Thursday, 14 June 2018

The number of the beast

The Christian bible reports ‘the number of the beast’ as follows:
This calls for wisdom: let anyone with understanding calculate the number of the beast, for it is the number of a person. Its number is six hundred and sixty-six.
-- Revelation 13.18 (NRSV translation)
This number is a tremendously popular icon in modern culture -- and in popular religion, too. Its popularity owes a lot to the repeating sixes: ‘6-6-6’. This snippet from Doctor Who relies on it, for example:

Doctor: To generate that gravity field, and the funnel, you’d need a power source with an inverted self-extrapolating reflex force of 6 to the power of 6 every 6 seconds.
Rose: That’s a lot of sixes.
-- The Doctor, about to encounter the Devil (Doctor Who, ‘The impossible planet’, 2006)

This six-ishness depends on a decimal counting system. And not just that: it has to be a Hindu-Arabic-style decimal system. You have to take a very specific kind of notation for granted.

Not enough of a myth for you? Well, how about this one. A fairly popular story floating around these days is that the number of the beast wasn’t actually 666, but 616. This is supposedly proved by the earliest papyrus of Revelation, which was published in 1999; the idea was popularised in a 2003 BBC documentary.

Well, none of that’s true. Not quite true, anyway. So, yes, there are some catches.
  1. Hindu numerals started to appear in the 7th century in India; Revelation was written about 600 years earlier in the eastern Mediterranean. Now, some ancient Mediterranean cultures did have decimal numerals -- Greek Ionic numerals, and Hebrew numerals -- but they were place-dependent. They had different symbols for 6, 60, and 600. You used different symbols depending on whether you were writing a number in the ones column, the tens column, or the hundreds column. The Greek notation for 666 is χξϛ, not ϛϛϛ; in Hebrew it’s םסו‬, not ‬‬‬‬‬‬‬‬‬ו‬ו‬ו‬. Repeating the symbol for ‘6’, ϛϛϛ or ‬‬‬ו‬ו‬ו‬, would be meaningless.
    A spectrogram: the computer game Doom (2016), premised on a demon invasion from hell, has the numerals ‘666’ embedded in part of the soundtrack. This wouldn’t work in any ancient numeral system.
  2. A prophetic interpretation, which sees the text as a genuine prophecy of a specific later time, is never going to have as much staying power as a historical-critical interpretation. This is an ancient Greek text which tells us explicitly that it contains a coded message: you’d need some very powerful and specific reason to imagine that the message should work just as well for, say, a 21st century American (who doesn’t speak ancient Greek) as for a 1st century Christian (who did). That way lies madness. But in a 1st century Greco-Jewish context, as we’ll see, the number is meaningful and based on well-attested practices.
  3. The 616 variant is certainly ancient: Irenaeus knew about it in the late 2nd century, though he strongly preferred 666. As we’ll see below, both readings are based on the same reasoning. But, so far as we can tell, neither number is older than the other. 616 is moderately well supported, but it definitely doesn’t have earlier support than 666.

The text

Here’s what the number of the beast looks like in two early manuscripts of Revelation:

Left: codex Sinaiticus, quire 90 fol. 6r (4th century). Right: Chester Beatty Library P. Bibl. 3 (= P47 Gregory-Aland, Trismegistos 61628), fol. 7r (3rd century).

The left-hand image is from one of the most important manuscripts of the New Testament, the famous codex Sinaiticus, written in the 4th century. Its version of verse 18 reads:
ὧδε ἡ σοφία ἐστίν· ὁ ἔχων [[ουσ]] 'νοῦν' ψηφισάτω τὸν ἀριθμ'ὸ(ν)' τοῦ θηρίου, ἀριθμ'ὸς' γὰρ ἀν(θρώπ)ου ἐστίν· ἑξακόσιαι ἑξήκ'ο(ν)'τα ἕξ.
(Key: [[...]] = crossed out by the scribe, '...' = inserted above the line, (...) = abbreviation.)

There are a few differences from the modern critical text. The scribe has made one error, crossed it out, and written νοῦν (‘understanding’) in the margin. The phrase ‘and its number is’ (καὶ ὁ ἀριθμὸς αὐτοῦ) is missing here. And most importantly, the number itself is written out as words, ἑξακόσιαι ἑξήκοντα ἕξ (‘six hundred and sixty-six’), not as the numeral χξϛ (‘666’).

The right-hand image is the earliest known copy of the passage: a fragmentary papyrus written in the mid-3rd century and currently held in Dublin, Ireland. Here verse 18 reads:
ὧδε ἡ σοφία ἐστίν· ὁ ἔχων νοῦν ψηφισάτω τὸν ἀριθμὸν τοῦ θηρίου, ἀριθμὸς γὰρ ἀνθρώπου ἐστίν· ἐστὶν δὲ χξϛ.
This is a closer match to the modern text. But again, the words immediately before the number are different -- ἐστὶν δὲ (‘and it is’), rather than the modern critical text καὶ ὁ ἀριθμὸς αὐτοῦ (‘and its number (is)’) --; and this time, the number is written with numerals, χξϛ (‘666’).

‘The number of a person’

It’s fairly widely known that the number was designed as a reference to the Roman emperor Nero.

(1) Isopsephy. The most specific point is that the number 666 is plainly a case of isopsephy. Ancient Greek and Hebrew used letters of the alphabet as numerals: nine letters for 1 to 9, nine for 10 to 90, and nine for 100 to 900, for a total of 27 numerals. In isopsephy, the number of a name or phrase is obtained by treating its letters as numerals and adding up their values. In a Jewish context, this practice is known as gematria.

We have independent evidence of isopsephy in relation to Nero. Suetonius (Nero 39) quotes a Greek epigram where Nero’s name, transliterated into Greek,
Νέρων = Nerōn ‘Nero’
= 50 + 5 + 100 + 800 + 50
= 1005
is equated with
ἰδίαν μητέρα ἀπέκτεινε = idian mētera apekteine ‘killed his own mother’
= 10 + 4 + 10 + 1 + 50 + 40 + 8 + 300 + 5 + 100 + 1 + 1 + 80 + 5 + 20 + 300 + 5 + 10 + 50 + 5
= 1005
The epigram calls this a neopsēphon, a ‘new calculation’ -- the same word as in Revelation 13.18 psēphisatō ‘let (him) calculate’. Isopsephy is attested in Judaeo-Christian contexts too, notably in a set of poems called the Sibylline Oracles; and a 1st-3rd century CE Jewish apocalyptic text known as 3 Baruch uses equations based on Greek words transliterated into the Hebrew script.
  • Sib. Or. 1.324-9 uses the number 888 = Ἰησοῦς, ‘Jesus’ in Greek.
  • Sib. Or. 5.12-44 gives a list of Rome’s rulers numbered from their initials: e.g. 12-13 ‘he will be the very first king, who will sum twice ten / with the start of his name ... he will have his first letter from 10’, that is, 20 = Κ(αῖσαρ), 10 = Ἰ(ούλιος), for Julius Caesar. The text carries on with Augustus = α = 1, Tiberius = τ = 300, Gaius = 3 = γ, Claudius = κ = 20, Nero = ν = 50; then three rulers with short reigns; then Vespasian = ο = 70, Titus = τ = 300, Domitian = δ = 4, and Trajan = τ = 300.
  • 3 Baruch ch. 4 discusses a dragon in connection with 360 rivers, where ‘dragon’ δράκων > דרקון drqwn, which adds up to 360; and a cataclysm involving 409,000 giants, where ‘cataclysm’ κατακλυσμός > קטקליסמס qtqlsms, or 409. (Bear in mind that the Hebrew script doesn’t have vowels.)
  • Koester 2014 has more evidence on the use of isopsephy/gematria in ancient graffiti, inscriptions, and epigrams.

