Wednesday 30 August 2017

Roman plagiarism of Greek gods

Left: Poseidon (Melos, late 2nd cent. BCE, National Archaeological Museum of Athens). Right: Neptune (Trevi fountain, Rome, 18th cent. CE).

One of the first things a student of classical Greco-Roman mythology learns is that the Greeks and Romans talked about the same gods, but with different names --

Greek Roman god of ...
Zeus Jupiter sky, lightning, king of gods
Demeter Ceres agriculture, grain crops
Ares Mars war
Poseidon Neptune sea, earthquakes
Aphrodite Venus sex, erotic desire

and so on. A very few gods, like Apollo, keep the same name. Every myth course starts with a table like this.

So if stories about Zeus, Hermes, and Herakles are the same thing as stories about Jupiter, Mercury, and Hercules ... and the Greeks came first ... that means the Romans stole their mythology from the Greeks, right?

Well, if it were right we wouldn’t be here talking about it. Plenty of sources out there on the web point out that some of the Roman pantheon were native Roman gods, like Quirinus, Bellona, and Terminus. But even they’re seen as exceptions to the rule. Those few are the genuine Roman gods -- not like those other mainstream gods ...

Now, it is true that many of the resemblances between the Greek and Roman pantheons are no coincidence. But the Romans didn’t replace their own gods, and they didn’t steal Greek religion either.

... except when they did. I have to add a caveat: there are a few cases where the Romans really did appropriate Greek gods or Greek religious cults. But not as many as you think, and not the gods you’re thinking of. Let’s look at some reasons for the resemblances.

1. Indo-European religion

Names of gods, and in some cases maybe other aspects of them, can sometimes be traced back to prehistoric origins. This works the same way that historical linguists reconstruct family-trees of different languages. In this way of looking at things, you could imagine the Greek and Roman gods sort of as cousins.

To an extent, at least. You have to be careful: prehistoric Indo-European links aren’t as pervasive as people sometimes imagine.

For gods with related names, we only have a couple of really good examples. The best known is Zeus-Jupiter. Both of these names can be traced back to a reconstructed Proto-Indo-European name *dyēus-ph2tēr, where *ph2tēr = ‘father’. The name also shows up in India as the Vedic god Dyauṣ Pitā, and in a few other places too. In Zeus’ case, the ‘father’ dropped off (though he’s still regularly called ‘father of gods and men’) and the *dy- metamorphosed into either dz-, sd-, or di-, depending on the dialect and on his grammatical role in a sentence. In Jupiter’s case, the *d- disappeared, and the -ēu/ēw- got assimilated with the following -p-, resulting in -upp- followed by a neutral vowel: Iuppiter. In other grammatical forms, the -pater disappeared too, resulting in the stem Iov- (originally pronounced yōw-). That’s why even in English he can be called either Jupiter or Jove.

Another pair that’s obviously related is Hestia-Vesta. There’s some evidence to suggest a prehistoric Greek goddess called Eleuthera, and one of Jupiter’s cult titles in Latin is Liber: both mean ‘liberator’ in their own languages. [correction:] we know of Latin vegetation gods called Liber and Libera: all of them come from an Indo-European root meaning ‘growth, increase, the people’. [end of correction] And there’s some sort of connection between Greek Eos, Latin Aurora, Vedic Uṣas, and Lithuanian Aušrinė, but in their case it’s questionable whether they derive from an Indo-European goddess or from an Indo-European word meaning ‘east, dawn’: Eos, Uṣas, and Aušrinė don’t have much in common, and we know basically nothing about Aurora.

But for most gods, the name game doesn’t work. ‘Athena’, ‘Demeter’, ‘Dionysus’, ‘Hera’, ‘Hephaistos’, and ‘Hermes’ all appear to have non-Indo-European roots -- mixed with Indo-European elements, in the cases of Demeter and Dionysus (-mētēr ‘mother’, dio- ‘of Zeus’) -- and ‘Apollo’ and ‘Poseidon’ have uncertain origins. On the Roman side, ‘Juno’ and ‘Minerva’ both have ‘Minerva’ has a known Etruscan precursor, which is to say, not Indo-European; ‘Ceres’, ‘Mars’, ‘Venus’, and ‘Vulcan’ come from Indo-European words, but we don’t find any related names among other Indo-European gods of grain, war, sex, and fire.

