Alexandria, the royal quarter, Egypt

The Library of Alexandria

From the third century BCE (the books and the reports are the question)

The Bibliotheca Alexandrina, a library opened in 2002 on the harbor. Not the ancient library. — The Bibliotheca Alexandrina, opened in 2002, beside the harbor. A new building. It is not the Ptolemaic library, and no ancient wall of that library has been identified under it. Vyacheslav Argenberg, CC BY 4.0. Shown unedited.

The Ptolemies gathered scholars and books in Alexandria, copied what visitors carried, and had the sky measured from a noon shadow. Later writers report fires and a huge collection. The reports do not agree on what burned, how many rolls there were, or whether the books we lack were unique to this house.

Why this is extraordinary

In the early Ptolemaic city a king kept a house of scholars, the Mouseion, inside the royal palaces, and with it a library that later writers treat as the largest Greek collection of its age. The scholars ate together. A priest stood over the house. The books, on the accounts we have, were bought, borrowed, translated, and copied, including books taken from ships in the harbor. Strabo, who walked the palaces after Julius Caesar's war, describes the dining hall and the walk and does not, in that description, use the word library. No trench used here has produced a room that can be named as that library. The modern city sits on the ancient one. The tilted building in the photograph opened in 2002. It is a new library on the harbor, not the old shelves.

People often sort the losses into about three fires, and a fourth report sits beside them. This room does not turn that habit into a count. Caesar's war in 48 BCE, the damage to the royal quarter under Aurelian in the 270s CE, the end of the Serapeum of Alexandria in 391, and a story written down in the thirteenth century about the conquest of about 642 are four reports. Each has a text, a date of writing, and an open edge. A warehouse of books by the water is not the same sentence as the palace library. A ruined quarter is not, by itself, a list of burned titles. A destroyed temple is not, by itself, a catalogue. A late story can carry an older memory, and it can also be a story composed centuries after the day.

What the house is said to have held is the other half of the same question. Ancient counts of the rolls do not agree with each other, and some of those counts do not agree from one manuscript to the next, so this room adopts none of them. Works we know by title do not survive: Eratosthenes' own book on the measurement, the tables of Callimachus, the anatomical books of Herophilus. A small set of Athenian plays survives out of the much larger bodies ancient lists give those poets. Other books associated with this city did survive, Euclid's Elements and Ptolemy's Almagest among them, because someone copied them onward. One reading is that fires removed copies that existed nowhere else. The other is that much had already spread, and what we lack is what later copyists stopped copying. Both stay.

Alexandria was also a center of work on the sky. Eratosthenes, named as head of the library, measured the Earth from a noon shadow. Cleomedes, who preserves the method, has sundials at Syene and at Alexandria, a distance of 5,000 stadia, and a circumference of 250,000 stadia. Strabo reports 252,000. The length of the stadion is not one ancient measure, so no kilometer figure is used here. No published alignment of a library wall is used either. No such wall has been identified to measure.

Diagram of Cleomedes' account of the noon shadow between Syene and Alexandria. — Cleomedes' report, drawn for this room. One fiftieth of a circle, 5,000 stadia, 250,000 stadia. Strabo's 252,000 is not drawn. No mile figure. Not a direction of a library wall. Diagram: Our Deep History.

Why it matters

If the fires removed copies that existed nowhere else, then Greek drama, medicine, and Eratosthenes' own book are missing because of those days, and a papyrus can still surprise us. If the books had already been copied to other cities, the missing works are the ones that stopped being copied, and a fire in one building is not the whole path of loss. The count of rolls, and the count of fires, change what "lost knowledge" can mean. Neither count is a number this room will pretend to own.

The pin is the royal quarter under the modern city, where Strabo places the palaces, and where the new Bibliotheca Alexandrina now stands. It is not a surveyed wall. The temple of Serapis on the hill of the column called Pompey's Pillar is the Serapeum of Alexandria, the place later readers give a smaller, outer library. That temple is not the bull galleries at Saqqara, which have their own room. Another court that sent for texts and kept them is the library at Nineveh. The writing there is clay, and the century is earlier. An Egyptian priest of this same early Ptolemaic world, Manetho, wrote a history of Egypt that survives in later summaries, weighed on the ancient sources page. His book is not a shelf-list of this library.