Revelation follows the same procedure as 3 Baruch. It transliterates the Greek Νέρων into the Hebrew script and calculates with Hebrew numerals.
‘Nero Caesar’ = נרונ קסר Nrwn Qsr
= 50 + 200 + 6 + 50 + 100 + 60 + 200
= 666
In favour of this interpretation:
  1. The word ψηφισάτω (‘let him calculate’) plainly implies isopsephy.
  2. The same reasoning explains the variant reading ‘616’. It is based on the Latin form of Nero’s name: Nrw Qsr. (See below.)
  3. A papyrus from Qumran contains Nero’s name in Aramaic in the form Nrwn Qsr, so the orthography inferred here is legitimate, though not common.
  4. Two ancient discussions retain a memory of the link between Revelation and Nero, though neither mentions this isopsephic interpretation. The Liber genealogus (5th cent.; 194-6 ed. Mommsen) is aware that the number of the beast refers to Nero. Victorinus of Pettau (3rd cent.; Commentary on Apocalypse 17: 11) knows of a link between Revelation and Nero, and (at 13.18) knows both the 666 and 616 readings, though he explains the two numbers differently.
  5. The Greek word ‘beast’ can also be transliterated into Hebrew as θηρίον > תריון trywn = 666. The form actually used in Revelation 13.18 is a possessive, θηρίου, which gives תריו tryw = 616. With both 666 and 616, the text is equating ‘Nero’ and ‘beast’ in the same way that Suetonius’ epigram equates ‘Nero’ and ‘killed his own mother’.
Against this interpretation, there really isn’t much to say. One objection might be that Caesar/Καῖσαρ would normally be written in Hebrew as קי‬סר (Qysr), not ק‬סר (Qsr) (for a total of 676). On the other hand, a number of 1st-3rd century documents do attest the spelling ק‬סר (Qsr). Other isopsephies have been suggested, but they’re not strong: an abbreviated form of the emperor Domitian’s name, Α. Και. Δομετ. Σεβ. Γε. (for αὐτοκράτωρ Καῖσαρ Δομετιανὸς Σεβαστὸς Γερμανικός) adds up to 666, while Γάϊος Καῖσαρ adds up to 616. The Domitian one with the abbreviation looks like a stretch, though. And while Gaius was very objectionable to Jews, he wasn’t especially interesting to early Christians. Nero is by far the strongest candidate.

(2) Nero as boogeyman. There is pretty good evidence that Nero was a figure of terror for late 1st century Christians. Early 2nd century sources blame him for the first persecution of Roman Christians, claiming that he made them scapegoats for the great fire of Rome in 64 CE. There is some doubt as to whether the persecution actually happened, but even if it didn’t, the 2nd century sources show that there was an early belief that Nero had persecuted the Christians. The sources are the Roman historian Tacitus (Annals 15.44.5-8; Latin text): he reports that Nero had Christians ‘clothed in the skins of wild beasts, and torn to pieces by dogs’, or ‘set up to be burned, so as to serve the purpose of lamps when daylight failed’. These gruesome tortures are quite likely fictional. Suetonius reports on a persecution under Nero too, though in his version it had nothing to do with the fire (Nero 16).

(3) Nero redivivus. Nero may have been a menacing figure for Christians, but overall he was a tremendously popular figure in the eastern empire. Suetonius claims that after his death some people acted ‘as though he was alive and was soon to return, to the great harm of his enemies’ (Nero 57). In other words, he was to have a second coming. We have reports of at least two, maybe three ‘false Neros’, who pretended that they were Nero, that Nero had never really died, and gained a significant following. (This wasn’t unique to Nero: Tacitus reports on other people being impersonated after their death, most notably Scribonianus in Histories 2.72.)

In 68 or 69 CE, shortly after Nero’s death, one false Nero appeared in Greece and, after landing on the Aegean island of Kithnos because of a storm, turned to piracy until he was captured and killed by Roman forces. (Source: Tacitus, Histories 2.8-9.)

A second false Nero, whose real name was Terentius Maximus, appeared during the reign of emperor Titus in 79-81 CE. He was active in Anatolia, Mesopotamia, and possibly Armenia, and sought refuge with the Parthians, apparently getting as far east as the river Euphrates. (Sources: Cassius Dio epit. 66.19.3 = Zonaras Epit. hist. 11.18; John of Antioch fr. 104 Müller.)

The third one is less well attested. The main evidence for him is Suetonius’ claim that there was a false Nero ‘twenty years after (his death), while I was a young man’ (Nero 57). This would suggest a date around 89 CE, during the reign of Domitian. However, Suetonius mentions the Parthians too: he states (1) that the Parthian king Vologaeses II (fl. ca. 78 CE) asked the Roman senate to honour Nero’s memory; and (2) that the Parthians supported the third false Nero ‘vigourously’ and ‘returned him reluctantly’. This sounds awfully like the second one, above: there could be a confusion of dates. One point in favour of Suetonius’ story is that Tacitus’ report, of the first false Nero above, refers to ‘results and attempts of others’, plural. That implies that he knew of at least three false Neros.

The neighbour of the beast: Pokey (Don McKellar) calls on a wannabe Devil (Earl Pastko) who can’t get his act together. (Highway 61, 1991)

Which number?

Some early copies give a different number. In particular, the Ephrem palimpsest (5th century) gives it as ἑξακόσιαι δέκα ἕξ: ‘six hundred and sixteen’.

This has a pretty strong heritage. As I mentioned above, Irenaeus knew both variants. And 616 works as an isopsephy of Nero’s name just as well as 666 does:
‘Nero Caesar’ = נרו קסר Nrw Qsr
= 50 + 200 + 6 + 100 + 60 + 200
= 616
The Latin form of his name, Nero, gives 616; the Greek form Nerōn (Νέρων) gives 666.

In recent years the 616 variant has enjoyed a lot of prestige, because of an ancient papyrus published in 1999. The papyrus, found at Oxyrhynchus and catalogued as P. Oxy. 4499, dates to the late 3rd or early 4th century. We have 26 fragments of the papyrus, containing bits of Revelation chapters 2 to 15. Fragment p is the relevant one.

P. Oxy. 4499 (= P115 Gregory-Aland, Trismegistos 65898), fragment p (Revelation 13.18-14.2)

This papyrus, all by itself, has generated two modern myths:
  1. This papyrus unambiguously supports the reading 616 (highlighted in yellow above: χιϲ).
  2. This papyrus is the earliest existing copy of the passage, or even the earliest existing copy of Revelation, and therefore its text carries more weight than any other manuscript.
These two myths, taken together, would suggest that 616 was definitely the original reading and 666 is a corruption of the text. However, they are both absolutely completely dead wrong.

First, here’s a transcription. I’ve copied the original editor’s text, with some minor adjustments because I think some of the letters are clearer than the original editor thought. (For technical side of how a papyrus text is represented, you can refer to this post from last month.)

Diplomatic text:
1               ]υ̣νψ̣ηφιϲ̣[
2                ]αρανου[
3                    ]ηχιϲ[
4            ]επ̣ιτοορ[
5              ]εχουϲα[
6               ]α̣υτουγε
7             ]α̣υτωνκαι
8              ]φω̣ν  ̣νϋ
Reading text:
1   [(13.18) ὧδε ἡ ϲοφία ἐϲτίν· ὁ ἔχων νο]ῦ̣ν ψ̣ηφιϲ̣[ά]-
2   [τω τὸν ἀριθμὸν τοῦ θηρίου, ἀριθμὸϲ γ]ὰρ ἀν(θρώπ)ου
3   [ἐϲτίν· καὶ ὁ ἀριθμὸϲ αὐτοῦ χξϛ] ἢ χιϛ. [ ]
4   [(14.1) καὶ εἶδον, καὶ ἰδοὺ τὸ ἀρνίον ἑϲτὸϲ] ἐπ̣ὶ τὸ ὄρ[οϲ]
5   [Ϲιών, καὶ μετ’ αὐτοῦ ρμδ χιλιαδε]ϲ̣ ἔχουϲα[ι]
6   [τὸ ὄνομα αὐτοῦ καὶ τὸ ὄνομα τοῦ π(ατ)ρ(ὸ)ς] α̣ὐτοῦ γε-
7   [γραμμένον ἐπὶ τῶν μετώπων] α̣ὐτῶν. (14.2) καὶ
8   [ἤκουϲα φωνὴν ἐκ τοῦ οὐ(ρα)νοῦ ὡϲ] φω̣νὴ̣ν ὑ-
8a  [δάτων ... ]
Here’s a translation of the standard text. Before we get on to the number, some technical notes (skip over them if you want):
1. In Roman-era papyri, there’s often no visual distinction between the numerals ϛ ‘6’ (the letter stigma) and ϲ ‘200’ (sigma). However, it’s easy to tell which is which by looking at their position.

