Fun With Science / Globe Deconstruction / Section p. 203 / Draft
The book’s closing chapter, read as what it is: the summary of the case, the standard it proposes, and the two lists it drew up — both of them his.
Pages 203–240 close the shape debate as the book frames it. They state a standard (“extraordinary claims require extraordinary evidence”), assign the burden (the globe side makes the positive claims; the sceptic’s job “is simply to demonstrate fallacy”), and supply a scorecard: three claims for one side, a hundred and three for the other (his numbering runs to 103; there are 102 entries, and this page counts the entries). This page takes the chapter on its own terms. The part of its charge that lands is accepted and met with a rule this site binds itself to. The rest — the assignment of the burden, the two lists, the teams — does not survive being written down carefully, and the chapter’s own standard, applied evenly, is the one this review has been using all along.
Asymmetry charge: partly right, and metBurden as assigned: does not holdBoth lists: his, and out of balanceTeams: his framing, not a descriptionHis standard, applied evenly, is this review’s
Where this lands
The claim (pp. 203–205): the globe is a long list of positive claims that owe extraordinary evidence; the sceptic owes nothing but a demonstration of fallacy; and the two models are not held to the same standard. The verdict. One part lands: a critic who says a result is wrong is making a claim and carries its burden (Truzzi’s own point), and this review binds itself to that — every mainstream figure it uses is checked against its source or an identity it must satisfy. The rest does not. The book’s three claims are positive claims and carry the same burden; “falsification is independent of replacement” excuses a claimant from supplying a model, not evidence. Both lists were written by one side: the three are the book’s strongest, the hundred and two are mostly not about the shape of anything, and the list the globe actually rests on is fourteen structural claims, each with a number, a test and a falsifier, twelve of them checkable from a garden. Nobody joined a team. The maxim itself says nothing about the claim; it reports how unlikely the speaker already found it. What does justify treating the two models differently is a track record: predictions staked in advance and met. That test binds the globe as hard as it binds the flat plane. Against evidence that converges across instruments and centuries, a falsification has to be specific — which measurement, by how much, and why the others agree with it — and the book’s own claims are answered on this site to exactly that standard.
A converging measurement moves less each time the instrument improves. A wrong one gets worse, because the error was in the thing, not in the ruler.
It is the book’s summary: the standard of proof, a list of fallacies to watch for, the scorecard, a debate card, four short pieces of evidence, and a programme. Read in order:
| Pages | What it contains | Where it is answered |
|---|---|---|
| 203–206 | The standard: extraordinary claims, the burden on the globe side, “simply to demonstrate fallacy”; a list of ten logical fallacies. | This page, §§2–8. The fallacy list needs no answer; it is a fair list. |
| 208–231 | “An optimized approach”: Team A’s three claims plus Rampion; Team B’s hundred and three; the ratio rule. | The Optimized Debate, Mapped, item by item; the globe’s own list on What the Globe Actually Claims. |
| 232 | The debate card (Levi Miller vs Samuel Blitz): is extraordinary evidence provided that objects cannot vanish bottom-up, that gas propels in a vacuum, that Jupiter’s black circles are shadows? | Bottom Up Observations (standard physics: they can, and the pond shows it); Rockets and Thrust, Measured in Flight; Jupiter’s Shadows. |
| 234 | “This is Earth. CGI, curve, spin, orbit & spacetime sold separately” — space footage shown in classrooms. | Footage authenticity is one of the two things this review does not assess. Curve, spin and orbit are claims 1, 2 and 5 on the globe’s list, each with its ground test. |
| 235 | “What the curve should look like left and right at 118,000 feet” (QR 70, a horizon-curve simulation; QR 69, high-altitude balloon footage). | Measured on its own page: the render is right for its 100° lens, the balloon frames cannot decide, and the rocket footage bows toward space in twenty of twenty-one orientations; see §9. |
| 236–237 | The GoFast rocket: the rocket “suddenly stops”; no despin mechanism visible. | Labelled “Speculation” by the book itself, so inside its own fence. The observation under it — what stops at 227 s — is identified on the same page: the roll, not the rocket. |
| 238–239 | Kampf’s experiment; the four-point programme (spread the three claims; test The Final Experiment; recruit deconstructionists; independent teams for Kampf’s law, low Earth orbit, and mapping). | Kampf on the rockets page, pending his results; the programme in §9. |
| 240 | “Why the lie?” — the transition into the speculation chapters. | Not a claim. The speculation chapters have their own pages. |
Miller’s argument on p. 203 is that the two models are graded on different scales. An observation that merely suggests the globe is taken as settled; the same weight of observation for a flat plane is dismissed. An anomaly for the globe is filed as an open problem; an anomaly for the flat plane is fatal.
