AskHistoryAI — Interactive Timeline of Everything

Did a Solar Eclipse Ever Stop a War?

Yes, according to our oldest source. Herodotus reports that in the sixth year of war between the Lydians and the Medes, day suddenly turned to night during a battle; both armies broke off and agreed terms. Astronomical retrocalculation places that eclipse on 28 May 585 BCE, making the battle one of the earliest events in Greek history fixable to a single day.

What Herodotus actually says

The account is short, and it is worth reading rather than paraphrasing. In Histories 1.74, Herodotus describes a war between Alyattes, king of Lydia, and Cyaxares, king of the Medes, that had run for five years with neither side gaining a decisive advantage. In the sixth year the two armies met again, and during the fighting the day became night. Both sides, he writes, broke off the battle and became eager for peace.

The settlement was brokered by two outside rulers, Syennesis of Cilicia and Labynetus of Babylon, and sealed the way such settlements usually were: a marriage, in this case of Alyattes’ daughter Aryenis to Astyages, the son of Cyaxares. The border between the two kingdoms was fixed at the River Halys. You can read the passage in full, free, at openscriptorium.com/books/herodotus and judge the wording for yourself.

Why the date can be pinned to a single day

Solar eclipses are the most precisely retrodictable events in the ancient record. The geometry of the Sun, Moon and Earth can be run backwards, and each total eclipse leaves a narrow track across the Earth’s surface that can be reconstructed. Only a handful of eclipses in the relevant century were total anywhere near Anatolia, and the one that fits the region and the reign dates is the eclipse of 28 May 585 BCE.

That is why the Battle of the Halys is so often called the earliest event in Greek history that can be dated to an exact day. The claim needs one caveat: it depends on Herodotus having reported a real eclipse at a real battle, and a minority of scholars have argued for a different eclipse, with 610 BCE the usual alternative. The 585 date remains the standard one, and it is the reason the event is known as the Eclipse of Thales at all.

The part historians actually doubt

Herodotus adds a second claim: that Thales of Miletus had foretold the loss of daylight to the Ionians, fixing it within the year in which it occurred. That is the detail specialists treat with real caution, and it is worth separating cleanly from the eclipse itself.

Babylonian astronomers of the period could predict lunar eclipses well, because a lunar eclipse is visible from an entire night-side hemisphere and recurs in a usable pattern. A solar eclipse is different in kind: totality falls along a narrow track, so predicting that one will be visible from a particular place requires a model of the Moon’s parallax that nobody in the sixth century BCE possessed. Historians of ancient astronomy, Otto Neugebauer prominent among them, have argued that Thales could not have made the prediction Herodotus credits him with. The usual reading is that a real eclipse was later attached to the most famous natural philosopher of the age.

The older eclipse that holds up ancient chronology

If you want the eclipse that genuinely does load-bearing work for historians, it is not the Lydian one. An Assyrian administrative list, the eponym canon, records under the year of the official Bur-Sagale that “the sun was eclipsed” in the month of Simanu. That eclipse can be identified with the one of 15 June 763 BCE.

Because the eponym lists are a continuous year-by-year sequence, a single fixed astronomical point converts the whole run into absolute dates. That one line is a substantial part of why Assyrian and, downstream, wider Near Eastern chronology can be stated in calendar years at all. It is a quieter story than an eclipse stopping a battle, and a far more important one.

What most scholars accept, and what stays contested

Frequently asked questions

Which war did the eclipse stop?

A six-year war between the Lydians under Alyattes and the Medes under Cyaxares, fought in Anatolia. Herodotus says the two armies were engaged when day turned to night, that both sides broke off, and that they then agreed peace with the River Halys as the border between the kingdoms.

Did Thales really predict the eclipse of 585 BCE?

Almost certainly not in the way Herodotus suggests. Predicting that a solar eclipse will be visible from a specific place requires understanding the Moon’s parallax, which was beyond sixth-century BCE astronomy. Babylonian methods of the period handled lunar eclipses well but could not do this. Most historians of ancient astronomy treat the prediction as a later attribution.

Is the Battle of the Halys really the earliest event with an exact date?

It is often called the earliest event in Greek history datable to a single day, which is a narrower claim. Older events elsewhere carry exact dates too: the solar eclipse recorded in the Assyrian eponym canon under the official Bur-Sagale is identified with 15 June 763 BCE, and it anchors Assyrian chronology.

Explore related content on this site

Sources and further reading