2. The supplements stick to the standard text fairly rigourously, except for (a) numerals replacing numbers written out as words, in the missing parts of lines 3 and 5; (b) some normal abbreviations in the missing parts of lines 6 and 8. However, the result is that the number of letters per line is erratic, ranging from 28 (line 3) to 35 (lines 2 and 6). These could be evened up if we imagine some textual variants: for example line 2 could omit τοῦ, by analogy with ἀνθρώπου later in the line, bringing the line down to 32 letters; line 5 might have written out πατρὸς in full but omitted the repetition of τὸ ὄνομα, bringing it down to 31. Line 4 is also longer than I’d like, at 34 letters.

OK, on to line 3 and the numeral. Take a look back at the image above, where the number of the beast, χιϲ ‘616’, is highlighted. Notice the letter immediately before the number? It’s the letter eta, η. That letter raises an important question of its own.

The problem is that there is nothing that eta can possibly correspond to in the standard text of Revelation 13.18. What is it doing there?

There are only a few things an eta before a word-break is likely to be:
  • certain forms of a-stem (1st declension) feminine nouns and e-stem (3rd declension contracted) neuter nouns
  • a verb in a specific form (the aorist passive 3rd person singular)
  • ἡ ‘the’, in a specific singular form
  • δή ‘indeed’
  • ἤ ‘or’
Revelation 13.18 has no occasion for any 1st or 3rd declension nouns or aorist passive verbs, and you can’t have a singular ‘the’ before the number 616. So it’s going to be δή or ἤ. The text is either
[ ... ] indeed 616.
[ ... ] or 616.
Now, what do you imagine has happened here? See if you can work out the correct answer before going on. Because there is a clearly correct answer.

The answer: the eta before the number of the beast can only be the word ἤ ‘or’. The papyrus originally quoted both numbers, 666 and 616, as alternatives. Line 3, in translation, had
[ ... and its number is 666] or 616.
So, no more talk of 616 being attested earlier than 666, please. We can be absolutely confident that they were both written in this papyrus.

And one last thing: there’s still that myth that P. Oxy. 4499 is the earliest existing copy of Revelation. Well, it isn’t. It’s the earliest existing copy of some parts of the book. But not this part.

The misunderstanding may be partly due to this phrasing used on the Oxyrhynchus Papyri website:
The newest volume of Oxyrhynchus Papyri contains a fragmentary papyrus of Revelation which is the earliest known witness to some sections (late third / early fourth century).
If you stop paying after the word ‘witness’ you could easily come away thinking this is the earliest copy of Revelation. Even if you carry on reading, you might think it’s implied that it’s the earliest copy of the ‘616’ passage. But it ain’t. That honour goes to P47, above, which is solidly 3rd century, and which has the reading χξϛ ‘666’.

Even if we didn’t have P47, we’d have Irenaeaus’ discussion of the passage -- and he’s earlier than both papyri. He knew both the 616 and 666 variants, though he didn’t know the reason for either number. Both of them are very satisfactorily explained by the equations
Νέρων Καῖσαρ = נרונ קסר Nrwn Qsr = 666 = תריון trywn = θήριον ‘beast’

Nero Caesar = נרו קסר Nrw Qsr = 616 = תריו tryw = θηρίου ‘of the beast’

Further reading

Monday, 28 May 2018

The new papyrus of the gospel of Mark

I give here a copy of the new papyrus of the gospel of Mark: Oxyrhynchus papyrus 5345, or P. Oxy. 5345 for short. This isn’t new work (mostly), but is simply intended to be slightly more accessible than the article that has now appeared in The Oxyrhynchus Papyri 83 (2018). I also give some basic guidance on how to read papyrological notation, and a condensed selection of notes on the text.

The Egypt Exploration Society has made their publication open access: the publicly available PDF copy of the article is here.


In December 2011 Scott Carroll, until 2012 the director of the Green Collection, announced that he knew of a New Testament manuscript older than any other that was publicly known. In February 2012 the papyrologist Daniel Wallace cited the existence of a papyrus of Mark dating to the 1st century, as supporting evidence in a public debate.

These announcements were important because a first-century copy of any New Testament text would be of great intrinsic interest. It wouldn’t have much impact on any argument about the historicity of any aspect of the New Testament: nearly all biblical scholars date Mark to the 1st century anyway. So while the existence of a first-century papyrus would be very exciting, it wouldn’t be shocking.

Still the oldest: P. Rylands 457
In May 2018, this month, Dirk Obbink and Daniela Colomo published the papyrus in The Oxyrhynchus Papyri. It is now dated to the late 2nd or early 3rd century. It is definitely not the earliest known New Testament manuscript (that honour still rests with Rylands papyrus 457, mid-100s). But it is the earliest known copy of Mark.

In itself that doesn’t add anything to our state of knowledge: we already had external testimony of the text’s existence before 200. Still, the 2018 publication has again generated intense interest. But misinformation spreads virally where reliable information is difficult to come by. The real news here isn’t the text in the papyrus: there’s nothing very important there. The news is how knowledge about the papyrus has been abused.

On Wednesday last week, on 23 May 2018, Carroll stated that Dirk Obbink tried to sell the papyrus to the Greens in 2011 and 2013; that the papyrus was ‘in his [Obbink’s] possession’; and that Obbink had stated ‘unequivocally’ that it dated to the first century. Also on 23 May, Wallace stated that he had been required to sign a non-disclosure agreement before being allowed to see the papyrus in 2012; that he had been informed that an unnamed organisation had already bought the papyrus; and that he had been told by the same source that ‘a high-ranking papyrologist had confirmed that FCM [the Mark papyrus] was definitely a first-century manuscript.’

And the following day, the Egypt Exploration Society stated that the papyrus was and still is in their collection; that they have never sold it or intended to sell it; that it was excavated by Grenfell and Hunt’s excavations at Oxyrhynchus in the period 1896 to 1906, probably in 1903; and the first-century dating was ‘a provisional dating when the text was catalogued many years ago’.

This talk of the papyrus being for sale, and that it was definitely 1st-century, could in principle be the result of miscommunication. But what the unnamed informant did to Wallace was both deceptive and unethical: lying to Wallace about having bought the papyrus; compelling him to sign a non-disclosure agreement -- these things show a deep, reckless dishonesty. I imagine I share few opinions with Wallace, but I must commiserate with him for the plight he was placed in.

The text: introduction

The text below is designed to be read with the font New Athena Unicode installed. It is likely that the sublinear dots (U+0323) will appear incorrectly aligned in any other font.

The most important conventions for presenting a papyrus text are as follows:
  • A diplomatic text is a letter-by-letter transcription with no modern orthographic marks: so no spaces between words, no punctuation, no capitalisation, and no accents or diacritics except where they appear on the papyrus.
  • A reading text adds modern orthographic conventions and fills in missing text where possible. (Note that the publication in The Oxyrhynchus Papyri 83 combines the diplomatic and reading texts: in my opinion, greater clarity comes from keeping them separate.)
  • Square brackets ] [ represent edges of the papyrus.
  • A dot underneath a letter or space represents places where the papyrus has visible ink, but not enough to be certain (without further evidence) of which letter was there.
  • In the reading text, round brackets ( ) represent where an abbreviation has been filled in.
  • Recto refers to the ‘front’ side of the papyrus, verso to the ‘back’.
Notes specific to this papyrus:
  • On orthography:
    • Lunate sigma ϲ is used throughout, rather than the modern forms σ (medial) and ς (final). The latter variants belong to Byzantine and later orthography; ancient papyri regularly use the lunate variant.
    • Iota adscript is not written in this papyrus: see verso 2.
  • Line numbering follows the edition of Obbink and Colomo, but I add an extra line above recto 1: see my note on recto 1.
  • The papyrus is a fragment of a leaf from a codex. That is to say: the book was in the form of multiple leaves bound at the margin, and not a scroll. That is unsurprising, as even the earliest Christian texts are nearly universally in codex format. Outside Christian texts, the codex was still the minority format before 200 CE.
  • Obbink and Colomo estimate that the leaf held 20 lines per side, and approximately 28 letters per line. These figures are entirely normal. They estimate a total written area of 9.4 × 12 cm, plus margins; the surviving fragment is 4.4 × 4 cm. The fragment comes from the bottom of the leaf. Part of the bottom margin is preserved, with about 1.5 cm space to the text. The layout is consistent with the top of the recto side beginning at Mark 1.1, possibly with a heading.

