Thursday, 16 March 2017

Pi Day

I'm a bit late for Pi Day, but hey, research takes time. It has become trendy to celebrate the 14th of March as 'Pi Day', because in American notation the date looks like '3/14'. These are the first three digits of the mathematical constant π, or pi, the ratio between a circle's circumference and its diameter. (Assuming we're talking about Euclidean geometry. Which we are.)

Outside the US, some people like to celebrate the 22nd of July instead -- because in everyone else's notation, that looks like '22/7', and 22/7 is a very good approximation for π.

A few incidental bits of trivia about π:
  • π is an irrational number. This means it cannot be expressed as a ratio of two integers. Put another way: the circumference and diameter of a circle are incommensurable. Or put yet another way: if you write out π in decimal notation, it will never ever repeat.
  • π is also a transcendental number. This means that it cannot be expressed as the solution to a polynomial equation with integer coefficients. That is: given an equation axn + bxn-1 + cxn-2 ... + zx0= 0, where a, b, c ... are integers, a transcendental number is any number that x cannot be.
  • π is widely suspected to be a normal number. This is not known for sure. A normal number is, roughly, one whose decimal expansion shows no patterns, where every digit is equally likely, and every finite sequence of digits is equally likely. This sounds pretty limiting; at present no one really has any idea how to prove that a given number is normal with 100% certainty. But if you look at it statistically, almost all real numbers are irrational; almost all irrational numbers are transcendental; and almost all transcendental numbers are normal. If you randomly pick a number on the real number line, the probability that it will be normal is 1. So, pretty good odds that π is normal, then.
  • If you know π to 39 decimal places -- 3.14159 26535 89793 23846 26433 83279 50288 4197 -- then you know it precisely enough to measure a circle the size of the observable universe to a precision finer than the width of an atom.
So much for interesting trivia of the day. What about myths? Give me modern myths about antiquity!

OK, here's one myth.
The first recorded algorithm for rigorously calculating the value of π was a geometrical approach using polygons, devised around 250 BC by the Greek mathematician Archimedes.
-- Wikipedia, 'Pi'
It is true that Archimedes used this method to calculate π. But it is not true that he devised the method. He just did it with a bit more precision than anyone had done previously. He made an advance, but it was an incremental advance, not something revolutionary. You can find Archimedes' full exposition in a surviving work, the Measurement of the circle.

The 'exhaustion method'. If you draw regular polygons inside and outside the circle, then the more sides the polygons have, the more closely they approximate the actual circumference of the circle. (source:

The illustration shows how the exhaustion method works. Using 96-sided polygons, Archimedes narrowed down the value of π to between 3 10/71 and 3 10/70 -- that is, he found that π is somewhere between 3.1408... and 3.1429...

But the method was already in use 200 years earlier. Antiphon of Athens (ca. 480-411 BCE), Bryson of Heraclea Pontica (ca. 400 to after 340 BCE), and Eudoxus of Cnidus (ca. 391-338 BCE) had all used a similar method to calculate π long before Archimedes came along.

Antiphon, the earliest of the bunch, only used inscribed polygons -- that is, he only drew one shape, inside the circle, but not outside. As a result he only had one bound for the value of π. We don't know much about Eudoxus' effort. We do know that Bryson guessed (wrongly) that π would be given by the arithmetic mean of the inner and outer perimeters; and that Antiphon and Bryson were working on the area of the circle, not its perimeter. It was Eudoxus who showed that the area and perimeter were linked by the square of the radius.

The New Pauly encyclopaedia reports (subscription needed) that it was Eudoxus, not Archimedes, whose influence led to the widespread use of exhaustion for all problems involving infinitesimals. Archimedes' work on π was just a refinement of Eudoxus.

Here's another myth, from a Time article published on 'Pi Day' this year.
However, not too many generations after [Archimedes'] lifetime, the world experienced a "real decline in math," according to John Conway, mathematics professor emeritus at Princeton University who once won the school's Pi Day pie-eating contest. "Math and science in general went into a great decline from roughly the year zero to the year 1,000, and then the Arabs developed lots of math after that, like trigonometry."
Oooh, do I detect a note of a renowned world expert saying something a little bit silly about another field? I think I do!

What, no love for all those Alexandrian mathematicians of the Roman era? No love for Heron, whose Metrica has recently been published in a new French translation? Or Menelaus, whose work on spherical geometry was foundational for Arabic, Hebrew, and western astronomers for over a thousand years? Not to mention Diophantus, whose work laid down the parameters for the modern study of polynomials, and whose notation foreshadowed the development of algebra?

And then there are many other figures who are, admittedly, lesser, but still made important contributions: Sporus of Poros, who demolished earlier mathematicians' reliance on a curve called the 'quadratrix' in problems to do with squaring the circle; Ptolemy, who in the early 100s CE gained the world record for closest approximation of π (3 + 8/60 + 30/3600, = 3.141666...); and commentators like Pappus, Theon, Hypatia, Proclus, and Eutocius, whose work on Euclid, Ptolemy, and Archimedes were colossally useful in helping later mathematicians to understand the impenetrable language of their predecessors.

I guess it is fair to speak of a decline in Greek mathematics -- but Archimedes was not the be-all and end-all. If there was a decline, it was after the time of Diophantus. Archimedes has a curiously inflated reputation. I suppose that's because there are lots of good stories about him: the story of his death ray; the dramatic story of his death that we find in Plutarch and Valerius Maximus; the story of the bathtub and the running around naked shouting 'Hēurēka!'; and the story of the Cattle Problem, whose solution involves a number with over 200,000 digits (ca. 7.76 × 10206544). Everything about him sounds tremendously exciting. But hey, let's not forget later giants like Hipparchus, Menelaus, and Diophantus, all right?

Tuesday, 7 March 2017

The library of Alexandria: vox populi

You may think I've already spent too much effort on the 'loss' of 'the library' of Alexandria. I make no apology: there is an obsession with the topic in popular culture.

This post isn't meant as a critique but as a sampler. I think it's worth having an awareness of what kinds of things people believe about the Alexandrian libraries. There is a gaping discontinuity between what a trained classicist is likely to think about this topic, and what your average viewer of Cosmos is likely to assume. I think it is salutary to have a reminder of that gap: improved communication of realities about antiquity can only be a good thing.

Relief from Neumagen, Germany, now lost, showing a slave at work in a 2nd century CE bookshop or library (source: Brower and Masen, Antiquitatum et annalium Trevirensium libri XXV [1670] vol. 1 p. 105)

Still, for the sake of clarity, I'd better be explicit about some points that are not popular knowledge.
  1. Libraries existed in the hundreds, maybe thousands, around the ancient Mediterranean. Any book whose survival depended on one specific library was already as good as lost. Books didn't disappear because of a single library, but because of the collapse of a whole system of knowledge exchange. (And, I believe, a format shift.)
  2. The royal archive at Alexandria was indeed burned in the Alexandrian War of 48/7 BCE. But other similar incidents are at best poorly attested, at worst illusions. (The supposed destruction in 389 or 391 CE was invented by Gibbon; the supposed destruction in 642 is a 13th century morality fable inspired by the Letter of Aristeas.) Libraries don't need calamitous tragedies to destroy them: time will do that all by itself. If you don't believe me, go visit Pergamon and see how many books are still on the shelves.
  3. The fetishisation of the Alexandrian libraries is driven by Gibbon, Carl Sagan, and (probably) Sid Meier's Civilization games. None of them is reliable, and the second and third are actively misleading. To get a more balanced picture, read an actual book about ancient libraries. Try especially Lionel Casson's Libraries in the ancient world (2001), and Yun Lee Too's The idea of the library in the ancient world (2010).
And now, for the sake of grasping how present-day people think about antiquity, I present a list of suggestions of what was lost in the 'destruction' of the libraries. The list is taken from a recent social media discussion; I've done a bit of categorisation to make things easier. I offer relatively little comment, meant more for clarification than as criticism.

