Victory messenger may have run in fatal August heat.
Astronomical detective work has revealed that the commonly accepted date of the Battle of Marathon may be wrong.
The 490 BC battle, in which the Athenian army defeated an invading band of Persians at the seaside plain of Marathon, was the first time that the Greeks had bested the Persians on land.
Historians argue that this victory was pivotal in the development of Western civilization because it preceded a flowering of Athenian culture that led to great advances in mathematics, drama, philosophy and astronomy.
"It was certainly an enormous morale boost for the Athenians because it proved they could stand up to the Persian Empire," says John Lazenby, a historian at the University of Newcastle upon Tyne, UK. "If the Persians had won at Marathon, it's very likely they would have conquered Greece."
Legend has it that after the battle ended, a runner was immediately dispatched to deliver the good news to Athens, roughly 26 miles away: the world's first marathon. Upon arrival the heroic messenger exclaimed, "Rejoice, we conquer!" and promptly expired. Most historians, however, think that this part of the tale is a myth.
The generally accepted date of the Battle of Marathon is 12 September 490 BC. This was proposed by the nineteenth century scholar August Boeckh, based on accounts written shortly after the battle by the Greek historian Herodotus.
Herodotus records that the Athenians called on the army of Sparta, based about 240 kilometres away, to help them fight the Persians. But the devout Spartans explained that they could not march to war during their religious festival of Karneia, which was due to end at the next full moon. Boeckh's calculations related this festival to the Athenian calendar to find the date of battle.
"We realised that Boeckh's method of dating, using the Athenian calendar, had a serious flaw," says Donald Olson, an astronomer from Texas State University, San Marcos, who argues his case in September's issue of Sky & Telescope. "The Karneia was a Spartan festival, so the analysis should have been done with a Spartan calendar."
Although both calendars followed the lunar cycle, they were not identical. Whereas the Athenian year began with the first new moon after the summer solstice, the Spartan year began with the first full moon after the autumn equinox. Olson and his colleagues calculated that in the year from 491 to 490 BC, there were ten new moons between the autumn equinox and the summer solstice, one more than usual. This happens occasionally because a solar year is not an exact multiple of a lunar month.
The evidence for the Marathon run is very bad. We don't have the faintest idea where Plutarch got the story from John Lazenby , University of Newcastle, UK
So, for that year, the Spartan calendar was running one lunar month ahead of the Athenian calendar. Olson says that this means the Battle of Marathon actually happened on 12 August 490 BC. He suggests that the summer heat of August may have pushed the runner into a state of heat exhaustion, perhaps explaining his reported collapse.
But Lazenby remains sceptical about this part of Olson's conclusions. "The evidence for the Marathon run is very bad," he says. "There's no record of a messenger dropping dead until the historian Plutarch, who lived over 500 years after the battle. We don't have the faintest idea where Plutarch got the story from."
He points out that the run is the sort of story that Herodotus, writing just after the time of the battle, loved telling. If it had happened he would have been unlikely to miss the chance to record the tale. "I can't believe he would have failed to pick this up from the Athenians at the time," says Lazenby.
University of Newcastle, UK
About this article
Cite this article
Peplow, M. Battle of Marathon date revised. Nature (2004). https://doi.org/10.1038/news040719-1