NA28 = Nestle-Aland, Novum testamentum graece, 28th edition
Obbink-Colomo = D. Obbink and D. Colomo (2018), ‘5345. Mark I 7-9, 16-18’, The Oxyrhynchus Papyri 83: 4-7.


0                        ]  ̣[
1                        ]μι̣[  ]  ̣[
2                        ]τ̣ω̣ν̣[  ]π̣[
3       ]ωεβαπτ̣ιϲαυμα̣ϲ̣υδ̣[
4        ]ιϲειϋμ̣[  ]ϲ̣π̣ν̣ιαγ̣[
5          ]ιναιϲ̣[  ]ϲημε[

0  ‘[ ... ἔρχεται ὁ ἰϲ]χ̣[υρότερόϲ]
1  [μου ὀπίϲω μου, οὗ οὐκ εἰ]μὶ̣ [ἱ]κ̣[ανὸϲ κύ]-
2  [ψαϲ λῦϲαι τὸν ἱμάντα] τ̣ῶ̣ν̣ [ὑ]π̣[οδημά]-
3  [των αὐτοῦ. ἐγ]ὼ ἐβάπτ̣ιϲα ὑμᾶ̣ϲ̣ ὕδ̣[ατι],
4  [αὐτὸϲ δὲ βαπτ]ίϲει ὑμ̣[ᾶ]ϲ̣ π̣ν̣(εύματ)ι ἁγ̣[ίω.’ καὶ]
5  [ἐγένετο ἐν ἐκε]ίναιϲ̣ [ταῖ]ϲ ἡμέ[ραιϲ ἦλθεν ... ]

Critical apparatus:
1 Obbink-Colomo ]μ̣  ̣[ ]  ̣[
3 prob. orig. ϋμαϲ ϋδατι: tremata not preserved in papyrus
4 NA28 ἐν πνεύματι

‘[ ... One stronger than I comes after me, and I a]m [not w]o[rthy to bend down and undo the clasp] of his [s]a[ndals. ]I baptised you with wa[ter, but he will bapt]ise y[o]u with the [holy] spirit.’ [And it happened, in th]ose da[ys (Jesus) came ... ]


1                   ]  ̣  ̣[
2         ]τ̣ηθαλ̣α[
3        ]ναυτοιϲδευτ̣ε̣ο̣π̣[
4       ]ϋμαϲγενεϲθαιαλι[
5        ]ϲ̣αφεντε[  ]τ̣αδικ[

1  [ ... τὸν ἀδελφὸ]ν̣ Ϲ̣[ίμωνοϲ ἀμφιβάλλον]-
2  [ταϲ ἐν] τ̣ῆ θαλ̣ά[ϲϲη· ἦϲαν γὰρ ἁλιεῖϲ].
3  [καὶ εἶπε]ν αὐτοῖϲ· δεῦτ̣ε̣ ὀ̣π̣[ίϲω μου, καὶ]
4  [ποιήϲω] ὑμᾶϲ γενέϲθαι ἁλι[εῖϲ ἀνθρώπων].
5  [καὶ εὐθὺ]ϲ̣ ἀφέντε[ϲ] τ̣ὰ δίκ[τυα ... ]

Critical apparatus:
1 Obbink-Colomo suggest ἀ̣δ̣[ελφὸν αὐτοῦ (omitting Ϲίμωνοϲ) or ἀδελϕὸ]ν̣ τ̣[οῦ Ϲιμωνοϲ
2 θαλ̣α[: Obbink-Colomo θαλ̣α̣[
3 NA28 αὐτοὶς ὁ Ἰησοῦς
4 Obbink-Colomo ἁλι̣[εῖϲ or ἁλε̣[εῖϲ

[ ... (Jesus saw Simon, and Andrew) the brothe]r of S[imon, fishing on both sides in] the se[a; for they were fishermen. And he sai]d to them, ‘Come aft[er me, and I shall make] you become fish[ers of people.’ And immediate]ly they released the nets [ ... ]


Recto 1. Obbink-Colomo express uncertainty over , on the grounds that some manuscripts omit κύψαϲ, and that would mean the line-lengths would be off. However, is clearly legible, with the same letter-form as in verso 4.

Above a dot of ink is visible at the edge of the papyrus which could be part of almost any letter near the start of recto 0 ἰϲχυρότερόϲ.

Recto 4. ἐν is missing in many later manuscripts as well. It makes little difference to the meaning. βαπτίσει ὑμᾶς πνεύματι ἁγίωι (‘he will baptise you with the holy spirit’), with a plain instrumental dative, means much the same as βαπτίσει ὑμᾶς ἐν πνεύματι ἁγίωι, with instrumental ἐν.

The abbreviation of πνεύματι (‘spirit’ in ‘holy spirit’) is a nomen sacrum (‘sacred name’). This is a feature of many Jewish texts and ancient Christian texts: the idea is to avoid writing the name of God in full, or even the word ‘god’, and instead use an abbreviation. A nomen sacrum regularly appears with overlining, as it does here.

Verso 1. The ink over the second α of 2 θαλ̣ά[ϲϲη looks to me like it makes most of a circle, which would indicate ε, ο, or ϲ. If it is Ϲ̣[ίμωνοϲ, that makes verso 3 a little short at 27 letters: I guess that is why Obbink-Colomo try to read the letter as δ or τ.

Verso 2. The second α of θαλ̣ά[ϲϲη looks clear to me.

Verso 3. Jesus’ name is omitted. This does not significantly change the meaning, and it is not necessarily another case of a nomen sacrum (see note on recto 4, above). That would be the situation if the papyrus had replaced ‘Jesus’ with an abbreviation. Instead, the name is simply absent, so that the meaning is ‘he said’ rather than ‘Jesus said’.

However, Obbink-Colomo suggest a possibility that the omission may be an error resulting from an earlier copy with a nomen sacrum for Jesus. If so, the earlier copy would have read ειπεν αυτοιϲ ο Ιϲ. Without spaces and capitalisation this would appear as ειπεναυτοιϲοιϲ. -οιϲοιϲ could have seemed to be an erroneous repetition, and the scribe of this papyrus might have ‘corrected’ it by omitting the second -οιϲ. Wallace also points out this possibility.

Verso 4. The ink after λ looks like a perfectly vertical stroke, and the blob of ink at the bottom of the stroke is similar to that in verso 3 αὐτοῖϲ, so it looks to me that ἁλι[εῖϲ can safely be read.

Wednesday, 23 May 2018

The authoress of the Odyssey

In the 1890s Samuel Butler argued that the Odyssey was composed by a young unmarried girl living in Trapani, western Sicily, in the 11th century BCE.

The basic idea is that the character of Nausicaa, in Odyssey books 6 to 8, is an authorial self-insertion. Nausicaa is a woman, young, single, and doesn’t live in mainland Greece or Ionia. Therefore the author of the Odyssey must be all those things too. (Demodocus the blind bard represents Homer, boring the author’s socks off with his interminable recitations of the Iliad.)

Trapani, Sicily (Source: Google Streetview)

Butler first presented the idea in a lecture on 30 January 1892, at the Working Men’s College in London; then in a series of magazine articles and pamphlets (1892a-e, 1893); and a few years later in a book, The authoress of the Odyssey (1897).

The reception of Butler’s idea

Butler’s argument has had great longevity, and at the same time, has persuaded almost no one. Butler himself predicted that he’d encounter an entrenched opposition (‘How can I expect Homeric scholars to tolerate theories so subversive’, etc., 1897: 3). He imagined he’d find academics insisting that epic poets must have been men. A much more serious problem is that he starts with a complete absence of evidence, and concludes something incredibly specific.

