Sabre-rattling between powerful enemies. Middle eastern conflict creating refugees. Ethnic strife. Violent retaliation by the marginalized. Senseless murder of innocent persons. Tensions between nations over harboring of terrorists. Bombastic leaders bent on conflict. And, no doubt, a population hopeful for divine intervention or some sign from heaven. Enter a total eclipse.

No; this is not some kooky end-of-the-world prediction or new age hope based on the total eclipse of the sun eagerly awaited by many this Monday, 21 August 2017. The above paragraph actually describes a situation in the mid sixth-century BC, in which a total eclipse did bring about peace—or so we are told.

The Event

Our source on this unusual event is Herodotus—the 5th century BC writer sometimes referred to as the “Father of History.” As a background for later events, he relates the following account from the early-6th century BC (paraphrased here).

Certain nomadic Scythians (from central or eastern Anatolia/Turkey) were displaced by local conflict and became refugees in the kingdom of the Medes (in northwestern Iran). The Scythians were well-treated by the great Median king Cyaxeres, who charged them with teaching their language and bow skills to a group of youths. On one occasion, Cyaxeres berated the Scythians for having returned from a hunt empty-handed and thus severely offended them. The latter responded by killing one of the youths and secretly preparing his flesh as a meat dish presented for the king and his guests. As the Medians ate the dish and learned the horrible truth, the Scythians fled for safety to Sardis, the capital of the kingdom of Lydia in west and central Anatolia. The Lydian King Alyattes refused Cyaxeres’ demands that the refugees be extradited, and war broke out between the kingdoms.  After five years of indecisive warfare with some battles won by each, the Lydians and Medians squared off again. After the battle was engaged, the “daylight suddenly turned into night” and both sides suddenly became motivated to end the conflict. Peace was formally concluded under the mediation efforts of the kings of Cilicia and Babylon with Aryenis, daughter of Alyattes wed to Astyages, son of Cyaxeres, to seal the pact (Herodotus 1.73-74). Herodotus adds that these nations have “the same form of oath as the Greeks, but for additional confirmation they make a shallow cut in their arms and lick each other’s blood” (1.74). Truly a lost bit of civil diplomacy, perhaps needed in our modern world!

2014-05-16 13-40-06c
Above: Karnıyarık Tepe, a huge tumulus (burial mound) at Bin Tepe, the necropolis-tumulus field near Sardis (photo © Daniel C Browning Jr); it was originally assumed to be the tomb of Gyges, but then suggested to belong to his son Alyattes, the Lydian king in our story; however, a larger tumulus nearby (for which I don’t have a photo handy) is claimed to be the tomb of Alyattes by the Sardis Expedition (see “Bin Tepe, The Tumulus of Alyattes, and Karnıyarık Tepe,”

Herodotus surely implies a solar eclipse as the means by which “daylight suddenly turned into night.” He further comments that the philosopher Thales of Miletus had predicted the event (1.74). Scholars have long observed that the only significant solar eclipse that appeared near potential battle sites in the possible chronological window occurred on 28 May 585 BC. Thus, the battle probably occurred on that date and, presuming the reliability of Herodotus, a solar eclipse led to the cessation of hostilities and a lasting peace between bellicose nations. We should be so lucky today.

Of course, the ancients had the wonder and “advantage” in seeing their eclipse as some celestial omen, whereas we know in advance the exact time, place, path, and duration of the event Monday, 21 August 2017. Consequently, we also know the cause and do not (except for fringe elements) connect it with divine displeasure or warning. Nevertheless, I will note that the precise predictability of modern eclipses attests to the accuracy and reliability of scientific measurement. And that fact strongly suggests that we should heed data based warnings on things like climate change and environmental damage caused by human activity.

Extra Stuff for the Nerdily Interested, part 1: Eclipse Predictability and the Role of Thales  

But, it is reasonable to question whether Thales’ prediction of the event might/should have lessened its impact on the warring nations of 585 BC. As it happens, however, there is serious doubt by scholarship that Thales made such a prediction.