Part of that is right. The standard usually cited to justify the difference — extraordinary claims, extraordinary evidence — does not justify it (§6). And a critic who says a result is wrong is making a claim of his own. Marcello Truzzi, who gave the maxim its modern form in 1978 (“extraordinary claims require extraordinary proof”), said so himself in “On Pseudo-Skepticism” (1987): a critic who asserts there is evidence for disproof “is making a claim and therefore also has to bear a burden of proof.” This review is that critic. It says, page after page, that specific claims are wrong for specific reasons, and each of those assertions carries its own burden. The rule below is how that burden is met.
What is not right is who the charge falls on. That is §3.
The practical form of that burden is a rule: a mainstream figure gets the same check his figures get before it is used — verified against its primary source, or against an identity the numbers must satisfy by definition. Every geomagnetic value this site publishes must obey H = F·cos I; a value that fails it is discarded whatever its provenance. That is one rule. It does not answer the general charge; the rest of the page does.
The chapter’s assignment of the burden rests on a sorting: the globe side makes positive claims, the sceptic only questions. The sorting is wrong on both ends.
The book’s three lead claims are positive claims. That objects vanish across flat water with nothing in the way, that gas cannot push in a vacuum, that four celestial observations show the Sun is not the illuminator — each states something about the world, each is graded on this site as stated, and each carries the burden the chapter assigns to the other side. “Falsification is independent of replacement” is true and beside the point: it means a falsifier need not offer a rival model, not that a falsifying claim needs no evidence. The Rampion observation is a fourth positive claim, and the Antarctica footage a conditional fifth. All five are on the map.
On the other end, the hundred and two were written by the same author for his opponents, and most are not about the shape of anything: galactic dynamics, dark matter, the asteroid belt, life aboard the space station. Sorted by what each bears on, eight concern the Earth, twenty-one the heliocentric arrangement, and forty-six of the hundred and two are structural at all. A side can be right or wrong about every particular fact on that list and the shape of the ground is untouched. The list the globe rests on is shorter and is written the way the chapter asks for evidence to be written: fourteen structural claims, each with its number, the observation that tests it, the observation that would break it, and how far a reader can check it without an agency. Twelve of the fourteen are checkable from a garden; two need the tools amateur astronomers own.
And there are no teams. “Helio-skeptic” describes an activity, not a shape, and scepticism is what both sides are supposed to be doing; “heliocentrist” is someone who accepts an arrangement on the evidence for it, not a member of anything. Both sides of this argument are trying to find out what is true. This site belongs to neither: it questions the book’s claims with the instruments it would use on anyone’s, concedes where the book is right, and shows its working. The chapter’s standard, applied evenly, is that method.
The claim was extraordinary once. Measuring the size of a planet from the length of two shadows in Alexandria was an extraordinary claim in 240 BC. So was putting the Sun eighteen lunar distances away from the angle of a half-lit Moon. The instruments were a stick and a protractor, and scepticism was the correct response.
What happened next is the part the maxim cannot describe. The result did not accumulate trust; it got measured again, with a better instrument, then by a method that had not existed before. Eratosthenes had two sticks. Picard had a quadrant and a triangulation chain. Maupertuis surveyed Lapland to settle an argument between Newton and Cassini that had nothing to do with the shape being flat. Struve had 2,821 km of arc and 265 stations. Then satellite geodesy, which needs no ground at all. Then a sealed box in an aircraft bay that finds its own latitude by sensing the planet turn underneath it, and refuses the alignment if the crew types in a position that disagrees. Different instruments, different physics, different centuries, different motives, no shared apparatus that could fail the same way twice.
And that convergence is load-bearing: it is inside things that would fail, visibly and expensively, this week, if it were wrong. That is what “ordinary” means here — not that people stopped asking, but that the answer is checked continuously by people who are not thinking about it at all.