And, in regard to Trapani, something impossible. Butler was infuriated when he was told that Trapani didn’t exist when the Odyssey was composed. He insisted that his argument proved the archaeologists wrong, so their ‘views must be reconsidered’ (1893: 13)! Just to be clear: Greeks didn’t start to colonise Sicily until the late 700s BCE, and they never held western Sicily. The west belonged to the non-Greek Elymi until Roman times.

Still, the idea gets trotted out every generation or so. First by the classicist Benjamin Farrington (1929); then in one of Robert Graves’ less-known novels, Homer’s daughter (1955); then by Andrew Dalby (2006). Its longevity probably owes a lot to Butler’s translations of Homer, which were the first English translations of Homer to use a modern prose style. Plus, Butler has celebrity status: he’s still famous for his classic satirical novel Erewhon (1872). Dalby’s version, too, has become better known than you might expect, thanks to his very successful publicity campaign.

Excursus: the authoress of the ‘Rediscovering Homer’ and ‘Andrew Dalby’ articles on Wikipedia
Dalby and his book Rediscovering Homer (2006) both have substantial Wikipedia articles devoted to them. At the time of writing Dalby’s is the only non-fiction book about Homer to have a Wikipedia article. So you might imagine this indicates that Dalby and his book have had a major impact.

And that would be wrong. Dalby is a prolific Wikipedia-editor. He created the book article himself. It’s strictly self-promotion. Outside the ‘Rediscovering Homer’ article, Wikipedia has fifteen citations of the book -- but fourteen of them were inserted by Dalby (1 2 3 4 5 6 7 8 9 10 11 12 13 14, between May 2006 and May 2007; no. 15 is someone else’s doing, and dates to 2010).

So these articles and citations don’t in any way indicate acceptance of Butler’s (or Dalby’s) ideas. Look to the reviews instead. Here’s what they say about Dalby: ‘inventive’ but ‘weak’ and unproveable (Kelly); ‘speculations ... inadequate argument and dicey assertions’ (Lateiner); ‘lively’ but ‘thinly grounded’ (Lentini, Classical Bulletin 82.2 [2006]: 221-3); ‘enjoyable’ but ‘unconvincing’ (Loney).

But wait, there’s more. The ‘Andrew Dalby’ article is self-promotion too. For that one, Dalby tried to be sneaky: he used a sockpuppet account called ‘Charles David Douglas’ to create it. The entire Wikipedia career of ‘Charles David Douglas’ lasted just over two hours, and in the surrounding five days Dalby’s main account edited most of the articles that ‘Douglas’ worked on, 1 2 3 4, including the ‘Andrew Dalby’ article, ten minutes after ‘Douglas’ created it; ‘Douglas’ displayed an intimate knowledge of Dalby’s employment history in the 1970s and 1980s, his unpublished essays, and even ‘informal’ work; and more than half the article was (and still is) publicity for Dalby’s books.

Incidentally, Wikipedia does have policies on conflicts of interest, spam, and sockpuppet accounts which look like they ought to prohibit all of this (links are to 2006-2007 versions of the policies). Not very effective policies, maybe?

Even setting Butler aside, gender in the Odyssey continues to be a hot topic. For lots of reasons. In the 1990s, literary criticism was popping with ideas about the topic, and that excitement hasn’t quite worn off yet. In 2017 Emily Wilson published a new high-profile translation of the Odyssey, its first translation into English by a woman (not the first into any modern language: a French one by Anne Dacier appeared in 1716). Also in 2017, Mary Beard’s Women & power: a manifesto opened by citing Odyssey book 1 as a classic example of women’s speech being suppressed.
I want to start very near the beginning of the tradition of Western literature, and its first recorded example of a man telling a woman to ‘shut up’; telling her that her voice was not to be heard in public. I’m thinking of a moment immortalised at the start of Homer’s Odyssey ... Penelope ... [finds a bard] singing about the difficulties the Greek heroes are having in reaching home. She isn’t amused, and front of everyone she asks him to choose another, happier number. At which point young Telemachus intervenes: ‘Mother,’ he says, ‘go back up into your quarters, and take up your own work, the loom and the distaff ... speech will be the business of men, all men, and of me most of all; for mine is the power in this household.’ And off she goes, back upstairs.
-- Beard 2017: 1
(= London review of books, March 2014)
Some people will want to make excuses for the Odyssey, arguing that its misogyny is a product of its time. While true, that doesn’t mean the misogyny isn’t there. Whether or not that damages the quality of the Odyssey is in the eye of the reader.

Mt Sunday (foreground right), upper Rangitata River valley, a stone’s throw from ‘Erewhon Station’. Butler spent the first part of the 1860s farming sheep here, and the opening section of Erewhon is based on it. More recently, Mt Sunday served as the set for Edoras in The Lord of the Rings films. (Source:

Butler on Penelope

Excuses or no excuses, Butler’s Authoress of the Odyssey makes a weird contrast with this Penelope scene. Because the core of Butler’s argument is that the epic has a firmly and exclusively feminine sensibility.
What, let me ask, is the most unerring test of female authorship? Surely a preponderance of female interest ... Hence if in any work the women are found to be well and sympathetically drawn, while the men are mechanical and by comparison perfunctorily treated, it is, I imagine, safe to infer that the writer is a woman.
-- Butler 1897: 105
‘The Odyssey takes more interest in women than the Iliad, therefore the Odyssey was written by a woman.’ No, that doesn’t make a lick of sense. Can he really have meant to say something that daft? Yes, yes he did. Here’s the next page.
Men seem unable to draw women at all without either laughing at them or caricaturing them ... I doubt whether any writer in the whole range of literature (excepting, I suppose, Shakespeare) has succeeded in drawing a full length, life-sized, serious portrait of a member of the sex opposite to one’s own.
-- Butler 1897: 106
Make no mistake: even though this is halfway through the book, it’s the foundation for the whole thing. His argument is this: in any literary work the characters, physical settings, and interests necessarily represent people, settings, and interests in the author’s everyday life. Nausicaa may be Butler’s motivation, but this is his rationale.

Butler in 1858, two years before he laid claim to farmland in inland Canterbury. (Source:
Let’s imagine his logic somehow worked, and follow up his line of thought. If the Odyssey really had ‘a preponderance of female interest’, with women ‘well and sympathetically drawn’, you’d assume the poet would be keenly interested in exploring what it is that makes Penelope tick. For example: after the incident with Penelope and Telemachus, you’d expect the narrator to track Penelope a bit, maybe follow her upstairs to her private quarters, maybe spend some time describing how she copes with her plight, and her efforts to keep peace with the army of violent suitors inside her house.

Nope. We don’t get any glimpse of Penelope’s inner life until she talks about it in book 19. Even there, many readers doubt she means anything she says. After the incident in book 1, the narrator stays by Telemachus’ side until near the end of book 4; book 2 focuses entirely on Telemachus, the suitors, and the male population of Ithaca, and the motivations driving each of them. (Men are ‘perfunctorily treated’? Hah!) Penelope only reappears briefly at the end of book 4. When we’re told about the problems caused by the suitors, it’s Telemachus talking to Nestor or Menelaus, or Eumaeus talking to the disguised Odysseus.

Butler’s argument also revolves around how he thinks the ideal Victorian lady -- personified as Nausicaa -- ought to feel about the world. At page 107 he invites his reader to ask any single woman that ‘he’ knows whether they like having men about the house. Chapter 4 is about the poet’s ‘jealousy for the honour and dignity of women’. Chapters 5 and 6 argue that the poet has ‘whitewashed’ Penelope out of feminine sympathy, and that in reality Penelope was ‘scandalous’, ‘an artful heartless flirt’. At page 136 he states, ‘The authoress of the Odyssey is never severe about theft’, with the assumption that that’s a feminine perspective. Chapter 7 is about ‘Further indications that the writer is a woman -- young -- headstrong -- and unmarried.’