Question: 'What books and knowledge did we definitely (and likely) lose in the library of Alexandria?'

Answers #1: non-Greek books

  • the works of emperor Claudius ('an Etruscan history and eight volumes on Carthaginian history, as well as an Etruscan dictionary and a book on dice playing') (1)
  • 'how to make Roman concrete and Greek fire' (2)
  • 'Carthage advances in science', especially their death ray (3)
  • 'the complete works of Julius Caesar' (4)
  • history of Carthage (5)
  • Egyptian music and hymns (6)
Some notes:
  1. We have one piece of testimony, the Letter of Aristeas, that the library acquired some non-Greek material. We have no indication of how much, why, or from which languages, other than Hebrew. (The Letter is about the creation of the Septuagint, which is why it picks out Hebrew.) If you want to speculate on which other source languages were represented in the Alexandrian library, native Egyptian material isn't a terrible candidate. But there's no reason to suspect that would include poetic material like hymns: according to the Letter, these acquisitions were all in Greek translations. Roman texts would be a terrible guess. A stronger candidate would be astronomical and mathematical texts from Achaemenid Persia.
  2. Julius Caesar and Claudius lived later than the destruction of the Ptolemaic royal archive.
  3. The 'Etruscan dictionary and book on dice playing' attributed to Claudius are fictional. Robert Graves made them up.
  4. The 'death ray' is presumably the legendary one associated with Archimedes, a Sicilian, not with the Carthaginians.
  5. We have most of Julius Caesar's historiographical output. The stuff we're missing is rhetoric and rhetorical theory (the De analogia, the Anticato, some legal speeches).

Answers #2: things from long after the library ceased to exist

  • how to make Greek Fire (Byzantine, not Egyptian) (7, 8, 9, 10, 11, 12)
  • how to make Damascus steel (introduced westward from India to Syria at some point after the 12th century) (13, 14)
  • the Key of Solomon (14th/15th century; it is extant) (15)
  • the maps that Piri Reis used as a source (early 16th century) (16)
Little comment needed on these, except to note the extraordinary popularity of Greek fire.

Answers #3: 'hidden knowledge'

  • blueprints of the pyramids (17)
  • the location of Atlantis (18)
  • a 'history of man going back 25,000 years' (19)
These contributors appear to be dead serious. I don't think it's worth engaging with them though.

Is this vision from Disney's Atlantis: the lost empire (2001) a real one? What secrets did Disney steal from Alexandria in their time-travelling black helicopters?

Answers #4: things that actually sound sensible ...

... until you pause to think that of course nothing here can possibly have existed in only one copy in only one library.
  • 'most of' Democritus' books (20)
  • history (21, 22)
  • lost plays by Euripides and Aeschylus (23)
  • Sappho (24)
  • the six lost poems of the Epic Cycle (25, 26, 27, 28, 29)
  • 'romances, musics, poem and so on' (30)
  • book 2 of Aristotle's Poetics (31, 32)
  • Heron's work on steam engines (33, 34, 35, 36, 37, 38)
  • Hellenistic tactical manuals and the works of Alexander's successors (39)
  • 'all of the world's knowledge on magic' (40, 41)
  • how the Colossus of Rhodes was built (42)
  • Archimedes (43)
  • works of Galen and Hippocrates (44, 45)
  • works of Plato (1)
  • the majority of Aristotle (46)
  • history prior to Herodotus (47)
  • Chrysippus and Cleanthes (48)
  • commentaries on the Iliad (49)
Some notes:
  1. Best to start by repeating that the destruction of one library didn't suddenly obligate every other copy of its books to cease to exist.
  2. All of Democritus is lost.
  3. Sappho still survived in the 7th century CE as a school text (p. Berol. 5006).
  4. On the Epic Cycle: the last indication we have of anyone having personally read intact copies of these poems dates to the late 2nd century CE, in Athenaeus and Pausanias. That's more than 200 years after the destruction of the royal archive at Alexandria, and 200 years before Gibbon's supposed destruction under Theodosius. The heyday of the Cycle was in the 3rd-1st centuries BCE. Some poems (AethiopisTelegony) may have disappeared as early as the 1st century CE; the last remaining pieces of the Cycle probably disappeared in the 200s CE.
  5. Poetics book 2 is never cited by any ancient source other than Aristotle himself. It may well have been lost before it ever left Athens, within decades of being written. (There are those who disagree: notably Richard Janko, in his work on the Tractatus Coislinianus.)
  6. We do, actually, have quite a lot of Galen, Hippocrates, Plato, Aristotle, and commentaries on Homer. And Galen spent most of his career in Rome, so that'd be the place to expect copies to be preserved.

Answers #5: and to finish off with ...

  • 'prior era philosophy, science, religious, mathematics, and historical texts that went against the then current era ideologies' (50)
  • 'about everything we ever had as human collective' (51)
  • 'They probably all still exist in the vatican archives' (52)
On the last one, I should perhaps mention that the Vatican Apostolic Library is entirely open to visiting scholars, so feel free to go pay a visit or at least browse the online catalogue. As so often, the confusion here is to do with the Secret Archive, which is (a) mostly open access; (b) for documents relating to the papacy, the Curia, and various religious institutions; (c) its oldest document is a collection of ecclesiastical formulae dating to the 8th/9th century.

A reading room at the Secret Archive, Vatican City (source:

There are depressingly few joke responses. One person suggests a book about the origins of Cthulhu; books on why aliens helped humans build the pyramids; an autobiography by Jesus. Aside from these, they all take the subject terribly terribly seriously. (I'd like to categorise the Atlantis one here too, but I'm very much afraid that one isn't a joke.)

Multiple respondents also pause to genuflect at the altar of Carl Sagan (53). If you want proof that Sagan is key to the fetishisation of the library, hey presto.

There isn't much point making fun of any of this. I'll admit it's sorely tempting in a number of cases: you can certainly say that it's making fun of them for me to write this post at all.

But ignorance is just a matter of not having done the right research yet. What's really worrying, because they have an impact on present-day society, are the ones espousing heavily teleological views of the history of knowledge, where knowledge is a quantity that changes as a function of time, as though it were a score that humanity has achieved --
  • 'It's impossible to say what subsequent research would have occurred had the library not been burned. Maybe we would have seen the microscope invented centuries earlier.' (54)
  • 'I like to compare knowledge to compound interest. The more knowledge you accrue, the more it returns.' (55)
These are the ones to worry about. Opinions like these have a potential impact on things like research funding and school curricula. They also affect how people think about, and interact with, societies that aren't as wedded as western elite culture is to post-Enlightenment ideas of cultural teleology.

Ignorance, in and of itself, is no problem. I have no quarrel with the other people posting their suggestions on what was lost. We can never expect to fix the misapprehensions of every layperson, and it's unreasonable to expect perfection. However, specialists can realistically aim to be accessible to the people that popularise ancient history -- the Carl Sagans, the QIs, the Wikipedias, the Snopeses. If they can be reached, there's a chance they can teach their readers and viewers to put a high value on accurate facts ... and avoid the alternative ones that we've seen here.

Wednesday, 1 March 2017

Getting the Iliad right

There are so many misconceptions and myths about antiquity in mainstream culture that it's a refreshing pleasure to see someone with a really solid grasp of their subject. (Especially when I've criticised the same author on a previous occasion for spouting absolute nonsense ...)