There are more options when we look at religious practice, divine functions, and story-elements. For example, Zeus and Jupiter both have cult titles built out of verbs: Zeus can be ‘Zeus the Gatherer’, ‘Progenitor’, or ‘Saviour’; Jupiter can be ‘Victor’, ‘Helper’, and ‘Adviser’. Zeus is a Greek sky god who was nursed by a goat as a baby, and gets called aigiochos, ‘goat-rider’ or ‘goat-driver’; Thor is a Norse sky god who drives a chariot drawn by goats. When Zeus enters a gathering of the gods in Homer, they all stand up for him; the same happens for Indra in the Indian Mahābhārata, and for Ea in one Hittite text.

For Roman mythological elements, it’s harder to find Indo-European parallels. This is partly because there’s a severe shortage of information about gods in early Rome. The Greeks had myth encyclopaedias, and scholarly investigations of their own mythology: we have a huge amount of information from them. But from the Romans, not so much. There are snippets in writers like Servius and Pliny, but nothing as systematic as the Greek mythographers. Partly it just comes down to the sheer amount of text that survives: we have about eight times as much textual sources from the Greek world as from the Latin world. Varro, one of the greatest scholars of Greco-Roman antiquity, wrote a long treatise on early Roman religion called Antiquities of divine matters, which ought to be a key source for Roman religion ... but, alas, it doesn’t survive.

2. Interpretatio romana

Most ancient cultures had different names and even different stories about the same gods. ... So as the Romans conquered the Greeks they adopted Greek Mythology and replaced the gods’ names with traditional Roman gods’ names. Similar process would occur when the Germanic tribes were Romanized.
-- ‘Farhan Elohovitch’,, June 2016
Ancient Mediterranean civilisations who were in touch with one another often drew parallels between their pantheons and equated them with each other. When the Greeks looked at Egyptian mythology, they saw some parallels between Egyptian gods and their own gods, so they ended up referring to the Egyptian gods by Greek names: Zeus = Amun, Apollo = Horus, Adonis = Osiris, and so on. Nowadays we call this interpretatio graeca. There was no danger of mixing the gods up, because the resemblances weren’t very deep, and because the different names were linked to different places. Ancient religion was often more about observing local customs than about belief: when in Egypt or Libya, you visit the oracle of Amun; when at Dodona in Greece, you visit the oracle of Zeus.

In the same way, the Romans referred to foreign gods using Roman names. When talking about the Greek Zeus, they’d call him Jupiter; when talking about the Greek Hephaistos, they’d call him Vulcan. And there was an awful lot of Greek myth floating around, so it’s not really surprising to find Romans telling Greek stories with Roman names. Again, this was more a terminological thing than a religious thing.

The Romans did this with other pantheons too. Julius Caesar’s account of Gaulish religion refers to the Gaulish gods by Roman names (Gallic wars 6.17): he says their chief gods were Mercury, Apollo, Mars, Jupiter and Minerva. These could be Lug, Belenus, Teutates, and Sul; but the interpretatio romana of Gaulish gods wasn’t nearly as one-to-one as it was with the Greek pantheon. The equations changed depending on time and place. Teutates could at times be equated with any of Mercury, Mars, and/or Jupiter, and the Gauls had many gods of healing to equate with Apollo.

It is fair to say that in the Roman-Greek case, the resemblances are relatively accentuated by comparison with the Greek-Egyptian or Roman-Celtic cases. Roman religion was unquestionably something very distinct from Greek religion; but elite Roman literature very often took stories about the gods from Greek sources. When Ovid tells stories about gods using Roman names, he’s telling Greek stories (with his own spin on them).. But that has basically nothing to do with Roman religion. Elite poetry and religion have very little overlap.

