Primer: Bryan G. Wallace, "The Farce of Physics" (radar, c+v light, dynamic ether)

This is a study guide for the Wallace-Byran folder (the folder name misspells "Bryan"). It rests on the arbiter's final verdicts (arbiter.md, including its "QE revision" section) and on the physicist's, mathematician's and contrarian's reviews, the deep reads of the whole book (deep/farce-d10910.md: Chapter 7 first, then Chapters 1-6 after the QE step), the QE report (qe.md) and the Investigator's notes (sites.md, research/). I redid the worked example in section 5 myself, from page images of Wallace's 1969 radar paper. example.py reproduces every number in it.

Source tags: [read] = taken from Wallace's 1993 book farce.txt (one long text file, so locations are given as line numbers, L); [read: page image p.N] = read from a rendered page of the 1969 paper (Spectroscopy Letters 2, 361-367), which I downloaded to a scratch folder from http://sciliterature.50webs.com/SpecLetters1969-p361-367.pdf; [web] = Internet source with URL; [memory] = recalled, with confidence; [calc] = computed in example.py.

Source gap (carried over from Steps 1-3). The folder holds one document: the 1993 manuscript The Farce of Physics, a plain DOS text file. It prints no equation of Wallace's "mass dynamics", no atomic model, no radar data table and no program listing. Those are in papers that are not in the folder: Foundations of Physics 3, 381 (1973); Spectroscopy Letters 3, 115 (1970), 4, 79 and 4, 123 (1971); J. Clas. Phys. 1(2), 17 (1982); Speculations Sci. Technol. 9, 9 (1986). Only one primary paper has been read by anyone in this review: the 1969 radar paper, as page images (by the Mathematician, the Contrarian and, for this primer, by me). So: - What this primer says about the radar argument rests on the 1969 paper and the book's Chapters 4-6 (deep-read after the QE step). - What it says about the unified theory (claims 3-5) rests on the book's prose and the deep read alone, plus standard physics. - Never read by anyone here: Wallace's 1971 radar paper; Shapiro, Ash and Tausner, PRL 17, 933 (1966); the 1972 rebuttal by Horedt and Ruck; Tolchelnikova-Murri's own papers; Standish's 1974 report; the Texas Symposium discussion record. Also unread, and named by the QE step as the next records to obtain: Shapiro's published reply to Wallace, Scientific Ethics 1(2), 10 (1985); Evans et al., AJ 70, 486 (1965); Ash, Shapiro and Smith, AJ 72, 338 (1967) itself; Moyer, Celes. Mech. 23, 33 (1981); McVittie, AJ 75, 287 (1970); Kotel'nikov et al., Astron. Zh. 53, 1270 (1976); Schmeidler, Sky & Telescope 27, 217 (1964). - The book's Step 1a digest covered only the opening of each chapter (the 300 KB file arrived as one "page"). The QE step found this and ordered deep reads of Chapters 1-6, which fill the gap.

The extracted text also lost every non-ASCII character. The deep reader decoded the original (code page 437): Jupiter "2½" times (not "2"), "1.2 and 1.3 µA", "90°". Lost superscripts make "104 electrons and positrons" ambiguous between 104 and 10⁴.

Resolved by the QE step. An earlier draft of this primer re-read p.364-366 of the 1969 paper and found that Wallace puts both radar equations at the moment the echo returns, and that by Wallace's own yardstick (Newcomb's curve N in Fig. 1) the constant-c curve E is the closer one. The QE step confirmed both readings against the page images, and the arbiter withdrew its "Internally inconsistent" verdict on claim 1(a) (arbiter.md, QE revision A). Sections 5 and 6 now use the revised verdict.

---

1. The idea in one paragraph

Bryan G. Wallace (1933-1997, St. Petersburg, Florida; he worked with physicists at Eckerd College) believed that light does not always travel at the same speed. He thought it behaves like a thrown ball: if the thing that emits or reflects it moves toward you at speed v, the light arrives at speed c + v. This is the old "emission" or "ballistic" theory of light, which Walther Ritz put forward in 1908. From 1961, radio astronomers bounced radar pulses off Venus and timed the echoes to fix the size of the solar system. Wallace argued that the distances they derived, which assumed a constant speed c, showed patterns tied to the motion of the radar station and of Venus. He said those patterns disappear if you assume c + v instead. Since Einstein's special relativity rests on light having the same speed for every observer, he concluded that the orthodox account of twentieth-century physics was "an elaborate farce" (hence his title). He did not reject all of it: "I certainly have no problem with E=mc2 since the atomic bomb is ample evidence that it is true. So in general, much of relativity theory is true, but many of the original arguments are not" [read: L4549-4552]. He accepted clock time dilation as real, but measured against an absolute frame; he rejected Einstein's two postulates and the equivalence principle [read: deep read Ch.6; arbiter QE revision D]. He then proposed his own replacement. In it, all matter is a fluid-like "dynamic ether" whose mass circulates at the speed of light inside each particle and unrolls into a straight-line photon when matter turns into light. He tested it with a computer model of electrons. Much of his book is about how journals, referees and funders refused his work.