The book's questions cluster at the edges of measurement: where an effect is small, an observation ambiguous, an instrument pushed past its range, or a photograph hard to read. Edges are where models break, and a legitimate place to look. But an edge case only threatens a model whose centre is untested, and here the centre is in daily industrial use.
| The edge the book examines | What runs on the same physics, every day |
|---|---|
| Whether an object vanishes bottom-first across a few hundred metres of pond — Claim #1, p. 2 | Radio and radar path planning. Every microwave link and air-traffic radar is sited with an explicit earth-bulge term and a four-thirds-radius refraction factor; get it wrong and the link does not close. |
| Whether water in a controlled channel curves — Q2, pp. 88–116 | Long-baseline construction. The Verrazzano-Narrows towers stand 1⅝ inches farther apart at the top than at the base across a 4,260-foot span, because vertical is not parallel. LIGO’s 4 km arms were laid straight through 1.25 m of departure from level. |
| Whether the Sun shrinks rather than sets — Q8, p. 51 | Solar arrays and daylight studies. Every photovoltaic installation and shadow study runs a solar-position algorithm built on a spherical Earth and a Sun at one astronomical unit. |
| Whether refraction invalidates shadow angles — Q12, p. 55 | Land surveying — the trade his refraction constant belongs to, where it is applied as a seventh of a curvature correction nobody in the trade disputes. |
| Whether a flight recorder would show a turn on the equator — Q3, pp. 117–123 | Inertial reference alignment before every departure, tens of thousands of times a day, in a procedure that computes latitude and rejects the crew’s entry if it disagrees. |
| Whether the atmosphere co-rotates — Q10, pp. 179–202 | Weather forecasting and gunnery. Every numerical weather model carries a Coriolis term going as sin φ; artillery firing tables carried it long before computers. |
| Whether the star field moves at one rate — Q5, p. 48 | Telescope drives, hobbyist to observatory, all tracking at the sidereal rate — one turn against the stars, 360° in 23h 56m 04s. |
| Whether Jupiter’s moon shadows align — Q6, p. 49 | Shadow-transit times published years ahead and checked from back gardens; the same moons fixed longitude before chronometers existed. |
| Whether the Moon’s silhouette appears before an eclipse — Q9, p. 52 | Paths of totality published years in advance, to the kilometre and the second, then walked into by millions of people who would notice. |
| Whether the rotational south pole shows landscape movement — Q11, p. 54 | The marker at Amundsen–Scott is repositioned every 1 January because the ice under it flows about ten metres a year. The pole stays put; the ice does not. |
| Whether magnetic declination is itself an invention — pp. 415–473 (a separate chapter; a shell page reproduces his four declinations and computes the flat-plane alternative) | Aviation charts, reissued on the World Magnetic Model’s five-year cycle, and compass swings on aircraft against surveyed ground marks. |
| Whether rockets need air to push against — Claim #2 and Q1, pp. 56–87 | Every satellite in orbit was placed there by an engine burning where there is no air, and is held there by thrusters firing in the same vacuum. |
Running the independence check on our own table. Twelve rows are not twelve independent confirmations. Telescope drives and inertial alignment both measure one quantity, the rotation rate of 15.04° per hour that follows from 360° in 23h 56m 04s: two instruments for one fact, counted once. The radio four-thirds factor and the surveyor's one-seventh descend from the same standard-atmosphere temperature gradient: a shared modelling assumption, counted once. That leaves ten independent facts, not twelve — still the number a flat plane has to make false at once, by different mechanisms, without any of the trades noticing.
The book asks for verification “by a handful of independent 3rd parties.” Most of the right-hand column is exactly that: shadow transits checked by amateurs with telescopes they paid for, curvature offsets run by survey crews, a tower plumbed vertical by ironworkers who find the top wider than a flat plane would allow, a sidereal drive in a back garden — and the shortest version is a stick, a level lake and an afternoon. Everything in the left-hand column started as a question asked when the answer was genuinely unknown; the division is between a question taken to a measurement and a question that stays a question.
“Extraordinary claims require extraordinary evidence” sounds like a standard of evidence. It is not. It says that a claim the hearer finds unlikely needs more to shift them, which is true of every claim, and nothing about which claims are the unlikely ones. Used bare, it reports how improbable the speaker already found the claim, in the grammar of a fact about it. Its coiner, Marcello Truzzi, kept it in 1987 and added the half usually dropped: a critic’s negative claim can be extraordinary too, and then bears the heavier burden. Jefferson applied it in 1808 to a stone said to have fallen from the sky, asked that the testimony be weighed, and the stone had fallen.
