Maxwell: Science Contrarian review

Reviewer role: Science Contrarian (PhD-standard; judges by experiment, not consensus; Mallove is a standards model, not a persona). Inputs: facts.json, claims.md (Step 1), plus my own reading of the source PDFs via text layer (pdftotext). I did NOT view equation page images this step; equation content below rests on the OCR text layer, which is garbled for formulas, plus the prose around them. Where I rely on prose only I say so. Source tags: [read] document itself; [web] URL; [memory] model recall with confidence.

Honest framing first

This project is almost entirely mainstream: Maxwell's field theory is the foundation of classical electromagnetism. A contrarian has little to "rescue" in the physics. The real contrarian interest lies in (i) the 1865 evidence for light = EM wave, which was weaker than textbooks suggest, (ii) the ether, which the equations outlived, and (iii) the Heaviside story (an outsider's work blocked, then vindicated), which needs careful separation of genuine obstruction from legitimate criticism.

Corrections to Step 1

1. claims.md #7 and facts.json lump three different events as "blocked or dismissed by officials and journals". Hunt (Physics Today 2012, [read] h. text, sections "Inductive loading" and "Gaining recognition") says: - Preece (Post Office chief engineer) declared the Heaviside brothers' joint 1887 loading paper "worthless" and blocked it. He did this by controlling Arthur Heaviside, a Post Office employee; Oliver was not refused by a journal here. [read] - Electrician: editor Biggs declined to print Oliver's caustic letters about Preece (fearing libel), was then "abruptly removed" (Hunt says Biggs "later hinted" this was due to his support for Heaviside, which is a hint, not proof), and the new editor cancelled the series saying "no one read them". The cause is therefore unproven. [read] - Royal Society 1893: third operator installment was refereed and rejected because pure mathematicians objected to his handling of divergent series. That is a rigour objection, not suppression of inductive loading. Hunt notes the first two installments were published. [read] Step 1's "the Royal Society rejected a third operator paper" is right but should not be grouped with the Preece obstruction as one pattern.
2. facts.json (Maxwell 1865) says a "time-varying electric and magnetic disturbance" propagates. In the paper itself, the plane-wave derivation (par. 95-96) is explicitly: "This wave consists entirely of magnetic disturbances, the direction of magnetization being in the plane of the wave." [read] The electric part enters via displacement current and is not described as a co-equal E-wave in this passage. This matters historically: the modern symmetric E-B wave picture is Heaviside-era.
3. Step 1 says the claim of "twenty equations in twenty variables" is stated by Maxwell; I verified it and counted: 18 components (F,G,H; alpha,beta,gamma; P,Q,R; p,q,r; f,g,h; p',q',r') plus e and the potential = 20 variables; equations 3 each of (A)-(F) plus (G) and (H) = 20. Confirmed [read]. Note: these 20 include constitutive relations (Ohm, elasticity) and equation (D) with motional EMF, so they are NOT a four-equation set in disguise. Heaviside's four are the curl/div subset.
4. claims.md #4 says Maxwell "found" the match. Accurate, but Maxwell's own wording is "agrees sufficiently well" and in the introduction "strong reason to conclude" [read]. It is an argued identification, not proof. See claim 4 below.
5. Step 1 flagged that Poynting preceded Heaviside; confirmed (Hunt: Poynting's abstract Jan 1884) [read].
6. Not previously noted: Belrose (2014) is largely a paraphrase of Hunt plus one antenna case study; it adds no independent evidence on the credit question. Its NEC-4 "support" for Heaviside's field view is an interpretive illustration, not a test (see claim 6).

---

Claim 1: Faraday's lines of force can be put in exact mathematical form via an incompressible-fluid analogy (analogy, not hypothesis)

Read: the paper's opening says one must avoid both bare formulae and physical hypotheses, and adopts "physical analogies" so as to "obtain physical ideas without adopting a physical theory" [read: MaxwellOnFaradaysLines.pdf, opening pages].

a. Math: Internally consistent as far as I can tell from prose; the analogy maps flow velocity to force/flux and pressure to potential. Equations not checked on page images. b. Experiment: The formal identity of electrostatics, steady currents and magnetostatics (Laplace/Poisson equations) is established. Nothing here conflicts with experiment. c. Tested: Not a testable physical claim; it is a representational method. Its output (field equations) was later tested extensively. d. If true, what changes: nothing; it is already absorbed.

