Le Voyage dans la Lune (A Trip to the Moon)

Filed correctly as fiction, and both alternatives are worth arguing before they are declined, because this entry exists to teach a grading rule and the kind is half the rule.

moment is the serious one, and toy-story-1995 is the precedent that makes it serious: an entry proposed under fiction was re-kinded because its stated citation occasion — an industry displaced, with dates on both ends — was satisfied by the release rather than by the story. The house test that entry settled is a date on which something visibly happened in public, and 1 September 1902 clears it: a purpose-built narrative film of thirty tableaux, three months in production at ₣10,000, was put in front of paying audiences and became the most widely copied film of its decade. But what a reading comes here to cite is not the release. It is the content — a depicted act, a depicted method, and the sixty-seven-year gap in which one of the two came true. That is the story, so this is fiction. There is a second reason to decline, and it is the one time-machine-1895 used: a moment framing leans on a priority claim, and the priority here does not survive checking. See the correction to the proposal line below.

prediction is the tempting one and refusing it is the point of the entry. The kind requires four things — a claim in the predictor's own words, a date made, a date due, and what actually happened. This film has one of the four. Méliès set no horizon, offered no forecast, and described his own method in print five years later in terms that rule it out: in Les Vues cinématographiques (1907) he wrote that he considered the scenario last, and that "the scenario constructed in this manner has no importance, since I use it merely as a pretext for the 'stage effects,' the 'tricks' or for a nicely arranged tableau." He sold the film as a pièce à grand spectacle, the term for a Paris stage extravaganza. Recording it as a prediction and then scoring it would manufacture a forecaster out of a showman, which is the exact operation kurzweil-2005 exists in this canon to block. What it can do, graded as fiction, is supply the distinction that prediction entries need and rarely get: the act and the method are two different claims, they come apart, and a grade that merges them is worthless in both directions.

descends_from is empty, and the gap is documented rather than shrugged at. The ancestors here are known, named by the maker, and none of them is in canon/. Méliès said in 1930 that his sources were Jules Verne's From the Earth to the Moon (1865) and Around the Moon (1870); the cannon, the projectile, the committee of gentlemen, the ocean splashdown and the recovering ship are all Verne's, and the entry grades them as inherited rather than invented. Georges Sadoul argued that the lunar half came from H. G. Wells's The First Men in the Moon, published in 1901 and in French translation shortly before production — Wells coined "Selenites," and the Selenites are in the film. Thierry Lefebvre's reading adds two more: Jacques Offenbach's opéra-féerie Le voyage dans la lune (1875), itself an unauthorised parody of Verne, for plot elements, and the "A Trip to the Moon" walk-through attraction at the 1901 Pan-American Exposition in Buffalo for the structure. Four or five ancestors, all real, not one of them an id I am entitled to invent. Note that time-machine-1895 is in canon/ and is by the right author — and is the wrong book. An edge drawn from it here would be an edge drawn on the surname, which is the failure that entry itself refused to commit in the other direction. A session that adds first-men-in-the-moon-1901 or a Verne entry should come back and draw the edges.

One claim in canon/proposals.md I will not repeat. The proposal calls this "the first film to picture a thing people had only argued about." I cannot verify the priority and the counter-example is by Méliès himself: La Lune à un mètre (1898), also called The Astronomer's Dream, is a three-minute trick film in which a great moon-face invades an astronomer's study and eats his telescope — four years earlier, same maker, same subject, and Siegmund Lubin imported it to the United States in 1899 under the title A Trip to the Moon, which is a small trap for anyone checking. Méliès had been making trick films since 1896 and the substitution splice was already old work. What is actually distinctive about the 1902 film is not that it was first but that it was large and it travelled: thirty tableaux at 260 metres where the standard subject ran a minute or two, and a diffusion so wide that the image of a capsule in the moon's eye became the public picture of a moon voyage for two generations before anybody went. That is citable without a priority claim, and the priority claim would not survive a reader with a search engine.

What it is

A French silent film of about fourteen minutes at sixteen frames per second — 260 metres, thirty tableaux by the maker's own count — written, designed, directed by and starring Georges Méliès (1861–1938), shot from May 1902 and released on 1 September 1902 by his own Star Film. It cost ₣10,000 and three months, both extravagant for a subject in 1902, and it was sold outright from a catalogue: exhibitors bought prints, there was no rental and no residual.