3. Genuine adaptations

We know that the ancient Greeks had a massively entertaining sets of gods and goddesses. So it’s no wonder that when Rome conquered Greece, they replaced their own dull pantheon with renamed versions of Zeus, Athena, and the others. But not all Roman gods were Greek copies ...
-- Rob Bricken, io9, Jan. 2015
Some Roman gods genuinely were borrowed, inherited, or copied from the Greeks. Apollo is the most famous case, but there are some other significant ones. The divine twins, Castor and Pollux, were extremely important very early on, in the battle of Lake Regillus at the very beginning of the Roman republic in 495 BCE; and they were definitely not home-grown. They were borrowed from the Greeks, either directly or via the Etruscans. This goes for some important heroes, too, especially Hercules, Aeneas, and Ulysses: Ulysses had a significant role in Roman legend as early as the sixth century BCE, and evidence for Aeneas pops up quickly afterwards (during the reign of king Tarquinius Superbus the Romans founded a colony named after Circe at the supposed location of Circe’s island; one 5th century Greek source, subscription needed, tells us that Aeneas and Ulysses founded Rome together, and a Hellenistic poem has them meeting up near the future site of Rome). Later on, Dionysus-Bacchus became another major import.

But a couple of cautions are warranted. First, it wasn’t a one-way street. Greeks colonised Italy, but Italians visited Greece too: one inscription from ca. 500-470 BCE shows ‘Tyrrhenians’ (Etruscans? native Campanians?) dedicating an offering to Apollo at Delphi. The Latin demigod Faunus shows up in Greek as ‘Agrios’, alongside Latinus, in a passage that got added to the Hesiodic Theogony probably in the late Archaic period.

In Rick Riordan’s third series of demigod novels (The Lost Hero, 2010) the Roman gods are more variations on a theme than a truly distinct pantheon: Ares and Mars have different personalities, but the same face.
And second, the borrowings weren’t just about the Greeks. The cults of Isis, Cybele, Mithras, and Christ were ‘borrowed’ in exactly the same way as Apollo or Bacchus. But you don’t hear people complaining that the Romans ‘stole’ Christianity, or ‘replaced’ their own gods with Christ. With the relationship between Roman and Greek religion, the fashion is to assign agency to the Romans: the Romans stole, or borrowed, or adapted the Greek gods. Yet somehow, when it comes to Christianity, you’ll often find Christianity cast as the agent: Christianity expanded, it rose to prominence, it took over the Roman empire. (Did I say ‘somehow’? Well, actually there’s no real uncertainty: blame Gibbon.)

Saying ‘the Romans stole their gods from the Greeks’ makes about as much sense as saying ‘Peter Jackson stole New Zealand cinema from Hollywood.’ It’s nonsensical to talk about theft or appropriation when you’re talking about cultural hegemony. For 6th-century BCE Italians, the Greeks had that hegemony, in much the same way that Hollywood has it in 21st-century New Zealand.

Later, especially in the Pyrrhic War of the early 3rd century BCE, it looks very much as though the Romans took pains to reject Greek cultural hegemony, or at least western Greek. That appears to be when they came close to erasing Odysseus-Ulysses from Roman legend. Aeneas was left as the only ancestral figure: traces of Odysseus-Ulysses remained only outside Rome. His and Circe’s son Telegonus was the legendary founder of Tusculum, and in some sources Caere and Praeneste; in another source, Odysseus himself was buried at Cortona. These cross-overs between Italian geography and Greek mythical figures aren’t an act of Roman cultural appropriation: they’re a collaboration between Greek writers and Italian legends -- just as The Lord of the Rings trilogy is a collaboration between Hollywood and ‘Wellywood’.

Tuesday 15 August 2017

Volcano Day

‘Pompeii ... we’re in Pompeii! And it’s Volcano Day!’
Doctor Who, ‘The fires of Pompeii’ (2008)
Warning: this blog post is wrong!