2. What you need to know first

Kinematics of a round trip (radar ranging). A radar sends a pulse at time t_e, it reflects off a target at t_refl, and the echo returns at t_r. The measured quantity is the round-trip time t = t_r − t_e. If the target moves during the trip, "the distance" is ambiguous until you say which moment you mean: emission, reflection or reception. The whole radar debate in this project turns on that bookkeeping. Read: any first-year mechanics text on relative velocity. Kleppner and Kolenkow, An Introduction to Mechanics; Taylor, Classical Mechanics.

Galilean relativity and velocity addition. In Newton's mechanics, velocities add. A ball thrown at u from a cart moving at v goes at u + v relative to the ground. The laws of mechanics look the same in every frame moving at constant velocity (the Galilean principle of relativity). Ritz's emission theory applies this to light: light leaves its source at c relative to the source. Read: Kleppner and Kolenkow (relative motion); Resnick, Introduction to Special Relativity, ch.1 (the historical background, including emission theories).

Special relativity (SR) and its two postulates. Einstein (1905) assumed (1) the laws of physics are the same in all inertial frames, and (2) light travels at c in every inertial frame, whatever the motion of its source. Postulate 2 is what Wallace attacks. Its consequences include time dilation and the Lorentz transformations. You need only the idea that "c is the same for everyone" and how it differs from "c relative to the source". Read: Griffiths, Introduction to Electrodynamics, ch.12 "Electrodynamics and Relativity"; Taylor and Wheeler, Spacetime Physics.

The Doppler shift. A wave reflected from a target moving toward you at radial speed v comes back with its frequency raised by about Δf = 2vf/c (for v ≪ c). Radar astronomers measure Δf to get v; Wallace writes this as v = dc/2f, with d the Doppler shift [read: page image p.364]. Read: any intro physics text (Halliday, Resnick and Walker, chapter on sound/waves).

Planetary orbits and ephemerides. An ephemeris is a table of where a planet will be at each date, computed from orbital elements (size, shape, orientation and timing of the ellipse) and the masses of the planets that perturb it. Simon Newcomb's tables (1895) were the standard for Venus until radar. The astronomical unit (AU, about 1.496 × 10⁸ km) sets the distance scale; before 1961 it was known only to about a part in 10⁴ [memory, medium]. Read: Danby, Fundamentals of Celestial Mechanics; Murray and Dermott, Solar System Dynamics (later chapters are advanced).

Mass-energy and photon momentum. A particle has rest energy E = mc²; a photon has energy E = hf and momentum p = E/c. In electron-positron annihilation, e⁺ + e⁻ → 2γ, each photon carries 0.511 MeV and momentum m_e c, in opposite directions. Read: Krane, Modern Physics; Serway, Moses and Moyer, Modern Physics.

Spin angular momentum. The electron carries intrinsic angular momentum ħ/2 = 5.27 × 10⁻³⁵ J·s (ħ = h/2π is the reduced Planck constant). A classical mass m moving at speed u on a circle of radius r has angular momentum L = m u r. Checking a model's spin against ħ/2 is a quick test. Read: Krane, Modern Physics (atomic structure chapter); Griffiths, Introduction to Quantum Mechanics (chapter on angular momentum).

Ionospheric/plasma delay. Radio waves crossing ionised gas are slowed by an amount proportional to 1/f² (f = radio frequency). A delay that changes with frequency is the fingerprint of plasma; a change in the speed of light itself would not depend on frequency. Read: Chen, Introduction to Plasma Physics, chapter on waves in plasmas.