Lindybeige, a.k.a. Nikolas Lloyd, talks about the Homeric Iliad in a video published to YouTube almost exactly a year ago today. No nonsense this time. Instead, we have a discussion that is well-informed, accurate, and also, apparently, interesting -- at least to the 400,000-odd people who have watched it. So let's take a moment to celebrate popular media getting antiquity more or less right!
I think that a lot of people buying this [holds up a copy of the Iliad], buy it, start reading, get a little bit confused, and realise, 'Oh! We're already deep into the war when this starts,' and they go alllll the way to the end, they slog through it, and are so disappointed! No wooden horse! That's right! It ends before Achilles dies, so we don't get him being shot in the heel or any of that, and it ends before anyone even has the idea of making a wooden horse.
-- Lindybeige, 'The Iliad' 0:40-1:07

Alexander Ivanov, Priam asking Achilles to return Hector's body, 1824 (Tretyakov Gallery, Moscow)

Lindybeige's mini-lecture spends its first 5 minutes (1) dispelling a popular misconception about the Iliad (see quotation above), and (2) giving some basic information about the epic and its historical context. The remaining 9 minutes are literary criticism, highlighting two ideas: (3) that the poem walks a tightrope between celebrating warfare, and celebrating the human tragedy caused by war; and (4) the resolution to Achilles' personal narrative in the final third of the epic.

His points are well chosen, and there are no major inaccuracies or misrepresentations. Sure, there's room for disagreement on something as subjective as the core meaning(s) of a literary work -- people come up with wild readings sometimes: there are people who actually think characters like Humbert Humbert and Walter White are heroes -- but it's clear that the themes Lindybeige emphasises are important ones.

As fourteen-minute lectures go, you could do a lot worse. Would I be happy for Lindybeige to do a guest lecture in a course I was teaching? Mm, not necessarily. But I would give his essay a decent mark.

Let's look at his account of the historical context of the poem. He highlights the Iliad's importance in antiquity by comparing it to the Bible -- not an exact analogy, but a traditional trope, and one that makes its point -- and mentioning, entirely correctly, that there are stories of Alexander taking his personal copy with him on campaign,* and that at Alexandria scholars spilled a great deal of ink over the Iliad (and here's the main end-product of that scholarship).
* Alexander's copy of the Iliad: see Plutarch Alexander 8 (= Onesikritos BNJ 134 F 38), 26. Cf. Plutarch On the fortune of Alexander i.327f, where he also takes the Odyssey on campaign.

I especially like his comments about modern popular culture's fetishisation of the library of Alexandria:
You can tell how middle-class you are by how aggrieved you are and how much you wince every time someone mentions the fact that the library of Alexandria burnt down. Ahh! Grr! Oh if only it hadn't! Agh!
(I could wish the same about the Palatine library in Rome, the Athenaion at Pergamon, and hundreds and hundreds more. Any of them would involve multiple miracles.)

Anyway, the next bit is what's most likely to raise the hackles of those who have been taught about the Iliad in a certain way. was an epic poem that would be performed over several nights by a poet. Quite often we are told that they would beat a stick to a strict rhythm, as they spoke the rhythmic words. And it would take a few nights for them to get all the way through this. And yes, they wouldn't have a script to work from: they had to memorise the entire lot, a feat that was made possible by an oral tradition and the fact that there are a lot of standard phrases and repetitions within the rhyme itself.
-- Lindybeige, 'The Iliad' 2:12-2:43
There are a few things here that some classicists would probably want to see phrased differently, but I'm on Lindybeige's side. The word 'rhyme' is just a trivial slip (rhyme was almost unheard-of in ancient poetry): we'll skip that.

The real qualm I'd expect a classics student to have is with the word 'memorise'. American scholarship on Homer in the last 30 years or so has strongly disliked the idea that Homer was memorised and transmitted. They tend to prefer to talk about recomposition. The idea is that bards supposedly improvised the poem afresh every time they performed it. That idea is common in America, but in the rest of the world it's more common to think of the epic as informed by a sophisticated tradition of recomposition-in-performance, without necessarily being produced directly within that tradition. The archaic language of the Iliad belongs firmly to the first half of the 600s BCE, but the earliest likely date for its transcription is in the second half of the 500s. To bridge that gap, there's a much stronger case for verbatim or very-nearly-verbatim transmission than you'd think if you just read American books. I'm not certain which books or people have shaped Lindybeige's views: but he's English, and I gather he's based in Newcastle, so there's that.

Then imitate the action of the rhapsode; stiffen the sinews, summon up the hexameters! Lindybeige illustrates using a staff to beat time (2:20)

I'm especially pleased to see no reference to singing. Lindybeige describes a performer 'beat[ing] a stick to a strict rhythm'. As an onscreen caption makes clear, he's talking about rhapsodes, who declaimed epic, not about bards who sang with a musical accompaniment. 'Singing' often appears in Homeric epic as a metaphor for performance, but all external evidence points strongly to rhapsodic declamation. This is a position that many professional classicists would contest, but I'm firmly on Lindybeige's side here. One of Homer's rivals, Hesiod, describes performance in their genre as follows:
And [the Muses] gave me a staff, a branch of lovely laurel
that they'd plucked, a marvel: and they inspired me with a voice
divine, so that I would popularise things that will happen and did happen,
and they told me to hymn the race of the blessed ones, who are eternal,
and always sing of themselves both first and last.
-- Theogony 30-4 (ca. 700 BCE)
Rhapsode with staff,
declaiming an epic episode
set at Tiryns; from the cover
of my own book (BM E270,
ca. 490-480 BCE)
Hesiod talks about 'hymning' (hymnein) and 'singing' (aeidein), but the staff shows that he's thinking about rhapsodes. For my money, I'd say 'singing' is a conventional poetic image, not a literal reality. (See also my Early Greek hexameter poetry, 2015, pp. 76-7, with more sources.)

Lindybeige is also nicely cautious about the date of the Iliad: 'somewhere between the 700s and 500s BC', he says. Modern scholarship dates the Iliad anywhere from ca. 800 (Powell) to the mid-500s (Jensen) -- for what it's worth, in my view the most powerful evidence points to ca. 670-650 -- so this is a fair reflection of an open question.

So much for the historical aspects. The rest of the video is occupied with literary exegesis. Lindybeige illustrates how the Iliad simultaneously celebrates war and shows an extraordinary sensibility to the human suffering caused by war. He gives a full reading of Iliad 11.218-247, the death of Iphidamas, killed by Agamemnon, and his touching backstory ending in 'bronze sleep ... far from his wedded wife'. In the last minutes, he turns to Achilles' fury. The very first line of the epic highlights this theme, but Lindybeige confines himself to talking about the last third of the poem, where Achilles is enraged at Patroclus' death, rampages on the battlefield, fights a river-god, kills, captures, and sacrifices Trojans, buries his friend, but cannot find peace anywhere -- until the night-time visit from king Priam in book 24 (also a genuinely powerful moment in Petersen's film Troy (2004), with Peter O'Toole as Priam, as Lindybeige points out).

'Perhaps more than anything, the Iliad is about this scene.' (10:00-11:25)

Do these points give an exhaustive account of the literary merit of the Iliad? Of course not. But they're a very good selection. Lindybeige's discussion of war and humanity neatly encapsulates two divergent modes of interpretation which focus on the Iliad as praise poetry and as tragic, respectively -- the Gregory Nagy school and the Aristotle school, you might say. And his treatment of Achilles' fury, while not the deepest or most thorough, nonetheless draws out an aspect of how the Iliad develops that theme throughout books 17 to 24, and does so without mistakes.