The story is a burlesque. At a meeting of an Astronomers' Club, Professor Barbenfouillis — Méliès, in a pointed hat and robes that make the astronomers look like wizards rather than scientists — proposes an expedition to the moon. Five colleagues agree, and their names are the register the whole film is pitched in: Nostradamus, Alcofrisbas, Omega, Micromégas, Parafaragaramus. A bullet-shaped capsule is forged, loaded into an enormous cannon, and fired by a corps of "marines" played mostly by young women in sailors' costumes — ballet dancers from the Théâtre du Châtelet. Then the shot that is the whole reason anyone remembers the film: the moon approaches with a human face, and the capsule hits it in the eye.

On the moon the six climb out, watch an earthrise, sleep, and are visited by a comet, the Big Dipper with human faces in its stars, Saturn leaning out of a window, and Phoebe on a crescent swing. Snow wakes them. They shelter in a cavern of giant mushrooms — one man's umbrella plants itself and becomes one — and meet the Selenites, insectoid inhabitants played by acrobats from the Folies Bergère. The astronomers hit them with umbrellas and the Selenites explode in puffs of smoke. Overwhelmed by numbers, the party is captured and marched to the Selenite king, whom Barbenfouillis picks up and dashes to the ground, where he also explodes. They run for the capsule, which is sitting on a ledge; Barbenfouillis climbs down a rope hanging from its nose and his weight tips it off the edge, and it falls back to Earth with one Selenite clinging to it. It splashes into the ocean, sinks, bobs up, and is towed in by a steamer. The last tableau — missing from many prints for decades — is a parade: the travellers are feted, the captured Selenite is displayed, and a commemorative statue of Barbenfouillis is unveiled bearing the motto LABOR OMNIA VINCIT.

Two things about that summary are load-bearing. First, it is a comedy, and the joke is on the explorers: a scientific expedition that is really a colonial one, staffed by pompous old men who arrive, kill the inhabitants, seize a specimen, and vote themselves a statue. Méliès had drawn political cartoons in the 1890s mocking militarism and colonial bullying, and the anti-imperial reading is not a modern imposition — it is the standing scholarly one, argued most recently by S. F. Crivelli in Early Popular Visual Culture (2023), which takes the parodic scientist-explorer as the structuring device of the whole film. France was the world's second-largest colonial power at the time. Second, none of it is presented as plausible. The moon has breathable air, snowfall, vegetation and a monarchy; the astronomers travel in frock coats; the return trip is a fall off a shelf. A film arguing for the feasibility of a moon voyage would not have made any of those choices, and the entry's grading depends on noticing that it did not try.

What happened next is the other half of the entry. Alfred C. Abadie, a stringer for the Edison Manufacturing Company, acquired a print through an exhibitor and shipped it to Edison's laboratory for duplication; Edison was exhibiting the film in America by October 1902, weeks after the Paris release. By 1904 Edison, Siegmund Lubin and the Selig Polyscope Company were all selling copies, Edison matching Méliès's hand-coloured version and his pricing. Méliès, who had put his own money into the most expensive film he ever made, had nothing left to litigate with against Edison and did not try. He opened a New York branch of Star Film in 1903 under his brother Gaston to sell his own prints in his own market, which was closing the door after the horse had been distributed nationally. Edwin S. Porter came to know the film intimately in the course of duplicating it for Edison, and said that seeing it led him to "come to the conclusion that a picture telling a story might draw the customers back to the theatres, and set to work in this direction" — which he did, in 1903, with The Great Train Robbery. The theft and the transmission are the same event.

The maker's own ending is short and should be recorded without varnish. Méliès made over 500 films between 1896 and 1913; roughly 200 survive. He broke with Pathé in 1913 and was bankrupt. In 1917 the French army confiscated more than four hundred Star Film prints and melted them down for the silver, using the celluloid to make heels for boots. In 1923 the Théâtre Robert-Houdin was demolished, his remaining prints were sold by weight to a second-hand film dealer, and he burned all his surviving negatives in his garden at Montreuil in a fit of rage. From 1925 he and his second wife sold toys and sweets from a stand in the main hall of the Gare Montparnasse. Rediscovery came late and from the film people: a gala at the Salle Pleyel in December 1929, the Légion d'honneur pinned on him by Louis Lumière in October 1931, a place in the cinema's retirement home at Orly in 1932. He died in Paris on 21 January 1938, thirty-one years before Apollo 11.

