Monday, 31 December 2018

Top posts of 2018

This is a small-time blog, but I do see it cited here and there, so it’s not without its impact, in a minor way. Here are the big hitters from 2018 -- though I should perhaps also mention that a post from January 2017, ‘Salt and salary’, continues to be one of the most heavily visited, and would be occupying third place for 2018 if I counted it here.
  1. Not ‘the oldest written record of the Odyssey’ (12 July). Just because the Greek Ministry of Culture and Sports says it, doesn’t make it so. It is, however, pretty damn cool. Because it’s a clay tablet and it’s sloppily written and it’s at Olympia and everything about that is fantastic.
  2. The citation problem (21 September). Lazy people writing outside their own field don’t bother to look over existing research: OK, fine, in other news, water is wet. What’s really worrying, with well known journals like PLoS ONE and PNAS, is that apparently the editors and referees couldn’t care less.
  3. Ancient Greeks climbing Mount Olympus (25 April). Yes, they did climb the mountain. No, it isn’t a difficult climb. No, we don’t know if they expected to see gods up there. Yes, they did have at least one cult-site up at the top.
  4. Why are there no Romans named ‘Quartus’? (16 April). ‘Quintus’, ‘Sextus’, and ‘Decimus’ come from the same origins as ‘Marcus’, ‘Maius’, and ‘Junius’. There’s no ‘Aprilus’, though. Maybe because it’s Etruscan? Also, there’s a boo-boo in the Cambridge Latin Course.
  5. The not-so-cryptic oracle of Delphi (28 June). The oracle was a decision engine, not a prediction service. When the oracle spoke, it said ‘yes’ or ‘no’, ‘this’ or ‘that’. It didn’t try to confuse people.
  6. Fake quotations (27 September). ‘In quotations, truth is the first casualty.’ -- Albert Einstein. ‘A room without books is like a quotation without a source citation.’ -- Abraham Lincoln.
  7. Cosmos #1. Eratosthenes (20 November). We live in a world where people rely on Megan Fox and Giorgio Tsoukalos for their history. Some people go for Carl Sagan instead, and I guess there’s no denying he’s a step above them. But that’s a bit like upgrading from a clown car to a Trabant. I mean, come on.
  8. Paying the iron price: Spartan money (19 February). Maybe, just maybe, the Spartans did indeed use iron rods as money. But ‘money’ doesn’t mean what you probably think it means.
  9. Concerning Yule (18 December). Present-day Christmas customs are nearly all modern, but not pagan. Most Christmas customs are too recent to be based on Yule. The Yule log originated as a Christmas log, it seems.
  10. Thunderbirds in Atlantis (6 August). The 2015 series, not the 1965 one.
Also showing up in the stats --
  • Typically 30%-45% of people reading this blog are in the USA. UK readers accounted for anywhere up to 15% in the first half of the year, but dropped to 7%-9% in the second half. Other countries pop up in the stats inconsistently. NZ readers are typically around 1.5% to 4%.
  • Desktop computers provide the steadiest stream of readers. But when a post gets popular, it’s because it’s being read on iPhones. Nearly all the volatility in the visitor stats comes from iPhones. Android users have their moments too, but usually they track pretty closely with desktops, just a fair way behind. Linux users are normally around 4%-7% of the audience, but were more prominent in the first half of the year (10.5% to 14.5% between March and June: truly, this has been the year of the Linux desktop. That is, unless they were all bots, which is, let’s face it, likely).
  • This blog is not exactly big-time, so when bots and link farms come calling, they distort the stats quite a lot. This has been particularly noticeable in December, with unusual bursts of activity especially from Russian and Moldovan IP addresses. But there were also similar bursts of activity in March, May, and August.

Tuesday, 18 December 2018

Concerning Yule

HO. HO. HO.

Yule and Christmas have very different flavours. Yet it’s widely imagined that Christmas is derived from Yule, or that modern Christmas customs originated as Yule customs.

That idea is often motivated by anti-Christian sentiment. If Christmas is derivative, the idea goes, then that licenses a skeptic to treat it, and the people that celebrate it, as dishonest. But you don’t need to be a Christian (or a Neo-pagan, for that matter) to acknowledge that Christmas and Yule are very separate things.

Our earliest evidence on Yule and our evidence on Christmas come from different times and different places. Christmas originated as a Mediterranean festival, first attested in the 4th century but with a backdrop reaching back to the 2nd century. Yule pops up from the 6th century onwards as an East German and North German season of the year. There’s only the faintest trace of Yule in modern Christmas customs.