After I posted it, the archaeologist Sophie Hay alerted me to an article, Abdy 2013, which entirely tears out the rug from the central piece of evidence discussed here. The coin which I refer to as ‘minted no earlier than September 79 CE’ actually comes from an earlier issue. The inscription, illustrated and reported below as containing the phrase Imp XV ‘(recognised as) general 15 times’, actually reads Imp XIIII ‘general 14 times’, and there is no PP at the end. The poor state of preservation of the coin misled some very skilled numismatologists ... and I am no numismatologist.

Coin struck from the same reverse die as the coin found at Pompeii. In the case of the Pompeii coin it was originally thought that the number after IMP must be XV, because the capricorn’s tail would block any longer number. In this better preserved coin, however, notice that the XIIII actually crosses the tail. According to Abdy 2013, the Pompeii coin actually dates to July or August 79 CE.(Source: Abdy 2013, Plate 18)
I shall leave the post up, unedited, as a monument to my capacity for being completely wrong!

-- PG, ca. nine hours after post first went up

The eruption of Mt Vesuvius in 79 CE, which wiped out the city of Pompeii, is famously recorded in an eyewitness report by Pliny the Younger, Letters 6.16. Pliny tells the story of how his uncle, Pliny the Elder, the famed collector of facts, factoids, and oddities, launched a rescue expedition, in the course of which he died from causes apparently related to poor respiration.

Let’s talk about the date of the eruption. According to mainstream tradition, it happened on the 24th of August, 79 CE. This is the date Wikipedia reports, and the date that appears in every edition of Pliny’s Letters. Wikipedia also adds a caveat, ‘(probable)’, and has a substantial section devoted to doubting the date. Still, it leaves a final impression that the evidence is inconclusive. Readers will probably come away feeling free to go on quoting the date as 24 August -- maybe with a ‘(probable)’ warning at most.

In fact the evidence is entirely conclusive: the 24 August date is very definitely wrong. It’s just that among all the discussion, it may be difficult to see what the really decisive piece of evidence is. That’s also the case in the only academic discussion that the Wikipedia discussion cites, Rolandi et al. 2007. Rolandi et al. present loads of information, but their aim isn’t to weigh up which piece of evidence is the most important. But here’s a hint: it’s not the one in the headline of the article (‘the southeast tephra dispersion’). Judging from the Wikipedia article, it looks like it isn’t easy for a layperson to make out the smoking gun amidst the clouds of volcanic sulphur.

The really decisive piece of evidence is this: a coin, which was found in a hoard belonging to victims of the eruption, and which was minted no earlier than September 79 CE.

That’s the one item of evidence that points unambiguously to a later date for the eruption. Everything else is supporting evidence. Climate conditions and the locations of pyroclastic fall deposits; textual evidence in manuscripts of Pliny’s letter; food items found at Pompeii; styles of dress found on the bodies of the dead ... those things are all well and good, sure. But they’re all secondary to the direct and nearly explicit evidence of that coin. They’re helpful to the extent that they make the coin’s evidence more plausible, not because they’re more important than that coin.

(For similar assessments see Beard 2008: 17-19; Lazer 2009: 79-80. They mention all the other evidence, but like me, they both conclude that the coin is ‘[m]ore clinching’ (Beard); ‘This contentious issue may well have been resolved by the ... silver denarius’ (Lazer). If you want another second opinion that’s easier to get at, you can also try this 2013 blog post by ‘Roma Invicta’.)

The traditional date

The traditional date comes from Pliny. Here’s the relevant bit in his letter:
erat Miseni classemque imperio praesens regebat. nonum Kal. Septembres hora fere septima mater mea indicat ei apparere nubem inusitata ...

(My uncle) was at Misenum, on duty in command of a fleet. On the ninth day before the kalends of September, at roughly the seventh hour, my mother pointed out to him that an unusual cloud was appearing ...
-- Pliny, Letters 6.16.4
‘The ninth day before the kalends of September’ is a standard Roman way of reporting the date. The ‘kalends’ were the first day of the month, so the ninth day before that (counting inclusively) was the 24th of August.