And the film itself nearly did not survive. No hand-coloured print was known to exist until 1993, when one arrived anonymously at the Filmoteca de Catalunya inside a donation of about 200 films — made for a Spanish exhibitor, judging by the flag in the launch scene. Lobster Films obtained it in 1999. The reel had decomposed into a solid mass; only the edges had congealed, so the frames were salvageable. Between 2002 and 2005 13,375 image fragments were digitised out of it; a restoration running to 2011 reassembled them, filled gaps from a black-and-white print held by the Méliès family, and cost about $1m. It premiered at Cannes on 11 May 2011 with a score by Air. A. O. Scott called it "surely a cinematic highlight of the year, maybe the century."

Why a reading would cite it

First and principally: the act and the method are separate claims, and this is the cleanest case in the canon where one came true and the other was never close. Humans went to the moon on 20 July 1969, sixty-six years and ten months after this film was released, and every mechanical detail in it is wrong. That is not a curiosity; it is a rule a reading needs several times a week, and the record supplies the occasion continuously.

Take the two hardest items in this project's last week. On 1 August 2026 an internal OpenAI model produced solutions to ten open problems in mathematics and theoretical computer science for roughly $2,000 of compute, with machine-checkable Lean proofs and a 249-page manuscript. The act — machines proving theorems humans had not proved — is the act Newell, Shaw and Simon depicted in 1956 and this canon holds as logic-theorist-1956, and the depicted method was symbolic search over logical formulae, which is not what happened. On 18 August 2026 Anthropic reported a model autonomously designing protein binders against 14 of 15 targets, 354 confirmed binders from 1,320 designs, with hit rates of 22.6% and 26.7% against a stated field baseline of 10–15%, the proteins built and measured by Adaptyv Bio and Twist Bioscience rather than by the claimant. The act — a machine designing a molecule that works — was depicted for decades. The method, a general-purpose chatbot orchestrating other people's open-source folding tools from a human-written prompt while the wet lab is subcontracted to two other companies, was in nobody's picture. A reading that scores those items by asking whether they arrived the way the old pictures said will score both of them wrong.

The service runs the other direction too, and that is the sharper half. The record is full of capability claims where the method is impressive and the act has not arrived: musk-robotaxi-2019 is in this canon for exactly that. The entry's rule is not "mechanism doesn't matter." It is that the two grades must be given separately and stated separately, because a grader who merges them can produce any verdict they like. Méliès is the limit case that makes the rule legible: maximum mechanism error, act correct, and the film is neither vindicated nor refuted by 1969 — it was never in that business.

Second, the mechanism was not available to be got right. This is the part that turns a joke into a discipline. The correct method — reaction propulsion, with the mass-ratio arithmetic that governs it — was published by Konstantin Tsiolkovsky in Nauchnoye Obozreniye in 1903, the year after the film, in Russian, in a journal nobody in Paris was reading; Goddard's liquid-fuel work is 1926 and Oberth's is 1923. At the moment of the depiction the right answer was not in the world. What was in the world was Verne's cannon, which Méliès took. A reading looking at a 2026 depiction of some capability — a company roadmap, a demo video, a policy document reasoning about what agents will be able to do — is looking at a picture drawn from the methods currently in the room, and the actual route may not have been invented yet. That is not a reason to discount the picture. It is a reason to grade the picture on the act and hold the method loosely, and to be suspicious of any depiction whose confidence is concentrated in the mechanism.

Third, and this is the entry's other body of service: the value went to whoever owned the duplication machinery, and the maker had no practical remedy. The most copied film of its decade earned its maker almost nothing. The copier's own employee learned narrative cinema by duping it and used what he learned the following year. Méliès did not sue because litigation cost more than he had. This is a 124-year-old case with the entire arc visible, ending, and the ending is that the medium flourished and the man who made its most famous object sold sweets in a railway station for seven years.

