# An 18th-century machine could play chess on its own, beating Napoleon and Benjamin Franklin

**Verdict: Contradicted.** The Turk was never a thinking machine. Built by Wolfgang von Kempelen in 1770 and later run by the showman Johann Maelzel, it was a cabinet illusion: a strong human chess player hid inside on a sliding seat and moved the wooden figure's arm through a lever mechanism, tracking the game by magnets under the board. Skeptics deduced this from the 1780s onward, Edgar Allan Poe argued it in 1836, eyewitnesses saw operators climb out, and the full method was published after fire destroyed the machine in 1854. Rated debunked.

Category: Hoaxes & Pop Culture · Era: Enlightenment · First circulated: 1770, when Wolfgang von Kempelen unveiled the automaton at the Habsburg court of Empress Maria Theresa in Vienna · Believed by: Paying audiences across Europe and America for over 80 years, and some who watched it play; from early on, though, many observers suspected a concealed human, and no serious authority holds today that it was a genuine automaton
URL: https://theconspiratory.com/theory/mechanical-turk

## Summary
In 1770 the Hungarian official and inventor Wolfgang von Kempelen presented the court of Empress Maria Theresa with an astonishing device: a life-sized figure in Ottoman dress, seated at a cabinet, that appeared to play, and usually win, a game of chess against any human challenger. The Turk toured Europe and later the United States for more than eight decades, beating Benjamin Franklin in Paris and, by legend, Napoleon Bonaparte in Vienna. It was not a machine that could think. Hidden inside the cabinet, behind a clever arrangement of dummy gears and sliding panels, sat a living chess master who followed the game through magnets beneath the board and worked the figure's arm with a pantograph lever. Doubters suspected as much from the start, Edgar Allan Poe laid out the case in an 1836 essay, and after a Philadelphia fire destroyed the Turk in 1854 the operators' own account was published in full. This case file lays out how the illusion worked, how it was exposed, and why a fake automaton became a founding parable of artificial intelligence. The verdict is debunked.

## The claim
That the Automaton Chess Player, or Mechanical Turk, built by Wolfgang von Kempelen in 1770, was a genuine self-operating machine: a clockwork device that could perceive a chessboard, calculate strong moves, and defeat human opponents on its own, without a person guiding its play.

## Origin and timeline
- 1770: Wolfgang von Kempelen, an official in the service of the Habsburg court, unveils the Automaton Chess Player before Empress Maria Theresa in Vienna. He had reportedly promised, after watching a conjuror's act, to build a more impressive illusion of his own. The figure is a carved man in Turkish robes seated behind a cabinet with a chessboard on top.
- 1770s: Kempelen opens the cabinet before each match to show a dense array of gears and machinery, then invites a challenger. The Turk moves its own arm to play, nods at check, and usually wins. Kempelen, who seems to regard the device as a distraction from his serious engineering, is said to be reluctant to exhibit it.
- 1783–1784: Under pressure from Emperor Joseph II, Kempelen tours the Turk through France and Britain. In Paris it plays at the Cafe de la Regence and at Versailles; among its opponents is Benjamin Franklin, then the American minister to France. Crowds are captivated, and pamphleteers begin arguing over whether a human must be concealed inside.
- 1789: Joseph Friedrich zu Racknitz publishes a book proposing, with detailed engravings, that a person of ordinary size could hide in the cabinet and operate the figure. His reconstruction is not exact, but the central claim, that a human sits inside, is essentially correct decades before it is confirmed.
- 1804–1805: Kempelen dies in 1804. His son sells the automaton to Johann Nepomuk Maelzel, a Bavarian showman and inventor (later associated with the metronome), who restores it, improves the act, and turns it into a full-time touring exhibition across Europe.
- 1809: During the Wagram campaign Napoleon Bonaparte plays the Turk at Schonbrunn Palace near Vienna and loses. In the best-known version of the story the emperor deliberately attempts illegal moves to test the machine, and the Turk responds by sweeping the pieces from the board, delighting onlookers.
- 1826–1837: Maelzel brings the Turk to the United States, exhibiting it in New York, Boston, Philadelphia, and beyond. His concealed operators in this period include the strong European player William Schlumberger. American writers, including Edgar Allan Poe, study the exhibition closely and publish their own attempts to explain it.
- 1836: Poe publishes the essay 'Maelzel's Chess-Player' in the Southern Literary Messenger, arguing from the machine's behavior that it cannot be a true automaton and must be directed by a hidden human mind. His reasoning is imperfect in its mechanical detail but correct in its conclusion.
- 5 July 1854: By now owned in Philadelphia and no longer touring, the Turk is destroyed in a fire at the National Theater (the former Chinese Museum). Soon after, those who had known the secret, including the physician John Kearsley Mitchell and, in 1857, his son Silas Weir Mitchell, publish a full account of how the illusion actually worked.

