In 1900, divers discovered the wreck of an ancient cargo ship in the waters off the Greek island of Antikythera. Amongst the artefacts subsequently recovered from the wreck were the fragmented and corroded remains of what transpired to be an ancient astronomical computing device, dubbed the Antikythera Mechanism. Constructed of numerous bronze gears, encased in a wooden frame, the mechanism functioned as an analog computer, used to predict the positions of the sun, the moon, and even possibly the planets. Dated to somewhere between 100BC and 150BC, the sophistication of the mechanism, and, in particular, its use of differential gears, is such that nothing of comparable complexity can be found for at least the next 1000 years. (See http://www.ams.org/featurecolumn/archive/kyth2.html and linked pages for some illustrations, animations, and explanations).
In the November 30th issue of Nature, new research on the Antikythera mechanism was published by an international team led by Mike Edmunds and Tony Freeth (Cardiff University, Wales). This research employed X-ray tomography to reconstruct in greater detail the structure of the device, and the inscriptions upon it.
This latest research, suggests philosopher Nicholas Rescher today (http://philsci-archive.pitt.edu/archive/00003078/01/Anaxmander_and_Antikythera_Mech.doc), indicates that the Antikythera mechanism was based upon Anaximander's pre-Aristotelian model of the universe. Anaximander postulated that the stars, the moon, and the sun, were circles of fire, hollow like chariot wheels, and visible down tube-like passages which resemble the spokes of a chariot wheel. This contrasts, of course, with the Aristotelian model of the universe as a set of rotating crystal spheres. Rescher points out that "the one major piece of the Antikythera-mechanism that has survived intact is just exactly the chariot wheel that lies at the core of Anaximander’s cosmology." (See the photo at http://skytonight.com/news/4776976.html).
The Antikythera mechanism was capable of predicting the positions of the sun and the moon, and the occurrence of lunar and solar eclipses, with high levels of accuracy. Despite its function as an empirically successful analog computer, the assumptions it represented were no closer to the truth than the assumptions represented in an astronomical computing device based upon the Ptolemaic model of the solar system. It is a trite, but salutary reminder to those physicists and scientists who claim that the precision of the predictions made by modern physical theories are evidence of their truth.