In previous years, in 2015 and in 2017, I wrote long posts about supposed links between Christmas and pagan Roman customs and festivals. The short answer is: there aren’t any.
  • Christmas has nothing to do with Mithras. Neither does Christianity in general. The supposed similarities are all imaginary, made up out of thin air, mostly in the 1990s.
  • Christmas isn’t based on Saturnalia. Saturnalia is on 17 December, and ancient Christians celebrated it alongside Christmas for a long time. We haven’t inherited any customs from Saturnalia -- it’s just too far in the past.
  • Christmas isn’t based on Sol Invictus. We have only one indication of a Sol Invictus festival on 25 December; it dates to 354 (not 274, as often claimed); it was celebrated in only one place (Rome); and it’s no older than Christmas, which is attested in the same document.
  • The date of Christmas is linked to the winter solstice, indirectly. Ancient Judaeo-Christian custom reckoned that prophets and saints died on the same date they were born or, in later times, the date they were conceived. Jesus supposedly died at the spring equinox, so by custom, that was also the date of his conception. That put his birth nine months later at the winter solstice. Evidence of Christian interest in the link between Jesus’ death and the equinox goes back to the 150s, so Christmas has its background in that period, even if we can’t be sure it was celebrated at that time.
  • The solstice is on 21 or 22 December these days, but in the Julian calendar, it was traditionally reckoned to be 25 December. 1st century pagan sources are very clear on this. That’s in spite of the fact that when the Julian calendar was first instituted, in 46 BCE, the solstice had already drifted a few days out of synch with that date. The solstice was on 25 December in the retrojected Julian calendar in the 4th century BCE, so that’s probably when the traditional date was fixed by Greek astronomers. (See this post, section 4, for more details.)
My older posts didn’t cover Yule. So let’s have a go now. Here’s a compressed timeline for quick reference.


The earliest evidence

The earliest source to mention Yule is a calendar of saints’ days dating to the 500s. This text, in Gothic, is in a palimpsest held at the Biblioteca Ambrosiana in Milan. It contains the phrase fruma jiuleis, which means either ‘first part of Yule’ or ‘before Yule’. It’s often reported that the phrase is a gloss of the word ‘November’, equating the Roman month to a Gothic season: Landau (2006) has shown that ‘November’ doesn’t appear in the manuscript, though he accepts, on other evidence, that fruma jiuleis probably does refer to November or December anyway.

The earliest reference to ‘Yule’: Gothic jiuleis, in a 6th century palimpsest. A palimpsest is a manuscript that has been recycled by partly polishing off the original text: the calendar is in the earlier layer of text. At bottom is a filtered version of the image. At the left, in the box, is where the word Naubaimbair (November) was believed to be by an early 19th century scholar: in the view of more recent scholarship, that reading isn’t supportable. Source: Landau 2006, figures 3 and 8.

(Let me just repeat that this is a calendar of saints’ days. There’s no mention here of Christians killing people for celebrating a season observed in a Christian calendar, no mention of ‘woodland spirits, feasting, male fertility’, no Christmas trees. Why on earth would anyone expect any of these things? Well, some people do. Observe:)


Where does the word jiuleis itself come from? Its linguistic origins are disputed. Landau argues (2009) that it’s derived from the biblical Jubilee (via Greek Ἰωβηλαῖος), and that already in the Gothic calendar it’s used as a nomen sacrum to refer to Jesus. That neglects the fact that some later forms of Yule in other languages display a velar fricative: Old English geohhol, Old Finnish (loanword) juhla. It’s more usual to infer an Indo-European origin (e.g. Koivulehto 2000). On the other hand, Landau is right to point out that Gothic jiuleis appears in a firmly Christian context, and centuries before any evidence of a non-Christian festival. I don’t think we have enough evidence to draw a firm conclusion on this point.

Our next reference to Yule appears in Bede’s De temporum ratione (‘on time-reckoning’), written around 730 in northern England, two centuries after the Gothic palimpsest. At §15 Bede lists off names of the lunar months in the English calendar. He states that giuli corresponded to two months, not one, namely December and January, and mentions that the English calendar was reckoned as starting on 25 December.

By the time of the Old English Martyrology, around the late 800s, 25 December itself is referred to as ‘the first Yule day’ (þone ærestan geohheldæg: Martyrology 25 Dec.), and December and January are known as ‘former Yule’ and ‘after Yule’ respectively (ærra geola, æftera geola: Martyrology start of Dec., 1 Jan.).

