Monday, 13 July 2020

Stripping myths down to a historical core (part 2)

Part 1 | Part 2

Myths contain stuff that has been transmitted from generation to generation, century to century, to some extent even millennium to millennium. But old doesn’t mean real. It doesn’t even mean inspired by something real.
Left: a cyclops. Centre: the skull of a dwarf elephant, a species that lived on Mediterranean islands up until about the mid-Mesolithic. Right: a euhemerised cyclops designed for A Total War Saga: Troy (forthcoming August 2020). Was the mythical cyclops inspired by the skull’s nasal cavity, as famously argued by Mayor 2000? Well, it’s possible. But the nature of myth is that there’s no expectation that it needs to be based on anything real.
Also, some stuff is better at surviving than other stuff. If you want to understand the potential of myths to preserve historical information, it matters which kind of information you’re talking about.
Note. This would be my main concern about the kind of model of oral history that Echo-Hawk 2000 proposes. Echo-Hawk’s ideas about oral traditions are, not exactly all-or-nothing, but they’re not fine-grained. Human bodies are real; eyes are real; but the fact you can put those two things together doesn’t mean you can say the Cyclops has a historical basis.
For example, Homeric epic shows an outstanding survival rate for place names. This includes at least one town that had been abandoned since the time of the Mycenaean palace culture (Eutresis, in Boeotia: TH Ft 140, Iliad 2.502). Narrative tale-types, poetic devices, linguistic formulae also do impressively well, though very unevenly. Some elements are recent, others are extremely ancient. Some story elements and theological concepts are evidently inherited from Bronze Age Mesopotamia (via the Hittites or the Phoenicians); a handful of linguistic and metrical elements go back to Proto-Indo-European (ca. 4000 BCE [edit: or 3000 BCE: see the response from Timo below]); and one or two tale-types might be even older still — like the story of a hero who escapes a cave by blinding a one-eyed giant, a tale-type found in Europe, Asia, and Africa.

But material culture, etiquette, military tactics, burial practices, legal and political framework, kinship, marriage, inheritance customs ... not so much. With just a couple of exceptions, what we have in these categories looks like contemporary practices, usually dating to the first half of the 600s BCE or a bit earlier, heavily altered by false archaisms to give them an artificial flavour of age. And while some of Homer’s place names are genuinely archaic, his geography and topography are a mess.

How about a historical Trojan War? Who knows. But in terms of popularity, in the Archaic period the Trojan War had a similar standing to the story of Jason and the Golden Fleece. In terms of prevalence in the epic tradition, it was on a similar level to the war between the gods and the Titans. These comparisons aren’t encouraging.

With myths, just because some elements are old, that doesn’t mean any particular aspect of the myth is real. There’s no principled reason to infer, purely from internal evidence, that a myth originated in historical events. If the extant evidence shows a really good survival rate for some kind of stuff, as with Homer’s toponyms and language, it’s reasonable to expect that other toponyms and linguistic forms stand a decent chance of being old, even if we don’t have external corroboration. But if there’s an area where there’s no track record — like, um, almost everything else — then it’s bad methodology to assume it’s a faithful representation of anything at all.
Lara Croft, tomb raider and arch-euhemerist. In her career Lara has visited Atlantis, a Polynesian settlement in Antarctica, Niflheim, and Yamatai, and she has handled the Spear of Destiny, Excalibur, and Mjöllnir. In one game it is ‘revealed’ that the Midgard Serpent of Norse myth was actually the movement of the earth’s tectonic plates.