Photo tour

7 photographs, unedited. On a phone, swipe the row. Choose one to look closely.

A nineteenth-century artist's drawing of a columned library hall with readers and shelves of rolls. — A nineteenth-century drawing of the library's hall, by O. Von Corven. An artist's reconstruction. It is not a plan of a dug room, and no ancient wall has been matched to it. O. Von Corven, Public domain

A tall ancient column on a ruined platform, with the modern city of Alexandria behind it. — The column called Pompey's Pillar, on the hill of the Serapeum of Alexandria. The inscription honors Diocletian. This is the temple later readers give a daughter library. It is not the royal library, and it is not the bull galleries at Saqqara. Daniel Mayer, CC BY-SA 4.0

A marble bust of a bearded god with a grain measure on his head. — A Roman marble bust of Serapis in the Vatican, with a grain measure on his head. A copy of a Greek type. It was not lifted from the Serapeum of Alexandria. The god is the one that temple housed. Jastrow, Public domain

A gold coin in the Pergamon Museum, Berlin. The file's description reads the portraits as Ptolemy II and Arsinoe II, and the inscription as "of the siblings." A portrait of that court, not a coin struck for the library. MatthiasKabel, CC BY-SA 3.0

Fragments of a Greek papyrus with lines of verse. — PSI 1092, in the Biblioteca Medicea Laurenziana, Florence. A papyrus of the first century BCE with Callimachus's poem on Berenice's lock. A later copy of a court poem about a constellation, not a roll lifted from the royal shelves. Unknown photographer, Public domain

A page of a Greek manuscript with a folio number 143. — Folio 143 of BnF Grec 2392, a fifteenth-century Greek copy of Ptolemy's Almagest. Not Ptolemy's own hand. He worked in Alexandria in the second century CE. This page is how a book from that city reached a later reader, by being copied. Bibliothèque nationale de France, Public domain

No ancient shelf label reading "from the ships" is in a free photograph here. Galen's label is a sentence in a commentary.
The Pinakes do not survive as a book, so there is no photograph of the catalogue.
Drawings of the Earth measurement in circulation add a mile figure and a well. The diagram in this room does not.

What is still unexplained

These sit beside the consensus account, with the same visual weight. None of them is marked solved.

Open Palaces, and no dug library Strabo places the Mouseion inside the royal palaces: a public walk, a hall with seats, and a large house where the men of learning ate together and held property in common. A priest set over them was once named by the kings and, in his day, by Caesar. He does not, in that passage, say library. A satirist of the third century BCE, Timon of Phlius, already mocked scholars as bookish men fed in the Muses' cage, so the staffed house is early, not only a Roman memory. The quarter is called the Bruchion. Harbor work and digging in the modern city have mapped pieces of the ancient waterfront. None of the sources used here names a cleared room as the library. Twenty lecture halls at Kom el-Dikka, built most plausibly in the late fifth and early sixth centuries CE, with some use into the seventh, show that teaching went on in the city centuries later. They are not the Ptolemaic shelves. A newer hypothesis moves the Mouseion toward that central ground. Strabo's palaces stay by the harbor. The catch is that both placements are readings of texts and of a city that has been rebuilt. The photograph of the tilted roof is the library opened in 2002.

Open Ships, copies, and a catalogue The Letter of Aristeas, a Jewish Greek book probably of the second century BCE, has Demetrius of Phalerum buying and transcribing books for the king, with more than 200,000 rolls already in hand and a hope of 500,000. He tells the king the laws of the Jews are missing, and that they need translating. Galen, in the second century CE, repeats an older medical writer, Zeuxis. In that story a Ptolemy ordered the books of everyone who sailed in to be brought to him. The books were copied. The owners received the new copies. The originals went to the library with a note that they were from the ships. Galen says the books were piled first in other buildings, and only later placed in the library. The same king, he says, took the Athenian state copies of Aeschylus, Sophocles, and Euripides against a deposit of fifteen talents of silver, kept the old rolls, sent back magnificent copies, and told Athens to keep the money. Later readers call that king Ptolemy III. Galen's text, in the translation used here, says "this Ptolemy" and does not number him. The catch is the gap in time. Galen is explaining odd marks in a medical book, and he says Zeuxis's own commentary had become hard to find. No earlier account used here confirms the ships or the fifteen talents. The practice can be a memory of real collecting, and it can be a story told to explain a label.