## The evidence, claim by claim
- Claim: The cabinet was opened before each game to reveal a mass of clockwork, so spectators could see there was no room for a person inside.
  Evidence: The reveal was part of the trick, not a refutation of it. The operator sat on a sliding seat and shifted position as different doors and drawers were opened, so that machinery was always shown where an observer expected to look and the hidden player was always in a section that was closed. Racknitz worked out the essentials of this in 1789, and the operators later confirmed it: the visible gears were largely a decorative screen, and the interior offered ample space for a seated adult.
- Claim: The figure genuinely calculated its moves, since it played strong chess and beat skilled opponents like Franklin and Napoleon.
  Evidence: The Turk played well because a strong human was playing. Maelzel and Kempelen recruited accomplished chess masters to sit inside, among them Johann Allgaier (credited with the game against Napoleon) and William Schlumberger in the American years. The quality of play tracked the skill of whoever was hidden that day, and the machine could and did lose. A device that calculated on its own would not need a rotating cast of expert operators, nor would its strength rise and fall with which human was available.
- Claim: No hidden person could have followed the game or moved the pieces so precisely from inside a sealed box.
  Evidence: The mechanism for exactly this was reconstructed from the operators' accounts. Each chess piece contained a small magnet, and beneath the board hung a corresponding set of magnets that told the concealed player, on a small internal board, where every piece stood. The operator read the position by candlelight, chose a move, and worked a pantograph lever linked to the figure's arm, which reproduced his hand movements above the table. Ventilation and a smoke-masking candle handled the practical problems of sitting inside.
- Claim: If it were a fraud, someone would have caught it; instead it fooled experts for more than eighty years, which suggests there was really no one inside.
  Evidence: It did not fool careful observers for eighty years so much as entertain audiences who could not prove what they suspected. Printed exposures appeared within two decades (Racknitz in 1789), a Baltimore account in the 1820s described spectators glimpsing a man inside, Robert Willis published a skeptical analysis in 1821, and Poe argued the case in 1836. The secret held commercially not because it was undetectable but because Maelzel controlled the viewing conditions and never let a challenger inspect the closed compartments during play.
- Claim: After the machine burned in 1854 the truth was lost, so we cannot really say it was a hoax rather than a lost invention.
  Evidence: The opposite happened: the fire released the secret rather than burying it. With the object gone and no more money to be made from it, people who had been inside or in on the trick spoke openly. John Kearsley Mitchell, who had owned the Turk, and his son Silas Weir Mitchell published detailed descriptions of the sliding seat, the magnet indicator, and the pantograph arm. Their account matches the earlier deductions of Racknitz and Poe, leaving no genuine mystery about how the Turk did what it did.
- Claim: Even granting a human sometimes helped, the underlying device might still have contained real chess-playing automation for its era.
  Evidence: There is no evidence of any move-selecting mechanism, and the era's technology could not have built one. Working chess play requires evaluating positions and choosing among legal moves, something no gear train of the 1770s could do; the first machine to play even a restricted endgame came only in 1912 (Torres Quevedo's device), and general chess computing is a 20th-century achievement. The Turk's cabinet held linkages to move an arm on command, not logic to decide what move to make. Every documented function points to a human decision-maker inside.

