Research Briefs
This site's other axis is software: benchmarking research tools, building small calculators, testing what a claim on a vendor page actually holds up to. This axis applies the same habit to a different kind of claim — the ones that circulate as trivia, get repeated with total confidence, and rarely get checked against the thing they're actually based on.
A brief exists because chasing a claim to its source is more useful than reading the tenth article that repeats it. Most competing pages on these topics blend what's actually confirmed with what merely sounds right into one paragraph. We don't. Separating those two is the entire point of this format — everything else here follows from that one rule.
How a brief is built
Every brief on this site follows the same fixed order, not as a style choice but as a check on us:
- Hook — why the claim is worth checking, in two or three sentences, without overselling it.
- The claim — the exact thing being tested, stated as one precise sentence.
- Source trail — where the claim actually comes from: the original source if we could reach it, how it spread, and what form it takes now.
- What's confirmed — only what we verified against a primary source ourselves, each item with a link and the date we checked it.
- What isn't — what we could not verify, what a source hedges that its popularizers don't, and any number that turned out to have no citation behind it at all.
- Verdict — two or three sentences. We state a conclusion when the evidence supports one, and we say plainly when it doesn't.
"We could not confirm this" is a legitimate verdict here, and often the more interesting one — it's certainly more honest than repeating what everyone else repeats.
Source rules
A figure only ships with a primary source and a checked-on date attached: the original paper, the court's own opinion text, the institution's own page, the manufacturer's own spec. A mirror or an aggregator is not a primary source, and being widely repeated is not the same fact as being true — when we can't reach the original, the brief says so on the page instead of citing a copy as if it were the thing itself. Every brief also names, separately, the sources it tried to reach and couldn't, rather than quietly dropping them.
Before we write one
Being able to trace a claim's sources is not, by itself, a reason to publish it. Before writing a brief we check who currently ranks for the question it would answer: if the search results are Wikipedia and major press with no opening for a small independent site, we reject the topic and say so, no matter how interesting the claim is. We've done this — a brief on NASA's "space pen" was researched and rejected at exactly this stage, because every result for it was a heavyweight science or fact-check publisher with no small site holding a page-one position. A rejection with a reason attached is a normal, useful outcome of this process, not a failure of it.
What's here so far
Was Walt Disney's body really cryogenically frozen?
No — he was cremated two days after his death, and every source agrees. But the debunking itself has a soft spot: the founding 1972 quote used to disprove the myth claims the first human cryopreservation happened "two weeks" after Disney died, when an independent record puts it at 28 days — and two well-known sources trace the legend's origin to different named people without ever mentioning each other's candidate.
Was Van Halen's 'no brown M&Ms' clause really a genius safety test?
The clause was real, confirmed by an authenticated scanned rider. But that rider is dated 1982, two years after the incident it's used to prove; a widely quoted "exact" version of the clause doesn't match that same document at all; and no two sources agree on what the resulting damage actually cost — one primary account gives two figures, and two other sources each give a third and fourth.
Does a duck's quack really not echo?
False — confirmed independently at least three times between 1998 and 2003 (a newspaper column's own field test, a university acoustics lab, and a TV myth-busting show). But nearly every current source credits only the celebrated 2003 university study, with no mention that a satirical column ran the identical test, with an actual duck, five years earlier.
Did Alexander Graham Bell really steal the telephone from Elisha Gray?
A patent examiner's sworn 1886 confession that Bell bribed him is real and still gets cited today — but reading the actual confession, not a summary of it, shows Wilber himself took money commissioned by Bell's own lawyer for an earlier, pro-Bell affidavit before recanting it. And the notary named in that primary text doesn't match the name Wikipedia's own cited historian gives for who witnessed it.
Did Shakespeare really hide his name in Psalm 46?
The word count is real — the 46th word of Psalm 46 (King James Version) is "shake" and the 46th word from the end is "spear," confirmed here by counting the actual text ourselves. What it's supposed to prove doesn't hold up: those same two words sit in the same two verses of English Bible translations dating back to 1535, decades before Shakespeare was even born.