3. The textbook view

3.1 Light speed and the source

Ritz's emission theory (1908) says light from a source moving at v travels at c + v. Special relativity says it travels at c regardless. For about fifty years the main evidence was astronomical: de Sitter (1913) argued that light from the two stars of a binary system would arrive scrambled if its speed depended on the star's motion, and it does not. That argument has a known loophole called extinction. Light passing through gas is absorbed and re-emitted, so after enough gas it "forgets" its original speed. Fox (1965), whom Wallace cites, made this point, and it is why the old binary-star tests were not decisive [read: page image p.361; physicist; arbiter]. The modern tests avoid the loophole:

• Alvager et al. (CERN, 1964). Photons from the decay of neutral pions moving at v/c ≥ 0.99975 were timed over a flight path. The emission theory predicts nearly 2c. They measured 2.9977 ± 0.0004 × 10⁸ m/s, so in a model "speed = c + kv" the fraction k is below about 10⁻⁴ [web: https://arxiv.org/pdf/0912.3818, per physicist and arbiter].
• Brecher (1977). X-ray pulses from binary stars give k < 2 × 10⁻⁹, and X-rays are much less affected by extinction [web: https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.39.1051; arbiter corrected the citation to PRL 39, 1051].
• Cassini (2003). Radio links passing near the Sun matched general relativity's extra delay to 2 parts in 10⁵: γ = 1 + (2.1 ± 2.3) × 10⁻⁵ [web: https://www.nature.com/articles/nature01997]. This is the same kind of measurement as Wallace's, done forty years later.

3.2 How radar ranging is reduced (constant-c model)

Symbols: t = round-trip time (s); c = 2.998 × 10⁸ m/s; D = Earth-station-to-Venus distance; v = rate at which that distance shrinks (positive while Venus approaches, negative while it recedes).

If light always travels at c in the frame where you do the calculation, then: - The pulse reaches Venus at the half-way point in time. The distance at the moment of reflection is exactly D_refl = t c/2. - During the second half of the trip (t/2) Venus keeps moving, so the distance at the moment of reception is D_rec = t c/2 − v t/2.

Real reductions (Lincoln Laboratory, JPL) do this in a frame centred on the solar system's centre of mass. They evaluate the transmitter's position at emission and the reflector's at reflection, iterate, add the relativistic gravitational delay, and fit the orbital elements and the AU to all the data at once [physicist; memory, high]. Section 5.3 shows that, to first order in v/c, the speed of the solar system through space cancels out of a round trip, so only the relative velocity v enters. Tolchelnikova-Murri (Pulkovo Observatory) makes this cancellation the centre of her argument that the 1961-62 radar did not test relativity either way [web: https://ritz-btr.narod.ru/tolchel.html, Investigator's summary].

3.3 Mass, photons and spin

In standard physics a photon has no rest mass, but it carries energy E and momentum E/c. Its "mass" E/c² is a bookkeeping quantity. In e⁺e⁻ → 2γ, energy and momentum balance with nothing "created" [physicist; deep reader M2]. The electron behaves as a point particle in high-energy scattering and in magnetic-moment (g − 2) measurements: g − 2 bounds its radius below about 10⁻¹⁸ m, and electron-positron scattering at the LEP collider gives r < 2.8 × 10⁻¹⁹ m at 95% confidence [web: https://arxiv.org/pdf/hep-ph/0002172, as checked by the arbiter]. One mainstream picture comes close to Wallace's: Zitterbewegung ("trembling motion", Schrödinger 1930). In Dirac's electron theory a free electron's position oscillates at speed c on the scale of the reduced Compton wavelength ħ/(m_e c) = 3.86 × 10⁻¹³ m [memory, high; deep reader M3].

3.4 Cosmology

The mainstream picture is an expanding universe, about 13.8 billion years old, supported by: - the redshift-distance relation; - the cosmic microwave background (CMB) with its almost perfect blackbody spectrum; - the abundances of the light elements; - the stretching of supernova light curves by the factor (1 + z), where z is the redshift. "Tired light" (photons losing energy on the way, with no expansion) does not predict this stretching.

The supernova stretching was first clearly measured by Goldhaber et al. in 2001 (35 supernovae at redshifts 0.30-0.70; stretch factor b = 1.07 ± 0.06, where b = 1 is the expanding-universe prediction) and again by the Dark Energy Survey out to z ≈ 1 [web: https://www.osti.gov/servlets/purl/836363; https://arxiv.org/pdf/2406.05050, as checked by the arbiter]. That is after Wallace wrote his book (1993). When he wrote, a refereed paper he relied on, LaViolette (ApJ 301, 544, 1986), had found a tired-light model fitting four cosmological tests better than an expanding model without evolution; it did not compare against expanding models that include evolution [read: L1240-1305; deep read Ch.2]. Stars shine by fusion, which is seen directly in solar neutrinos; tired light also has trouble with the CMB's blackbody spectrum [memory, high; not web-checked].