The only real criticisms I can imagine being levelled at Lindybeige are about all the other things going on in the Iliad which he doesn't mention. The rampages of the other Greek heroes; how the poet plays on competing poetic traditions; the narrative of divine withdrawal and return; the humanity and failings of Hector; the use of folkloric themes. And so on. But who has time for all that? This is 14 minutes long. And it's perfectly good for that length. It's a delight to see something this competent in the popular arena. Congratulations.

Postscript: for more multimedia coverage of Homer and the Iliad, try the BBC World Service's programme 'The Iliad: beauty, brutes, and battles' (Dec. 2016), with Bettany Hughes talking to Stathis Livathinos, Antony Makrinos, Folake Onayemi, and Edith Hall. I haven't listened to it yet, and the cast-list makes it sound like it's more about reception than about the Iliad itself, but I'm looking forward to it!

Friday, 17 February 2017

Dying and rising gods: are they a thing?

What are dying and rising gods? Well, if you select a bunch of ancient mythological stories where gods apparently die then come back to life, you can (if you want) define those gods as a class based on a common pattern. I'll call them 'DRGs' for short. DRGs have been around since Frazer's The golden bough (1st edition 1890): the usual suspects are figures like Adonis, Osiris, Dionysus, Attis, and of course Jesus.

Osiris: fresco from the tomb of Sennedjem, Deir al-Medina, Egypt, 13th cent. BCE (source: Caution: Osiris is not a dying and rising god, he's a dying god. See below.

These days DRGs are especially hot stuff among Jesus mythicists. These are folks who argue that the central figure of Christianity is derivative, plagiarised, and distorted from a variety of other mythologies. The payoff, presumably, is that it licenses anyone to critique present-day Christianity for the same flaws. So DRGs aren't a neutral anthropological category for studying the past: they're a hot political weapon.

Here's a fairly well-known mythicist, Richard Carrier:
... almost all the dying-and-rising gods award their followers a handsome afterlife with a baptism through which the follower emulates the death and resurrection of the savior. To claim this isn’t astonishingly similar to Jesus is simply lying at this point.
-- comments on a debate about the historicity of Jesus, Oct. 2016
Notice how Carrier gestures at 'all the dying-and-rising gods' without naming any? For good reason. Classifying Frazer's canon of DRGs as 'rising' would be tenuous, to put it mildly. Carrier knows that if he actually names names, he'll have to add pages and pages of provisos and caveats.

Setting mythicists aside, in the last 50 years the dominant tendency has been to reject DRGs as a non-category. The work of Jonathan Z. Smith (1969, 1987), a scholar of the history of religion, has led many people to reject Frazer. Smith has some good points, but there has been a qualified counter-reaction to him, too. The most prominent current work on the subject, Tryggve Mettinger's The riddle of resurrection (2001), is more reserved -- but there's still plenty to disagree with: I'm not going to draw on Mettinger much here.

The problem with DRGs is that the entire concept was designed specifically with Jesus in mind. If your methods for deciding who is and is not a DRG revolve around 'In what ways are they like/unlike Jesus?', then of course you're going to end up seeing 'death and resurrection' elements behind every bush. You're also going to end up paying less attention to ways in which they are related or unrelated to one another.

Certainly there are gods who can be said to have died and risen. If you really want, you can declare that that's a pattern. But you can find patterns practically anywhere you look. Humans are really good at seeing patterns in noise.
And (Pythagoras) engraved an epigram on the tomb, titling it 'Pythagoras, to Zeus', which began:

      Here lies dead Zan, whom they call Zeus.
-- Porphyry, Life of Pythagoras 17
This snippet refers to a supposed tomb of Zeus on Crete. Yes, Zeus himself. The tomb is almost certainly a fiction, derived from Euhemerus' Sacred history (early 3rd century BCE), but it was widely believed to have existed. Many modern scholars, too, have been lured into believing the tomb was a real tourist attraction, even if not a real cult site.
On Zeus' tomb, see Winiarczyk 2013: 33-41. Winiarczyk's list of ancient testimony (nn. 81, 82, 83) includes '23 pagan writers (14 Greek and 9 Roman), 28 Christians (19 Greek and 9 Roman) as well as 19 later authors (16 Byzantine, two Roman and one Syriac).' Testimony appears only later than Euhemerus, beginning with Callimachus Hymn 1.8-9 (1st half of 3rd cent. BCE) and Dionysius Scytobrachion (mid-3rd cent. BCE, reported in Diodorus 3.61.2). The tomb is variously placed on Mount Ida (Varro, Porphyry, Cyril), Mount Dicte (Nonnus), or at Knossos (Malalas).

And yet, funnily enough, I've never heard anyone suggesting that Jesus was plagiarised from Euhemerean Zeus. That isn't because Euhemerean Zeus wasn't a DRG, it's because Zeus doesn't fit our preconceptions about what a parallel to Jesus ought to look like. In Zeus' case it's blindingly obvious that it's a fringe belief, and more importantly, the resemblance is an isolated motif: it doesn't represent any systematic, deep commonality between the two figures. (We also have a lot more information about Zeus than about any of the others, which means it's harder to get away with misinterpreting what we do know.)

A fairly typical catalogue of imagined DRGs

The truth is a lot muddier than either Carrier or Smith would like. It's at least true that DRGs are rarer than most people think, and they're not remotely uniform. Best practice is going to be: (1) pay attention to ancient testimony about syncretism; and (2) think of a better set of categories for gods than just 'DRG = similar to Jesus, non-DRG = dissimilar to Jesus'.

Syncretism is when two gods are merged in some way. I'll use it in an extended sense, to mean any of: (1) one god being completely absorbed into another (e.g. Selene being absorbed into Artemis); (2) two separate gods being equated without merging fully (e.g. Jupiter = Zeus = Amun); (3) a motif being transferred or copied from one god to another. For example, the story of Orpheus getting torn to shreds reappears in one variant of Dionysus, in a poem attributed to Orpheus. But the two aren't equated with one another anywhere. On the other hand, the Orphic Dionysus does get equated with the Egyptian god Osiris in one late source.

And here's a sample taxonomy of 'DRGs':
  • Gods who die
    Examples: Osiris (Egyptian/Syrian), Adonis (Greek/Syrian), Castor + Polydeuces (Greek/Latin), Attis (Phrygian/Lydian)
  • Gods who die once and return to life once
    Examples: Inanna (Babylonian), Baal (Ugaritic), Telipinu (Hittite), Theseus (Greek), Orphic Dionysus (Greek), Jesus (Hebrew/Greek)
  • Time-sharing gods, i.e. gods who spend part of the year in the world of the dead, and part either in heaven or on earth
    Examples: Dumuzi and his sister (Babylonian), Adonis (Greek/Syrian), Persephone (Greek), Castor + Polydeuces (Greek/Latin)
(Frazer calls Dumuzi 'Tammuz', the Hebrew form of his name: he pops up in the Bible once at Ezekiel 8:14.)