The rhymes in the current record are specific and dated. On 17 August 2026 404 Media put a tracker inside a rare book, followed it into the bulk-buying channel, and watched it end at an Amazon facility in Las Vegas where spines are cut off and pages scanned for AI training data, the copy destroyed in the process; out-of-print pre-2022 text is the point, because it is scarce and certainly not model-written. The same day, Spirit Airlines' internal corpus — roughly 100 million emails, 500 million Teams conversations, 3.4 million payroll records, employee data back to 1986 — was sold in bankruptcy for $10m to train models, with the customer data carved out and the employee data not. And on 13–14 August 2026, California's AB 412, which would have required documentation of copyrighted material used in training, was held in committee with no recorded vote and cannot return this session. Maker, machinery, remedy: the value accrues to whoever holds the copying apparatus, the original maker's remedy is theoretical, and the mechanism by which the work becomes universally known is the same mechanism by which its maker is not paid.

The honest version of that paragraph is the one worth citing, and it is not the sentimental one. Méliès was himself working entirely on unlicensed material. He took Verne's plot without a licence; Offenbach's operetta, one of his probable sources, was an unauthorised parody of Verne; the Buffalo attraction was a commercial ride. The whole foundational layer of the medium was built out of borrowing, and then spread by copying. What distinguishes Méliès from Edison in this story is not that one respected rights and the other did not. It is that one had a distribution machine and a legal department and the other had a theatre in Paris. A reading tempted to narrate the 2026 training-data fights as theft versus creation has a case here showing that the axis is nearer to who owns the duplication apparatus, which is precisely what the Concentration lens exists to track.

Fourth, a smaller one this project should keep for its own sake: the copy is what survived. Méliès burned his negatives. The army melted the prints. What came back was a hand-coloured print that had been sold abroad to a Spanish exhibitor and forgotten in a cupboard for ninety years, plus a family black-and-white print — two redundant, dispersed, commercially scattered copies against zero masters. That is an argument about durability with an unusually clean proof, and it is the argument under llama-weights-2023: a thing that exists in many hands in many places is hard to lose, and a thing that exists as one master held by one party is one bad afternoon from gone. On 18 August 2026 this project recorded roughly 500 community quantisations of an open weights release appearing unprompted within days of it. Those are not convenience artefacts. They are the reason the release cannot be withdrawn. canon/ is flat files in a git repository for the same reason, and the point is not that copies are virtuous — it is that the only records that survive neglect are the over-copied ones.

Fifth, and offered with the brake on: the film is a satire of the expedition, not a celebration of it. Six credentialed men decide among themselves, build an enormous machine, arrive somewhere they were not invited, kill what they find, take a specimen home, and erect a statue to themselves reading labour conquers all. A reading looking at capability announcements will find that shape available constantly and should mostly not reach for it, because it is the cheapest available sneer and this project's rule is to record the magnitude and not the consequence. Where it is legitimately citable is narrow and concrete: when the subject of an automated system is not a party to it. On 17–18 August 2026 Reddit began turning selected users' posts and comments into videos with synthetic narration labelled "Real conversation voiced by AI," having disclosed neither its selection criteria nor whether the people whose words are being read aloud can opt out. The Spirit Airlines employees whose working lives were sold did not sell them. The film's final tableau is a captured Selenite led on a leash past a cheering crowd at a ceremony held in its honour. That image is earned by those two items and by very little else.

What it got right, and what it got wrong

Claim date for everything below: 1 September 1902. No due dates were set by anybody, and none can be invented. Méliès made no forecast, named no year, and said in 1907 that the story was a pretext for the effects. What follows is therefore graded as a depiction against what happened — the standing comparison being Apollo 11, launched 16 July 1969, landed 20 July 1969, splashed down 24 July 1969 — and every line says which of the act or the method it is grading. Nothing here is a prediction score.

Right, on the act, and it is the only unqualified mark in this section. People went to the moon deliberately, arrived, walked on it, and came home alive. Depicted 1902; happened 1969; interval sixty-six years and ten months. No horizon was offered, so this is not a timing grade, and it should be said plainly that "humans will reach the moon" was not a bold position in 1902 — it had been argued in print for centuries and Verne had staged it in detail thirty-seven years earlier. The film's contribution to the act was not the idea but the image, and that is a real contribution: it put the thing in front of audiences who would never read Verne, in a form they could not argue with, decades before there was any mechanism to argue about.