The use of Yule for month names is perhaps more suggestive of a season than a festival. Bede does mention something that looks like a pagan festival, though. He tells us that the New Year in the English calendar, that is 25 December, was called modranicht or mothers’ night. Not mother’s night, as it’s often reported: Old English modra is plural. Now, Bede can’t be trying to cast modranicht as a fixed date in the Julian calendar, or equate modranicht and Christmas in any religious sense. What he’s saying is that modranicht was the New Year; the New Year was reckoned as starting on the winter solstice; and the solstice is 25 December, which also happens to be the date of Christmas. (See above on the solstice being traditionally reckoned as 25 December in the Julian calendar.)

Evidence about Yule customs appears from the late 800s onwards, in Old Norse texts. At this point we also start to see distinctly pagan features. I don’t just mean Norse sagas: the sagas have tons of references to Yule (Old Norse jól), but they’re half a millennium after Bede. The earliest references are in poetry. The first is in the Hrafnsmál (raven’s song), reliably dated to the second half of the 800s. Stanza 6 refers to the custom of drinking a toast at Yule. Another less direct reference appears in the Helgakviða Hjörvarðssonar (poem of Helgi Hjorvarth’s son), which is a patchwork of multiple sources, probably mostly dating to the 900s. This poem mentions the custom of drinking a toast too, along with a vow made over the pledging-cup, at stanza 32. The Helgakviða doesn’t name the festival: jól only appears in the prose frame-narrative, written later; it also refers to a ‘sacred boar’ (sónargölt-).

One of the earliest references to Norse Yule customs: the Codex Regius manuscript of the Helgakviða Hjörvarðssonar, fol. 46. The first two highlighted passages are in the prose written around the verse as a frame-narrative: they refer to Yule (iola), the custom of making a vow over a pledging-cup (bragar fvll), and bringing in the sacred boar (sonar gꜹltr). The third highlighted phrase belongs to a verse section, and refers to the bragar fvll at the king’s toast (konvng borno).

The upshot of all this is that Christmas goes back centuries earlier than any of our evidence for Yule; the very earliest evidence for Yule is already in a Christian context; and Yule customs don’t show up until much later. Also, as I mentioned at the start, Christmas has its origins very firmly in the 2nd-4th century Mediterranean, while our only evidence for Yule is East/North Germanic.

With all this in mind it would be very weird to see Christmas as based on Yule in any sense. Christmas is a (2nd-)4th century Mediterranean festival; Yule is a season in eastern and northern Germanic calendars, linked to pagan customs by the 9th century, but of disputed origins.

Yule customs

Could we at least say that Christmas absorbed aspects of Yule over time? Well, what do we know about Yule customs?
  • Making vows over a toast, attested from the 8th century onwards in Norse texts, as we saw above. In modern times, vow-making has become linked to the Gregorian new year (New Year’s resolutions), not Christmas. And fair enough too: Bede does after all cast Yule as a season centred on the New Year, not as a religious festival.
  • A sacred boar is attested as early as the Helgakviða Hjörvarðssonar (see above). It hasn’t left much trace in modern Christmas customs, but it has left some: most prominently, the 15th century ‘Boar’s head carol’, though even that isn’t exactly well known these days. An adapted version of the custom does appear in some Neo-pagan celebrations of Yule, and in Scandinavia, and apparently a number of the more extravagant American universities lay on a boar’s head celebration. But it’s not exactly mainstream. Wikipedia claims that the modern western Christmas ham is based on the Yule boar, but doesn’t do the legwork to demonstrate continuity. Is there a line connecting the mediaeval boar’s head to the Christmas ham? Well I suppose there might be. Tracing it is beyond me, I’m afraid.
  • Feasting: a parallel for sure, but this is hardly distinctive to any one festival.
  • Spirits and hags coming out to wreak havoc: this happens at Yule an awful lot in Norse sagas. As a modern Christmas custom? Not so much. Not in the English-speaking world anyway. (Unless you want to count Tim Burton’s The nightmare before Christmas (1993) ... but it’s clearly not related, even though it is a great film.)
We’ve got one custom that has stuck to the modern western New Year, not Christmas; one doubtful case (the boar/Christmas ham); one that is typical of nearly every festival that has ever existed; and one that definitely is not represented in (most people’s) modern Christmas customs.

In Grettis saga, Yule is spent fighting berserks or getting all your bones broken by a draugr. In Tim Burton, the Pumpkin King steals Santa’s sleigh. Not much in common, really. It may be that there’s more of Yule in The Elder Scrolls games than in modern Christmas customs.