Now, these categories aren't rigid. There are loads of things that mess them up:
  • There are overlaps: some dying gods are also time-sharing gods. But not all of them.
  • Things also get messy if we take ritual context into account. For example: for the deaths of Baal or Asclepius, we don't know of any ritual context; in the cases of Adonis and Osiris we know only of ritual contexts, with no external narratives.
  • Things also get messy if we take into account other factors like whether our sources link the stories to fertility. For example, when Telipinu disappears into the ground, barley and wheat stop thriving, and farm animals stop reproducing. We see comparable elements in Inanna + Dumuzi, but not in the cases of Castor + Polydeuces or Jesus. The case of Adonis is unclear.
  • Smith and Mettinger tend to get bogged down in whether a given divinity is really a god, or just a hero. Personally, I think this isn't a valuable distinction. But it's another factor that can mess up the categories, if you want.
This is why it's essential to pay attention to what the sources tell us about syncretisms. Which gods get equated with which in ancient sources? And who is giving us that equation: is the equation part and parcel of cult practice, or is it imposed by someone outside that religious context?

We can't possibly do an exhaustive study of all these gods, but let's at least take a look at the three that I think are the most difficult: Adonis, Osiris, and Dionysus.

Adonis relief, Rome, 118-125 CE
(source: Brill's New Pauly)


Adonis appears first in the Greek world. In the 7th-6th centuries BCE he pops up in fragments of the Catalogue of Women, Sappho, and Epimenides. The earliest account of his full story comes from an epic poem, Panyassis' Heraclea, where he is a mortal Assyrian prince who becomes the object of a romantic dispute between Persephone and Aphrodite. The dispute gets resolved. Later, Adonis is killed by a boar while hunting. The end. No resurrection.
References: Catalogue of Women fr. 139 Merkelbach-West (7th-6th cent. BCE); Sappho frs. 58, 140a, 168 Lobel-Page (early 6th cent.); Epimenides fr. 57 Bernabé (late 6th cent.?); Panyassis fr. 27 Bernabé (2nd half of 5th cent.).

No later sources for the Adonis story add a resurrection. However, some sources relating to the Adonis cult point in that direction. He's regularly depicted as dividing the year between Persephone (in the underworld) and Aphrodite (in heaven), but a couple of late sources state that he continues to do this after death: so in some contexts, at least, he's a 'time-sharing' DRG.
Adonis time-sharing after death: Orphic Hymn 56.8-11 (ca. 2nd cent. CE?); Cyril of Alexandria Comm. on Isaiah lxx.441.3-19 (late 4th cent.).

Some DRG-hunters have laid emphasis on a passage in Lucian, which describes an Adonis festival where he is mourned as dead, but then the next day the celebrants parade a figurine of him into the open air and 'they say allegorically' (μυθολογέουσι) that he is alive. That carries no weight by itself: Lucian is a satirist, a master of blending biting satire with pure farce; in addition, his description isn't much like what we know of the older Adonis cult at Athens.
Lucian On the Syrian goddess 6 (2nd cent. CE); similarly Origen Selecta in Ezechiel 800a Migne.

Smith thinks that's the earliest sign of Adonis' time-sharing returning to life, but he's wrong. Half a millennium earlier, we have Theocritus (3rd century BCE) telling us about Adonis' annual time-sharing, and not between Hades and heaven but between Hades and the world of the living:
λύσασαι δὲ κόμαν καὶ ἐπὶ σφυρὰ κόλπον ἀνεῖσαι
στήθεσι φαινομένοις λιγυρᾶς ἀρξεύμαθ' ἀοιδᾶς·
'ἕρπεις, ὦ φίλ' Ἄδωνι, καὶ ἐνθάδε κεἰς Ἀχέροντα
ἡμιθέων, ὡς φαντί, μονώτατος. ...'

We'll undo our hair, let our bodice hang down to the ankles,
and with our chests exposed we'll begin the clear song:
'O dear Adonis, you are absolutely the only demigod who comes
both here and to Acheron, they say ...'
-- Theocritus Idyll 15.134-7
This is Theocritus' poetic rendering of a ritual lament in the Adonis festival at Alexandria, Egypt. We don't have anything similar for Adonis in Athens. But it appears that the Alexandrian Adonis, at least, was understood to be time-sharing between the afterlife and the world of the living.

Now to complicate things. Since this motif is missing in Athenian evidence, it looks like some elements of Egyptian and Syrian Adonis worship were adapted from rites relating to Osiris. After Lucian's description of the Adonia at Byblos, he immediately goes on to state that there's a question over whether this rite is actually about Adonis or Osiris. There's a similar question over the Alexandrian Adonis -- here's Theocritus' song again:
And in the morning, with the dew, we'll come together
and bear him outside, towards the waves splashing on the shore
-- Theocritus Idyll 15.132-3 (see also scholion ad loc.)
The practice of a procession to throw a figurine of the god into the sea is better attested for the cult of Osiris than that of Adonis. It makes most sense to understand this motif being transferred from Osiris cult after the Adonis cult had already migrated to the east. Obviously Osiris is where it's all happening, so let's turn to him now.


In Egyptian myth the god Osiris is killed, and his body is dismembered and the parts separated all over Egypt. Isis then gathers the body parts together and reassembles them. They have sex and bear a son, Horus. Later, Osiris becomes judge of the dead.

Except there's a crucial gap here. Isis has sex with Osiris' dead corpse. He doesn't come to life.

Isis, in bird-form, has sex with Osiris. Doesn't he look lively!
(Relief from mortuary temple of Seti I, ca. 1279 BCE; source: Wikipedia)

All over the web you'll the above image, and others like it, described as Isis 'reviving' Osiris. Well, ahem, I guess that's kind of true: getting his corpse to cooperate in impregnating her involves a very specific kind of reanimation. (Honestly, these people that say she's 'fanning life' into him ... what does it look like she's sitting on?)

To be fair, there is a set of Ptolemaic reliefs at Dendara where Isis descends upon Osiris' erect penis, then Osiris rises up from his bier, makes a boat journey, and is given an ankh. You could be excused for interpreting this as 'resurrection' ...

Reliefs from Osiris chapel (1st cent. BCE), temple of Hathor, Dendara. (1) Isis about to land on Osiris' erect penis to impregnate herself; (2) Osiris arises from his bier; (3) Osiris travels in a boat to the netherworld, and on arrival is presented with an ankh. (Source: Mettinger 2001: 173-4. Images are reversed where necessary to make a left-to-right narrative.)

... unless you're aware that in Egyptian iconography it's dead people that carry ankhs. The ankh symbolically completes Osiris' journey into the afterlife. This isn't a dead god coming back to life, it's a god becoming good and properly dead.

Osiris' role as judge in the afterlife isn't that of a living god who just happens to live in the world of the dead: he's a dead god among the dead. The story of his body parts being dispersed is a paradigm for the spread of Osiris-worship all over Egypt; the story of Isis gathering and reassembling them is a paradigm for the process of mummification. That's Osiris' role in extant textual material, too. In funerary rites the deceased is regularly given Osiris as a forename: in the pyramid texts the pharaoh Unis becomes 'Osiris-Unis', Teti becomes 'Osiris-Teti', and so on. That's the context for prayers like the following, which at first glance seem to talk about 'Osiris' coming to 'life' --
Atum, this Osiris here is your son, whom you have made revive and live;
he will live and this Unis will live, he will not die and this Unis will not die ...
-- pyramid of Unis (5th Dynasty, 25th cent. BCE); Allen and Der Manuelian 2005: 34
This formula, repeated and addressed to a different god each time, isn't about Osiris coming to life: it's about Unis (now Osiris-Unis) transitioning to the afterlife. Osiris' involvement is because of the central place of his death in the Egyptian understanding of death.

Osiris is a dying god, not a DRG. And that makes perfect sense for the god in charge of the afterlife. But that didn't prevent Osiris from being equated with some time-sharing gods, as we've already seen -- and also with some DRGs, as we'll see next.


