The origins of Chess could probably be traced back to India. But it was the rest of the world that sought to perfect it. While the game is normally played between two living beings, it is a game where a machine could theoretically become the ‘perfect player’.
Today’s chess Artificial Intelligence (AI) programs can beat the world’s greatest players. But the quest to automate chess stretches back to the late eighteenth-century.
In 1809, Napoleon tried and failed to cheat at chess. You can’t blame the Emperor, however, as his curiosity had gotten the better of him. For his opponent was extraordinary for the time.
“The Mechanical Turk” was an automaton, created by the Hungarian engineer Wolfgang von Kempelen. It was a sight to behold; a masterfully crafted Turk (brown-skinned, mustachioed and wearing a Fez) perched before a box roughly four square feet in size, that served as his chess board. His right arm controlled the pieces over the board, and his left held a smoking pipe – elegantly poised as if the Turk couldn’t care less.
The visual spectacle of the whole affair would make you think it to be a joke. How could an ‘automaton’ play chess? So Napoleon might have thought when he tried to make an illegal move.
But the Turk shook its head and moved his piece off the chessboard. When he tried to cheat a third time, it swept the pieces off the board.
Once the Emperor started playing seriously, he was beaten in under 19 minutes. The Turk added Napoleon to a list of vanquished foes that included Benjamin Franklin and many of the prominent chess masters of the time.
It was intended as a spectacle, nothing more. Kempelen first created it as entertainment for the Hapsburg Arch-Dukess, Maria Theresa. It ended up becoming the device he was (in)famous for – It was operated across the world until a fire destroyed it. Throughout this, he kept the Turk’s biggest secret – a man inside the box.
The Turk’s head motions, hands, and even expressions were controlled by a lithe figure who sat within the cramped container below. Much effort was spent in keeping alive the facade. The machine was opened up before every match, moved from place to place, and even swept over with magnets. But an ingenious system of a sliding cushion kept the operator hidden. A magnet-proof magnet allowed him to see the moves being played above.
Kempelen’s device was still a marvel, for the Turk’s success rate was not to be laughed at. It was not unbeatable, but even the world’s best chess player reported his match as “his most fatiguing game of chess ever”.
Kempelen only exhibited the machine until 1804. Thereafter, his son sold it to a musician and tinkerer, Johann Nepomuk Mälzel. Mälzel utilized famous chess players as operators and built up the obscurity around the Turk. His tour of America inspired wonder – coming as it did in the midst of an industrial revolution. Everyone from the Scientific American to Edgar Allan Poe wrote of the Turk, building an early debate on the role of a machine to man.
The Turk may have been a hoax, but it has a powerful legacy in automation. The inventor, Charles Babbage, thought the Turk a hoax but was inspired by its idea. He included chess as a possible function of his never-completed ‘Analytical Engine’. Most notably, Edward Cartwright saw the Mechanical Turk in 1784 – and decided that inventing the machine that could weave was easier than making one that could play chess. He made the power loom soon after. The Turk may have been a fake, but it inspired very real advances in technology.
There were more automatons to come. Ajeeb (1868-1929) and Mephisto (1878-1889) were similar to the Turk in construction, though Mephisto could be operated remotely by a human. The first true automaton was ‘El Ajedrecista‘, debuted at the Paris World Fair by Leonardo Torres. While the machine couldn’t play a complete chess match, it could guarantee victory in an ‘endgame’ – a scenario in chess where there are only three pieces on the board. Nevertheless, it is called the first ‘computer’ game in history.
From this point on, the rise of computing power in the 1940s accompanied the first wave of creating full-fledged, chess-playing machines. In 1950, Alan Turing wrote the first chess program, but it was an algorithm with no machine to play.
By the end of the 1960s, machines were playing – the first famous match being between the computers of the Institute for Theoretical and Experimental Physics in Moscow and Stanford University.
In 1983, ‘Cray Blitz’ became the first computer program to beat a Chess Grand Master. In 1984, it was the World Computer Chess Champion. From that time, chess programs have grown ever-more sophisticated.
With the rise of IBM ‘Deep Blue’, only Grand Masters remain able to beat, even draw, a match with a top-end chess-playing computer. But the resilience of human intelligence keeps the pursuit of an ultimate chess machine alive. Grandmaster Gary Kasparov has played against AI multiple times, writing of his experience battling a computer – and the limitations of both sides.
Beating a human is no longer the ultimate goal in chess-algorithms. The question of “Solving chess” is the new one. For it requires a machine that can calculate every possible move on the entire chess board – the number of which exceeds that of atoms in the observable universe. It’s a task that modern machines cannot deliver on – but one that future quantum computers might be able to.
The frightening thing about the perfect chess-machine is that a human could never beat it. Imagine having these thoughts as you stare into the dark eyes of the Mechanical Turk in the eighteenth century. It is no wonder that Napoleon played a match with the Turk being blindfolded. Some of the implications of automation can make even the mightiest of us feel very small, indeed.
Copyright Madras Courier 2017. All rights reserved. You may share using our article tools. Please don't cut articles from madrascourier.com and redistribute by email, post to the web, mobile phone or social media.Please send in your feed back and comments to firstname.lastname@example.org