Open Tables, and a translation of the Law The Byzantine encyclopedia called the Suda gives Callimachus a work, the Pinakes, in 120 books: tables of people distinguished in every branch of learning, and of what they wrote. The books themselves do not survive. Later scholars quote pieces. One reading is that the Pinakes were the library's catalogue. Another is that they were a bio-bibliography that used the shelves, not the shelf-list itself. Callimachus is the cataloguer in that tradition. A place for him as head of the library is not in every later list. The names those lists do give for the heads include Zenodotus, Eratosthenes, Aristophanes of Byzantium, and Aristarchus of Samothrace. The Letter of Aristeas sets the translation of the Jewish Law under Ptolemy II, with seventy-two scholars finishing in seventy-two days on the island of Pharos. The name people use, Septuagint, means seventy. The Letter has the scholars working as a company. Later writers add that translators working apart produced the same words. Philo, in Alexandria, also tells of a translation made under Ptolemy II. The Greek Law exists. The catch is that the Letter is not a diary of the king's own year, and Demetrius's career sits mainly under the first Ptolemy, so the office the Letter gives him may be a later arrangement of the story. The buying, the copying, and the existence of the Greek Law can be weighed apart from that name.

Open Caesar's fire, 48 BCE During the Alexandrian War, Caesar set fire to ships in the harbor. His own account describes that fire and does not mention books. Plutarch says the fire spread from the dockyards and destroyed the great library. Dio says the docks and the storehouses of grain burned, and books with them. One English version of Dio calls those books the library. The Greek is also read as storehouses of grain and of books, which need not be the palace collection Plutarch names. Galen had already described books waiting in piles in other buildings, so a waterfront store of rolls is a possible object. It is not proof that those were the only rolls. The numbers disagree. Seneca, citing Livy, speaks of 40,000 books burned. Orosius gives 400,000 in the reading most manuscripts carry, and 40,000 in others. Aulus Gellius says nearly 700,000 volumes, acquired or copied under the Ptolemies, were burned in that war, and some copies of Gellius give 70,000. Strabo, a generation later, describes the Mouseion as a living house and does not say the library is gone. Suetonius says Domitian, restocking burned libraries in Rome, sent scribes to Alexandria to transcribe and correct books. That is the late first century CE. It shows that Rome still thought authoritative books were there to copy. It is not a list of what the scribes found.

Open The Bruchion under Aurelian Ammianus Marcellinus, writing in the late fourth century, says that under Aurelian the quarrels of the citizens became deadly, the walls were destroyed, and the city lost the greater part of the Bruchion, long the home of distinguished men. The recovery of Alexandria from Palmyra is placed about 272 CE. Ammianus does not write the sentence "Aurelian burned the library." The reading that the royal library ended then is an inference from the quarter where the palaces stood. The reading that the books had already gone in 48 BCE, or that part of them survived the street fighting, is just as open, because the text names the district and the men, not a shelf. In the same digression Ammianus still finds teachers in the city, and geometry, music, and the study of the motions of the heavens, though he says the numbers are few. A living school is not a surviving palace library. It does keep the city as a place of study after the Bruchion was torn.

Open The Serapeum of Alexandria, 391 In 391 the bishop Theophilus, under the edicts of Theodosius against pagan cult, brought down the Serapeum of Alexandria, the temple of Serapis in the Egyptian quarter. Rufinus, Socrates, Sozomen, and Eunapius describe that day. They describe the temple and the image. They do not, in the accounts used here, describe a catalogue of books burned with the statue. Ammianus, writing shortly before 391, says the Serapeum held libraries beyond calculation, and in the same passage he ties 700,000 volumes to Caesar's war. John Tzetzes, in the twelfth century, reports a second, outer library of 42,800 rolls. Later readers place that outer library in this Serapeum and call it a daughter library. Orosius, after 391, says he saw book chests in temples emptied in his own time. He does not, in that sentence, name this temple. So a smaller collection at the Serapeum of Alexandria is a real report. Whether it was still on the shelves in 391, and whether that day's destruction took the books, is open. The column in the photograph is called Pompey's Pillar. Its inscription honors Diocletian, around the end of the third century CE. It stands on the temple hill. It is not a column of the royal library, and this hill is not Saqqara.