Even today you may hear people repeating that Dionysus was a late addition to the Greek pantheon, imported from Thrace. That's because that's what the classical-era Greeks believed. It ain't true, and we've known that since about 1960. That's when Dionysus started popping up in Bronze Age Linear B tablets (1, 2, 3), which makes him one of the very earliest-attested Olympians. The fact that his cult existed all over Greece also supports his claim to be a divinity of long standing.

In the standard story, Dionysus is born twice (and doesn't die): once from his mortal mother Semele, then again from Zeus' thigh -- Semele dies when Zeus reveals himself to her in his full glory, so Zeus has to think of something quickly to save the unborn child. Dionysus goes on to become the god of wine and drama, and to play an important but nebulous role in mystery cults.

However, there are a couple of fringe variants of Dionysus who do die. In particular, an Orphic variant of Dionysus dies in a mythical episode called the Titanomachy, the battle between the Olympian gods and the Titans. The best known version is in Diodorus --
The mythographers have reported a tradition that he also had a third birth: they say the god was born of Zeus and Demeter, and that he was torn apart by the Giants and boiled down. His body-parts were reassembled by Demeter and he was born anew, from scratch; and they link this story to various natural forces.
-- Diodorus of Sicily 3.62.3 (= Orphica fr. 59.iii Bernabé)
You'll notice Diodorus makes the Giants the culprits, not the Titans: Titans and Giants regularly get mixed up in Orphic sources, though they're not compatible. (The Titanomachy belongs near the very beginning of the chronology of myth; the Gigantomachy comes much later, in the time of Heracles.) Here's a summary of our sources on the death of Orphic Dionysus:

fr. 59.i fr. 59.iii fr. 59.v fr. 327.ii
Philodemus On piety 16.1 Gomp. Diodorus 3.62.3 Servius on Georgics 1.166 Proclus Hymns 7.11-15
cause of death torn apart (διασπάω) torn apart (διασπάω), boiled (καθέψω) torn apart (discerpo) divided up (μερίζω)
culprit Titans Giants Giants Titans
reassembler Rhea Demeter -- Athena (preserves heart)
verb for revival ἀναβιόω 'come to life again' γεννάομαι 'be born' -- ἀνηβάω 'grow young again'
source(s) cited Euphorion's Mopsopia, Orphic writers 'mythographers' Orpheus (and Varro for story of Osiris) --
notes -- -- equates Dionysus with Osiris Dionysus reborn(?) from Semele

Another fragment, fr. 59.iv (= Cornutus Compendium 30) reports that Dionysus was torn apart by the Titans and reassembled by Rhea, but doesn't mention a resurrection. There are other references to Dionysus being torn apart, but without details.

The Olympian Dionysus, the god of state religion, emphatically did not die. But the Orphic Dionysus, the Dionysus of mystery cults, was evidently a genuine DRG. Diodorus reports that the sources available to him are 'inconsistent, numerous, and bizarre', and he complains that he can't sort them out.


I'm being loose with the word 'syncretism': some people might prefer to distinguish syncretism from a phenomenon called interpretatio graeca. The latter refers to a system for equating Greek gods with specific gods in other pantheons, e.g. Zeus = Jupiter = Amun, Apollo = Horus, etc. I'm lumping that together with syncretism because equations like these also pop up within the Greek pantheon (Titans = Giants) and in some edge cases (Samothracian Kabeiroi = Cretan Kouretes = Laconian Dioskouroi(?)).

The Dionysus sources, above, are chocker with syncretisms. Servius equates Dionysus with Osiris, and Typhon with Set; there's the Titans and the Giants; and there's the Orpheus crossover in the idea of Dionysus being torn apart. And the boiling of Dionysus sounds awfully like the trick Medea uses to kill king Pelias: Medea persuades him that she can rejuvenate anyone by cutting them up and boiling them, and of course Pelias dies. Even so, we have three sources -- all earlier than 400 BCE -- reporting that she did in fact perform the procedure successfully on Aison or Jason. So this may be another element of motif-borrowing or syncretism.
Medea's rejuvenation of Aison/Jason: Nostoi fr. 6 West; Simonides fr. 548 PMG; Pherecydes FGrHist 3 fr. 113ab (source for all three = hyp. Eur. Medea).

That's just the tip of the syncretism iceberg. Remember how Lucian is doubtful whether Adonis at Byblos is really Adonis or Osiris? As well as that, Origen and Cyril equate Adonis with the biblical Tammuz (= Dumuzi); Isis can be equated with either Demeter (Herodotus) or Persephone (Archemachus); Typhon is regularly Set. There are loads of other identifications floating round, most incompatible with one another.
Adonis = Osiris(?): Lucian De dea Syria 7. Adonis = Tammuz: Origen Selecta in Ezech. xiii.797d-800b Migne (repeated in Theodoret, Procopius, and ps.-Nonnus); Cyril Comm. on Isaiah lxx.440d-441b Migne. Dionysus = Osiris: Varro ap. Servius on Georgics 1.166 (= Orph. fr. 59.v Bernabé). Isis = Demeter: Hdt. 2.59, 2.156. Isis = Persephone: Archemachus BNJ 424 F 6. Typhon = Set: Hdt. 2.156; Varro ap. Servius on Geo. 1.166; Plut. De Isid. 351f, etc., etc.

But we also have to pay attention to who is making these equations. The ambiguity over Adonis and Osiris comes to us from pagan sources, Theocritus and Lucian: in Lucian's case, he is a Syrian writing about Syrian religion. Herodotus and Archemachus are all about trying to make Egyptian religion and Greek religion mutually intelligible, in part because shrines to Egyptian gods were starting to appear in Greece at the time.

But the equation Adonis = Tammuz, in contrast, appears only in Christian exegetes. They're coming from a hostile position, and for them it is emphatically not an ecumenical matter. They're trying to explain a hostile reference to Tammuz in the Bible, not to make it easier for Christians and Adonis-worshippers to talk to each other. We can be absolutely certain that Adonis-worshippers themselves would never have made this equation.

A genuine case of a derived god whom worshippers happily equated with other gods: Serapis (a.k.a. Osiris-Apis), invented by the Ptolemies and equated by Plutarch with Pluto or Osiris (De Isid. 362a-b), and in this case, found in a Mithraeum in London. (2nd cent. CE, Museum of London; BM listing; source:

The upshot is that while ancient practitioners of the pagan religions happily conflated Osiris with either Adonis or the Orphic Dionysus, lumping Tammuz/Dumuzi in with them is purely a Christian imposition, which Frazer happens to have copied. (Tammuz appears in ancient Greco-Roman texts only in discussions of Ezekiel 8:14.)

And, we should probably note, no one ever tried to equate Jesus with any of these figures -- at least, not until Frazer came along. Motif-borrowings are possible, but the more extreme ideas of Carrier and his ilk are a huge stretch. You don't have to believe cult leaders are fictional in order to disbelieve their religion!

