[http://www-groups.dcs.st-and.ac.uk/]

Thales, considered by Aristotle to have been the founder of the Milesian School, was the only presocratic thinker to be acclaimed a sage. The claim that he predicted the 585 eclipse has therefore exercised the minds of philosophers attempting to unravel the origins and progress of philosophical and scientific thought and method, as well as arousing the interest of historians of astronomy and ancient historians in general. Opinion is divided between those who think that Thales did make a prediction and those who think the whole tale was the result of later association of a great name with a legendary event.

Several authorities have put forward forceful arguments that Thales would not have been capable of predicting this or any other eclipse with such temporal or geographic accuracy. The early idea that Thales somehow had access to Babylonian astronomical data and science that enabled him to predict solar eclipses has failed to stand up because, it is now generally accepted, the Babylonians themselves did not have sufficient knowledge to predict solar eclipses, making further discussion unnecessary (O'Connor and Robertson 1999). Neugebauer wrote despairingly:

"Concerning the prediction of a solar eclipse in -584 (May 28) by Thales a few remarks may be made here though I have no doubt that they will remain without effect.

In the early days of classical studies one did not assume that in the sixth century B.C. a Greek philosopher had at his disposal the astronomical and mathematical tools necessary to predict a solar eclipse. But then one could invoke the Astronomy of the "Chaldeans" from whom Thales could have received whatever information was required. This hazy but convenient theory collapsed in view of the present knowledge about the chronology of Babylonian astronomy in general and the lunar theory in particular. It is now evident that even three centuries after Thales no solar eclipse could be predicted to be visible in Asia Minor - in fact not even for Babylon." (Neugebauer 1975, 604)

Herodotus, however, brings the prediction into his account and there is considerable body of other ancient tradition, which strongly suggests that Thales was indeed believed to have made a prediction that turned out to be correct (e.g. Diogenes Laertius, derived from Eudemus, and Apollodorus). The strength of this ancient tradition might be taken to infer that Thales' method, whatever its scientific basis or philosophical value, was indeed a method and not just a lucky guess, a stab in the dark. Presumably Thales predicted one particular eclipse, and only to the year 585 B.C. at that, because the tradition would hardly have been so strong had he made numerous eclipse predictions of which only one happened to be correct. Herodotus says that the eclipse was "foretold", i.e. was somehow announced or proclaimed in advance. Panchenko, arguing that the prediction is too well certified to reject as legend and, reasonably, arguing that Thales gained knowledge of Egyptian (rather than Babylonian) astronomy, has advanced an elaborate theory concerning chance intervals between eclipses that enabled Thales to predict the either the eclipse of 582 or that of 581 (Panchenko 1993; Panchenko 1994). Panchenko further suggested that it was for this prediction that Thales was acclaimed Sage in the same year. According to this reconstruction, Thales’ methodology depended on observed eclipses and the date of the 585 eclipse was crucial to the cycle that Thales computed. This cycle was the result of a coincidental and apparent sequence. Panchenko argued for a series of two intervals between consecutive eclipses thus:

17 lunations to Feb 13, 608, 18 lunations to July 30, 607: = 47 lunations.

17 lunations to Dec. 14, 587, 18 lunations to May 28, 585: = 47 lunations.

He also predicted a further eclipse before end of the 4^th year from 585, which was the eclipse that did in fact take place on March 16, 581.

The fact that the cycle has no real basis but was computed from a succession of coincidences does not, of itself, detract from the development of rational thought in the early 6^th century B.C., nor therefore from Thales’ place in the development of philosophy. But what is the relationship between modern ingenuity and historical reality? Panchenko’s elaborate calculations have been demolished by Stephenson and Fatoohi, who have demonstrated that only the 585 eclipse could have had any impact in Asia Minor (Stephenson and Fatoohi 1997).

An earlier attempt at finding a solution to the problem of how Thales made the prediction was made by Hartner, but this too contains special pleading and has found little acceptance despite the skill with which it was constructed (Hartner 1969). Hartner presumes that Thales had discovered the Exeligmos cycle from study of precise records that would have been kept at Miletus. From these putative records, it is assumed, Thales was able to predict an eclipse at Miletus on May 18, 584 BC, an eclipse that was, as it turned out, of insufficient magnitude to have been noticed in Ionia. On the basis of further assumptions concerning irregularities in the calendar Hartner suggests that the 585 eclipse, which would have been a surprise to Thales himself, was fitted into the prediction for the year in which it did in fact take place. To quote the harsh judgement of Mosshammer:

"Hartner’s ingenious approach only serves to demonstrate how utterly fictional the story of Thales’ prediction is" (Mosshammer 1981, 147).

Detractors have also argued that because Thales' disciple Anaximander clearly did not understand the true nature of solar eclipses, Thales' prediction must be considered spurious. This argument appears, however, to be weak since Thales’ prediction may have been correct although based on a false premise.

In summary, the strength of the ancient tradition indicates that Thales did indeed predict an eclipse for the year in which it happened. The basis on which the prediction was made is unknown, and the accuracy appears to have been the result of chance circumstances.