Pointing at the coin isn’t going to be enough: we also need to account for Pliny’s testimony. If he says it was the 24th of August, that’s always going to be more explicit than a coin, right?

Well, this is where the manuscripts come in. Looking at actual surviving manuscripts of Pliny shows that it’s extremely unlikely that nonum Kal. Septembres is what Pliny actually wrote. Most manuscripts of Pliny are pretty incoherent about the date, and only one manuscript gives a nice clear-cut text that actually makes sense, and that’s the one that reads ‘24th of August’. Some other manuscripts read novem, which could be either ‘nine’ (‘nine days before the Kalends of September’), which would be an unusual phrasing, or more probably an abbreviation for November, Novem(bres). It may well be that confusion between Novem(bres) and novem has produced nonum (ninth); the phrase hora septima ‘at the seventh hour’ could be responsible for the introduction of Septem(bres).

Realising that this piece of testimony doesn’t actually have much of a leg to stand on is an important component of this argument. It’s still not the decisive point, though. The manuscript readings just remove evidence for the 24 August date; it’s the coin that proves the eruption didn’t happen on 24 August.

Mount Vesuvius looms over the remains of Pompeii

Weighing up the evidence

Here’s the break-down of reasons for doubting the 24 August date, and the different ways in which each reason matters. I’ll give the evidence in the order that Wikipedia mentions it:
  1. Manuscript readings in Pliny Letters 6.16.4 show that the traditional date is poorly supported. They are not in any sense contrary evidence: they don’t disprove the 24 August date. Their role lies in making the positive evidence for 24 August decidedly weak.
  2. Evidence of heavy clothing found on casts of some eruption victims is mildly supportive of an autumn date. This is very far from compelling, since there are plenty of other explanations (there’s always variation in seasonal weather; the people were fleeing their homes, and may have planned for being without shelter). But it is mildly interesting supporting evidence for doubting the August date. Not remotely in the same league as the next two items, however ...
  3. Archaeological finds of autumnal crops, including fruits, hemp, etc., suggest an autumn date for the eruption. These are historically the reason why the traditional date was first doubted, by Carlo Maria Rosini, who excavated Pompeii in the late 1700s. They were decent evidence, and somewhat compelling, but not quite strong enough to counter the Pliny manuscripts that do read ‘24th of August’. Now that stronger evidence has come along, these finds are demoted to being high-quality supporting evidence.
  4. The coin mentioned above, found in 1974, and with a detailed argument published in 2006 by Grete Stefani, director of the Office of Excavation of Pompeii. This coin is the first unambiguous evidence that Pompeii was not buried before September at the very earliest. It is clear-cut, absolutely decisive, and extraordinarily difficult to refute.
  5. Dispersal of pyroclastic deposits. An article published by Rolandi et al. (2007) relies on seasonal wind patterns to argue against the August date. As with item 2, above, this isn’t decisive (I’m not aware of a place that has no variation in seasonal winds ... but then, I live in Wellington), but it is still interesting supporting evidence.
The Wikipedia article demotes the most compelling piece of evidence to fourth place, and devotes more than half of its discussion to the least compelling ones, items 1 and 5. It’s not surprising people are confused.

On the other side we have the evidence in favour of the 24 August date:
  • Pliny Letters 6.16.4: though the manuscript tradition is inconsistent, we do have the date 24 August supported there. Just not strongly.
  • Cassius Dio 66.21.1 states that the eruption took place ‘in the very time of summer-waning’, or late-summer/early-autumn. This would normally put the event between ca. 6 August (the setting of the constellation Lyra, to Pliny the Elder Natural history 18.59) and 25 September (the autumn equinox). Well, it’s consistent with the 24 August date, at least. On the other hand, Cassius Dio also reports that the eruption was preceded by omens of giants stalking the countryside and flying overhead (66.22.2) ...
Yeah, these points are pretty weak. They certainly don’t stand up to the overwhelming evidence of the coin, and the supporting evidence from archaeological finds of seasonal crops.