Does a champagne cork really break the sound barrier?
Correct, but not for the reason most retellings give: it's the escaping gas that goes supersonic, not the cork, and two independent peer-reviewed papers agree on that. What they don't agree on is the gas's actual speed — and the popular '50 mph' cork figure doesn't match the eye-injury group's own current page, which says 34.
Did ergot poisoning cause the Salem witch trials?
A psychologist's 1976 hypothesis blaming a rye fungus is real, peer-reviewed science, not an internet invention. But the same journal published a systematic rebuttal eight months later, and most retellings written for general readers — including a university's own 2024 article — repeat the hypothesis with no mention the rebuttal exists.
Did Victorians really believe a train could make a woman's uterus fly out?
Real doctors on both sides of the Atlantic did publish on railway travel and miscarriage — a dedicated Lancet commission chapter in 1862, plus independent American articles in 1844 and 1872. But the vivid "50 miles an hour, uterus flies out" line everyone quotes traces only to a 2011 magazine interview about modern tech panic, not to any of them — and the one other outlet that went looking hit the identical dead end we did.
Did a YouTuber really crack Coca-Cola's secret formula?
Coca-Cola's own current pages say the formula is known to "a few people," not the famous two who can't fly together — a detail traced here to a 1919 marketing stunt, not a security policy. A YouTuber's viral January 2026 mass-spectrometry "crack" is a real, independently reported event, but his own reported account describes matching a taste profile with legal substitutes for the one ingredient nobody outside Coca-Cola can buy — not recovering the document in the vault.
Do chameleons really change color to camouflage?
The "actually it's communication, not camouflage" correction traces to a real 2008 study — but of one South African genus, not chameleons broadly. The same lab's own second paper that year, and a fresh 2025 study on an unrelated species, both found real camouflage matching — and one site pushing the broader claim cites an unrelated paper about tongue speed to back it.
Was the champagne coupe glass molded on Marie Antoinette's breast?
The glass predates her by about 90 years, closing the myth on chronology alone. But the "real" object correction pages reach for instead — a genuine breast-shaped porcelain bowl she did own — carries the identical molding legend, one its own maker, Sèvres, denies on its current page as "pure fantasy." At least one small debunking site gets that second part wrong anyway.
Were medieval chastity belts ever actually real?
Historians agree there's no credible medieval evidence for them, backed by a named academic monograph. But the two UK museums that still display objects labeled "chastity belt, 16th century" hedge on that dating in their own catalogue text — and neither has actually corrected the title on its own page.
Does "rule of thumb" really come from a wife-beating law?
No record of the rumored 1782 English ruling exists, and the phrase itself predates it by a century. But a real 1868 American court case did cite a switch-no-bigger-than-a-thumb standard from unnamed "old authorities" — and even the sources that cover that case skip the detail that rejecting the standard as a legal test didn't change who walked free.
Was Napoleon really attacked by an army of rabbits?
The story traces to exactly one eyewitness memoir, not published until 49 years after its author's death — and reading that memoir directly confirms the core sequence in specific detail. What it doesn't confirm is the number doing the clicking today: the source says a thousand rabbits, not the "almost 3,000" that several current retellings repeat with no source of their own.
Did the Titanic's band really play until the ship sank?
All eight musicians did keep playing on deck and did all die. But a survivor who stayed until nearly the end wrote an entire passage in his own 1913 book denying their last song was "Nearer, My God, to Thee," a different survivor named a different song within weeks, and even the people who insist on the hymn can't be shown to agree which of three unrelated tunes sharing its name they mean.
Did Australia really lose the Emu War, and by what score?
The 1932 military campaign against Western Australian emus is real, and every source agrees it failed. But the number everyone repeats to prove it, 986 kills for exactly 10 rounds each, traces to no primary document this session could reach, carries a disputed 2,500-more-died add-on Wikipedia itself flags as "likely inflated," and doesn't match the total National Geographic reports today.