Part 2. Euhemerism

‘Myths are usually based on some version of the truth,’ Lara Croft says at the start of the 2013 Tomb Raider reboot. Lara’s sentiment is a nice illustration of a deep-running human impulse: the drive to dig through fantastic elements from a myth, to find a kernel of historical reality buried inside.
Note. Actually, in Lara’s specific case, her argument does potentially hold water. It’s her phrasing that’s the problem. She isn’t actually talking about myths, but about Yamatai, a place named in ancient Chinese historical records. And the Records of the Three Kingdoms may or may not be accurate, but it isn’t at all the same kind of thing as a mythological allusion like Callimachus’ Minotaur, or an allegory like Plato’s Atlantis, or even an anachronism like Strabo’s and Plutarch’s stories of inspired speech resulting from psychoactive gases at Delphi.
Well, if you’re designing a game, like Tomb Raider or Total War Troy, you have that freedom. But for history, don’t go relying on fictional characters for your methodology.

The methodology is called ‘euhemerism’. Maya Georgieva, director of Total War Troy, explicitly invokes euhemerism for the game’s design.
Euhemerism and authenticity
One of the key pillars of Total War games is authenticity — the ambition of creating the game’s sandbox in a manner that feels natural and true to its source while also accommodating all the eventualities that didn’t necessarily occur in history.

Maya acknowledges this, saying: ‘It is an important challenge — to capture the spirit of the source while also providing the necessary gameplay freedom that allows for anything to happen. So it was quite a relief when I realised our grounded approach to myths is not alien to the classical Greeks’ understanding of their legendary past — to the contrary, it is actually very fitting with their own mindset.’
Maya Georgieva in interview, May 2020
Euhemerus of Messene was the (probably pseudonymous) author of the Sacred Record, written sometime around the early 200s BCE. The book doesn’t survive, but we have second-hand reports. The Roman poet Ennius wrote a Latin version, the Sacred History, which was also very influential. On his travels, Euhemerus supposedly
saw a temple dedicated to Zeus that stood in the middle of an idyllic landscape (the whole island was exceptionally fertile). Inside the temple an inscription on a pillar told of the history of the cult of the gods of Olympus: these had been mortals who had been deified because of their extraordinary services to human civilization.
New Pauly s.v. ‘Euhemerus’ (see originally Diodorus of Sicily, Library 6 fr. 1.1–10 = BNJ 63 F 2)
It’s almost certainly Euhemerus that wrote about a gravesite of Zeus on Crete as well.

The basic idea of euhemerism is to take a myth, strip away the fantastic bits, and treat the result as history. It was a natural extension of earlier rationalisations of myths, and in particular of the gods. Here’s Aristotle on the subject:
It has been passed down from the ancients, even the most ancient, and left to later people in the form of a myth, that (heavenly bodies) are gods, and that divinity encompasses all of nature. And the other stuff got tacked on, mythically, to persuade the masses, and for the sake of laws and expedience: for example, they say that these (gods) have human shape, and are like other animals. The rest follows on from that and from other similar things. If we strip away this material, and take only the basic principle — that they consider the primordial beings to be gods — then we would regard that as a divinely inspired statement.
Aristotle, Metaphysics 1074b (12.8)
Euhemerism, or more generally rationalising interpretations of myth, had an instant and universal appeal. Ancient Christians loved it too: they drew on Euhemerus to reject all pagan gods and myths as distorted pictures of ordinary mortals, not actual gods.
Note. On Euhemerus and the grave of Zeus, see Winiarczyk 2013: 33–41. For a Christian writer playing Euhemerus as a ‘pagan gods are fake’ card, see e.g. Lactantius, Divine Institutes 1.11. (The irony of Lactantius trying to pull this was completely lost on him.)
Georgieva gives several illustrations. Hephaestus’ fall from heaven might have been a metal-rich meteorite; the monster Typhoeus might be a volcanic eruption; the Chimaera’s fiery breath could be natural gas vents; the Minotaur could be a man wearing a bull’s skull over his face, Centaurs could be skilled horse-riders with shaggy clothes, the Cyclops could be a man wearing a dwarf elephant skull. And so on.

The assumption behind euhemerism is that, if a myth has been passed down, it must have started somewhere. There must be a source, a first mover, a seed. And that seed is supposedly still there under all the impossible elements.