But wait, I hear the protesters saying, what about the Yule log? That’s a pagan custom that has survived to the present day!

Oh no it isn’t.

Oh yes it is!

Oh no it isn’t.

The Yule log, it is usually claimed, is first attested in 1184. That’s kind of true. But I doubt anyone who has claimed this in the last 50 years actually knows what the evidence for this is. They certainly haven’t checked the original source. I had to go to an 1899 book just to find out what the source is. And, it turns out, it’s been drastically misrepresented.

The source is an edict from a Christian bishop outlining the prerogatives of the Christian parish priest of Ahlen. These prerogatives include ‘a tree at the Nativity of the Lord’ -- not Yule -- ‘to be brought for his own fire at the festival’ (& arborem in Nativitate Domini ad festivum ignem suum adducendam esse, Kindlinger 1790: 210). So
  • the 1184 source doesn’t mention or allude to Yule;
  • it explicitly and specifically links the log to Christmas;
  • Yule sources belong to Britain and Scandinavia, but the 1184 edict is about Westphalia, in western Germany.
See further Tille 1899: 90-96, who cites more examples of early Christmas logs, and shows evidence that the fires are more utilitarian than religious.

In Britain, the earliest attestation is much later: the log first appears in the early 1600s, in a poem by Robert Herrick. He too calls it a ‘Christmas log’, not a ‘Yule log’. Don’t read too much into the fact that it dates to Protestant times: Herrick loved to troll Puritans.

Did the Yule log start out as a Christmas log, and only get rebranded as a ‘Yule log’ in later centuries? It looks that way to me. The best counter-argument Tille can find (1899: 88-89) is a letter written in 742 by a missionary in Germany, St Boniface, which tells of a disagreement over celebrating the New Year with pagan customs. People of formerly pagan German tribes had noticed
that on the first day of January year after year, in the city of Rome and in the neighborhood of St. Peter’s church by day or night, they have seen bands of singers parading the streets in pagan fashion, shouting and chanting sacrilegious songs and loading tables with food day and night, while no one in his own house is willing to lend his neighbor fire or tools or any other convenience. ...
-- Boniface, letter to Pope Zacharias (trans. Emerton)
Tille takes this as an indication that native German customs involved a sacred fire too. That’s a pretty thin argument. It’s not impossible: there’s a 400 year gap between this and the Christmas log of 1184, but it is the same part of the world -- Ahlen is in Westphalia, Boniface refers to western and southern German tribes. But even if Tille is right, we don’t have corroboration for Yule in Germany. I’m not inclined to believe the Yule log was originally a Yule log.

OK, how about Christmas trees -- the kind you decorate, not the kind you burn? They’re a Yule custom, surely? That’s another no: Christmas trees are another German custom. Even in Germany, we only start to see them in the 1500s, and they didn’t become popular outside Germany until the 1800s. (Famously, they were popularised in Britain by Prince Albert in 1840 -- though Queen Charlotte did set one up at Windsor in 1800.)

Gift-giving? Well, Christmas charity to the poor goes back to the 1200s or thereabouts, but gifts within the family are much more recent. Santa is based on an ancient Christian saint, St Nicholas, but St Nicholas had nothing to do with Christmas until Luther tried to suppress the cult of the saints in the early 1500s.

Is this all just Christian apologetics?

No. If we say there’s barely any trace of paganism in Christmas as practised in the English-speaking world, that isn’t the same as saying that there’s any authenticity about modern Christmas customs. At least, not ‘authentic’ in the sense of customs that have survived since antiquity.

There’s virtually nothing pagan about modern western Christmas customs. But at the same time, nearly all modern Christmas customs are exactly that -- modern.

The Christmas log, Christmas trees, and gift-giving all come from late mediaeval or early modern Germany. Santa went through several phases, starting out as St Nicholas with no connection to Christmas, then metamorphosing into the Christkind and Sinterklaas, before re-coalescing into Santa. Christmas trees only spread to France around 1830, and England in 1840 (1800 if you’re a Queen Charlotte fan). Santa’s flying reindeer were invented for the poem ‘’Twas the night before Christmas’ in 1823. Advent calendars only started to become popular in the early 1800s, Advent wreaths in 1839, Christmas cards in 1843 -- and it was also in 1843 that Charles Dickens published A Christmas carol.