4. Where Wallace departs

Wallace's strongest point of method is fair and worth remembering. A fit that assumes c can absorb some of a deviation from c into its adjustable parameters. So "the fit came out well" is not, by itself, a test of c against c + v. Shapiro wrote this himself in general terms, and Wallace quotes him [read: page image p.363]. Wallace also concedes the limits of his own case: "Although analysis to date presents strong evidence against c and for c+v, I don't think it can be considered reasonably conclusive until a full c+v investigation is made" [read: page image p.367].

5. Worked example: what Wallace's two radar equations actually compare

This example rebuilds the core of the 1969 paper with real 1961 numbers. Everything is in example.py.

5.1 The setup, in Wallace's words

From p.364 [read: page image p.364]: - "Equation (1) was used to calculate the distance from the radar station to the surface of Venus for the c theory: D_E = tc/2 − tv/2. Here t is the radar beam's transit time; v = dc/2f the relative radial velocity, positive during approach and negative during recession; d is the Doppler shift; f is the frequency; D_E = tc/2 during the instant of reflection which is t/2 in the c theory but not in the c+v theory." - "Equation (2) gives the distance for the c+v theory and is based on the fact that c + (c+2v) = 2(c+v). The second term of both equations (1) and (2) corrects the distance to the time the beam returned to the transmitter. If it is made positive it will correct the distance to the time the beam left the transmitter. D_G = t(c+v)/2 − tv/2 = tc/2."

So Wallace knew that the constant-c model puts tc/2 at the moment of reflection, and he refers both equations to the moment of reception.

5.2 Derive both equations yourself

Put the radar station at rest and let Venus start at distance D₀, approaching at v. The pulse leaves at c in both theories, since its source is at rest. The closing speed is c + v, so the pulse reaches Venus after t₁ = D₀/(c + v). The distance at reflection is D_refl = D₀ c/(c + v).

Now the return leg. There are three different rules, and emission theories have used more than one of them. Tolman (1912) discussed several, including the "new source" rule in which the mirror re-emits at c relative to itself [web: https://en.wikisource.org/wiki/Some_Emission_Theories_of_Light].

[calc: example.py §1, sympy; the first-order forms use t ≈ 2D₀/c.]

Two lessons: - Eq. (2) follows exactly from Wallace's stated rule. A Newtonian ball bouncing elastically off a heavy wall approaching at v does leave at its incoming speed plus 2v, so the rule is not arbitrary for a corpuscle theory. The arithmetic mean in "c + (c+2v) = 2(c+v)" is only a shortcut; doing the legs properly gives the same answer to within terms of order v²/c², about 10⁻⁹ here. The Mathematician's objection (c) to the averaging is therefore a second-order point. The Mathematician's other point stands: the c + v return rule gives tv/4, so the prediction depends on a reflection rule that the emission family has never agreed on. - Under the "original source" rule, a radar on Earth sees exactly what the c theory predicts. Radar ranging tests c + v only if the reflection rule is pinned down. This is one reason Tolchelnikova-Murri and Wallace can both look at radar and reach different conclusions.

The Mathematician and the arbiter first wrote that Eq. (2) gives the reflection-epoch distance while Eq. (1) gives the reception-epoch distance, an "epoch mismatch". With the c + 2v rule and Wallace's own sentence on p.364, both are reception-epoch distances. The book says the same ("for the time the beam returns to the transmitter") [read: L3810-3821]. The QE step confirmed this and the arbiter withdrew the mismatch (arbiter.md, QE revision A).

5.3 Does the solar system's own motion matter?

Under the c theory, let the station move at u_E and Venus at u_V through the frame in which light has speed c (both along the line of sight), so v = u_E − u_V. Writing each leg out: - Return leg: c t₂ = D_rec − u_V t₂. - Distance at reflection: D_refl = D_rec + v t₂. - Outward leg: c t₁ = D_refl + u_E t₁.

To first order this gives

t = (2 D_rec / c)(1 + v/c) [calc: example.py §2],

which contains only the relative speed v. Inverting it gives D_rec = tc/2 − tv/2, which is Eq. (1). The absolute speeds cancel to first order. So Wallace's use of the relative velocity in Eq. (1) is legitimate, and this is the cancellation Tolchelnikova-Murri stresses.

