Trojan Bull

Over the past 24 hours, news organisations around the world have been reporting (with varying degrees of excitement) a claim that astronomers have proved the historical basis of Homer. According to two researchers, who have written up their findings in the US-based Proceedings of the National Academy of Science, a passage in the Odyssey can be shown to refer to a total eclipse on April 16, 1178 B.C. Since the passage in question describes events ten years after the Trojan War, the fall of Troy in turn can now be dated precisely to 1188 BC - a date that ties in well with the findings of archaeology.

The End of an Odyssey: Homer's epic is finally pinned down, declared the Independent, while the Times's headline was almost equally clear-cut: Scientist settles debate on Solar eclipse in Homer's Odyssey. The New Scientist, tellingly, was more circumspect: Is an eclipse mentioned in ancient Odyssey poem? it asked. And the Guardian's take was positively non-committal. Celestial clues may end ancient debate about eclipse in Odyssey, it announced. Or indeed they may not.

Ian Sample's report in the Guardian, though, starts off on a distinctly odd note. He begins offering up the ominous-sounding quote from the Odyssey itself, in which the seer Theoclymenus describes how Penelope's suitors will meet their end: The sun has been obliterated from the sky, and an unlucky darkness invades the world. Sample then claims that

the line set the stage for an argument that after hundreds of years may finally have been settled: did the bloody massacre at the hands of Odysseus and his son take place during a real eclipse?

A more basic question, surely, is "Did the massacre take place at all?" It is, after all, just a poem. One might as well wonder whether Theseus really slaughtered a bull-headed man-beast in the Cretan labyrinth, or whether Samson actually killed ten thousand Philistines armed only with the jawbone of an ass.

Still, at least Sample credits his readers with some familiarity with the dramatis personae of the Odyssey, and the rudiments of the plot. (In the ten years since the fall of Troy, Odysseus has been missing, presumed dead, while back home in Ithaca his wife Penelope has been fighting off the attentions of various suitors. I'm sure you knew that.) Over in the Telegraph, science editor Roger Highfield seems to think the findings somehow prove the existence of the Trojan Horse:

the exact date when the Greeks used the Trojan horse to raze the city of Troy has been pinpointed for the first time using an eclipse mentioned in the stories of Homer, it was claimed today. Scientists have calculated that the horse was used in 1188 BC, ten years before Homer in his Odyssey describes the return of a warrior to his wife on the day the "sun is blotted out of the sky".

The original news agency reports made no mention of the horse.

Once you get past the hype, does this story actually amount to anything? Let's start with the researchers: Marcelo Magnasco, head of the Laboratory of Mathematical Physics at Rockefeller University, New York, and Constantino Baikouzis of the Proyecto Observatorio at the Observatorio Astronómico in La Plata, Argentina. Impressive qualifications, then. What they seem to have been engaged upon is trying to stand up an old hypothesis. It was noted in the 1920s that there had been an eclipse on April 16, 1178 BC which would have been visible in Greece. Since that was fairly close to traditional dates for the Trojan War, and the passage in the Odyssey appears to refer to an eclipse (though it may of course be entirely metaphorical) the link was made. Magnasco and Baikouzis's contribution is to track down four other apparent astronomical references in the poem which, taken together, seem to point to the same date.

According to Magnasco,

Not only is this corroborative evidence that this date might be something important, but if we take it as a given that the death of the suitors happened on this particular eclipse date, then everything else described in The Odyssey happens exactly as is described.

Everything else? That would presumably include that part where Circe turns Odysseus' shipmates into swine. Or the part where the hero escapes from the one-eyed cannibal giant Polyphemus by hiding underneath a sheep. Or indeed the part where he spends several years holed up on an island with only a supernaturally beautiful goddess for company. Then there were all those long, hand-wringing, heart-melting conversations he had with the goddess Athena...

The worst thing is that Magnasco is well aware that his claims are a bit far-fetched, or at the very least tentative. He is quoted in the Telegraph as saying, apropos the Trojan War,

Under the very large assumption that there was an Odysseus, there were suitors that got massacred, that it indeed took 10 years for Odysseus to get back ... yes, in that case the fall of Troy would have happened 10 years before the death of the suitors, thus in 1188BC. The current dating of the destruction layer of Troy VIIa is around 1190 plus/minus a few years.

If everything is true, then it's all true. Otherwise, though, it's probably bollocks. The Magnasco/Baikouzis evidence is certainly open to severe doubt. Of the four astronomical phenomena they rely on, three are fairly uncontroversial, being references to the phase of Venus, the visibility of the constellations Pleides, Bootes and Ursa Major, and the occurrence of a new-moon (a pre-requisite for an eclipse). Unfortunately, the most indicative "clue", in terms of tying down the precise date, is also the one most open to doubt. This is a description of the god Hermes moving from east to west, which they take as a coded reference to the retrograde movement of the planet Mercury. This interpretation, however, makes no sense whatever. I checked the passage, and it clearly refers (as do all Hermes' appearances in the poem) to the god's role as the messenger of Zeus. He is merely fulfilling a function of the plot, visiting the island where the divinely beautiful Calypso is holding Odysseus as a kind of sex-slave, and ordering her to let him return home.

It's a poetic device, not a piece of astronomy. And with that, the whole theory collapses: as Magnasco has admitted, forget Mercury and there are more than a dozen other dates in the 135-year interval that the researchers studied that might fit equally well. "This is a risky step in our analysis," he is quoted as saying. Indeed. If Homer had intended to describe the movement of the planet mercury, he could have done so. Otherwise, you would have to assume that every single time he mentions Hermes he was encoding an astronomical secret; in fact, that the whole poem was coded astronomy. It plainly isn't.

So is there any mileage in this story? Well, the dates do fit: if Homer did describe an eclipse, then there was an eclipse at about the right time (assuming the Trojan war took place then), and a memory of the eclipse might have entered the cycles of legends concerning the heroes who fought at Troy. But that is as far as it is possible to go. The trouble with the latest theory is that it requires a great deal more: precise information about a particular eclipse being recorded and transmitted over a period of four or five hundred years, a time when writing was unknown in Greece. And even that doesn't address how a specific eclipse, and the astronomical circumstances in the month leading up to it, came to be associated with a fictional journey of an almost certainly legendary hero from an imaginary country to a fairytale kingdom, where he metes out poetic justice to some made-up suitors.

The story of the Odyssey is not true. Whatever else they may be, The Iliad and the Odyssey are first of all stories: narrative poems whose content is largely or perhaps entirely fictional. Had Homer shared the conventions of the modern publishing industry, the first edition of the former poem might have had the tagline "inspired by real events"; but the Odyssey would certainly have carried the disclaimer that "any resemblance to actual persons is purely coincidental." Odysseus is a mythological hero with less historical justification than King Arthur, though he might be comparable with Beowulf. It's pointless to try to pin down the date of an event which never occurred.

Still, if you can get "scientists", "prove" and "ancient legend" into the same headline you've got it made.