And there’s nothing wrong with that. If Christmas customs are modern, well, so what? So are Neo-pagan customs relating to Yule and the solstice. There’s no rule that customs have to be ancient. Kwanzaa dates to the 1960s. The midwinter festival in my part of the world, Matariki, dates to the 2000s. They’re still real festivals.

The only bits of Christmas that are ancient are the bits that happen in church. A number of Christmas carols are pretty old: a handful are even ancient. Sizeable chunks of the liturgy are ancient.

The story of the nativity is certainly ancient. (Not that that implies it’s true, mind.) It’s a mash-up of the 1st century gospels of Matthew and Luke. Matthew has the star, dreams and prophecies, the wise men, the massacre of the innocents, and the flight to Egypt; Luke has the annunciation, the census, no room at the inn, the manger, and the shepherds. The idea of combining these two stories in a mash-up is ancient too. Even some non-canonical parts of the story are ancient. The Protevangelium of James (early 2nd century) gives us the virgin birth, as opposed to the virgin conception, and the idea of Jesus being born in a cave. The ox and ass standing by, a standard feature of modern nativity displays, appear regularly in ancient iconography and in some ancient Christian writers.
The ox and ass are premised on Luke’s manger, and by analogy with Isaiah 1.3 and the Septuagint version of Habakkuk 3.2; see e.g. Benedict XVI 2012: 69. For ancient sources, see: Origen, Homily on Luke 13, xiii.1832c Migne; Prudentius Cathem. 11.81-84. (Also, at this point I have to mention that Prudentius is the author of my personal all-time favourite Christmas carol, ‘Of the father’s heart begotten’, Cathem. 9.10-24.)

If a 2nd century Christian were to time-travel to 2018, they’d definitely recognise the story and motifs in a Christmas pageant, and in films like Ben-Hur (1959) and The star (2017). They wouldn’t recognise anything else about Christmas in its modern form. But then again, neither would an 8th century Northumbrian who was used to celebrating modranicht. Nor would a 9th century Norse person who was used to jól.

References

Wednesday, 12 December 2018

Cosmos #3. Hypatia and the library

This is the last of three annotated transcripts of segments on ancient Greek science from Carl Sagan’s Cosmos (1980). See my introduction in part 1 on the impact Cosmos has had, its extraordinary influence in propagating some myths, and in creating others; and my introduction in part 2 on how some misinformation is a direct result of Sagan’s choice to set up science and religion as antithetical to one another.

This final episode is about Alexandria, and the idea that knowledge is something to be treasured. Sagan is right about that. But he’s wrong when it comes to his moral condemnation of the loss of knowledge. He wants to blame someone, and religion is on his radar. The loss of knowledge isn’t a good thing, but it’s also a historical inevitability.

Episode 13. ‘Who speaks for earth?’

YouTube link. First broadcast 21 December 1980.

Carl Sagan:
One of the great revelations of the age of space exploration is the image of the earth, finite and lonely, somehow vulnerable, bearing the entire human species through the oceans of space and time.

But this is an ancient perception. In the 3rd century BC our planet was mapped, and accurately measured, by a Greek scientist named Eratosthenes, who worked in Egypt. This was the world as he knew it. Eratosthenes was the director of the great library of Alexandria, the center of science and learning in the ancient world.

An ancient atlas: Ptolemy’s latitude and longitude data rendered onto Ptolemy’s own projection by Hans van Deukeren. Ptolemy’s atlas followed in the footsteps of several earlier ones, especially Marinus of Tyre, but Eratosthenes was the grandfather of ancient cartography. Eratosthenes’ atlas probably didn’t look much like this, though. Much or most of Ptolemy’s data came from Roman-era surveys; we don’t know what projection Eratosthenes used, but it wasn’t this one; and it’s possible Eratosthenes didn’t arrange his map with north at the top.

Aristotle had argued that humanity was divided into Greeks and everybody else, who he called ‘barbarians’; and that the Greeks should keep themselves racially pure. He taught it was fitting for the Greeks to enslave other peoples. But Eratosthenes criticized Aristotle for his blind chauvinism. He believed there was good and bad in every nation. The Greek conquerors had invented a new god for the Egyptians, but he looked remarkably Greek. Alexander was portrayed as pharaoh in a gesture to the Egyptians. But in practice, the Greeks were confident of their superiority. The casual protests of the librarian hardly constituted a serious challenge to prevailing prejudices. Their world was as imperfect as our own.

Sagan is basically right about Aristotle -- certainly in his criticism of Aristotle’s thinking on slavery, at least, which is awful enough that you will want to shower after reading it (Politics 1.2 = 1253b-1257a).