Open A report from about 642 Arabic writers of the early thirteenth century say that when Amr ibn al-As took Alexandria, about 642 CE, a scholar asked for the books, Amr wrote to the caliph Umar, and the books were destroyed. Abd al-Latif, writing about 1200, has a short remark that Amr burned the library by Umar's order. Ibn al-Qifti and Bar Hebraeus tell a longer story. In it the scholar is John the Grammarian, later readers identify him with John Philoponus, and the caliph's letter says that if the books agree with the Book of God they are not needed, and if they disagree they are not wanted, so they are to be removed. The books are then said to have heated the baths of Alexandria for six months. Inside that speech the collection under Ptolemy II is given as 54,120 books. The English wording of the letter is not one fixed sentence from edition to edition. The shared core is the one just given. One reading weighs the centuries of silence, the conquest narratives closer to the event that do not tell this story, and the sixth-century date of Philoponus's other work. Another reading weighs the names, the letter, and the baths as a report that could have been old when it was written down, and treats silence as a gap in what was copied. This room does not drop either reading.

Open How many rolls, and what was only here Set the counts side by side and they do not become one number. More than 200,000, with a hope of 500,000, in the Letter of Aristeas. 400,000 mixed rolls and 90,000 unmixed, plus 42,800 in an outer library, in Tzetzes, who writes in the twelfth century and ties the tables to Callimachus. 40,000 burned, in Seneca's citation of Livy. 400,000 or 40,000, depending on the manuscript of Orosius. Nearly 700,000 or, in some copies, 70,000, in Gellius. 700,000 in Ammianus, tied to Caesar. 54,120 inside the thirteenth-century speech. Plutarch reports a charge, brought by Calvisius, that Antony gave Cleopatra the libraries of Pergamon, 200,000 volumes. That is a hostile list, not a bill of lading. One reading takes it as a real gift into Alexandria after 48 BCE. Another reads it as one slur among slurs. This room adopts none of these figures as the size of the collection. On the losses, the Pinakes do not survive as a book. Eratosthenes' own treatise on the measurement does not survive. The anatomical writings of Herophilus, who worked in this city, are known because Galen quotes them. Seven plays of Aeschylus and seven of Sophocles survive complete, and about eighteen of Euripides, out of the much larger bodies ancient lists give them. The editions of Homer made here shaped the text later copies carry. Euclid's Elements and Ptolemy's Almagest survived by being copied. No inventory used here shows the palace shelves packed up and delivered, as one collection, to Constantinople or anywhere else.

Open A noon shadow, and no library azimuth Eratosthenes worked at Alexandria in the third century BCE and is named as head of the library. His own book on the measurement of the Earth does not survive. Cleomedes reports the method. At Syene, modern Aswan, at noon on the summer solstice, the pointers of the sundials cast no shadow, and this was said to hold across a region 300 stadia wide. At Alexandria, at the same noon, the shadow marked one fiftieth of a circle. The distance between the cities is given as 5,000 stadia. Fifty times that distance is 250,000 stadia, his figure for the whole circle of the Earth. Strabo reports 252,000. The two figures differ by 2,000 stadia. The length of a stadion is not one published ancient meter, so this room does not turn either figure into kilometers. The method needs a round Earth, parallel rays from a distant sun, Syene on the summer tropic, and the two cities on one meridian. Syene is close to the tropic, not exactly on it, and it lies east of Alexandria, not due south. The diagram draws only Cleomedes' fraction and his 250,000. It does not draw Strabo's figure, and it does not draw a well. A well at Syene is a later telling. Cleomedes describes sundials. No published alignment of a library building is weighed here, because no wall of that building has been identified. The sky evidence is the scholarship. Ptolemy's Almagest, written in this city in the second century CE, uses star observations made here in the third century BCE under the names Timocharis and Aristyllus. Callimachus's poem on Berenice's lock tells of a queen's offering placed among the stars, the constellation still called the hair of Berenice. That is court sky lore, not a bearing of a lost room. Archimedes reports that Aristarchus of Samos set the sun at the center and the Earth in motion. The book of Aristarchus that survives is about sizes and distances, and it does not use that framing. This room does not give Aristarchus a post in the Mouseion.