Contrarian note: Maxwell's explicit refusal to commit to a hypothesis is a model of good epistemic practice, and contrasts with his 1865 ether commitments.

Verdict: Established (as a mathematical method). The "fluid" itself is Not testable as stated, and Maxwell said so himself.

Claim 2: Induction is governed by a point-wise vector (electro-tonic state, now vector potential A) whose rate of change gives EMF

Read: "We may conceive of the electro-tonic state at any point of space as a quantity determinate in magnitude and direction... This representation involves no physical theory, it is only a kind of artificial notation" [read: Summary of the Theory of the Electro-tonic State]. I did not verify the EMF = -dA/dt relation in this paper from equations; the identification with the modern A and the relation E = -dA/dt - grad(phi) is [memory: high], corroborated by the relation quoted in Hunt's sidebar (Hunt, Physics Today 2012 [read]).

a. Math: consistent; the line integral of the electro-tonic intensity around a closed curve equals flux (Stokes), which the paper states as a proposition [read]. That is B = curl A, modern. b. Experiment: Faraday induction, transformers, generators all agree. c. Tested: Faraday's law is the best-tested relation in EM. The physical reality of A (as opposed to a gauge-dependent bookkeeping device) was a real dispute; the Aharonov-Bohm effect (1959, observed by Chambers 1960 and later Tonomura 1986) shows A has measurable consequences. [memory: high, not web-checked; the verdict does not depend on it]. d. Nothing needs to change.

Contrarian note: the "A is physical" view that was dismissed by Heaviside/Hertz (who eliminated potentials) was later partly vindicated. That is a genuine case of a "rejected" idea being right in a precise sense. Many alternative-science works (potentials-based theories) lean on this; the correct reading is that A is physically meaningful up to gauge, not that new energy sources follow.

Verdict: Established.

Claim 3: Field energy resides in space around conductors; with displacement current the field obeys a closed set of equations (20 in 20 variables)

Read: Maxwell lists A-H, "twenty of these equations in all, involving twenty variable quantities" and derives the intrinsic energy as a volume integral over all space of (1/2)(alphamualpha + ...) [read: dynamicaltheoryo00maxw.pdf, intro and pars. 70-71]. Equation (A) is explicitly "total current, compounded of [displacement] and [conduction]" [read].

a. Math: Counting consistent (verified above). Equation (H), continuity, follows from the others (a modern reader sees redundancy); not an inconsistency. Gauge choice is implicit (Maxwell works with A and a scalar potential directly). OCR garbling prevents me from checking each equation's signs; I did not. b. Experiment: Displacement current is required by charging capacitors and by wave propagation. Energy density in fields is confirmed by radiation pressure, antenna power and light's energy. Hertz 1887-89 measured waves in air with speed equal to light [web: https://en.wikipedia.org/wiki/Heinrich_Hertz] (the page states he established these were EM waves obeying Maxwell's equations; a secondary source). c. Tested: Yes, by Hertz (1887-89) and thousands of later experiments; replicated. Note that in 1865 there was NO direct test of displacement current. Its only evidence was the wave-speed argument (claim 4). d. Nothing needs to change.

Contrarian note: the claim that energy is "in the field" rather than "in the wire" is not an extra assumption relative to circuit theory but is distinguishable (see claim 6).

Verdict: Established.

Claim 4: EM disturbances travel as transverse waves at v (ratio of EM to ES units), measured 310,740,000 m/s, matching light, so light is EM

Read: pars. 95-97. Maxwell: v = 310,740,000 m/s (Weber-Kohlrausch); Fizeau V = 314,858,000; Foucault V = 298,000,000; aberration-derived V = 308,000,000. "the velocity of light deduced from experiment agrees sufficiently well with the value of v". He also notes: "The only use made of light in the experiment was to see the instruments... No use whatever was made of electricity or magnetism" in Foucault's [read]. He also derives the n^2 = (dielectric capacity)(magnetic capacity) relation and says it holds if elasticity is the same in dense bodies [read, intro].