5.4 Real numbers: the 1961 Venus pass, and Fig. 1

Facts used. - Venus was at inferior conjunction (between the Sun and Earth, closest to us) on 10 April 1961, about 23:50 UT, which is Julian Date (JD) 2437400.49 [web: search result summary 2026-10-04; it agrees with the zero crossing of Wallace's Fig. 1]. - Lincoln Laboratory's Millstone radar observed Venus from 6 March to 18 May 1961 [web: https://ui.adsabs.harvard.edu/abs/1962AJ.....67..181P]. - For the model I use circular, coplanar orbits: Earth at 1 AU, Venus at 0.7233 AU, periods 365.256 d and 224.701 d. This is a crude model and should match the real figures to roughly 10-20%.

Step 1: the size of Wallace's effect. The difference between his two equations is D_G − D_E = tv/2. Near conjunction D = 0.277 AU, so t = 2D/c = 276 s. At JD 2437440 (30 days after conjunction), D = 0.453 AU, t = 452 s and v = −12.3 km/s (receding). Then:

tv/2 = 452 × (−12.3 km/s)/2 = −2,774 km [calc].

Compare this with the radar accuracy Wallace quotes, ±1.5 km [read: page image p.362]. The two theories differ by about a thousand times the measurement error. So round-trip radar, with Wallace's reflection rule, is a very sensitive test, not a weak one.

Step 2: turn it into Fig. 1's units. Wallace plots heliocentric radius vectors, meaning Venus's distance from the Sun, in units of 10⁻⁶ AU. Near inferior conjunction Venus lies between Earth and the Sun. Making the Earth-Venus distance larger moves Venus closer to the Sun, so the sign flips. To convert, project the shift onto Venus's radius direction:

[calc: example.py §3; the Fig. 1 column is my reading of the page image at 300 dpi, read: page image p.366]

The match is good, sign and shape included. G − E is negative before conjunction, zero at conjunction (where v = 0) and positive after. It peaks about a month later, then shrinks as the line of sight turns away from Venus's radius direction. The gap between Wallace's E and G curves is the term tv/2 and nothing else. That is what it should be: the figure displays the predicted difference between the two theories. Whether that difference is observed depends entirely on the yardstick the curves are compared against.

Step 3: the yardstick. Wallace compares both curves with N, Newcomb's perturbed radius vectors. He reasons that E "should fit the N curve within the maximum possible estimated error of the radar data", since Newcomb's masses agree with the Lincoln fit. "The radar data presents evidence against the c theory because the N − E differences are far larger than any possible error" [read: page image p.365]. Reading Fig. 1 [read: page image p.366, my reading]: - E lies about 0.5 to 2 units (75-300 km) below N throughout. - G departs from N by up to about 7 units (about 1,000 km) near JD 2437440.

So by the yardstick Wallace chose, the constant-c curve E is the closer of the two. Wallace's support for c + v comes from a different comparison: "The points on the G curve represent values from an ephemeris I made using Cowell's method of numerical integration of orbits and Newcomb's values for planetary masses. Note the close fit between Newton's Laws and his c+v corpuscular theory" [read: page image p.365]. Cowell's method is step-by-step numerical integration of the equations of motion. The dots lie on G only between about JD 2437395 and 2437420. The paper does not say how the starting position and velocity (the initial conditions) of that integration were chosen. If they were set from the G data themselves, a smooth orbit through about seven points of a smooth curve is expected and tests nothing.

This reading of the figure differs from the Mathematician's summary ("G lies near N, E does not"). The QE step and the arbiter re-read the figure and agree with the reading here (E about 1-2 units below N; G up to about 7-8 units away). One caveat cuts the other way: Newcomb's Venus tables were themselves uncertain by an amount the paper does not state, so N is a coarse yardstick [arbiter QE revision A].

Step 4: the AU. Wallace uses the Lincoln Laboratory AU of 149,597,850 km and argues that "since it was determined during inferior conjunction, the c value should be close to the c+v value because the relative solar orbital velocity would have been zero" [read: page image p.365]. The table supports him on this point: at conjunction tv/2 is only 35 km against 4,000 km a month and a half later. The arbiter's remark that using the c-fit AU makes his comparison "partly circular" is therefore weaker than it sounds, though not empty, since the AU fit used the whole pass.

