Modern Magic: A Practical Treatise on the Art of Conjuring

The door and drawer having been closed, the exhibitor wound up the works with a key inserted in a small opening in the side of the chest, placed a cushion beneath the arm of the figure, and then challenged any one of the company present to play a game with it. It was observed that in playing the automaton always selected the white pieces, and had the first move. Owing to a curious mistake of the inventor, the figure moved the men with his left hand. The error, when found out, could not afterward be rectified. Its hands and fingers opened, and then grasped a piece, which it conveyed to the proper square. In taking a piece, the same motion was made by the arm and hand as before; it, however, conveyed the piece off the board, and then placed its own piece upon the vacant square. While and after his opponent made a move, the figure paused for a few moments as though contemplating its own. It intimated with a nod of the head when it gave check to the king. During the time the arm was in motion, a low sound of clockwork running down was distinctly heard. The works were wound up at intervals by the exhibitor, who otherwise did nothing but walk up and down the room. As we find that the automaton both lost and won--in Kempelen’s time it very seldom lost--and that each game was different to the others, it necessarily follows that these phenomena are inconsistent with the sole effects of mechanism. Various conjectures have been offered as to the mode of communication between the figure and the intelligence which directed it. A plausible and probable explanation was given in 1821 in a pamphlet called, “An Attempt to Analyze the Automaton Chess Player.” In this brochure it is shown that in spite of the apparent display of the interior of the chest and the figure, there yet was ample space left unopened for the concealment of a person of ordinary size behind a false back to the narrowest division only. This is shown in the accompanying illustrations.