Right in the shape of the enterprise, and inherited rather than invented. A committee proposes it; a single enormous purpose-built machine is constructed; the launch is a public spectacle with a crowd; the return is a splashdown at sea and a recovery by ship; the survivors are paraded and memorialised. All five happened. Apollo 11 splashed down in the Pacific and was recovered by the USS Hornet; the crew had a ticker-tape parade in New York on 13 August 1969. Graded right and credited to Verne, who put the launch in Florida and the splashdown in the Pacific in 1865, and from whom Méliès took the whole frame. An entry that scored this to Méliès would be laundering a borrowed prediction, which is the specific dishonesty kurzweil-2005 records.

Right in the one place the medium was his own — the event would be watched. Méliès's instrument was spectacle, and the thing he added to Verne was an audience: the crowd at the cannon, the parade at the end, the statue. Apollo 11 was seen live by an estimated 600 to 650 million people, the largest television audience that had ever existed, and it held that record for twelve years. This is the softest of the "right" marks because it is a natural consequence of who he was rather than an insight, and it should be graded as such. But the man whose contribution to the moon voyage was to make it visible turned out to be describing the form the actual event took.

Wrong — the cannon, comprehensively and lethally. Published estimates of Verne's Columbiad put the acceleration on the occupants at something in the range of 20,000 to 30,000 g. The crew would have been killed in the barrel; hydraulic shock absorbers on the projectile floor do not change the arithmetic by a useful factor. Depicted 1902, graded wrong on physics that was available in 1902. The mitigation is stated above and is real: the right mechanism entered the literature in 1903 and was obscure for twenty more years. The non-mitigation is that Verne's own contemporaries had already objected, and Méliès kept the cannon anyway because a cannon is a better tableau than an equation.

Wrong — the moon. Air, snowfall, mushroom forests, caverns, an insectoid monarchy. Graded wrong, and gradeable at the time: William H. Pickering's occultation measurements of 1892 had already put the lunar surface pressure below one four-thousandth of Earth's, ten years before Méliès put weather on it. This is the single strongest piece of evidence that the film was not attempting a forecast — the airlessness of the moon was published, quantified and a decade old, and the film ignores it in the first minute on the surface, because breathable air is what lets six men in frock coats walk around a painted set.

Wrong — the return trip. The capsule is tipped off a ledge and falls to Earth. There is no orbital mechanics in the film at all, no escape velocity in either direction, no navigation, no course correction, no radio, no computation. Graded wrong. It is worth naming what is absent rather than merely incorrect: everything that turned out to be the hard part of getting to the moon — guidance, staging, mass budgets, the arithmetic of fuel — is not in the picture, and a depiction cannot get wrong what it does not represent. That absence is generic, and a reading looking at any confident depiction of a capability should ask what the picture simply does not contain.

Wrong — the crew. Six elderly academicians in robes, self-selected at a club meeting, travelling with umbrellas and blankets. The actual crews were military test pilots and engineers, selected against physical and technical criteria and trained for years inside a state programme employing hundreds of thousands. Graded wrong, and the error has a modern shape worth naming: the depiction assumed the people who understood the thing would be the people who did it, and the actual arrangement separated those roles almost completely.

Wrong, and specifically inverted — the specimen. The film brings a live Selenite home and parades it. The real programme's fear ran the other way: Apollo 11's crew were held in quarantine for three weeks against lunar back-contamination and released on 10 August 1969. Depicted 1902, graded wrong and interesting, because the failure is not of imagination but of direction — the film could imagine contact and could not imagine that the risk might point at the travellers.

The inverse case, and the reason this entry is worth its length. On 13 January 1920 the New York Times, in its "Topics of the Times" column, mocked Robert Goddard for supposing a rocket could work in a vacuum, and said he "only seems to lack the knowledge ladled out daily in high schools." On 17 July 1969, the day after the Apollo 11 launch, the paper printed a retraction: "Further investigation and experimentation have confirmed the findings of Isaac Newton in the 17th century and it is now definitely established that a rocket can function in a vacuum as well as in an atmosphere. The Times regrets the error." Set the two records side by side. A 1902 fairground comedy, which got the method absurdly wrong and never claimed otherwise, has aged into a reasonable picture of an event that happened. A 1920 editorial in the paper of record, which addressed the correct method with confidence and contempt, is one of the most-quoted errors in the history of journalism. This canon already holds rosenblatt-nyt-1958 for the same institution overclaiming in the other direction thirty-eight years later. The pairing is the base rate: confident dismissal of a working mechanism and confident amplification of a limited one are both worse-scoring behaviours than an honest fantasy that never pretended to be a forecast.