5.5 Scale checks on the book's numbers

• "3 ms against 10 µs, 30,000%": 3 × 10⁻³ / 10 × 10⁻⁶ = 300 = 30,000%. A 3 ms round-trip time equals 450 km of one-way distance, or 3.0 × 10⁻⁶ AU [calc]. The two numbers come from different eras: 3 ms is a 1961-66 residual, while 10 µs is the precision the book attributes to "the best fit the Lincoln Lab could generate" [read: farce ch.4; contrarian; arbiter correction 3].
• "260 km" between two stations on opposite sides of the Earth [read: page image p.362]. The equatorial rotation speed is 0.465 km/s, so two opposite-limb stations differ in radial speed by up to 0.93 km/s. With t = 276 s, the difference in tv/2 between them is 128 km, while t × Δv = 257 km [calc]. Wallace's 260 km is t·Δv, twice what his own equations give. The Contrarian noticed the factor of 2; it does not affect the rest.
• Frequency dependence. Wallace says the 30-day term at 440 MHz is "about five times larger" than at 1295 MHz [read: page image p.363]. Plasma delay scales as 1/f², giving (1295/440)² = 8.7. A change in light speed would give a ratio of 1. The observed 5 points toward plasma, which Wallace partly accepts ("part of the 30-day variation is due to intervening plasma"). It is a pointer rather than a proof, since the two data sets come from different epochs and stations [mathematician; arbiter].

5.6 A direct test of source dependence

If light from a source moving at v travels at c + v, photons from pions moving at 0.99975c should travel at about 5.995 × 10⁸ m/s. Alvager et al. measured (2.9977 ± 0.0004) × 10⁸ m/s [web, §3.1]. This crude check gives |k| < about 2 × 10⁻⁴; the published analysis gives about 10⁻⁴ [calc: example.py §5]. This test does not involve reflection or ephemerides at all.

5.7 What the example shows

• Wallace's algebra is sound for his stated reflection rule, and his two curves are referred to the same moment.
• The two theories do differ by a first-order amount (hundreds to thousands of km) that radar of ±1.5 km could easily see. Wallace was right that radar is in principle a strong test.
• The paper does not make that test. It compares an unfitted pair of curves with two different yardsticks: Newcomb's tables, which favour E, and Wallace's own integration, which agrees with G. The paper does not report the uncertainty of either. A real test fits both models, with orbital elements, the AU and the masses free, to the same multi-station data. Nobody has published that (arbiter: "Never properly tested").
• Even a perfect radar result would depend on which reflection rule the emission theory adopts. Source-velocity experiments that need no reflection rule (Alvager, Brecher) find no c + v effect.

6. Evidence table

Verdicts are the arbiter's: its "Corrected claim list" as revised in its "QE revision" section (arbiter.md, H). Rows marked "kept on record" carry material that bears on a claim but gets no verdict of its own; it stays here whatever the verdict, with its source, so you can weigh it.

The record goes dark. Wallace's last research paper was in 1986 ("my ninth, and last research paper to date") [read: L5290 area, deep reader]. Later came letters, two 1990 journal pieces, Leningrad conferences (1989, 1991) and this book. He died on 19 March 1997 [web: NPA wiki, per Investigator]; no obituary has been found. No one has published a c + v reanalysis of the radar data. The arbiter's competing explanations, all kept here: 1. Specialists judged the work wrong on its merits (Shapiro's letter, the Horedt-Ruck rebuttal, the later source-independence tests, and Shapiro's 1985 published reply, which exists but is unread). 2. Informal hostility closed publication channels (Robinson's letters, the PRL reversal). 3. Lack of data access (Shapiro's and Hollingsworth's statements, as relayed; McVittie's 1970 complaint about Shapiro's secrecy is a second voice on this, without any military claim). 4. Funding ended after 1978, and age. 5. Classification of tracking data (Wallace's own "Star Wars" inference, formed after reading a 1983 Science article, 14 years after the events). Of the 15 items he offers, none quoted ties secrecy to the speed of light or planetary radar, and no other evidence was found; it is not excluded.

The records that would decide between them: Shapiro's 1985 reply in Scientific Ethics (the first to obtain), Lincoln Laboratory and AFCRL (Air Force Cambridge Research Laboratories) reports and card decks, the NSF file, the Eckerd College archives, the full Horedt-Ruck text and the Texas Symposium discussion.

7. What a learner should take away

Solid. - Wallace's radar algebra, as printed in 1969, is correct for his stated reflection rule. His two equations really do differ by tv/2 at the same epoch, and you can reproduce his Fig. 1 gap from orbital geometry (§5.4). - His methodological point is right: a fit that assumes c cannot, by itself, test c against c + v. Shapiro said as much in general terms. - A peer-reviewed 1967 paper did report unexplained station-dependent radar residuals, and a 1976 Soviet radar paper reported Venus residuals against planetary theory of up to 3500 µs (about 500 km). Neither is specific to c + v. - Light really is bent by the Sun, by the amount general relativity predicts, whatever one thinks of the 1919 and 1922 plate reductions (claim 7). - His crossed-beam scattering experiment was real laboratory work, and it agreed with standard theory. - Jupiter does give off more heat than it receives.

