E = hf is a legitimate teaching question (the cycle/radian/2 pi confusion is real), but the answer is a convention, not a new physics.
• Where a [memory] claim was used above and not web-checked: Fermi LIV limits, KATRIN bound, Williams-Faller-Hill, GW170817 numbers, g-2 value and precision. The verdicts for claims 1, 2, 4 and 7 do not depend on these; claims 3 (the g-2 number) and 5 (later-event follow-up) do partly depend on them, and are marked [memory].