To be (somewhat) brief, it is fairly certain that Thales could not have been aware of the Saros Cycle, a period of 223 lunar months that was known to late Chaldean/Neo-Babylonian astronomers to predict lunar eclipses—and also relevant for solar eclipses. But, while lunar eclipses are visible throughout the night sky on occurrence, solar eclipses are location specific. Thus, solar eclipses one saros apart occur some 8 hours later and the visible path is therefore 120 degrees to the west—i.e., one third of the way around the globe! So, even if Thales knew the Saros Cycle he could not have known the appropriate previous solar eclipses or their location for calculation. Much more useful for predicting local solar eclipses is the Exeligmos Cycle of 54 years, 33 days, or 3 saroi, in length. The three saroi (each saros moving the center of the eclipse 120 degrees west) in a exeligmos bring the center of an eclipse all the way around the earth to the same approximate longitude. We now have evidence that the Greeks learned the Exeligmos Cycle by the beginning of the first century BC; but, alas, almost 500 years after Thales. This evidence comes from the long-enigmatic Antikythera mechanism. This device, a “computer” discovered in ancient shipwreck debris in 1902 and dated to the beginning of the first century BC, is now digitally reconstructed using high-tech scans of the oxidized remains. At least 30 connected bronze gears within the device rotated in clock-like movement to calculate celestial and calendrical phenomena, including lunar and solar eclipses (it deserves a post of its own, but Wikipedia has a good and well-cited article on it).

Above: the Antikythera Mechanism, main fragment, “front” (left) and “back” (right); in a special exhibit in the National Archaeological Museum, Athens in 2013 (photo © Daniel C Browning Jr)

Returning to Thales; there is no evidence that he or anyone else in the sixth century BC had the knowledge of the above cycles to accurately predict an eclipse. This has created a long and vigorous scholarly debate.1 “Scholarly debates,” it should be noted, often last for years [or, saroi or exeligmoi!] as painstaking research is published in various articles. This case reminds me of many such debates on biblical issues, with deconstructionists refuting all elements of the story and believers searching for ways to preserve it.

In one parallel to biblical debates, there is some question as to the exact text and meaning of Herodotus’ report. The easiest reading asserts that Thales predicted only the year in which the eclipse occurred, but not the specific day or even month. It is something of an all or nothing issue: knowledge of the Saros or Exeligmos cycles would make precision prediction as easy as the year—given accurate earlier observational data. The assumption that Thales could have been privy to theorized Babylonian records was convincingly refuted. But there have been other attempts to preserve faith in Thales. One such scheme theorizes that local records of observed eclipses were kept at Miletus (Thales’ home) and that record by coincidence suggested eclipses occurred in sets of three with a consistent pattern of lunations between the three. This local pattern, while untrue as a rule, led Thales to predict the 28 March 585 BC as a coincidence.2 An older variation of this theory, using slightly different presumed observations, holds that Thales was led to predict the eclipse of 18 May 584 BC, but in more general terms. The 28 May 585 BC, it is supposed, would have surprised Thales as too early, but impressed others as a correct prediction.3 Whether Thales lucked out by using a fortuitous but false cycle, or stumbled into the acclaim by an earlier-than-expected fulfillment, or didn’t make a prediction at all, he was acclaimed for his astronomical knowledge by contemporary and other pre-Socratic philosophers, according to Diogenes Laertius (Lives of Eminent Philosophers, 1.1).

Pliny the Elder, writing in the first century AD, affirms that Thales predicted a solar eclipse in the 4th year of the 48th Olympiad (Natural History 2.53), which corresponds to 585/4 BC. The convergence of Pliny’s date for the event and the known full solar eclipse across Anatolia at the time convince most scholars that the battle occurred as described on 28 May 585 BC, regardless of their position on Thales’ role.