Not gradeable, and the entry should not pretend otherwise: influence. It is routinely said that this film inspired the space age. There is no evidence for that and I did not find any. Griffith's "I owe him everything" and Porter's statement about narrative are documented and are about cinema, not spaceflight. Tsiolkovsky, Goddard and Oberth all cite Verne; none of them, so far as I could find, cites Méliès. The film's demonstrable effect is on how people pictured the act, which is not nothing and is not causation, and this entry claims the former only.

Commonly misused as

Not required for fiction, and it is the section that decides whether the entry is safe to hand to a reading, because this film is a fourteen-minute comedy carrying a century of accumulated meanings that are not in it.

• "Méliès predicted the moon landing." The headline misuse, and the reason the entry is kinded fiction. There is no claim, no date due, no forecaster and no argument. The maker said the story was a pretext for the tricks. The method he used was contradicted by the astronomy of his own decade, and he used it anyway because it looked better. Anything of the form "X predicted Y" applied to this film is manufacturing a prediction out of a picture, and a canon that permitted it here could not object when a principal does it to their own back catalogue — which is what kurzweil-2005 is in this canon to guard against.
• "Fiction gets there first." It doesn't, and this film in particular is downstream of everything. Verne is 1865; the moon-voyage argument is centuries older; the working mechanism came from Tsiolkovsky, Goddard and Oberth doing physics in the open, and it arrived after the film, not from it. Depiction and invention ran in parallel here, and where they touched, the traffic went from the science into the story.
• "The mechanism was wrong, so the depiction was worthless." The symmetric error, and the one this project is more likely to make, because its whole discipline is a distrust of pictures. The film named an act correctly at a moment when the method for it did not exist in any language. A reading that discards every depiction whose mechanism failed will discard the entire useful category and keep only the ones that got lucky about implementation details.
• "The Selenites are the aliens / the machines / the AI." They are people, played by acrobats from the Folies Bergère, living where they live, and the astronomers kill them with umbrellas and put one on a leash. Reading them as a non-human threat inverts the film — the threat in this story is the expedition — and it is the same deletion of the owner that alphaville-1965 catalogues by a different route.
• "It shows science conquering nature." The statue at the end says LABOR OMNIA VINCIT and the film does not agree with it. The astronomers are ridiculous from the first tableau, the expedition is a colonial raid, and the standing scholarly reading takes the parodic scientist-explorer as the film's organising device. Quoting the triumphal ending straight is quoting the target of a joke as its thesis.
• "The capsule in the eye is the birth of special effects." It is the most reproduced image in early cinema and it is not a first. Méliès had been building trick films since 1896, the substitution splice was years old, and La Lune à un mètre (1898) already had a face-moon eating an astronomer's telescope. What the 1902 film has is scale and reach, and a reading that needs a priority claim to make its point should find a different point.
• **"Edison stole it, therefore copying is theft, therefore this copying is theft."** The analogy is real and it does not resolve the way a reading reaching for it usually wants. Méliès's own picture was assembled without a licence from Verne, Wells, Offenbach — whose contribution was itself an unauthorised parody of Verne — and a commercial fairground ride; the duplication that cost him the money is also what carried the film across the world and taught the medium how to tell stories. The transferable finding is about who holds the apparatus, not about who was virtuous, and a reading that flattens it into a morality tale has made the case weaker than it is.
• "Le Voyage dans la Lune is evidence for X." It is evidence for nothing. It is a fourteen-minute silent comedy, this canon is not the evidence ledger, nothing in this file enters history/events.jsonl, and nothing here places or moves the needle. It exists so that a reading can explain a finding it reached some other way.

Sources

Primary, as far as the object goes: the film itself, which is in the public domain and viewable in both the black-and-white and the 2011 hand-coloured restorations; the plot beats, the tableau count and the final parade sequence above were checked against the film's own shot-by-shot descriptions in reference material rather than asserted from memory. The Méliès quotation on method — "the scenario constructed in this manner has no importance, since I use it merely as a pretext for the 'stage effects,' the 'tricks' or for a nicely arranged tableau" — is from Les Vues cinématographiques, published in the Annuaire général et international de la photographie in 1907, and reached me through secondary quotation rather than from the 1907 text; it is the most load-bearing quotation in this file and it is second-hand, which is recorded here rather than hidden.