And there’s absolutely no reason to expect that reasoning to work. In Euhemerus’ own case this is perfectly clear. The Olympian gods as fantasy-coated versions of historical kings may sound neat, but we, nowadays, know that ‘Zeus’ is a reflex of a Proto-Indo-European root *dyēus, meaning ‘sky, day’, and that his function as a sky god has analogues in several other ancient cultures who spoke Indo-European languages. He definitely wasn’t a mortal king, he didn’t live on the island of Panchaea, and there wasn’t a real gravesite on Crete. In a similar way, the classical Greeks believed Dionysus was a recent arrival in their pantheon from Thrace; now we know that isn’t true either, because he’s there in Linear B tablets. He was baked into the Greek pantheon all along.

There’s no need for a first mover, there’s no expectation that there’s a historical kernel. It can just as easily be encrustations all the way down. And it probably is.

Euhemerism means choosing any pseudo-historical explanation of a myth, rather than admitting that we don’t have good enough data to draw any conclusions. I’m not saying a myth is never based on historical events. For all I know, the Lelantine War may have been a real thing. As a matter of fact I think it probably was. But assuming the war happened, simply because we don’t have any competing high-quality evidence about relations between Chalcis and Eretria around 700 BCE, is garbage reasoning.

Euhemerism isn’t a tool for turning bad data into good data. It’s an excuse to carry on using bad data.
People can go to great lengths to resist the idea that a story was just ... made up. This tweet (July 2020) treats an image from a 2013 Photoshop contest as evidence of a cover-up conducted by the Catholic Church and scientists. He gets very angry at the people pointing out that it’s a photoshop. This doesn’t bode well for the author’s state of mind, but it’s also unsurprising: it’s part-and-parcel of the human habit of expecting a kernel of reality even in material that’s totally invented. (Thanks to David S. Anderson for publicising this.)

Phlegon on giant bones

There are always going to be people who find the Trojan War appealing, simply because Troy is real — even if the same logic does imply that The Avengers (2012) is a docudrama. And Georgieva is absolutely right that euhemerism is in keeping with the way some Greeks saw their own myths. It’s just that it’s hopeless as a method for finding out anything real.

And that’s why I find it perplexing when I see bits of Phlegon of Tralles treated selectively as reliable reports. Phlegon, who worked for the Roman emperor Hadrian for a while in the 2nd century CE, wrote a short book called On Wonders which is full of truly weird stories. One part has attracted more attention than any other: a section where Phlegon reports on discoveries of giant bones.

Adrienne Mayor, in her 2000 book The First Fossil Hunters, documents widespread Greco-Roman interest in the remains of species that no longer existed in their environment. Many subsequent writers have taken that as licence to treat Phlegon’s account as true to reality — so long as he’s talking about giant bones.
The huge bones were evidently prehistoric fossils which, compared to the skeletal remains of normal size, were in antiquity identified as relics of extinct races, usually those of giant heroes.
Doroszewska 2016: 129

Unlike many of the strange phenomena Phlegon reports in Mir., the existence of such giant bones can be easily confirmed and explained. As Mayor has persuasively demonstrated after considering a wide variety of evidence, the bones interpreted as the skeletons of giants by the people of the ancient Mediterranean were actually fossilized bones of giant mammals left in Europe after the last Ice Age (e.g. mammoths).
Shannon-Henderson 2019 (on Phlegon, On Wonders §11–16)
Just for reference, Phlegon also tells stories like:
  • People return from the dead, and a decapitated head delivers prophecies in verse (On Wonders §1–3).
  • Spontaneous sex changes of both mythical and contemporary figures (§4–9).
  • Contemporary children born with multiple heads, extra limbs, and growing to maturity; women giving birth to a monkey, a dog-headed child, and snakes; men giving birth (§20–27).
  • The king of Saune, supposedly a city in Arabia, captures a live centaur, which ends up embalmed and on display in a palace in Rome (§34–35).
But why assume Phlegon’s stories are based on anything real at all? If there’s some reason to think that he’s telling a distorted version of some real phenomenon, then bones of large extinct species could be a good candidate. But Phlegon’s material doesn’t encourage me to expect that anything should be taken as reflecting reality in any particular way. I don’t see why we should even be looking for the city of ‘Saune’, which doesn’t appear in any other source.
Giant bones reported by Phlegon, On Wonders 11–19. Immediately after this bit, Phlegon reports on a child with four heads ‘and a corresponding number of other limbs’ who grew to adulthood during the emperor Nero’s reign.
Even in the bit about the giant bones, it’s clear that Phlegon is describing complete skeletons, and there’s no hint that they’re anything other than human-shaped. Even if they were non-human, there are other details which make it hard to reconcile them with mammoths — let alone dinosaur, let alone complete skeletons of the very largest dinosaur species.