Butler argues his point with a classic case of victim-blaming. If Penelope really wanted to get rid of the suitors, he says, ‘[o]ne would have thought all she had to do was to bolt the doors as soon as the suitors had left for the night’ (page 126). Or again: ‘The reader will have noted that on this occasion the suitors seem to have been in the house after nightfall’ (page 129).
A man ... would not have made the suitors a band of lovers at all. He would have seen at once that this was out of the question, and would have made them mere marauders, who overawed Penelope by their threats ...
-- Butler 1897: 128
The weird thing is that the suitors are mere marauders. Very very obviously so. They try to assassinate Telemachus. They threaten violence constantly. They commit actual violence throughout books 17 and 18. They force unwilling people into a fist-fight, with promises of torture and mutilation for the loser. They’re brutal monsters.

But no. For Butler, the suitors are romantic lovers: their morals are imperfect, but their love is pure.

And Penelope is leading them on. Butler doesn’t use the word ‘slut’, but if he did, it would be completely in place. He’s an archetypal ‘nice guy’. If Penelope had been able to keep out the suitors with a bolted door, Butler would be the first to complain that she was ghosting her former lovers.

I find it gobsmacking that a noted feminist writer has characterised Butler’s book as ‘[t]he most elegant, ingenious, imaginative and imagination-respecting of feminist gestures’ (Brophy 1985: 814). It’s worrying to see any critic regarding him as a ‘careful reader’ (Kendal 2016: 1). The Odyssey has loads of misogyny, as Mary Beard points out -- but maybe not quite as much as Butler has.

Penelope’s perspective, and that of the slaughtered maids: Margaret Atwood’s The Penelopiad, from a trailer for the 2013 stage production by the Nightwood Theatre, Toronto. (Source: YouTube)

Penelope’s perspective

In the real Odyssey, we only get occasional fragmentary glimpses of what it is that drives the women of the poem. Often that’s exactly the point: we’re not supposed to see what makes Penelope tick. People tell stories about her, they describe her, they interpret her: everything is from someone else’s viewpoint.
‘A night will come when a hateful marriage meets me,’ (says Penelope,)
‘my end, when Zeus takes away my joy.
But ... this isn’t how suitors do things -- it never has been --
when they want to court a noble woman, a wealthy man’s
daughter, and compete with one another.
They bring oxen and fat sheep,
as a feast for the girl’s family. They give fine gifts.
They don’t devour someone else’s livelihood for free!’
    So she spoke. And long-suffering noble Odysseus laughed
because she coaxed gifts out of them, and she charmed their hearts
with soft words while her mind intended something quite different.
-- Odyssey 19.274-283
Penelope flatly declares that she’s going to give in and marry one of the suitors. Is it really a trick? Can we be sure? If we can, it isn‘t because of anything she says. It’s only because Odysseus is there to explain that she didn’t really mean what she said.

So this is a little more involved than your everyday misogyny: it’s essential to her character. In the Odyssey there’s no such thing as Penelope’s perspective, only perspectives on her. Entire books have been written about how there is no inner psychology there, and how her character is created by what other characters think about her motives, and by her role in the plot. (It’s a pity some of those books are notoriously difficult to read: the underlying idea is really solid.)

If Butler had really been a careful reader, he might have argued that this characterisation of Penelope, with her character as a kind of blind spot in the story, shows a more powerful insight into gender relations than any supposed whitewashing could. (More powerful than any of his actual arguments, anyway.)

When the poet keeps Penelope’s inner life obscure, it isn’t easy to tell if it’s because the poet is being misogynistic -- a creature of his time, if you prefer --, or because the whole point of Penelope is that her inner life is inscrutable.

It’s a bit of both. Take the ‘soul-summoning’ scene in book 11: Agamemnon’s ghost advises Odysseus as follows. (a) You must never under any circumstances trust Penelope; (b) Penelope is a really good egg and she’ll always stand by you; and (c) you must never ever trust Penelope ever. Mixed messages much?
‘Never be nice to your wife!
Don’t tell her everything you know.
Say some things, keep some hidden.
But your wife won’t kill you, Odysseus ...
Still ... arrive in your home country secretly, not openly.
Because there’s no trusting women these days.’
-- Odyssey 11.441-456
What a testimony to the Odyssey’s ‘female interest’; what a great benchmark for virtue. ‘Your wife is so virtuous that she isn’t actually going to murder you.’ Well ... yay, I guess? Here we’ve got someone telling us how to think about Penelope, while at the same time stressing that women are scary.

Agamemnon’s words, and other similar passages, tend to suggest the obscurity of Penelope’s inner life isn’t just an insight into the feminine condition in a patriarchal society: it’s also a refusal to engage with the feminine condition.

I don’t think that says anything about the author, though. That would be the same kind of fallacy as when Butler insists that the epic is transparent and the author’s identity is plainly visible through it. Butler’s argument says a lot more about Butler than it does about Homer.
He seems to have rewritten an ancient epic as a Victorian novel ... Along with other Victorians, he is skeptical of Homer worship, but he resents the German classical scholars who deny the great poet his identity. His ‘authoress,’ an inventive restoration of that identity, has the appearance of an exotic, but she is very much at home in the Victorian tradition. ... Butler regenerates a classic as the offspring of a modern temperament: the artist as a young Victorian lady.
-- Booth 1985: 865-867
Butler’s modern critics seem to sway between regarding him as a crackpot (Whitmarsh 2002 -- but maybe I’m interpreting Whitmarsh uncharitably) and an iconoclast (Beard 2007). I think Booth, above, is closer to the mark: Butler wasn’t just an iconoclast, he was a Victorian gentleman’s image of what an iconoclast ought to look like.

That’s not to say he didn’t mean his argument seriously. His dedication to it over several years strongly suggests he was entirely serious. But that doesn’t mean we have to go re-writing the history of Sicily at the drop of Butler’s stovepipe hat.


Wednesday, 25 April 2018

Ancient Greeks climbing Mount Olympus

Why didn’t ancient Greeks just climb up Mount Olympus and notice that there were no gods pottering around up there?

Mount Olympus as seen from the Dion Archaeological Park, Greece

This is a pretty common question on Q&A websites. The answers are usually wrong. Here’s a selection (all sic):
Quora, Jan. 2017:
Why didn't the Greeks climb Mt Olympus (Mytikas) and see that there weren't any Gods up there?
Answers: ‘it’s a really hard mountain to climb’; ‘hubris’; ‘some did climb Mt Olympus and felt the presence of the gods’; ‘they were afraid to climb it’.

Quora, August 2016:
Did the ancient greeks ever scale to the summit of Mount Olympus?
Answers: ‘a really difficult mountain to climb’; first climbed in 1913; ‘would be kind of taboo’; there were actually three Mount Olympuses, ‘one in sparta, one in thay and one outside athens’.

Stack Exchange, June 2016:
Did the ancient Greeks ever climb Mt. Olympus?
Answers: one correct answer, but also plenty of ‘it’s impossible to ascertain for sure’; ‘people didn’t climb mountains until the 19th century’; ‘Olympus was first climbed in 1913’.

The Straight Dope, February 2009:
Why did the Ancient Greeks not ever climb Mount Olympus?
Answers: ‘it would be blasphemy’; anyone who did climb ‘probably would have "seen" [the gods]. A sceptic's account would have been ignored’; they probably did and ‘concluded theiy were in the wrong place’.

‘The home of the Greek gods [was] on the Mytikas peak’; ‘in all regions settled by Greek tribes, the highest local elevation tended to be ... named [Olympus]’; ‘Ancient Greeks likely never tried to climb the two main peaks’.
Mt Olympus’ highest peaks:
  1. Mitikas (2918 m)
  2. Skolio (2911 m)
  3. Stefani (2909 m)
  4. Skala (2866 m)
  5. Agios Antonios (2817 m)
The answers that focus on the idea of gods as ‘metaphors’, or a distinction between the divine and visible worlds, are the best ones here. They’re a pretty good take on the topic.

But even they miss a reasonably important point. And that is that we know, perfectly well, that ancient Greeks did climb up Mount Olympus. Plutarch (2nd century) and St Augustine (4th-5th century) report on annual pilgrimages up the mountain. Ceramic plates have been found dotted around the various plateaus and passes at the top, similar to ones found at Dion, a small city to the northeast which was the home of the ancient sanctuary of Zeus Hypsistos (Zeus the Highest). The clearest evidence comes from the Agios Antonios peak, where burned sacrifices were offered to Zeus and various other religious offerings deposited. Agios Antonios is somewhat separated to the south of the highest cluster of peaks, running Skolio-Skala-Mitikas-Stefani in a curve from southwest to northeast.