## Why people believe it
- The Enlightenment was infatuated with automata. Craftsmen like Jacques de Vaucanson built mechanical ducks and flute players that seemed almost alive, so a chess-playing machine did not sound absurd to audiences who had watched clockwork perform other lifelike feats. The Turk arrived exactly when people were primed to believe machinery might do anything.
- The presentation was engineered to disarm suspicion. Kempelen and later Maelzel opened doors, held a candle to the interior, and displayed whirring gears before every game, staging a demonstration of transparency that was itself the core of the deception. Being shown the inside felt like proof, which is precisely what the trick required.
- Chess carried an aura of pure intellect. Because the game was seen as a contest of reason, a device that played it well seemed to be thinking, and observers extended to the whole cabinet the intelligence that actually sat, breathing, in its lower half.
- Endorsement by famous opponents lent it authority. Stories that the Turk had beaten Napoleon and played Franklin traveled far ahead of the machine and made doubt feel like ignorance. If emperors and philosophers had faced it, surely it was more than a puppet, the reasoning went, when in fact their participation only advertised the show.
- Showmanship controlled what anyone could test. Maelzel never let a skeptic dismantle the cabinet during play or sit a suspected operator down for inspection, so suspicion could not harden into proof. In the absence of access, the safest public posture was wonder, and wonder sold tickets for decades.

## Open questions
- The identities and full roster of the concealed operators are only partly documented. Names like Johann Allgaier, Boncourt, Aaron Alexandre, Jacques Mouret, and William Schlumberger recur in the sources, but which player sat inside for which famous game, including the encounter with Napoleon, rests on later testimony and is not always firmly established.
- Some celebrated episodes are hard to verify in contemporary records. The vivid detail of Napoleon attempting illegal moves and the Turk sweeping the board, for instance, comes down through retellings of varying reliability. None of this touches the settled core, that no automation played the games; it concerns only the color of particular anecdotes.
- Kempelen's exact original mechanism is not fully recoverable, since Maelzel rebuilt and modified the interior and the machine later burned. The general method (sliding seat, magnetic indicator, pantograph arm) is well attested, but the precise 1770 configuration is reconstructed rather than directly documented.

## Sources
- The Mechanical Turk: AI Marvel or Parlor Trick?, Encyclopaedia Britannica: https://www.britannica.com/story/the-mechanical-turk-ai-marvel-or-parlor-trick
- Debunking the Mechanical Turk Helped Set Edgar Allan Poe on the Path to Mystery Writing, Smithsonian Magazine (2015): https://www.smithsonianmag.com/smart-news/debunking-mechanical-turk-helped-set-edgar-allan-poe-path-mystery-writing-180964059/
- How a Phony 18th-Century Chess Robot Fooled the World, HISTORY (A&E Television Networks): https://www.history.com/articles/how-a-phony-18th-century-chess-robot-fooled-the-world
- Untold History of AI: When Charles Babbage Played Chess With the Original Mechanical Turk, IEEE Spectrum (2019): https://spectrum.ieee.org/untold-history-of-ai-charles-babbage-and-the-turk
- The Turk: The Life and Times of the Famous Eighteenth-Century Chess-Playing Machine, Tom Standage / Walker & Company (2002)
- Mechanical Turk, Wikipedia: https://en.wikipedia.org/wiki/Mechanical_Turk
- Amazon Mechanical Turk, Wikipedia: https://en.wikipedia.org/wiki/Amazon_Mechanical_Turk
- The Mechanical Turk: The Automaton Chess Player, Chess.com: https://www.chess.com/blog/iiAscari/the-mechanical-turk-the-automaton-chess-player

Rated by The Conspiratory, a neutral, sourced encyclopedia of conspiracy theories. Full page: https://theconspiratory.com/theory/mechanical-turk