Three things follow, and all three go Miller’s way. “Extraordinary” is measured against some background of belief, and nobody says whose. The usual repair — “most fringe claims turn out false” — depends on which pile the claim is filed in, and a person chooses the pile. And a rule that treats every anomaly for the globe as an open problem and every anomaly for the flat plane as fatal cannot be moved by any observation: it looks like scrutiny and works like a verdict. This review could fall into that; no number is accepted here for where it came from.
What the maxim does not do is settle anything about the globe, in either direction. A round Earth in motion was an extraordinary claim when made, against every appearance, and was then backed by an extraordinary quantity of evidence: centuries of navigation, survey and eclipse prediction, laid out in the fourteen claims with the checks a reader can make. That is not authority or popularity; it is a record of predictions staked and met, the only thing that earns a model its standing. Showing that some small thing is not as the textbook has it — a refraction coefficient, a shadow timing — is not showing the model false, and the maxim cannot make it so. Nor is the model immune; our prior on the flat model is not zero. Blackwell and Dubins showed that two people with different priors who weigh the same evidence converge, unless one has set a probability to nothing. Evidence can move this review; what would is listed at the end of this page.
There is a version of the maxim that does work, and it runs the other way. Each of the fourteen structural claims sits on measurements that agree across different instruments, different centuries and different trades that would notice a failure the same week: the surveyor’s curvature correction, the radio engineer’s earth-bulge term, the airliner’s inertial alignment, the eclipse timetable. Evidence that robust is not falsified by a doubt. It is falsified by a specific claim: which measurement is wrong, by how much, and why every other measurement that agrees with it is wrong in the same direction. “Refraction could explain it” is not a falsification of the hidden-height claim; a measured temperature gradient that the solver turns into a different picture would be. “Nobody has proven sunlight is straight” is not a falsification of the orbit; a measured aberration of zero would be. The book asks for evidence “verified in multiple ways by independent third parties”; that is what the right-hand column of the table in §5 is, and it is why a falsifier of any one row has to explain the others.
The obligation is symmetric, and this site accepts its half. Every claim page states the number, shows the working, names what would change its verdict, and the fourteen claims each name the observation that would end them. The specificity demanded of a falsifier is the specificity offered by the thing being falsified.
If the maxim cannot license asymmetric appraisal, one candidate can, and it never mentions surprise. Imre Lakatos appraised rival research programmes on their track records rather than their plausibility. A programme is theoretically progressive when each successive version “must predict novel and hitherto unexpected facts” — novel meaning predicted neither by any rival programme nor by conventional wisdom; empirically progressive when some of that novel content is subsequently corroborated; degenerating when its modifications merely accommodate anomalies after the fact and generate nothing new to test.
The criterion has no term for how strange a claim feels, no reference class, no base rate. It is symmetric by construction, and it is settled by records of prediction, which either exist or do not, rather than intuitions, which are always available in whatever quantity is required. Being the established model earns nothing. If the globe programme were reduced to absorbing anomalies without staking risky new predictions, it would be degenerating, and this review would be obliged to report it.
Scored, one dated prediction each. The globe programme's entry is in the table above: paths of totality and Jovian shadow-transit times, published years ahead to the kilometre and the second, then met, and renewed every year. For the flat programme we know of no dated, quantitative prediction that the globe does not also make. The book's stated goal on p. 205 — “simply to demonstrate fallacy in the globe model” — is, in Lakatos's terms, a programme that has not yet staked one. That is where the book places itself on the one criterion that does not beg the question, not a verdict on anyone's sincerity; the first item under “What would change our mind” is the invitation to move.
A companion argument this site leans on elsewhere needs stating carefully, because stated carelessly it hands over a weakness. It runs: the flat model cannot be adopted alone, because it requires a long list of independently established results to be false, each established by a different method, community and purpose. The word doing the work is independently, and two confirmations that share an instrument, a calibration chain, a datum or a modelling assumption are one confirmation wearing two coats. The fourteen claims on this site were not chosen for strict independence and should not be read as if they were; several share a method, and the audit above found two dependencies in this page’s own table. But fourteen is a reader’s list, not a census. The independent lines of evidence for the Earth’s shape alone — geodetic survey, star altitudes, ship and aircraft navigation, eclipse geometry, radio and satellite ranging, time zones, the Coriolis record, gravity surveys, and on — run far past fourteen, each with its own instruments, its own community and its own reasons for existing. The flat plane does not need to find one of them wrong. It needs every one of them wrong, independently, and at the same time.