Wrong, and why the mistakes are instructive. - The yardstick problem. Two curves that differ by a predicted amount prove nothing until you compare them with an independent truth, using the same freedom for each model. By Newcomb's tables, the yardstick Wallace chose, the c curve fits better. His c + v support came from his own integration with unstated starting conditions. Always ask: "fit to what, with how many free parameters?" - The reflection-rule problem. An emission theory must say how a moving mirror reflects light. The three choices give tc/2 − tv/2, tc/2 − tv/4 and tc/2. A theory that can choose its rule after seeing the data is hard to test, and the CERN pion experiment sidesteps the question entirely. - Epoch bookkeeping cuts both ways. Wallace read JPL's practice of evaluating positions at signal reception (Moyer 1981) as a hidden c + v law. A constant-c light-time calculation must evaluate each endpoint at its own event time anyway, so the practice alone does not show c + v (the reviewers' reply; Moyer unread). The reviewers, in turn, first misread which epoch Wallace's own equations used. Write down which moment each distance refers to before arguing about it. - Spin as a quick check. Mass circulating at c on Wallace's 5.7 × 10⁻¹⁵ m radius gives 1/34 of the electron's spin. One line of arithmetic tests a model against a known number. - Mass budgets. A pion of 104 or 10⁴ electrons and positrons cannot have the pion's mass without an enormous, unexplained binding energy. - Energy is not optional. A picture with conservation of mass and momentum but no energy law runs into ½mc² versus mc².

Genuinely open. - No neutral joint refit of the early multi-station Venus data with a free ballistic parameter has been published. - What did Shapiro say at the Texas Symposium, what does PRL 17, 933 actually test, and what did his 1985 reply in Scientific Ethics argue? - Did the 12:00 UT reduction (Evans et al. 1965) and the fitted parameters of Ash, Shapiro and Smith (1967) remove a real signal, as Wallace charges, or only reduce data to a standard epoch? Both papers can be read. - What do Schmeidler's 1964 plot of the 1922 eclipse, and the modern re-measurements of the 1919 plates, show about how those early results were reduced? - What did Tolchelnikova-Murri's 2001 paper and Clemence's IAU remark establish about the 1961-62 reductions? - What caused the USSR-versus-US station discrepancy and Standish's 1974 systematics? - Is the gravity in Wallace's program derived, or a fitted constant? - Have Arp-type discordant redshifts had a neutral test?

8. Check yourself

1. A radar pulse makes a round trip of t = 300 s to a target approaching at v = 5 km/s. Under the constant-c theory, what is the distance at the moment of reflection, and at the moment of reception?
2. Using Wallace's c + 2v reflection rule, what is D_G − D_E for that same pulse? Using the "mirror as new source" rule (return at c + v)?
3. Why does the gap between Wallace's G and E curves vanish at inferior conjunction, and why does it change sign there in Fig. 1?
4. The 30-day residual term is about five times larger at 440 MHz than at 1295 MHz. What ratio would a plasma delay predict, and what ratio would a pure c + v effect predict?
5. Show that mass m_e circulating at speed c with one turn every 1.2 × 10⁻²² s has about 1/34 of the electron's spin. What radius would give exactly ħ/2?

9. Further reading

Mainstream. - R. C. Tolman, "Some Emission Theories of Light", Phys. Rev. 35, 136 (1912): the reflection rules of emission theories [web: https://en.wikisource.org/wiki/Some_Emission_Theories_of_Light]. - J. G. Fox, "Evidence against emission theories", Am. J. Phys. 33, 1 (1965): the extinction loophole; cited by Wallace. - T. Alväger, F. Farley, J. Kjellman, I. Wallin, Phys. Lett. 12, 260 (1964): the CERN pion test [memory, high for the citation]. - K. Brecher, Phys. Rev. Lett. 39, 1051 (1977). - I. I. Shapiro, M. E. Ash, W. B. Smith, Astron. J. 72, 338 (1967); G. H. Pettengill et al., Astron. J. 67, 181 (1962): the radar papers Wallace analysed. - J. V. Evans and T. Hagfors (eds.), Radar Astronomy (McGraw-Hill, 1968): includes Shapiro's chapter on fitting. - C. M. Will, Theory and Experiment in Gravitational Physics (Cambridge): how ranging tests relativity. - Griffiths, Introduction to Electrodynamics, ch.12; Taylor and Wheeler, Spacetime Physics. - I. Langmuir, "Pathological Science" (1953 talk, transcribed in Physics Today, Oct 1989): the criteria Wallace applies. - I. Shapiro, Scientific Ethics 1(2), 10 (1985): Shapiro's published reply to Wallace (not read in this review). - T. D. Moyer, Celestial Mechanics 23, 33 (1981): JPL's light-time equations, eq. (A6), the subject of Wallace's Moyer argument. - P. A. LaViolette, "Is the universe really expanding?", Astrophys. J. 301, 544 (1986): the tired-light comparison Wallace relied on. - G. Goldhaber et al., Astrophys. J. 558, 359 (2001): supernova light-curve time dilation [memory, medium for the volume and page; the result is web-checked above].