The strongest honest case: two completely independent measurements (a capacitor discharge through a galvanometer; a rotating mirror) agree on one number to within a few percent. Chance coincidence is improbable. Further, polarization (transversality) is a second independent match.

Quantitative check against the modern defined value c = 299,792,458 m/s [memory: high; defined value since 1983, consistent with [web] https://en.wikipedia.org/wiki/Speed_of_light]: - v (310.74) vs c: +3.65 %. - Fizeau (314.858): +5.0 %. - Aberration (308.0): +2.7 %. - Foucault (298.0): -0.6 %. So the stated "agreement" is within a spread of about 5 %, and the electrical number was off by 3.6 %. The agreement was a good fit but, in 1865, not conclusive; Maxwell's phrase is appropriately hedged. Also v from Weber-Kohlrausch is derived from Weber's c = sqrt(2) times light speed; 439,450,000 m/s / sqrt(2) = 310,740,000 m/s, which I computed; the sqrt(2) relation is supported by [web: https://en.wikipedia.org/wiki/Speed_of_light] (statement that Weber-Kohlrausch's constant equals sqrt(2) times light speed). [memory: medium] for the exact 439,450 figure.

a. Math: The wave derivation (par. 95, equations 70-71, V^2 = ...) follows from the equations; Maxwell's result that only transverse disturbances propagate is derived in the paper and he honestly says electromagnetic science and optics are "at a loss" on normal vibrations [read, par. 100]. I did not verify the algebra from page images. b. Experiment: Consistent with the modern value (see above); the discrepancies reflect the 1850s measurement errors, not a flaw in the theory. c. Tested: Hertz (1887-89) produced, detected and measured the speed of radio waves, finding it equal to light [web: https://en.wikipedia.org/wiki/Heinrich_Hertz]. Replicated by Lecher, Lodge and others [memory: medium, not web-checked; the verdict does not depend on it]. d. If true, nothing new needs changing; this is the mainstream.

Verdict: Established. The 1865 evidence alone: "Plausible, strongly suggestive" (a 5 % agreement of two numbers); confirmation came with Hertz.

Claim 5: All of this happens in an all-pervading ethereal medium (later abandoned, equations kept)

Read: introduction and throughout: Maxwell speaks of "the elasticity of the medium", "affections of the same substance", and the medium "when not encumbered with any matter denser than air" [read]. Hunt and Belrose do not discuss the null experiments.

a. Math: The equations do not need the ether; they are internally consistent without it. Maxwell's mechanical model is not in the 1865 equations (he said he deliberately kept to the dynamical standpoint, though he does use "electric elasticity"). b. Experiment: Michelson-Morley (1887) found no ether-wind at the sensitivity then available [memory: high; the Speed of light page fetched did not detail this, so not retagged web]. Special relativity (1905) dispensed with it. Not all of it is "dead": the vacuum has measurable properties (permittivity, permeability, Casimir effect, polarization) and quantum field theory has a vacuum; but a mechanical, rest-frame-defining aether is excluded. c. Tested: Michelson-Morley and later versions, Kennedy-Thorndike, Ives-Stilwell [memory: high]. The contrarian caution: the null result excludes the simple stationary-ether model; it does not exclude all ether-like constructions (Lorentz ether theory is empirically equivalent to special relativity by construction). Alternative-science claims of "ether drift" positives (Miller's 1920s data) exist; I have no source in this folder and make no claim about them. d. Not applicable: the claim has already been dropped by mainstream.

Verdict: Historically superseded but instructive.