(Wikipedia does report the archaeological finds on Agios Antonios, to be fair. But the article takes the additional leap that the Agios Antonios finds somehow imply that the ancients didn’t climb up the higher peaks -- and that obviously makes no sense at all.)

The weather station on Agios Antonios, where remnants of ancient sacrifices to Zeus were found in 1961. (Source:

The textual evidence comes from ancient discussions of the fact that cloud cover is often lower than high mountain peaks. One ancient interpretation of this was that clouds and wind are confined to lower altitudes: and this leads to some factoids about mountain-tops. Here’s Plutarch:
For people who have placed ash on top of some mountains, or have left it behind after sacrifices there, have when investigating many years later found that it was still lying as they left it. ... Plutarch reports that letters, too, remained from one ascent of the priests to the next on Olympus, in Macedonia.
-- Plutarch fr. 191 Sandbach, reported by Philoponus, On Aristotle’s Meteorologica i.82
And Augustine:
In that air [at high altitudes] they say that clouds to not gather and no stormy weather exists. Indeed where there is no wind, as on the peak of Mount Olympus, which is said to rise above the area of this humid air, we are told, certain letters are regularly made in the dust and are a year later found whole and unmarred by those who climb that mountain for their solemn memorials.
Where Augustine’s Latin has ‘letters’ (litteras), Plutarch’s Greek has grammata, which can mean either ‘letters’ or ‘writings’. ‘Writings’ is the correct interpretation -- we’re not talking about letters scrawled into the ashes, like Augustine thought! -- and indeed the archaeological finds on Agios Antonios do include some inscriptions, including two dedications to ‘Olympian Zeus’.

The distinction Augustine is making between different types of air is a standard feature of ancient cosmology. As sea level is thick, humid aēr; higher up is the clear, fiery aithēr. (These are the Greek terms.) These were regarded as distinct layers or spheres surrounding the earth, with the fieriest layer of aithēr in the neighbourhood of heavenly bodies like the sun. Augustine mentions in a few other places that rain doesn’t fall on the summit of Olympus, and in one place he attributes it to ‘one of the pagan poets’. The setting on Olympus, and Augustine’s misunderstanding of grammata, both suggest that the story originates in a Greek source. The most likely ‘pagan poet’ is Homer, Odyssey 6.41-46, which reports that Olympus
is not shaken by winds, nor ever wet by rain,
and snow does not come near, but pure cloudless aithrē (= aithēr)
is spread out, and a white brightness plays on it.
The story is fairly likely to have come to Plutarch and Augustine from a commentary on Homer: maybe the 1st century BCE scholar Didymus.

The peaks of Mt Olympus, rendered in Google Earth looking from the west, with slopes exaggerated by a factor of 1.5. Notice the walking tracks up gentle slopes from the south-west (there are more tracks on the other sides).

The main archaeological evidence was discovered in 1961, when the Aristotelian University of Thessaloniki was building a meteorological observatory on Agios Antonios. Excavators found a thick layer of ash, with ancient ceramic vessels, inscriptions dedicated to ‘Olympian Zeus’, and coins. Most of the evidence on that spot appears to date to the 300s CE, when Christianity was well established in the Roman empire, and not long before pagan religion was banned by emperor Theodosius. One older coin has also been found, from the 200s BCE, which may or may not suggest a long-standing tradition.

Now, there are some half truths among the answers above. For example it’s true that, as Wikipedia reports, the first recorded climb of Mitikas was in July-August 1913. But even Wikipedia emphasises that that’s just the first recorded climb. But it does at least appear to be true that no modern-style mountaineers had been up Mitikas before that date.

However, there are also outright falsehoods. Here are some corrections:
  • Mt Olympus is not a hard climb. It is free of snow for five months of the year, it is not high enough for oxygen deprivation to be a concern, temperatures are normally above freezing in summer, and Agios Antonios in particular is a basic walk for someone who’s moderately fit. A return trip can in principle be done in a single day from the trailhead. People even do organised runs up the mountain. Mitikas and Stefani are more challenging, in that they actually involve some climbing.
  • There is no evidence to suggest any taboo or blasphemy.
  • There is no evidence to suggest that people who did climb to the top believed they had seen or ‘felt’ the gods there. (It’s not impossible, but we have no testimony on the subject.)
  • It’s true that there are several Mt Olympuses, but they’re not relevant to the question, and they aren’t generally the ‘highest local elevations’. The Olympus near Athens, out of Anavyssos, is under 500 m high; the one just outside Sparta is only about 150 m. (Not even a very significant hillock, by Greek standards!)
  • There is no evidence to suggest that it was specifically Mitikas that was regarded as the home of the gods.
The 16th century chapel on Profitis Elias (source: YouTube)

On that last point -- whether one peak in particular was especially sacred to Zeus -- even Agios Antonios isn’t the strongest candidate. That honour should probably go to the northernmost prominence, Profitis Elias (2803 m).

First: Elias, the Greek form of Elijah, supplanted Zeus in many parts of the Greek world, and is strongly associated with mountain-tops. Several peaks that were called ‘Olympus’ in antiquity are now named for Elias.

Second: the archaeological evidence mentioned above comes from Agios Antonios, the main southern peak, but the urban sanctuary of Zeus Hypsistos was on the northeast side of the mountain, in the city of Dion. From Dion, Profitis Elias is the nearest (14 km) and most prominent peak. Physical remains of the sanctuary of Zeus at Dion were discovered in 2003, and included a nearly complete cult statue of Zeus enthroned. (Dion itself is named after Zeus: oblique forms of ‘Zeus’ take the stem Di(w)-, e.g. Dios ‘belonging to Zeus’.)

Third: in the mid-1500s, St Dionysius of Olympus built a chapel dedicated to Elias on Profitis Elias, which is still standing. Reportedly the chapel was built on the site of an older ruin. I’m guessing that report comes from records kept by the Monastery of Agios Dionysios, on the north-eastern slope of the mountain. We don’t know how old the ruin was, unfortunately, and it seems unlikely that anyone will ever find out: no one’s going to go digging up a 16th century chapel of intense interest to a nearby monastery, and the highest-altitude chapel in Greece to boot. But even without physical remains being dug up, there’s a fairly strong suggestion of a long-standing religious significance to the site.

Sanctuary of Zeus Hypsistos, Dion (source: Wikipedia)

Just as an interesting end-note: we have on record an ancient measurement of the height of Mt Olympus, and it’s intriguingly accurate. Plutarch quotes an inscription at Pythion, on the west slope of the mountain, which stated that one Xeinagoras, son of Eumelus, measured the height of the mountain as follows:
The sacred height over Apollo’s Pythion
    of Olympus’ peak, in vertical measure,
is ten full stadia, and in addition
    a plethron less four feet in size.
Eumelus’ son placed a measure of the distance,
    Xeinagoras. Farewell, lord, and grant your blessings.
We’ve got no idea what methodology Xeinagoras used, but the measurement isn’t too shabby. A plethron is 100 feet, and a stadion is 600 feet, so this comes out to exactly 10.16 stadia. The length of a stadion is not exact, but usually ranged between 181.3 and 192.25 m (figures quoted by the New Pauly); conversion to Roman measurements allowed greater precision, which made the stadion between 184.4 and 185.1 m. Taking the mean of those two figures, 184.75 m, Xeinagoras’ measurement comes out as 1877 m over the elevation of Pythion. Pythion is at about 700 m, so this gives a total of 2577 m for the summit of Olympus. It’s definitely not perfect -- 341 m short of the elevation of Mitikas, 226 m short of Profitis Elias -- but not half bad considering that Xeinagoras probably didn’t even have access to Roman surveying techniques, let alone modern ones.

Further reading

Monday, 16 April 2018

Why are there no Romans named ‘Quartus’?

Quintus, Sextus, and Decimus were common Roman personal names, or praenomina. They come from numbers: they mean ‘fifth’, ‘sixth’, and ‘tenth’.