However, ‘Eratosthenes criticized Aristotle’ is fantasy. We do not know what Eratosthenes thought about racism or even about Aristotle. The surviving fragments of Eratosthenes’ Geographica do not mention Aristotle.

But the Ptolemies, the Greek kings of Egypt who followed Alexander had at least this virtue: they supported the advancement of knowledge. Popular ideas about the nature of the cosmos were challenged, and some of them discarded. New ideas were proposed and found to be in better accord with the facts. There were imaginative proposals, vigorous debates, brilliant syntheses, and the resulting treasure of knowledge was recorded and preserved for centuries on these shelves. Science came of age in this library.

The Ptolemies didn’t merely collect old knowledge. They supported scientific research and generated new knowledge. The results were amazing. Eratosthenes accurately calculated the size of the Earth, he mapped it, and he argued that it could be circumnavigated. Hipparchus anticipated that stars come into being, slowly move during the course of centuries, and eventually perish. It was he who first cataloged the positions and magnitudes of the stars in order to determine whether there were such changes. Euclid produced a textbook on geometry which human beings learned from for 23 centuries. It’s still a great read, full of the most elegant proofs. Galen wrote basic works on healing and anatomy which dominated medicine until the Renaissance. These are just a few examples. There were dozens of great scholars here, and hundreds of fundamental discoveries.

Some of those discoveries have a distinctly modern ring. Apollonius of Perga studied the parabola and the ellipse, curves that we know today describe the paths of falling objects in a gravitational field, and space vehicles traveling between the planets. Heron of Alexandria invented steam engines and gear trains; he was the author of the first book on robots.

Imagine how different our world would be if those discoveries had been explained and used for the benefit of everyone, if the humane perspective of Eratosthenes had been widely adopted and applied. But this was not to be.

Let’s just repeat these two snippets, juxtaposed:
Euclid produced a textbook on geometry which human beings learned from for 23 centuries. ... Galen wrote basic works on healing and anatomy which dominated medicine until the Renaissance.
Imagine how different our world would be if those discoveries had been explained and used for the benefit of everyone ... But this was not to be.
You don’t often see a normally level-headed person contradict themselves quite this quickly.

Alexandria was the greatest city the western world had ever seen. People from all nations came here to live, to trade, to learn. On a given day these harbors were thronged with merchants, and scholars, tourists. It’s probably here that the word ‘cosmopolitan’ realized its true meaning, of a citizen not just of a nation, but of the cosmos. To be a citizen of the cosmos. Here were clearly the seeds of our modern world.

But why didn’t they take root and flourish? Why, instead, did the west slumber through a thousand years of darkness until Columbus and Copernicus and their contemporaries rediscovered the work done here? I cannot give you a simple answer but I do know this: there is no record in the entire history of the library that any of the illustrious scholars and scientists who worked here ever seriously challenged a single political or economic or religious assumption of the society in which they lived. The permanence of the stars was questioned. The justice of slavery was not.

Columbus and Copernicus did not ‘rediscover’ anything. Columbus’ ambitions were solely colonial. He went out of his way to reject Eratosthenes’ work on the size of the earth when his opponents tried to remind him of it in Salamanca, in favour of guesswork and incomplete reports by Marinus of Tyre, Marco Polo, and Pierre D’Ailly.

Copernicus should really be credited as a discoverer rather than a rediscoverer. His idol wasn’t any of the Alexandrian researchers, but Pythagoras, whom Sagan rightly casts as a mystic (see part 2).

Neither of them knew or worked with any ancient sources that were obscure to their contemporaries.

Science and learning in general were the preserve of the privileged few. The vast population of this city had not the vaguest notion of the great discoveries being made within these walls. How could they? The new findings were not explained or popularized. The progress made here benefited them little. Science was not part of their lives.

The discoveries in mechanics, say or steam technology mainly were applied to the perfection of weapons, to the encouragement of superstition, to the amusement of kings. Scientists never seemed to grasp the enormous potential of machines to free people from arduous and repetitive labor. The intellectual achievements of antiquity had few practical applications. Science never captured the imagination of the multitude. There was no counterbalance to stagnation, to pessimism, to the most abject surrender to mysticism. So when at long last the mob came to burn the place down, there was nobody to stop them.

Sagan’s thesis that slavery obstructed the development of ‘science’ -- though really it sounds like he’s talking about industrial engineering, not science -- is coherent, but hugely over-simplified. The connection between slavery and something as distantly connected as engineering is always going to be a complex one, dependent on a huge range of historical circumstances.