Did Coca-Cola invent Santa's red suit?
Coca-Cola's own site denies it, and pre-1931 red Santas are real and checkable — a 1902 Puck magazine cover confirmed against the Library of Congress's own catalog record. But the single piece of evidence nearly every debunking page leans on hardest, Thomas Nast's 1881 illustration, has a print-color question none of them address.
Did Van Gogh really cut off his own ear?
Archival finds in 2010 and 2016 give real, primary-document answers to two once-blurry details: how much ear was cut, and who actually received it. The detail nobody questions, that Van Gogh did it to himself, rests on one witness's account written years after the fact, and has never been independently confirmed either way.
Did Mrs. O'Leary's cow really start the Great Chicago Fire?
The cow-and-lantern story is confirmed fiction, and every source agrees the reporter who invented it eventually admitted it. But the sources can't agree when: three independently checkable accounts give three different years for his confession, decades apart, and two more disagree about who actually wrote it for him.
Is sushi-restaurant wasabi actually fake?
The '95 to 99 percent is fake' figure repeated by dozens of food sites traces to one person: the president of a company that sells real wasabi. But the world's dominant wasabi-tube brand independently backs the core claim on its own site, stating its international product is roughly seven parts horseradish to one part actual wasabi.
Did Thomas Crapper invent the flush toilet?
No — and that correction is decades old, tracing to a 1969 satirical book. But the correction itself has a blind spot: the actual patent for the device most tied to his name and his myth belongs to a real, documented, barely-mentioned plumber named Albert Giblin, and even the sources that admit this can't agree how the two men were connected.
Do Twinkies really last forever?
No — Hostess's own guaranteed shelf life is 45 days, and a 2020 mycology study found real mold on Twinkies eight years old. But one specific 1976 Twinkie, unwrapped and left on a Maine classroom chalkboard, just turned 50 with no mold at all — and nobody we found explains why one survived clean while the other didn't.
Are diamonds made from coal?
No, and every source we checked agrees on the age, depth and chemistry that rule it out. But the tidy replacement answer skips a live 2025 scientific fight: the clearest evidence that some diamonds carry once-living carbon sits inside an unresolved dispute over whether those specific diamonds are natural at all, or lab contamination.
Is the boiling frog story actually true?
The founding experiment everyone credits only got its 'frog stays put' result by removing the frog's brain first — its intact frogs escaped at 25°C, the opposite of the legend. Two later German experiments reported the classic result on intact frogs, but nobody alive has been able to check their work, and no source we found describes a modern replication of either.
Is the tongue map of taste zones a myth?
Every debunking page agrees the four-zone map is false, based on a 1940s textbook graph that distorted a 1901 thesis. The same 2022 paper that documents the distortion also finds real, if modest, extra bitter and umami sensitivity at the back of the tongue — the nuance every debunking page we checked leaves out of its own tidy correction.
Did only six people really die in the Great Fire of London?
The City of London's own monument site names six. But two national archives, examining the fire's own 1666 official record directly, found no death toll in it at all — and the historians who think the toll really was small can't agree among themselves whether it was four, six, eight, or just "single figures."
Does Switzerland really have a bunker spot for more than everyone?
The civil-protection office's own current page won't commit to a precise coverage number, only "more than 100 percent." The two figures everyone else repeats instead, 107% and 114%, are absent from it — one traces to 2006 data still being cited 18 years later, the other has no traceable source at all.
Did a freeze-thaw cycle really keep mosquitoes out of Iceland?
The October 2025 discovery that ended Iceland's mosquito-free status is confirmed directly by the country's own natural history institute. The explanation everyone adds for why it took so long — a freeze-thaw cycle — traces to no source we could find, and one small site skips it for four other unsourced theories instead.
Were London black cabs made tall so you could keep a bowler hat on?