Extra Stuff for the Nerdily Interested, part 2: Issues with the Battle

Currently a plethora of web pages offer information on the looming solar eclipse. In my opinion, the best and most trustworthy is the one by NASA . Even better for planning and historical research is NASA’s permanent database of eclipse information, developed at Goddard Space Flight Center and available here: I am rather proud when a former employer does something really well, and this is a spectacular resource.

NASA’s online version of Five Millennium Catalog of Solar Eclipses yields Google Map tracks of the totality path and clickable time data for every eclipse from “-1999” to “+3000,” including the 28 May 585 BC event of interest. One difficulty arises out of the time data from the eclipse: it occurred very late in the day, with the end of the eclipse path ending in Mesopotamia at sundown. Herodotus’ account is read by some to indicate that the armies had just engaged when “daylight turned into night.” They thus reject the story on the basis that armies would only engage earlier in the day with plenty of daylight remaining.4 I am not convinced Herodotus intended to limit the timing in this way.

NASA map 28 March -584
Google Maps screen shot from NASA eclipse data of 28 May 585 BC (
[NOTE: as of this writing, NASA informs me that they “had to redirect the site to because  server constraints. Most of its content can be found at the primary author’s personal site,” Presumably, the redirect will be lifted after the eclipse mania is over on 22 August 2017.]

As can be seen from the above map, the 28 May 585 BC area of totality covered a significant portion of Anatolia (Asia Minor; modern Turkey), where the Lydo-Median conflict certainly occurred. Anywhere in the region the eclipse would have been dramatic, but only a significant westward Median penetration into Anatolia—doubted by some scholars—could put the battle in the path of totality. This now seems less a problem with the recent potential identification of the Median outpost Pteria deep in the heart of old Hittite territory in north central Anatolia.5 The Iron Age site—at Kerkenes—is huge, but difficult to access . . . and clearly calling me to visit it! If I manage that, I will post about it here in my intended series on weird, significant, ultra-remote, or forgotten places: “You Don’t Get This On the Bus Tour.”

As for Alyattes versus Cyaxeres, Thales, and the eclipse that stopped a war—and to return to my biblical studies analogy above—I remain a “believer.” In part, I will claim, because the deconstructionist efforts have not convinced me; but in part, I will confess, because it is just more interesting that way. I am intrigued nevertheless, and will continue to study and research, and keep my belief honest. Perhaps to stretch my analogy (and move the bar from “belief” up to “faith”); faith makes life worth living, and thoughtful investigation makes faith worth having.

So (whether this is a logical conclusion of these ramblings or not); I will go to observe the total eclipse Monday with excitement and wonder . . . and I will hope for the best in our seemingly crazy world.


1 See, as a recent and representative overview, Miguel Querejeta, “On the Eclipse of Thales, Cycles and Probabilities,” Culture And Cosmos 15.1 (Spring/Summer 2011): 5–16.

2 Dirk L. Couprie, “How Thales was able to ‘Predict’ a Solar Eclipse without the Help of Alleged Mesopotamian Wisdom,” Early Science and Medicine 9.4 (2004): 321-37.

3 Willy Hartner, “Eclipse Periods and Thales’ Prediction of a Solar Eclipse: Historic Truth and Modern Myth,” Centaurus 14 (1969): 60–71.

4 Alden A. Mosshammer, “Thales’ Eclipse,” Transactions of the American Philological Association 111 (1981): 145-155

5 Christian Marek, In the Land of a Thousand Gods (Princeton: University Press, 2016), 113.

3 thoughts on “Can a Total Eclipse Bring World Peace?

  1. I have just purchased what I first believed to be a Mysian Electrum 1/6 Stater from circa 500BC. It has a Pegasus on the obverse and a sun on the reverse wedged between two quarter-circles with a horizontal bar between them. I could not figure out what it represented at first. Then I started looking at astronomical events. A depiction of an eclipse late in the afternoon, or early evening fits the narrative of the Lydian Battle of Halys during an eclipse perfectly. That would date my coin 85 years earlier than I thought and probably places the mint in Lydia at Sardis.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s