But only some numbers are represented. Why don’t we see Romans named ‘Primus’, ‘Secundus’, ‘Tertius’, or ‘Quartus’? Or for that matter ‘Septimus’, ‘Octavus’, or ‘Nonus’?

Neil Gaiman’s Stardust. (Sorry, too late for a spoiler alert.) Gaiman bucks Roman custom and has the princes of a royal family named for Latin ordinal numbers from ‘first’ to ‘seventh’. Here they appear as ghosts in the 2007 film based on the novel: left to right are Quintus, Tertius, Primus, Septimus, Secundus, Quartus, and Sextus. Septimus, in the middle, is one of the main antagonists in the story.

Today’s post isn’t really a debunking of a popular myth. An unpopular myth, maybe. I’m posting it because it’s something I just learned this morning, and it shocked me. It’s one of those things that’s staring you in the face all the time when you’re reading about the Greco-Roman world. So when I found out the true explanation, I felt a little bit betrayed -- as though it was something I ought to have known all along.

I guess it’s hard to find the time to get around to thinking about why Romans had names that meant ‘fifth’, ‘sixth’, and ‘tenth’. If you do think about it, you’re likely to make the same assumption that I did: that children were named for the order in which they were born. The 1st son would be Primus, the 2nd Secundus, the 3rd Tertius, and so on.

But that isn’t the case. If it were, we’d see corresponding names for the first to fourth children. And they just don’t exist. We do find Primus, Secundus, Tertius, and Quartus as regular cognomina -- official nicknames -- but not as personal names, and not at an early period. They start to pop up in the imperial period, and they’re not in Rome: they appear in Celtic contexts, which tends to suggest contamination from Celtic naming customs (Petersen 1962: 349 n. 6).

What’s the solution, then? Should we assume that the first four sons would get ‘real’ names, like Marcus, Titus, Publius, and so on? And when the parents got to their fifth child, suddenly they’d be all like ‘Hey let’s start using numbers now.’

Nope. The scholarly consensus is that these names originally came from the names of the months in which they were born.

If that surprises you, you’re not alone. I was startled too. It was originally suggested by the 1st century BCE scholar Varro, and it appears he was dead right. The standard modern discussion is 56 years old, Petersen 1962, but Petersen’s argument hasn’t been superceded. If anything the argument has been strengthened by parallels that have been found in other ancient Italian languages.

There are a couple of complications. First: Quintus and Sextus don’t sound like month names. However, prior to the lives of Julius Caesar and the emperor Augustus, the Romans had different names for the 7th and 8th months of the year: Quintilis (‘fifth-ilis’) and Sextilis (‘sixth-ilis’). (see this previous post from March for more details, and for discussion of why the months’ names don’t match up to their position in the year.)

Second: it’s not just the number-months. March, May and June aren’t named for Latin numbers, but children born in those months got names related to the months anyway: not Martus, but Marcus; Maius is a rare name, but it exists; and June gave Iunius both as a praenomen and as a gentilician (family) name.

And third, there’s a semi-regular pattern of praenomina ending in -us with corresponding family names ending in -ius: hence Marcus ≈ Marcius. This helps fill in the gaps with some of the numbers. We don’t see Octavus as a praenomen, but we do see Octavius as a family name {edit, six weeks later: also ‘Octavianus’ as a praenomen, but that looks like it’s modelled on ‘Octavius’, not on ‘October’}; Iunius is rare as a praenomen, but common as a family name; Maius appears both as a praenomen and a family name.

Month praenomen gentilician name
Martius (March) Marcus Marcius
Aprilis (April) -- --
Maius (May) Maius Maius
Iunius (June) Iunius (very rare) Iunius
Quintilis (July) Quintus Quinctius
Sextilis (August) Sextus Sextius
September Septimus (rare, archaic) Septimius
October -- Octavius
November -- Nonius
December Decimus Decius (Roman),
Decimius (Samnite)

Varro, who first came up with this explanation, was bothered a bit by the lack of any names corresponding to April. But not enough to prevent him from proposing it anyway, and not enough to put off modern proponents. (The fact that the name ‘April’ appears to come from Etruscan may have a lot to do with this gap in the table. Maybe one of the other traditional Roman praenomina, like Gaius and Publius and Titus, was related to an older Latin name for April? Who knows.)

The Cambridge Latin Course, volume 1, stage 11. I think the appearance of ‘Quartus Tullius’ here is an unintentional error. In the story he has a brother, Marcus, and Marcus at least was a real historical person: he served three terms as a duumvir in Pompeii. (The lesson? Even professional classicists can jump to conclusions.)

What are the arguments in favour of Varro’s theory? Well, first is the fact that the number-based praenomina start with Quintus and end with Decimus, and this constraint corresponds tidily to the fact that month-names also start at ‘fifth’ (Quintilis) and end at ‘ten’ (December).

Second is the fact that we find some related names in two ancient Italic languages related to Latin, Oscan and Faliscan, in ways that suggest they’re also related to month names.

Oscan was spoken by the ancient Samnites, in the Appenine mountains south of Rome, and we have a Samnite family named Decimius attested in Roman sources. In Rome itself the corresponding name was ‘Decius’. In a study of the ancient Samnites, E. T. Salmon (1967: 53) cites the Oscan names Mamerkis ≈ Marcus, Sepis ≈ Septimus, and Dekis ≈ Decimus, and states that Mamerkis is actually formed from the Oscan name for the month of March. I haven’t been able to confirm the last point, but it’s certainly true that Mamerkis is related to the god Mars, who was called Mamertis in Oscan.

Petersen cites some parallels from Faliscan and Oscan too. Marcius only appears as a family name in Rome, but Petersen points to an example of Marcius as a praenomen in a Faliscan inscription (1962: 352 n. 16).

And most strikingly of all, he cites an Oscan family name Sehsimbriis. This is transparently based on an alternate formation for the name of the month of August. In republican-era Latin, August was called Sextilis (‘sixth-ilis’); Sehsimbriis must reflect an Oscan formation which would correspond to Latin *Seximber.
Note, a day later: I’m no longer too sure of this. If Sehsimbriis did correspond to *Seximber, it’d be an artificial formation, based on analogy with September, November, etc.: the Latin for ‘six’ is sex, not *sexem/sexim. A strict formation ought to be just *Sexiber. (Or in Oscan maybe something like *Sehsikbri-: sehsik- appears to be the Oscan for ‘six’.)

So there we have it: some of the most common praenomina in Latin, Marcus, Quintus, Sextus, and Decimus, come from month names. If your name is Mark, you’re named after the month of March. The idea must surely be that in the earliest times, the names would have corresponded to the month in which the boy was born.

There are some caveats and provisos, mind. First, customs changed over time. The practice of naming children directly for months was long gone by the historical period. And in the imperial era, some number-based names were formed by analogy with the traditional names: as a result we start to see some innovations like Decimius as a praenomen, as in the name of the poet Decimius Magnus Ausonius.

Second, women’s names are different. Men’s names followed a regular pattern of praenomen plus gentilician name; women’s names were much less regulated. And among women we do see Prima, Secunda, Tertia, and Quarta as personal names alongside Quinta. So however exactly the month-name custom worked originally, it didn’t work the same way for women’s names. It could be that for women these names did originally indicate order of birth.

And third: we do see some other number-names popping up as gentilician names which do not correspond to month names. These names seem to come from non-Roman contexts. For example, the Roman name Pomponius is based on a non-Roman word for ‘five’. ‘Five’ in proto-Italic was *kwenkwe. Latin preserved /kw/ sounds relatively faithfully, and so ended up with the form quinque ‘five’. But in many languages, /kw/ transformed into /p/: so in Oscan the word for ‘five’ was pumperias or pompe. This or a related language must have provided the gentilician name Pomponius, basically an Oscan equivalent of Quinctius. In the same way proto-Italic *kwetwōr ‘four’ ended up as quattuor in Latin, but pettiur or pitora in Oscan: a form related to these must be the origin of the Roman gentilician name Petronius.


  • Petersen, Hans 1962. ‘The numeral praenomina of the Romans.Transactions and Proceedings of the American Philological Association 93: 347-354.
  • Salmon, E. T. 1967. Samnium and the Samnites. Cambridge: Cambridge University Press.