The Antikythera mechanism

To quote Tony Freeth(1), "If it had not been for two storms 2,000 years apart in the same area of the Mediterranean, the most important technological artifact from the ancient world could have been lost forever."  The knowledge of the existence, in the last centuries BCE, of an advanced technology had been completely lost to history; without this it would never have been recovered.

The bottom line is that this is an Orrery (a mechanical model of the Solar System), the oldest one in existence by about 1,500 years.

My interest in this was reignited when I recently saw a Science TV show mainly concentrating on the determination of the purpose of the gears.

(1) Decoding an Ancient Computer, Tony Freeth, Scientific American, December 2009, p76

======

Astronomy and Eclipses

Astronomical observations were made in many early civilizations, stone age and earlier.  The fundamental reasons were presumably originally just practical - to generate a formal cycle for the planting and harvesting of crops.  However, many years of observations were needed to do this correctly because there is not a whole number of days in a lunar cycle (month), nor days in a annual cycle (year), nor months in a year.  This would have required careful record-keeping and processing over multi-year periods in days before there were formal scripts and permanent records.  Although it's difficult to imagine physically and socially, we know it happened.  For major monuments, think of Stonehenge; for more practical simple artifacts, there is the Nebra Sky Disc (buried c.1600 BCE).

The Babylonians are known to have recorded their information intermittently in the 2nd millennium BCE, and then systematically (https://en.wikipedia.org/wiki/Babylonian_astronomical_diaries) since the latter part of the 8th century BCE.  As a consequence, they made remarkable advances in astronomy, including the determination of what is now known as the Saros cycle.  This is the repetition cycle of eclipses every 223 lunar months.  Even to speculate about a cycle this long, they needed access to data of at least one cycle of 18 years, presumably many more -- to allow for missing observations due to weather and local events and to the fact that some eclipses were not observable in the region -- i.e., probably a century of data or more..

In Greece in the 6th and 5th centuries BCE, two authors report that people still believed that an eclipse was a sign from the gods.  Herodotus reports that a solar eclipse caused a battle between the Medes and Lydians to be halted and a truce agreed.  It is likely that this was on 28 May 585 BCE.  Thucydides reports that the Athenian forces decided to delay their departure from Syracuse as the result of their fear of a lunar eclipse (on 28 August 413 BCE), and were annihilated.

Now forward to the 3rd century.  By this date, apparently, eclipses were believed, by some at least, to be natural phenomena and to be predictable.  Enter Archimedes (c.287-c.212 BCE).  It is believed that he designed and built a mechanism (in fact he may have built two according to Cicero) to model the motions of the sun, moon, and the five known planets, and included other dials to display other events including eclipses.  He apparently documented the workings in a manuscript in the Library of Alexandria.  Unfortunately, the fire at the library during the defence of Alexandria in 48 BCE by Julius Caesar led to the destruction of a large number of manuscripts, so this will never be provable.

However, the knowledge must have survived.  The non-uniform orbital speed of the moon, now called the lunar anomaly, was first determined by Hipparchus (c.190-c.120 BCE) in the 2nd century BCE, and it is included in the existing mechanism. 

======

The Antikythera mechanism (https://en.wikipedia.org/wiki/Antikythera_mechanism#/media/File:NAMA_Machine_d'Anticyth%C3%A8re_1.jpg)

Its discovery and provenance are detailed in two Wikipedia articles.  The wreck and its excavations are described in

  https://en.wikipedia.org/wiki/Antikythera_wreck

The description of the device itself in

  https://en.wikipedia.org/wiki/Antikythera_mechanism

As described in the first reference, the date of the wreck is estimated to be c.70 BCE.  The date of the mechanism could have been as early as 200 BCE, but could be as late as 87 BCE.  Unfortunately, it is therefore likely not one of Archimedes' original devices, but the "scientists who have reconstructed the Antikythera mechanism also agree that it was too sophisticated to have been a unique device."

The gear teeth are triangular.  This may also indicate Archimedes' influence.  The odometer (to measure distances) described by Vitruvius (c.80-c.15 BCE), but likely a design originally by Archimedes, was essential for Romans to measure and map their empire.  However, later reconstructions even by Leonardo da Vinci failed until finally replacing "the square-toothed gear designs of da Vinci made with the triangular, pointed teeth found in the Antikythera mechanism."

The device required the manufacture of gears with a prime number of teeth, since cycles of 53 and 223 were needed.  In addition, it had amazingly sophisticated pin-and-slot wheels to allow for the non-uniform speed of the moon.  Although it wasn't then known, this is due to its elliptical orbit.  This brilliant invention was probably also used in parts of the mechanism that did not survive; for the sun's true position and for the orbits of the outer planets Mars, Jupiter, and Saturn.

As stated, although "the engineering was remarkable for its era, recent research indicates that its design conception exceeded the engineering precision of its manufacture by a wide margin -- with considerable cumulative inaccuracies in the gear trains, which would have cancelled out many of the subtle anomalies built into its design."

However, until this device was found and analyzed, it was not realized that such engineering skill was present in ancient Greece.  The skills had been completely lost, and it was not until the 14th century CE (about 1,500 years in the future) before devices of this sophistication were again constructed.

Gareth Hunt, (updated) 18 April 2020