The main impact of slavery, insofar as it had anything to do with engineering, was economic. It’s a bit of a trite platitude these days that a slave economy had no need, and no room, for an industrial revolution. On that view, it’s true that it was impossible for industrial engineering to benefit society at large -- but not because of ‘mysticism’, or ‘pessimism’, or a failure of imagination; but because it was economically impossible.

There’s probably some truth to that platitude. But the subject of what the Romans didn’t invent involves speculation and complexities. OK, maybe they didn’t have an economic framework that would reward that kind of industrialisation. But as well as that, they didn’t prize fuel (coal); they didn’t know Boyle’s Law; hell, they didn’t even know Newton’s Laws, not even the First Law. Steam engines depend on all of these things, including the more sophisticated Third Law, that every force has an opposing force. Like it or not, the theoretical considerations are important. Heron probably had only a hazy idea of how his steam engine worked. For the ancients it was a curiosity, not a tool: you can’t scale an industrial tool if you don’t understand how it works.

We’ll hopefully all agree with Sagan that slavery is a bad thing, and that the monstrousness of slavery makes any society like that a model to avoid. But, as so often with Sagan’s historical arguments, there’s absolutely no reason to bring mysticism or superstition into it.

Let me tell you about the end. It’s a story about the last scientist to work in this place. A mathematician, astronomer, physicist, and head of the school of Neo-Platonic philosophy in Alexandria. That’s an extraordinary range of accomplishments for any individual, in any age. Her name was Hypatia. She was born in this city in the year 370 AD.

We do not know the date of Hypatia’s birth. ‘370’ is an estimate based on guessing her age at the time of her death in 415 (Sagan reports her death-date below accurately).

This was a time when women had essentially no options. They were considered property. Nevertheless, Hypatia was able to move freely, unself-consciously through traditional male domains. By all accounts she was a great beauty. And although she had many suitors, she had no interest in marriage.

The Alexandria of Hypatia’s time, by then long under Roman rule, was a city in grave conflict. Slavery, the cancer of the ancient world, had sapped classical civilization of its vitality. The growing Christian church was consolidating its power and attempting to eradicate pagan influence and culture. Hypatia stood at the focus, at the epicenter of mighty social forces. Cyril, the bishop of Alexandria, despised her: in part because of her close friendship with a Roman governor, but also because she symbolized, she was a symbol of learning and science which were largely identified by the early church with paganism.

In great personal danger Hypatia continued to teach and to publish, until in the year 415 AD, on her way to work, she was set upon by a fanatical mob of Cyril’s followers. They dragged her from her chariot, tore off her clothes, and flayed her flesh from her bones with abalone shells. Her remains were burned, her works obliterated, her name forgotten. Cyril was made a saint.

The glory you see around me is nothing but a memory. It does not exist. The last remains of the library were destroyed within a year of Hypatia’s death.

When Hypatia died, in 415, the Serapeum had been gone for 24 years. It was destroyed in 391. I won’t labour the story of exactly how she died, though there are some problems there too.

Sagan also does not mention that one of Hypatia’s best teacher-student relationships was with Synesius, a Christian, who became bishop of Ptolemais and Metropolitan of Pentapolis in 410.

As to the library itself: it’s hard to gauge how much of a library existed in the Serapeum in 391. There’s no mention of it in the accounts of the temple’s destruction (Theodoret, History of the church 5.22; Eunapius, Lives of the sophists 472). Ammianus, writing in the first part of the 300s, refers to a Serapeum library in the past tense, and gives the distinct impression that it was destroyed in Caesar’s invasion in 48 BCE (Ammianus Marcellinus 22.16.12-13).

But there are some sensible counter-arguments too. Alexandria was certainly still a centre of learning. The fact that Theodoret’s and Eunapius’ terse accounts don’t mention books doesn’t mean they weren’t there: it was first and foremost a temple after all. Ammianus is definitely confused about the multiple libraries in classical Alexandria, and appears to have mixed up the Serapeum with the royal library (it was the latter that was destroyed in Caesar’s invasion). Aphthonius, writing in the late 300s (or even later?) describes public book collections in rooms off the temple colonnades, though we can’t be sure his rhetorical exercise is an eyewitness account (Progymnasmata 48 ed. von Spengel; §12 tr. M. Heath). Eunapius talks of Alexandria as ‘a sort of sacred world, because of the temple of Serapis’, but he says this in the context of describing the hordes of students that thronged to become disciples of one Antoninus, a mystic-cum-philosopher, who supposedly foretold that the end of the Serapeum would coincide with the death of his philosophy (Lives of the sophists 471(a), 471(b)). In any case, the story of Antoninus makes it clear that the Serapeum was still an important element of Alexandrian cultural life.