To a reader of the book that sounds like a charge of vast conspiracy, and it is not one. The better picture is a fortress of concentric walls. Each wall was built by different people for a different purpose, and each must be breached on its own terms before the next is even reached; no single breach, however clean, touches the core. That is what “established” means when it is earned rather than asserted, and it is the reason a demonstration that one refraction coefficient or one shadow timing is not as the textbook has it, however sound, is a mark on one wall.
The 118,000-foot curve (p. 235). The claim is that balloon footage should show a visible left-to-right curve at that height, and does not. It is a geometry prediction about a camera at a stated height, which is exactly the kind of claim this site grades, and it is graded on The Curve at 118,000 Feet. The arithmetic is standard — the horizon dips about 6.1° at 36 km, and the bow of the horizon across a frame depends more on the lens than on the height — and the render on p. 235 turns out to be right for a 100° lens and to say nothing about any other. The balloon compilation (QR 69) carries no lens, no height per clip and no level, and its frames sit within a few pixels of straight; that cannot decide the question either way. The rocket footage the book links next can, because it rolls: its horizon passes through the centre of the frame in every orientation, where a flat plane’s horizon must image straight, and it bows toward space in every clean frame — along the bright limb by a median 26 px over a 700–800 px chord, against 19–26 predicted for a GoPro-class lens at 118 km and zero for a flat plane. The contrast the pages draw is refuted; the balloon frames are left open.
The GoFast rocket (pp. 236–237). Labelled “Speculation of GoFast rocket video” by the book itself, so inside the fence the author draws around his own speculation, and the speculation is left there. The observation under it is not speculation and is read on the same page: what stops at 227 s of the re-upload is the roll, from about a turn a second to nil inside a second, which is what a yo-yo despin does and what an on-board camera can see; whether the vehicle is moving is not something a camera 100 km up can show, and at the top of a ballistic arc the vertical speed is zero on every model. The despin hardware the book cannot find in the frame is out of frame: from 180 s to the end no part of the vehicle is in the picture, where the MAPHEUS-5 camera it offers as the counter-example looks along its own body throughout. That MAPHEUS-5 flight is also the chapter’s quiet collision with Claim #2: its second stage burns from 4 km to 43 km, through air thinning to a five-hundredth of sea level, to the speed a 253-km arc needs. Footage accepted for the horizon is accepted for that too.
The programme (p. 239). Spread the three claims: they are answered. Test The Final Experiment in Antarctica: not assessed here, by declaration. Independent teams to test Kampf’s law: welcomed, and the protocol that would make the result count is on the self-test page. Teams to test whether low Earth orbit is possible: the station’s orbit is observable from a garden — a timed pass gives the speed, the pass interval the period, the public elements the drag — and the map’s rows 96–98 say how. Mapping teams using tangible geometry: the visual-geometry pages are where that idea is tested.
Two of the epistemology citations are still second-hand. The Laplace formulation is taken from the standard English rendering rather than the French, and the Lakatos criteria are quoted as the Stanford Encyclopedia gives them, with its page references, rather than from the 1970 paper itself. The Truzzi 1987 text, the Jefferson letter and the Hume section have been read; the Hájek and Blackwell–Dubins citations are confirmed against the publishers’ records. And the concession is epistemological, not substantive: a different, prior-independent criterion still separates the two programmes, and the obligation it places on us is exactly as binding as the concession made to him.
Stated in advance:
One audited novel prediction. A quantitative prediction derived from Miller's model, stated in advance with an error bar, that the globe model does not make, then confirmed by a measurement taken by someone with no stake in the question. One such result would carry a likelihood ratio no prior of ours could survive, and would satisfy Lakatos's criterion on its own terms.
A broken independence claim. Two or more of the lines of evidence we have enumerated as independent shown to share an instrument, a calibration chain, a datum, or a modelling assumption. Each one reduces the exponent in the compounding product.
A model-dependent “independent” measurement. Something we have described as made for reasons unrelated to this debate shown to have assumed the geometry in dispute in its reduction from raw instrument output to published value.
Another double standard of ours. A mainstream number accepted on some specific page without the check we demanded of his.