The coin

Two coins of emperor Titus, one (top) showing Titus recognised as imperator (‘general’) fourteen times, the other (bottom, outlines enhanced) the coin discussed by Stefani 2006, showing Titus recognised as imperator fifteen times, and therefore dating no earlier than September 79 CE.
Top: Heads side reads Imp Titus Caes Vespasian Aug PM, ‘Gen(eral) Titus Caes(ar) Vespasian(us) Aug(ustus), p(ontifex) m(aximus).’ Tails reads TrP VIIII Imp XIIII Cos VII PP, ‘Tr(ibunician) P(ower) 9th time, (hailed as) gen(eral) 14th time, cons(ul) 7th time, f(ather of his) c(ountry).’
Bottom: Heads side same as above. Tails reads TrP VIIII Imp XV Cos VII PP, ‘Tr(ibunician) P(ower) 9th time, (hailed as) gen(eral) 15th time, cons(ul) 7th time, f(ather of his) c(ountry).’
The bottom coin is the important one. It was found in 1974 next to the so-called House of the Golden Bracelet, along with about 200 other coins that victims of the eruption took with them as they fled. No, it’s not the most beautiful coin ever designed. Titus has quite the neck there, doesn’t he? But it’s neatly unambiguous: as the inscription on the tails side says, when it was minted (or, arguably, just about to be minted), Titus had been recognised as imperator (‘general’) fifteen times.

How does it have a bearing on the date? It’s because we know that as late as 8 September 79 CE, official Roman documents were still referring to Titus as imperator for the fourteenth time. In particular, the emperor’s own office was still calling him imperator for the fourteenth time on 7 September.

This is not something that can be chalked up to news travelling slowly: changes in who was emperor and the emperor’s status were circulated around the empire very promptly. For example, we have papyri from Egypt reporting on new emperors within a month or so of their taking the position. And Pompeii is a lot closer to Rome than Egypt is. And with a coin, you also have to add in extra time for it to be minted and get into circulation and into someone’s purse.

The documents in question are two inscriptions. One is a military diploma found in Egypt dating to 8 September (line 17: a(nte) d(iem) VI Idus Sept(embres)); that could be blamed on communication delays between Rome and Egypt. The other is much more compelling: it is a letter on a bronze tablet sent from the office of the emperor himself and dating to 7 September (line 16: dat(um) VII Idus Septembr(es)). So unless we’re going to argue that the emperor’s own secretarial staff had somehow forgotten there had been a fifteenth acclamation as imperator, we have absolutely rock-solid evidence that Pompeii was still unburied when the letter was sent on 7 September.


As noted at the beginning of this post, this post is wrong! The coin highlighted as the central piece of evidence here was misidentified, and the misidentification was only realised in 2013. See Abdy 2013, and some further details at the beginning of this post.


  • Abdy, R. 2013. ‘The last coin in Pompeii: a re-evaluation of the coin hoard from the house of the Golden Bracelet.’ Numismatic Chronicle 173: 79-83.
  • Beard, M. 2008. Pompeii. The life of a Roman town. London: Profile Books.
  • Borgongino, M.; Stefani, G. 1999. ‘Intorno alla data dell’eruzione del 79 d.C.’ Rivista di studi Pompeiani 10: 177-215.
  • Lazer, E. 2009. Resurrecting Pompeii. London/New York: Routledge.
  • Rolandi, G.; Paone, A.; Di Lascio, M.; Stefani, G. 2007. ‘The 79 AD eruption of Somma: the relationship between the date of the eruption and the southeast tephra dispersion.’ Journal of Volcanology and Geothermal Research 169: 87-98.
  • Stefani, G. 2006. ‘La vera data dell’eruzione.’ Archeo 206: 10-13.
  • Stefani, G.; Borgongino, M. 2007. ‘Ancora sulla data dell’eruzione.’ Rivista di studi Pompeiani 18: 204‑6.