The one Parliament record anyone can point to is about a top hat, not a bowler — and the government's own reply that same day said the current regulations had no hat reference at all. No earlier edition of the rulebook was reachable to check either way.
Was Napoleon actually short?
Every debunking source agrees "5 foot 2" is a unit-conversion myth — and disagrees with every other debunking source on the corrected number. One states a suspiciously precise, entirely uncited figure of its own.
Is a tomato legally a vegetable?
The 1893 ruling, holding, and quotes are exactly as usually described — but its own opening sentence scopes it to one word in one tariff schedule, a qualifier that drops out of most retellings, including a major outlet's own headline.
Is the Eiffel Tower taller in summer?
The thermal expansion is real physics, confirmed independently. No source, official or popular, cites an actual before/after measurement — every centimetre figure in circulation, including the tower operator's own, is a calculation from an unstated assumed temperature swing.
Did Vikings wear horned helmets?
No horned combat helmet has ever been found, and the modern image traces to a specific 1876 opera costume. But the National Museum of Denmark's own page leaves the broader question — display or ritual use during the Viking period itself — explicitly open, while popular retellings quietly close it.
Does lightning really strike the same place twice?
The myth is false, confirmed by the National Weather Service's own page. But the number everyone cites to prove it — how often lightning hits the Empire State Building — doesn't match between the Weather Service's own figure and the building's own site, and a related Guinness World Records odds figure doesn't match the Weather Service's either.
What is the deepest hole ever drilled?
The Kola Superdeep Borehole is still the deepest hole by vertical depth. But Guinness World Records' own page gives a different year, cause, and temperature for why drilling stopped than the independent science coverage does — and "deepest" quietly gets confused with "longest" once oil wells enter the picture.
How much does the internet weigh?
The "about as much as a strawberry" figure is a real 2006 estimate, not a fabrication — but Discover Magazine's own 2007 follow-up, in the same outlet, got a number nine million times smaller using a different definition of what's being weighed, and a reported 2025 revisit called the original "just wrong" and landed on a third answer again.
What is the most expensive object ever built?
Nearly every source names the International Space Station and quotes "$150 billion" — but that figure traces, via Wikipedia's own footnote, to a 2010 blog post that no longer loads, while NASA's own Inspector General audits put the U.S. share alone tens of billions lower.
Is the Bermuda Triangle actually more dangerous?
NOAA and a 2013 WWF shipping-safety study both agree it isn't — but the single most-repeated piece of supporting evidence, that Lloyd's of London itself confirmed this, doesn't trace to any page Lloyd's has ever published.
Can you outrun lava?
Yes, by USGS's own numbers — even the fastest lava flow on record barely beats a human jog. But the 1980 Mount St. Helens eruption people picture when they ask this was actually two hazards at very different speeds, and USGS's own account is the only source we found that keeps them separate.
Why is the QWERTY keyboard arranged that way?
The 1878 patent nearly every retelling points to really does draw the keyboard layout — but its own 14 legal claims, read directly, cover paper carriages and ratchets and never once mention why the letters sit where they do. The "prevents jamming" story is a later reconstruction, not something Sholes's own patent says.
Do goldfish really have a three-second memory?
The myth is false, backed by a real 1966 study that almost every debunking page cites by year but strips of its authors and journal. But one popular counter-example — a fish released into open water that supposedly remembered a sound five months later — turns out to merge two real studies, about a different species doing a different thing, into an anecdote that matches neither.
What animal is fastest over distance, not sprint?
The pronghorn really does outlast the cheetah, backed by a real 1991 oxygen-consumption study — but Discover Magazine, a-z-animals.com, africa-safaris.com and Guinness World Records' own official record for the category each state a different sustained speed, and none cite the study the claim traces back to. Even the cheetah's side of the comparison has the same problem: the commonly quoted 70 mph is higher than the only speed anyone has measured a wild cheetah reach.