Postscript: actual dying and rising gods

Actual gods who die once and then resurrect once are much older, much more obscure, and have no clear link to 1st century CE Graeco-Jewish culture:
  • Baal, in Ugaritic myth. Baal battles Mot and is devoured. The Canaanite god El and Baal's sister Anat lament for him, and fertility ceases. Anat destroys Mot, El dreams of Baal's return, and Baal reappears, to general rejoicing (though apparently so does Mot).
    Source: Baal Cycle (early 14th cent. BCE), tablet 10 col. viii to tablet 12. See Smith, Parker, et al. 1997: 138-64.
  • Inanna (and Dumuzi), in Babylonian myth. Inanna (Ishtar) travels to the underworld, and instructs her minister to mourn her death and pray for her. Ereshkigal makes her a lifeless corpse, and Enki sends messengers to restore her to life and release her. In exchange for her release she is also required to provide a substitute, her husband Dumuzi. (Dumuzi and his sister, in turn, go on to become a pair of time-sharers, alternating between the underworld and earth every half-year.)
    Source: Sumerian Descent of Inanna (first half of 2nd mill. BCE), also in an Akkadian version; cf. allusion in Gilgamesh vi.46-7.
  • Telipinu, in Hittite myth. The god Telipinu, in a fit of rage, disappears into the ground: weeds cover him, and fertility ceases. The sun-god sends emissaries to look for him, and a bee stings him until he gets up. Telipinu is angry at being awoken, and tears up the landscape until Kamrusepa uses magic to soothe him.
    Source: CTH 324, a mugawar song (15th-13th cent. BCE). See e.g. Della Casa 2010.
It should be pretty clear, though, that these links are even more tenuous than any imaginary links to Adonis.

[Note, 20 Feb.: this post has been edited. The original form had some inaccuracies caused by a lack of specificity, especially in parts discussing Adonis or Inanna. There are also some corrections relating to Inanna/Dumuzi, thanks to comments from Theo (see below). I've tried to leave the alterations visible, but that hasn't been feasible in every place.]


Wednesday, 11 January 2017

Salt and salary: were Roman soldiers paid in salt?

A few weeks ago,we looked at myths to do with ploughing over cities and salting the earth. Today we're looking at a kind of companion myth. The basic idea is that Roman soldiers were paid in salt, or received an allowance of 'salt money'.

Salt money? (photo by Benreis; CC licence)

A few other ancillary myths tend to come along with it too. Take a look at these gloriously mangled pieces of misinformation:
I thought you might like to know just where your salary comes from. The word, at least. The source seems to be the Latin 'salarium' ('sal' being salt) which is a word tied to the payments made to soldiers in the early Roman salt trade. In those days, salt (regular ordinary table salt) was a prized and valuable commodity. If you've ever heard the phrase "you are the salt of the earth" or "worth your salt", both are referring to the high value of salt.
A soldier's pay -- consisting in part of salt -- came to be known as solarium argentum, from which we derive the word salary. A soldier's salary was cut if he "was not worth his salt," a phrase that came into being because the Greeks and Romans often bought slaves with salt.
(The blog post, in particular, has been uncritically copied, paraphrased, and plagiarised on many other parts of the web -- like this page offered up by the European Parliament's Terminology Coordination Unit.)

First, the accurate bits. (1) The English word 'salary' does indeed come from Latin salarium 'stipend, money allowance'. (2) Salarium does indeed appear to be linked to sal 'salt', via the adjective salarius 'pertaining to salt'. And there the accuracy ends.

Here's the simplest form of the myth.
The word "salary" comes from the Latin word for salt because the Roman Legions were sometimes paid in salt.
-- Wikipedia, 'History of salt'
Pure fantasy. There isn't the tiniest scrap of evidence to suggest this. At all, to any extent, ever.

The allure of this myth comes simply from the link between salarius and salarium. Naturally everyone wants to have the true explanation of what exactly the link is. Unfortunately no ancient source tells us one. And so we end up in the situation where people invent explanations for themselves.

Folks who propagate this myth don't usually try to cite sources, but when people do go looking for sources, they end up drawn to two pieces of ancient testimony. First is the 1st century CE writer Pliny the Elder:
honoribus etiam militiaeque interponitur salariis inde dictis ...
(Salt) is also related to magistracies and duty abroad, and that's where we get the word 'salaries' ...
And second, testimony about state taxes on salt. For example, the historian Livy reports how the Roman censors imposed a new tax in 204 BCE:
vectigal etiam novum ex salaria annona statuerunt. sextante sal et Romae et per totam Italiam erat; Romae pretio eodem, pluris in foris et conciliabulis et alio alibi pretio praebendum locaverunt. id vectigal commentum alterum ex censoribus satis credebant ... inde Salinatori Livio inditum cognomen.
(The censors) also imposed a new tax on the annual salt production. Salt cost a sixth of an as in Rome and throughout Italy; they set it to be offered at the same price in Rome, but more in town squares and marketplaces, and at other rates in other places. It was widely believed that just one of the two censors devised this tax ... As a result (the censor) Marcus Livius was given the nickname 'salt-dealer'.
-- Livy 29.37.3
Elsewhere Cato the Elder is quoted as talking about salinatores aerarii, treasurers of the salt taxes, as a specialised post in the 190s BCE (reported in Servius auctus, commentary on Aeneid 4.244). These passages, along with Pliny, are close as we get to a link between salt and money in any extant Roman sources.

The trouble with citing Pliny as a source for the myth is of course that Pliny doesn't say anything of the kind. The problem is exacerbated by Wikipedia, which bald-facedly re-writes Pliny, and has been quoted very widely:
the Roman historian Pliny the Elder, who stated as an aside in his Natural History's discussion of sea water, that "[I]n Rome...the soldier's pay was originally salt and the word salary derives from it...".
-- Wikipedia, 'Salary' (the addition of this line dates to 2004)
This is a mistranslation, just to be clear. Even the source cited doesn't contain this wording. And Pliny is writing about salt itself, not sea water. None of that has stopped this fake quotation being repeated in countless books and websites.

*Note, 18 Jan.: this error, and the other Wikipedia excerpt quoted above, have since been corrected. As of now, the reports in the 'Salary' and 'History of salt' articles appear to be more or less accurate.

Brine refinery at Fuerteventura, Canary Islands (source: tourist blog). Ancient Roman salinae worked in more or less the same way: see Pliny Nat. hist. 31.81-83.

If you take a global view, of course you're bound to find some times and places where salt could act as a means of storing value and facilitating exchange. The most famous example is Ethiopia in the modern era. Here's how it's reported by Ray's Travels, a classic 17th century piece of travel writing:
In trading, they make no use of coined money, as the Europeans do, but their money are pieces of fifteen or twenty Pics of cloth, gold, which they give by weight, and a kind of salt, which they reduce into little square pieces like pieces of soap, and these pass for money. They cut out that salt upon the side of the Red Sea, five or six days journeys from Dangala, as you go from Cairo, and the places where they make it are called Arbo.
-- John Ray, A collection of curious travels and voyages, vol. 2 (1st ed. 1693, 2nd ed. 1705), 1738 printing, p. 486
This 1949 book, this 1977 essay, and this 1994 book report that salt bars called amoléh continued to serve as an important medium for exchange -- one among many; others included Maria Theresa thalers, clothing, iron, gold, and cattle -- all the way up until the beginning of the 20th century. Reportedly the chief source of Ethiopian salt bars was the Afar depression, next to the Red Sea, a region that includes present-day Djibouti as well as slivers of Ethiopia, Eritrea, and Somalia.

Just bear in mind that this has nothing at all to do with Roman soldiers. The fact that salt could mediate exchange in 17th-19th century Ethiopia has no bearing on ancient Rome. Salt money might be a plausible thing in and of itself, but we have absolutely no reason to imagine salt currency in Rome. It's just that when you go hunting for something specific across the whole of human history, you're likely to find it.

A few more examples. This 2013 book claims that salt has also been used as 'money' (the word is tendentious: 'a medium for trade' and 'money' are not the same thing) in China, pre-Columbian Mexico, Borneo, and elsewhere. A person who uploaded this photo to claims it is a sample of salt currency from early 20th century Angola, held at the Royal Ontario Museum. And Wikipedia alleges that American soldiers were paid in brine during the War of 1812. This last one appears to be completely fictional, like the Roman case: apparently it's some kind of distant distortion of the British salt embargo during the war, and the development of several important brine refineries in the USA throughout the 1800s-1810s.