One of Phlegon’s complete skeletons is far, far longer than any land-based species that has ever existed outside the Americas:
[T]here is a certain island near Athens, and ... the Athenians wished to fortify it. When they were digging the foundations for the walls, they discovered a coffin measuring a hundred cubits, in which was a skeleton as large as the coffin. Inscribed on the coffin was the following:
       I, Makroseiris, have been buried upon a large island,
       after living five times one thousand years.
Phlegon, On Wonders §17 (tr. Shannon-Henderson)
A hundred ancient Greek cubits comes to 44.4 metres. The only land-based species ever to reach that length was Barosaurus, native to the Midwest of North America. No one should be imagining a complete Barosaurus skeleton in Greece. In a coffin, no less. And with an inscription in classical Greek. (Several others of Phlegon’s giant bones come in coffins too; one of them comes in a jar, labelled with the occupant’s name.)

I find it a bit perplexing that people can look at this, and still say ‘Yes Phlegon talks a lot of nonsense, but when it comes to these specific bits he’s actually based in reality.’ I mean ... why? If his material is that bizarre, why imagine any basis in reality at all?

As we saw above, the nature of the source makes a difference. When Lara Croft argues in favour of Yamatai, she’s drawing on sources that purport to be records of actual historical events. Even in cases like Ctesias and Megasthenes, who believed real dog-headed people lived in India, or Pliny the Elder, who believed the Alps were 75 km high, at least in principle there’s a possibility of sifting out material that reflects reality from distortions and misunderstandings.

Pliny, too, reports, on giant bones, and that should be taken much more seriously. But he doesn’t talk about ribs that are far bigger than any real rib bones that have ever existed. I think it makes no sense to take someone like Phlegon, seize upon the one bit that seems plausible (given a bunch of assumptions), and then say that bit is the one bit that isn’t completely made up.

Phlegon is an extreme case. But the point has a much more general applicability. Whenever you see a news-item on geomythology (myths supposedly based on geological events tens of thousands of years ago), or on Atlantis, or any claim that myths were once ‘true’ — well, that’d be a good occasion to stop and think to yourself:
OK, there’s a passing resemblance ... but why should we expect that the myth is based on anything real in the first place?


  • Croft, Richard J. 2015. Truth behind myth. Osiris Publications.
  • Doroszewska, Julia 2016. The monstrous world. Corporeal discourses in Phlegon of Tralles’ Mirabilia. Peter Lang.
  • Echo-Hawk, Roger C. 2000. ‘Ancient history in the New World: integrating oral traditions and the archaeological record in deep time.’ American Antiquity 65.2: 267–290.
  • Mayor, Adrienne 2000. The first fossil hunters. Paleontology in Greek and Roman times. Princeton University Press.
  • Ogden, Daniel 2013. Dragons, serpents, and slayers in the classical and early Christian worlds. Oxford University Press.
  • Shannon-Henderson, Kelly E. 2019. ‘Phlegon of Tralleis (1667).’ Die Fragmente der Griechischen Historiker Part IV. DOI link (retrieved July 2020).
  • Winiarczyk, Marek 2013. The ‘Sacred History’ of Euhemerus of Messene. De Gruyter.


