How Thales Was Able to Predict the Solar Eclipse of 28 May 585 B.C.

Abstract

In the year 467 B.C. at Aegospotamoi, a stone fell from the heaven. It is said that Anaxagoras, thanks to his knowledge of astronomy, was able to predict the fall of this famous meteorite (DK 59A1(10), DK 59A6, DK 59A10, and DK 59A11; also in Ammianus Marcellinus’ History of Rome 22.8.5, not in DK, but see remark on p. 9, line 12). Nobody, however, has ever been concerned to find out the method he could have used to predict this event. Taking for granted that it is impossible to predict the fall of a meteorite, the attribution of that prediction to Anaxagoras has been dismissed as an instance of the habit of crediting the Presocratics with all kinds of discoveries. In his Life of Dion, Plutarch reports that during Plato’s visit to Sicily, Helicon of Cyzicus predicted a solar eclipse. When his prediction came true, he was admired and rewarded by the tyrant (Life of Dion 19, see also Life of Lysander 12). Nobody has ever tried to determine the method Helicon might have used to accomplish such a remarkable achievement. Many scholars have, by contrast, tried to reconstruct the method that Thales might have used in predicting a solar eclipse, with which he is credited by ancient authors.1 The story is told by Herodotus: in the 6th year of the war between the Lydian king Alyattes and the Medes under Cyaxares, during a battle, all of a sudden the day became night (Histories I 74 = DK 11A5). The reports of other authors depend on his account (see DK 11A1, DK 11A2, and DK 11A5). The eclipse of 28 May 585 B.C., which was almost total at Miletus, is generally accepted as the one predicted by Thales. This date also matches best Pliny’s testimony that the eclipse took place in the 170th year after the foundation of Rome, 753 B.C. (Naturalis historia II, 53 = DK 11A5; see also Stephenson and Fatoohi 1997: 280).