'Roman soldiers were paid in salt' may be the simplest form of the myth, but it's also a secondary form. I've done some searching around in Google Books with date constraints, and this seems to indicate that people first started writing about the idea around the 1860s (here, for example).

The older, primary form of the myth is that soldiers were given 'salt money', that is, a monetary allowance for buying salt. This, too, is a modern invention. It isn't nearly as daft as 'soldiers were paid in salt', but it's still only a conjecture, unsupported by any ancient testimony.

The phrase 'salt money', or in Latin salarium argentum, is an invention of 18th and 19th century Latin dictionaries. The phrase was coined by dictionary-writers as their best guess for how salarium 'salary' came from salarius 'pertaining to salt'. Here's one of the two standard Latin-English dictionaries, Lewis & Short, on the subject:
B. sălārĭum, ii, n. (sc. argentum; cf.: calcearium, congiarium, vestiarium, etc.); orig., the money given to the soldiers for salt, salt-money; hence, post-Aug. (v. Dio Cass. 52, 23, and 78, 22), in gen., a pension, stipend, allowance, salary (cf.: honorarium, annuum, merces, stipendium)
-- Lewis & Short, A Latin dictionary (1879), p. 1618, 'salarius'
The key bit is in the first line. The supposed meaning 'salt money' ('sc[ilicet] argentum', i.e. 'with argentum implied') is not actually attested anywhere. It's inferred by analogy with some other, real, expressions: calcearium ('shoe money', from calceus 'shoe'); congiarium ('distribution of largesse', from congius 'half an amphora's worth'); and vestiarium ('clothing money', from vestis 'clothing'). Unlike salarium argentum, these terms actually do appear in various ancient sources, with the correct meanings.

Lewis & Short didn't invent the conjecture: it also appears in the older Latin-German dictionaries of Freund (1834) and Scheller (1804). It seems to have its origin in the 1st edition of Facciolati and Forcellini's Totius Latinitatis lexicon ('dictionary of the entire Latin language'):
Salarium, ii ...: proprie est annona salis, quae olim dabatur militibus.
'Salary' ...: strictly, the annual salt revenue, which was once given to soldiers.
-- Totius Latinitatis lexicon (1st edition, 1771), vol. 4 p. 15, 'salarius'
This was already a very muddled rendering of the evidence. Facciolati-Forcellini go on to cite Pliny, though as we have seen Pliny doesn't actually say this. It looks like what's happened is that they've conflated the Pliny passage with the Livy passage. Livy referred to a tax on the salaria annona 'annual salt production'. Annona can mean either 'annual production' or 'annual revenue', and Facciolati-Forcellini have taken Livy's phrase and used it with the other meaning: annona salis 'annual salt revenue'. Later on, Scheller and Freund took the idea and supported it with the analogies of 'shoe money', 'clothing money', and so on. And so the idea stuck.

All these dictionaries are engaging in conjecture. No ancient source ever actually uses salarium to mean 'salt allowance'. It's a guess. It isn't a terrible guess, but it's still a guess. One thing that weighs heavily against it is that even Pliny, who's trying to link salarium to 'salt' as closely as he can, doesn't try to get away with inventing 'salt money'.

The current standard, the Oxford Latin dictionary (1968), very properly avoids taking any view on the question. It just states that salarium comes from sal. Unlike the older dictionaries, it doesn't make any inferences about how or why the two words are related.

'Salt money' certainly isn't as ridiculous as the idea of paying soldiers in salt -- it does have parallels that make it at least a reasonable conjecture -- but there's still no evidence for it.

Sea water refinery in western France (source:

I don't have a perfect explanation for how the Latin word for 'salty' gave rise to the word for 'salary'. Of course I don't: that's why we have this myth floating around. We don't have the evidence to settle on a single explanation.

As I said above, 'salt allowance' isn't a terrible guess. But I strongly suspect it's much more metaphorical than that. Compare how the Greek word for a salary was opsōnion, literally '(money) for buying opson', where opson means 'fish, relish, sauce'. That doesn't mean Greek workers were given a 'fish allowance': it means that there was a generalised idea that wages went on traded goods like fish, and not on things like barley which land-owners would grow for themselves. Similarly, in Rome, grain allowances were a common thing; it could easily make sense to interpret salarium as 'everything-else-money'.

This interpretation is less specific, slightly metaphorical, and it's still just a conjecture. But I'd say it's more plausible, and certainly a more economical explanation, than inventing a specialised category of wages out of thin air.

We still haven't dealt with this:
A soldier's salary was cut if he "was not worth his salt," a phrase that came into being because the Greeks and Romans often bought slaves with salt.
Oh dear oh dear. This one has made it into Wikipedia too ('soldiers who did their job well were "worth their salt"'). Unfortunately for Time and for the thousands of people who have repeated this idea, the phrase 'worth one's salt' is definitely not Roman. It is first attested in the 1830s (; for sources, see OED under 'salt'). The thing about buying slaves with salt is fictional too.

And then there's 'salt of the earth', which comes up in the 2009 blog post I quoted at the start. I mentioned this in my previous post on 'salting the earth'. It's nothing to do with Roman soldiers: it's biblical, from Matthew 5:13 in the New Testament. This means that (1) it isn't a Roman phrase, but at closest, Helleno-Christian; (2) it's later than Pliny's mention of salarium; (3) it's about using salt as a fertiliser as much as anything else, as I argued in my earlier post.

World salt production in 2012. That year, China produced between 22.5% and 27% of the world's salt, well ahead of the USA, India, and Germany (in that order). (Generated using OpenHeatMap, based on Wikipedia figures)

Salt was certainly a significant strategic resource in antiquity. But calling it 'prized and valuable' is silly. Yes, it's the single most common preservative agent ever used, and it is by far the most common seasoning. The Roman salt trade was under state control from the earliest times (see e.g. Livy 1.33.9, 2.9.6); the Via Salaria or 'Salt Road' owed its name to its role in salt transportation; the Etruscan city of Veii owed much of its wealth to salt production; and access to salt even provoked a war between two German tribes at Bad Salzungen in the 1st century CE.

But 'prized and valuable' -- no. That suggests a special cultural status which isn't supported by any evidence. No one thought of salt as an heirloom, or used it for jewellery. No one talks about awarding salt as a prize for contests. There's no evidence anyone used salt bars as money -- not even as one among many forms of exchange, as in 19th century Ethiopia. Salt was not a prestige object.

Modern people who repeat these myths sometimes emphasise the high value of salt in the Roman world. Well, sure, the salt trade was valuable ... that's because it was traded in such high volume. But in 204 BCE, when Marcus Livius 'salt-dealer' imposed his tax on salt, Livy quotes the price of salt at a sextans: that is, one sixth of a copper as, or one 60th of a silver denarius (or in a civilian context, a sextans was one 96th of a denarius). Polybius, writing in the mid-100s BCE, quotes a foot-soldier's pay as 'two obols' per day, that is to say, one third of a denarius (Polybius 6.39.12).

In other words, a Roman pound of salt (ca. 330 grams) cost one twentieth of a foot-soldier's daily wages.

Important? Of course. Expensive by modern standards? Maybe, depending on the price of salt where you live. 'Prized and valuable'? No.

Actually that deserves more than a 'no'. It deserves a hearty laugh followed by a 'no'. Thus: 'Ha ha ha ha! No.' There, got it right now.