Claim 6: Heaviside recast Maxwell's equations to four vector equations in E and H, derived S = E x H (just after Poynting), and held that the real action lies in the field around conductors

Read: Hunt: Heaviside "rewrote 'Maxwell's equations' the 20 fundamental equations" (subtitle/lead), dropped quaternions ("antiphysical and unnatural"), used grad, div, curl; Poynting published the energy-flow theorem first (Jan 1884 abstract); Heaviside learned this late in 1885 [read, h2 text]. Belrose quotes Heaviside: "the real action lay not within conductors but in the field around them" [read, Belrose p.2-3] and argues NEC-4 modelling of KVOK tower antenna supports it.

a. Math: Heaviside's four equations are a correct reformulation of the field part of Maxwell's. Heaviside's version eliminates potentials; the constitutive and force laws are separate. Poynting's theorem follows from them. Consistent. b. Experiment: The Poynting vector's integral over a closed surface gives radiated power; confirmed by every antenna measurement. Note, though, a caution: S is only defined up to a curl term; only the integral over closed surfaces is operationally fixed. "The real action is in the field" is an interpretation. c. Tested: The four-equation set is physics; the energy-flow claim is tested in antennas and in Poynting's analysis of circuits. Belrose's NEC-4 result is a calculation compared with Smith's measured data per Belrose [read, Belrose]; I did not see the underlying data. It supports that NEC reproduces a measured pattern, not that the "field-not-wire" interpretation is uniquely true, since a circuit-current description also fits. d. Nothing changes.

Contrarian note: the credit question ("Heaviside, not Maxwell, wrote Maxwell's equations") is accurate historically, and Gibbs and Hertz also contributed to the vector form (Hunt: Gibbs similar path, published after Heaviside [read]; Hertz's contribution is [memory: medium]).

Verdict: Established (equations, S); the "real action is in the field" part is an interpretation: Not testable as stated, and I flag Belrose's framing as over-reading a modelling exercise.

Claim 7: Heaviside, a self-taught outsider, was blocked by officials and journals and later vindicated

Read: see Corrections 1. Additional specifics from Hunt [read]: Heaviside was elected FRS in 1891, received a government pension (1896), and inductive loading was later used "with great success" by AT&T, after Pupin patented it in 1899 and sold the rights for $500,000; Heaviside never patented.

The strongest honest case: documented, sourced, specific obstruction. Preece, the head of Post Office telegraph engineers, publicly campaigned against inductance in 1886-87 and blocked the paper. The vindication is also documented: Pupin's patent was sold to AT&T for $500,000 (Hunt's phrase "under somewhat shady circumstances" is the historian's judgment; Wiener's denunciation is the polemic of an interested party).

Rigour against the researcher: Heaviside's operator methods did rely on non-rigorous manipulation; the rejection at the Royal Society was on stated mathematical grounds, and rigorous foundations arrived later (Belrose cites Bromwich, Wagner; [read] in the article; I did not verify the dates). Heaviside's own words ("Shall I refuse my dinner because I do not fully understand the process of digestion?") show he knew the objection. He also was abrasive ("scienticulist"). Heaviside also could not have been "vindicated" on Pupin's priority by any court.

a. Math: n/a for the story; the loading result (distortionless line condition) is a correct transmission-line result [memory: high]. b. Experiment: Loaded lines worked in practice (AT&T) [read, Hunt]; I have not located a primary source. c. Tested: Operationally yes, by industry deployment. d. If true, what changes: nothing in physics; it concerns how institutions treat outsiders. It is the best-documented case in this folder of an outsider being obstructed by an institutional figure and later proven right on the merits, but it is also a case where one of the three obstructions (Royal Society) had a defensible basis.

Verdict: Established as history for the Preece and loading episode (sources: Hunt). Not supported as a general claim that "journals blocked him": for the operator calculus the objection was a rigour dispute, later resolved in his favour on results. Cause of the Electrician cancellation: Never established.

---

Summary table

Web checks made (budget 6): Hertz wave measurement [web: https://en.wikipedia.org/wiki/Heinrich_Hertz]; speed of light and Weber-Kohlrausch sqrt(2) relation [web: https://en.wikipedia.org/wiki/Speed_of_light]. The Michelson-Morley date and result remain [memory: high] because the fetched page did not cover them. The Aharonov-Bohm and Miller points remain [memory] and no verdict depends on them.