  1. You mention in passing that Proto-Indo-European is from ca. 4000 BCE. In the context of your interesting text it’s not really that important, but I would bring that date closer to us. Dating is often fiendishly difficult in historical linguistics whenever there is no reliable data directly to anchor on.

    When even good historical linguists estimate dates for languages such as Proto-Indo-European, Proto-Semitic, or Proto-Uralic, they tend to add millennia (or sometimes centuries) just to be sure — they want to give amply time for languages to bifurcate and develop. You may find for instance an estimation of Proto-Uralic for 6000 BCE! There’s no need for that early a date.

    Generally, Proto-Indo-European is put in around 3000 BCE, but even that might be too early. It could easily be as late 2500 BCE: the Indo-European settlement in Central Anatolia is often estimated to 1900 BCE, and the first attested Hittite texts are from the 16th century BCE.

    1. Thank you for this! I'm a fan of down-dating, generally, but this is one that had passed me by. I'm very happy to be corrected. Would you say that higher datings like 4000 BCE are implausible, or just unnecessary?

    2. Please don't take my dating above as an absolute truth, either! The point I'm making is rather the tendency of linguists to add millennia (or centuries) simply to give enough time (according to their gut feeling) for languages to diverge and develop, and there is a clear fear to be seen to go a little lower, so they rather go farther back in time — just in case. So my aversion to high datings are mostly that they are simply unnecessary.

      If we take the Anatolian languages and their spreading to the central Anatolian plain, 500 years is plenty of time for PIE to bifurcate there (and we would have almost 900 years to the earliest extant IE texts). As we know, there's no constant speed by which languages develop, but it varies immensely.

      Noble efforts have also been made to attach PIE (and Proto-Indo-Iranian and Proto-Uralic etc.) to some archaeological culture, but this leads nowhere. We just need to imagine what non-textual archaeological remains e.g. modern Central Europe might leave behind and then wonder if we could draw the language map on this basis. The answer is a resounding no.

      I remember that the Indo-Europeanist/Uralist Petri Kallio wrote an article on this question of dating some 15 years ago.

      By the way, in recent decades there has also been more emphasis on the fact there is tendency for people to stay where they are, and often merely their languages (and cultures) spread (by diffusion), not generally people in big numbers. Obviously there are many exceptions, but in general people don't want to move.

  2. Hesiod said the Cyclops were blacksmiths. Which could be where their single eyes makes sense. Blacksmiths were in danger to losing an eye due to hot metal sparks. (I suppose if you lost two you were out of work.) Hesiod or earlier storytellers needed titanic blacksmiths, so large beefy one eyed giants seemed like a natural form to go with. As Homer's sheep farmers, their appearance is less logically reasoned.

  3. So where do myths actually come from?

  4. nicely done. I find that in the classroom one must address Euhemerism sooner and a lot more than I would have expected. That's because it's out there. Mayor does what she does very well (now with Amazons also) and it's important that Greco-Roman myths were located often in known geography and spoke to cultural practices, but I like to tell students that our modern culture's fascination with the 'origin' (and you get this from relatives and at dinner parties all the time, like getting gold dust through fleece in running water = 'golden fleece), but I like to point out that this method essentially disrespects stories and literature in an of themselves. Same with the Trojan War: the Schliemannesque point of view would not glance at the Homeric epics if they were not based on a 'real' war. But it's clear to me that they were not -- and it would be your loss if you thus ignored poetry and myth.
    Jonathan Burgess

    1. It's a bit of a conundrum -- it's perennially popular, and I can't see it ever going away. I see it in the same spirit as ancient allegorical approaches to myth, like Theagenes or Heracleitus, but it seems there's something especially timeless about the appeal of euhemerism. I can't see it going away, because I can't see its appeal going away!

      I like your efforts to address it in Homer -- I used to make regular use of your resources on Odysseus' travels in classes some years ago!