Map showing the path of the eclipse of May 28, 585 B.C. Note that it ends near the supposed site of a battle between the Medes and Lycians, near the Halys River in northern Turkey. Credit: Image by Fred Espenak and Jean Meeus, NASA GSFC

“… just as the battle was growing warm, day was on a sudden changed into night. This event had been foretold by Thales, the Milesian, who forewarned the Ionians of it, fixing for it the very year in which it actually took place. The Medes and Lydians, when they observed the change, ceased fighting, and were alike anxious to have terms of peace agreed on.” 
— Herodotus, writing about 130 years after the battle in 450 BC

Science began on May 28, 585 B.C, according to the late polymath Isaac Asimov. That’s the date of the Battle of the Eclipse, when a solar eclipse so scared the Medes and Lycians on the sixth year of their war that they made peace. The story is that the eclipse was predicted by Thales of Miletus (c. 622-546 B.C), known to us as the first “natural philosopher” who used empirical data — science! — rather than ancient myths to understand the world.

For at least 400 years, astronomers have understood the mechanics of eclipses based on the moon’s orbit around Earth and Earth’s orbit around the sun. So when Herodotus, so-called “Father of History,” writes that a battle in present-day northern Turkey was interrupted by a solar eclipse, and since we know the approximate year the battle took place (from lists of kings), we can calculate exactly when and about where it occurred: late in the afternoon of May 28, 585 B.C, on the Halys River. But Herodotus said that Thales only got the year right, not the date. Which calls his account (along with many of Herodotus’ other tall tales) into question. Thales lived a couple generations before ancient Greek philosophers figured out that solar eclipses happen when the moon comes between the sun and Earth. So the only way he could have made the prediction would have been from Babylonian records showing that eclipses occurred on a regular basis and thus could be predicted.

And there’s the problem. We have baked clay cuneiform tablets from ancient Mesopotamian cities, including Babylon itself, showing that they had figured out the so-called “Saros” cycle” of approximately 6,585.3 days, from hundreds of years of observations of lunar and solar eclipses. Solar eclipses from the same Saros cycle repeat in the same geographic region every 54 years and 34 days, that is, every three Saros cycles. Assuming this is how Thales made his prediction, he would have known the actual day on which an eclipse would occur. Surely Herodotus, while giving credit to Thales, would have mentioned that?

Another problem is that the eclipse of 585 B.C. happened near sunset, an unlikely time of day for a battle, based on ancient records. Was Thales prediction dumb luck? Or did Herodotus, writing over a century later, confuse fact and fiction? (On other occasions, he attributes the outcome of wars to the will of the gods.) My guess is that he heard a vague account of a prediction by Thales and, good storyteller that he was, co-opted the Battle of the Eclipse to make a tale worthy of Dr. Asimov’s bold assertion. In the event, he gives the story a happy-ever-after ending, with the Lydian king’s daughter Aryenis marrying Astyages, son of the Medean king.

Back to the Saros cycle, which, in addition to being about 6,585.3 days long (18 years plus 11.3 days), is exactly 223 synodic months, a synodic (or lunar) month being the time between one full moon and the next. This was known to the designer of the Antikythera Mechanism, the fabulous “computer” dated to about 100 B.C. found by sponge divers in 2001 (“The Antikythera Mechanism,” April 22, 2010). One of the few easily discernable inscriptions include the Greek letters ΣΚΓ′, where Σ = 200, Κ = 20 and Γ = 3, i.e. 223. (The prime symbol ′ indicates that this is to be read as numbers, not letters.) So the mechanism used the Saros cycle to predict solar and lunar eclipses, representing it on an engraved spiral dial on its back panel together with a large gear wheel having (you’ve guessed it) exactly 223 teeth. l

Barry Evans (he/him, barryevans9@yahoo.com) has a lot of chutzpah challenging Isaac Asimov, author of more than 500 books.

Leave a comment

Your email address will not be published. Required fields are marked *