Did an apple really fall on Newton's head?
Newton's own account, in his friend's 1752 memoir digitised by the Royal Society itself, never says the apple struck him — that detail traces to a specific later writer, Isaac D'Israeli, decades on. A second, separate question hides behind it: the Royal Society's own blog declines to confirm the tree tourists visit today is really his, asking "who's to say it isn't?" rather than saying it is.
Did Marie Antoinette really say "let them eat cake"?
The line is on record in writing years before she ever set foot in France, credited by its own original author to an unnamed "great princess." Her name wasn't put on it until 1843, fifty years after her execution — by a writer whose own cited source dated the line to before she'd even arrived in the country.
Did Napoleon's soldiers shoot the nose off the Sphinx?
Two travelers sketched the Sphinx in the exact same year, 1737, sixty years before Napoleon arrived — and drew two different faces. The one showing an intact nose looks like a copy of an even older drawing, not an eyewitness record, and even the leading debunking page disagrees with its own cited source on the year the real damage happened.
Was Uncle Sam really a real person named Samuel Wilson?
Congress named a Troy, New York meatpacker the origin of "Uncle Sam" in 1961 — but a Navy midshipman's 1810 diary and Wilson's own hometown newspaper both used the term before, or without ever naming, the man Congress credited.
Did knights really need a crane to mount their horse?
One outlet blames Mark Twain by name for the crane image. Twain's own 1889 novel, read directly, says an armored man needs a person's help mounting — not a machine's. The actual crane/derrick line traces to a different, far less famous 1906 book instead.
Do you really swallow 8 spiders a year in your sleep?
The claim is false, and a real spider curator's own page says so. But the tidy origin story nearly every debunking page repeats — a named writer, a named 1993 magazine — can't be found by anyone who's looked, including the fact-checking site that first published it and later admitted the name was an anagram for "this is a big troll."
Do goldfish really grow to the size of their tank?
Pet-care pages say goldfish secrete a growth-stunting hormone into the water. One detailed hobbyist source flatly denies that specific detail and names five studies instead — and the two we could independently check are lab experiments that never test tank size, crowding, or water transmission at all.
Was there really a room called a vomitorium for vomiting?
The one ancient text that uses the word at all means a theater exit, not a dining room — settled. But the academic article that names the earliest known date for the modern mix-up also states, twice, the wrong publication year for its own key example, a six-year error no other source checked catches.
Are daddy long-legs the most venomous spiders in the world?
Two of the three animals called "daddy long-legs" have no venom at all, settling most of the myth outright. The one that does have venom can also bite through skin — the opposite of the "fangs too short" claim — and the university whose own faculty supplied venom for the 2019 study proving it can't agree with its own paper about which animal the famous TV test used.
Did ninjas really wear black?
Japan's own National Diet Library backs the standard "kabuki invented it" story — then names the one 17th-century manual behind its account, and that manual's own list of ordinary ninja colors includes black. A separate critic disputes the theatrical-origin story too, on different, checkable grounds.
Did Einstein really fail math in school?
Two separate schools' own grade records — one an actual surviving 1896 certificate, the other a 1929 fact-check of a since-destroyed one — both put Einstein at or near the top of his class in math. But the earliest dated sighting of the myth itself, a 1935 newspaper clipping, is a point where the rumor was already circulating, not where it began, and nobody we found names who started it.
Was the spinach-iron myth really caused by a misplaced decimal point?
The satisfying 'twist' behind the Popeye-spinach myth — a scientist misplaced a decimal point — has no documented source either. The doctor who made the decimal-point story famous later admitted, when asked directly, that he couldn't remember where he'd read it.
Are "rat kings" — rats fused together by their tails — actually real?
Unlike most of our briefs, this one confirms the strange claim: real specimens sit in named museums, and the same scientist who wrote the one peer-reviewed count personally examined a live case in 2021. But when asked how rare that case was, he cited Wikipedia rather than his own research.