Wallace's own work. - The Farce of Physics (1993 manuscript; this folder; also online at https://www.ekkehard-friebe.de/FP_C5_LL.HTM). - "Radar Testing of the Relative Velocity of Light in Space", Spectroscopy Letters 2, 361 (1969) [web: http://sciliterature.50webs.com/SpecLetters1969-p361-367.pdf]. - "Radar Evidence that the Velocity of Light in Space is not c", Spectroscopy Letters 4, 79 (1971), and his other papers (GSJ author page, https://www.gsjournal.net/Science-Journals-Papers/Author/2378/Bryan,%20Wallace). - "The Unified Quantum Electrodynamic Ether", Found. Phys. 3, 381 (1973): the equations missing from the book. - With Rhodes and Block, "Computer Simulation of Mass Dynamics in Electrons", J. Clas. Phys. 1(2), 17 (1982). - "The Great Speed of Light in Space Coverup", Scientific Ethics 1(1), 2 (1985): the article Shapiro answered.

Related and critical. - G. Horedt and H. Ruck, Spectroscopy Letters 5, 455 (1972), the published rebuttal. - L. Essen, The Special Theory of Relativity: A Critical Analysis (Oxford University Press, 1971): the caesium-clock builder's critique Wallace quotes. - Rapier, Spectroscopy Letters 4, 303 (1971), and Wegener and Øhrstrøm, Nuovo Cimento B 30, 291 (1975): papers citing Wallace (Investigator leads; not read). - G. C. McVittie, Astron. J. 75, 287 (1970); V. A. Kotel'nikov et al., Astron. Zh. 53, 1270 (1976); F. Schmeidler, Sky & Telescope 27, 217 (1964): sources Wallace cites, not read here. - S. A. Tolchelnikova-Murri, Galilean Electrodynamics 4(1), 3 (1993), and her 2001 paper (Russian). - G. B. Malykin, "Sagnac effect and Ritz ballistic hypothesis (Review)", Optics and Spectroscopy 109, 951 (2010).

---

Answers

1. Reflection: tc/2 = 300 × 2.998 × 10⁸ / 2 = 4.497 × 10¹⁰ m = 44,968,869 km. Reception: subtract tv/2 = 300 × 5 km/s / 2 = 750 km, giving 44,968,119 km. In the constant-c theory the pulse reaches the target exactly half-way through the trip, and the target keeps moving for the other t/2.
2. With c + 2v: D_G = tc/2 at reception, so D_G − D_E = +tv/2 = +750 km. With c + v: D at reception = tc/2 − tv/4, so the difference from Eq. (1) is +375 km. Same experiment, different prediction, depending only on the reflection rule.
3. At inferior conjunction the Earth-Venus distance is at its minimum, so its rate of change v is zero and tv/2 = 0. Before conjunction Venus approaches (v > 0) and after it recedes (v < 0), so tv/2 changes sign. Since Venus is then roughly between Earth and the Sun, a larger geocentric distance means a smaller heliocentric radius, which flips the sign once more in Fig. 1's units: G − E is negative before and positive after.
4. Plasma: (1295/440)² = 8.7. Pure c + v: 1, since the speed of light would not depend on radio frequency. The observed 5 lies between them. Plasma must be part of the term, and the different epochs and stations make the remainder uncertain.
5. Circumference = c × 1.2 × 10⁻²² s = 3.6 × 10⁻¹⁴ m, so r = 5.73 × 10⁻¹⁵ m. L = m_e c r = 9.11 × 10⁻³¹ × 2.998 × 10⁸ × 5.73 × 10⁻¹⁵ = 1.56 × 10⁻³⁶ J·s. ħ/2 = 5.27 × 10⁻³⁵ J·s, and the ratio is 33.7 (close to 1/(4α), α = fine-structure constant). The required radius is r = ħ/(2 m_e c) = 1.93 × 10⁻¹³ m, half the reduced Compton wavelength, which is the Zitterbewegung scale.