It’s still pretty vague. There’s enough for some people to believe some fashion of library still existed in 391. But it won’t convince everyone, and there’s certainly no exactness about what did or didn’t exist. If, say, Eunapius’ Antoninus really was a mystic-philosopher, in a sort of Pythagoras-cum-Apollonius-of-Tyana vein, and if that’s representative of what kind of teaching went on at the Serapeum, then it’d be hard to think very highly of its culture. (Who knows, maybe the Serapeum was a bad spot in Alexandria’s intellectual landscape! Hypatia’s mathematical expertise survived it by a few decades, anyway.)

It’s as if an entire civilization had undergone a sort of self-inflicted radical brain surgery so that most of its memories, discoveries, ideas, and passions were irrevocably wiped out. The loss was incalculable. In some cases, we know only the tantalizing titles of books that had been destroyed. In most cases, we know neither the titles nor the authors.

It is vanishingly unlikely that anything at all was lost in 391 that wouldn’t have been lost anyway. The ancient Mediterranean had thousands of libraries. Any book that depended on a single library for its survival was already doomed, because no ancient library has survived to the present. To imagine the survival of any ancient library from antiquity to the present day is to imagine a miracle.

We have reports of many ancient library fires. The case of Alexandria is famous because the library was big during the Ptolemaic era -- and because Sagan has publicised it -- not because its collection was unlike anything else. When 1st century CE Egyptian scribes made the sole surviving copy of Aristotle’s Constitution of the Athenians (BL pap. 131), they certainly didn’t do it in Alexandria: you don’t make a 200 km hike for a job done on the cheap (the papyrus is recycled).

Alexandria has a place of honour in any account of the development of human knowledge. But the story of the loss of ancient knowledge is a story of economics, not of library burnings. In that story, Alexandria doesn’t even warrant a footnote.

We do know that in this library there were 123 different plays by Sophocles, of which only seven have survived to our time. One of those seven is Oedipus rex [Oedipus the king]. Similar numbers apply to the lost works of Aeschylus, Euripides, Aristophanes. It’s a little as if the only surviving works of a man named William Shakespeare were Coriolanus and A winter’s tale, although we had heard that he had written some other things which were highly prized in his time -- plays called Hamlet, Macbeth, A midsummer night’s dream, Julius Caesar, King Lear, Romeo and Juliet.

History is full of people who out of fear, or ignorance, or the lust for power have destroyed treasures of immeasurable value, which truly belong to all of us. We must not let it happen again.

This last bit lays bare the fundamental flaw in Sagan’s thinking. He sees the loss of knowledge as a crime. Where there’s a crime, there must be a criminal.

But that isn’t how it works. The default state of knowledge is not to be preserved. Don’t believe me? Go and try to read a file off a 5½" floppy disk. Hell, try to read a file off a CD-ROM that you burned ten years ago. Preserving knowledge doesn’t mean storage, it means copying, endlessly and without pause. If the reproduction technology is laborious or expensive, as it was in antiquity, it’ll be that much harder.

It doesn’t take much for the process to fail. Even in antiquity, you can find people in the 2nd century CE bemoaning the fact that they can no longer find a copy of one of Cicero’s speeches. As much as half the Epic Cycle may have been lost by the time of Augustus. We don’t know of any ancient writer who ever got to see a copy of book 2 of Aristotle’s Poetics (other than Aristotle himself).

Now, that’s trying to preserve knowledge for just a handful of centuries. If you want it preserved for 1600 years, you have to imagine thousands of people collaborating, with no lasting supervision, no political continuity, no continuity in funding, and no mutual agreement. The fact that anything has survived that long is amazing.

Even nowadays, a couple of decades of neglect would be enough to ensure the destruction of any modern library, acid-free paper or no acid-free paper. Ever heard how the video tapes of the Apollo 11 moon landing were lost? Or conversely, have you heard of the Archimedes palimpsest, and how new technological developments have allowed people to read heavily damaged texts that survived nowhere else? But were you aware that most of the damage to that book was done in the 1900s, not in the mediaeval period? That’s just how it is. Stuff gets lost, stuff gets damaged. If you wait long enough, every library disappears. Alexandria isn’t a tragedy. It’s a miracle that it happened at all, let alone that it lasted 7 centuries.