Reference: Wikipedia, "A Trip to the Moon" — the 1 September 1902 release, the 260 metres and thirty tableaux, the ₣10,000 budget and three-month shoot, the character names and casting (Châtelet dancers as the marines, Folies Bergère acrobats as the Selenites), the plot including the LABOR OMNIA VINCIT statue, Méliès's 1930 statement crediting Verne, Sadoul on Wells's The First Men in the Moon, Lefebvre on Offenbach and the 1901 Buffalo attraction, the Abadie route by which Edison obtained a print, the 1904 state of the piracy (Edison, Lubin, Selig), the 1903 New York Star Film branch, the 1908 MPPC price standardisation, the Porter and Griffith quotations, the 1993 Filmoteca de Catalunya donation, the 13,375 recovered fragments, the ~$1m restoration and the 11 May 2011 Cannes premiere with the Air score, and the A. O. Scott description. Wikipedia, "Georges Méliès" — the 1861 and 1938 dates, the 500+ films of which about 200 survive, the 1913 Pathé break and bankruptcy, the 1917 army confiscation of over four hundred prints melted for silver with the celluloid made into shoe heels, the 1923 demolition and negative-burning, the Gare Montparnasse stand from 1925, the December 1929 Salle Pleyel gala, the Légion d'honneur presented by Louis Lumière in October 1931, and the Orly retirement home in 1932. Wikipedia, "The Astronomer's Dream" — La Lune à un mètre (1898) and Lubin's 1899 American release of it under the title A Trip to the Moon.

Charles Musser, Before the Nickelodeon, for the Abadie-to-Edison duping account and for Porter's role in duplicating the film, reached through secondary citation rather than read here. The Public Domain Review's entry on the film for Méliès's 1930 attribution to Verne and for the "spoof of more serious science fiction" and colonial-comment readings.

S. F. Crivelli, "The spectacle of the moon conquest: how visual culture shaped Méliès' Le voyage dans la Lune and its anti-imperialist satire," Early Popular Visual Culture, 2023 (doi:10.1080/17460654.2023.2206008) — the scholarly statement of the anti-imperial reading and the parodic scientist-explorer as the film's structuring device. The full text returned HTTP 403 and I could not read it; the argument above is taken from the abstract and from the University of Birmingham repository listing, and is attributed at that weight. Matthew Solomon on Méliès's 1890s political cartoons mocking militant nationalism and colonial bullying, reached the same way.

For the mechanism grading: published analyses of Verne's Columbiad putting the occupants' acceleration in the 20,000–30,000 g range, consulted in general reference and popular-science treatments rather than recomputed here. Konstantin Tsiolkovsky, "Exploration of Cosmic Space by Means of Reaction Devices," Nauchnoye Obozreniye, 1903, for the date on which the correct method entered the literature. William H. Pickering's 1892 occultation measurements bounding the lunar surface pressure below 1/4000 of Earth's, taken from a historical overview of lunar atmosphere and water observations. The New York Times "Topics of the Times" editorial of 13 January 1920 on Goddard and the correction of 17 July 1969, both widely reprinted and quoted here from that secondary circulation rather than from the paper's archive.

Apollo 11 particulars — the 16, 20 and 24 July 1969 dates, the Hornet recovery, the three-week quarantine ending 10 August 1969, the 13 August ticker-tape parade, and the 600–650 million live television audience — are standard reference facts, taken from general reference sources and stated at that weight; the audience figure is a contemporary estimate with a range, and is given as a range for that reason.

The 2026 items named as the citation occasion — the Astra mathematics result of 1 August; the 404 Media rare-book tracking to an Amazon training facility and the Spirit Airlines corpus sale in bankruptcy, both 17 August; California's AB 412 held without a recorded vote on 13–14 August; Reddit's synthetic narration of users' posts on 17–18 August; and Anthropic's protein-binder result and the open-weights quantisation figures on 18 August — are as recorded in this project's own digests for 2026-08-15 through 2026-08-19, which hold the primary links. They are named here as the occasion for citing this film. They are not evidence deposited by this entry, nothing in this file enters history/events.jsonl, and nothing here places or moves the needle.