Location and dating

Location: Alexandria, on the Mediterranean. The pin marks the royal quarter under the modern city, the ground where the palaces are placed and where the Bibliotheca Alexandrina, opened in 2002, now stands. It is not a surveyed ancient wall. The Serapeum of Alexandria is the separate hill of the column called Pompey's Pillar.
Period label: Current consensus — open to revision The gathering belongs to the early Ptolemies, from Ptolemy I and Ptolemy II in the third century BCE onward. The reports of loss are 48 BCE, the Bruchion under Aurelian about 272 CE, the Serapeum of Alexandria in 391, and a story about the conquest of about 642 first written down in the thirteenth century. No one of those dates is adopted here as the end of every book.
How this dating was done: The dates are the kings, the wars, and the writers, not radiocarbon of a library wall, because no such wall is in hand. A writer's century is the date of the report. It is not, by itself, the date of the event. Strabo is after Caesar and describes a living Mouseion. Ammianus is before 391 and describes both a damaged Bruchion and libraries he places in the Serapeum. The Arabic story is about five and a half centuries after 642.
What could change it: A foundation text that names the year the library opened, or a dug room whose fittings can be tied to rolls, or a catalogue that is the Pinakes rather than a quotation from them. Any of those would move the placement. A manuscript of Gellius or Orosius does not, by itself, choose the size, because the copies already disagree.

Published ranges, drawn as bars. A bar is a claim about a span, not a point.

Early Ptolemaic gatheringContested

Alexandrian WarContested

Bruchion under AurelianContested

Serapeum of AlexandriaContested

Arabic reportContested

Eratosthenes at AlexandriaContested

305 BCE1 CE642 CE

Early Ptolemaic gathering. Ptolemy I through Ptolemy III. No foundation inscription is in the sources used here.
Alexandrian War. Ships burned. Whether the palace library burned is the open part.
Bruchion under Aurelian. Ammianus names the quarter. He does not name a library fire.
Serapeum of Alexandria. The temple's destruction is described. A library that day is the open part.
Arabic report. Told in the thirteenth century about the conquest. The event and the telling are not the same date.
Eratosthenes at Alexandria. Named as head of the library. Cleomedes does not date the noon shadow to a year.

What is actually there

Strabo describes the Mouseion inside the royal palaces and does not, in that description, say library.
No excavated room in the sources used here has been identified as the ancient library.
The Bibliotheca Alexandrina in the photograph opened in 2002. It is a new building.
Galen describes originals kept, copies returned, and a label "from the ships," and he is writing centuries later.
The ancient counts of rolls do not agree, and some of them split inside one author's manuscripts.
Cleomedes reports 250,000 stadia. Strabo reports 252,000. No kilometer figure is adopted.
Twenty lecture halls at Kom el-Dikka belong to the late Roman city, not to the Ptolemaic shelves.

Open questions

Question	Status
Where, under the modern city, did the Mouseion and the royal library stand?	Open
Did Caesar's fire take the palace library, books in waterfront storehouses, or both?	Open
Did a daughter library at the Serapeum of Alexandria still hold books in 391?	Open
Does the Arabic report of about 642 preserve an older account of a burning?	Open
Were the works we lack unique to this house, or had copies already left the city?	Open
Which length of the stadion, if any, should be applied to 250,000 or 252,000?	Open. This room does not choose, and it does not print a kilometer.

The current narrative

Where publication stands today. It is the best-supported reading, and it is open to revision. It is not a verdict.

The current narrative — open to revision

Weigh each report on its own

The Mouseion and a royal library in Ptolemaic Alexandria were real. The size of the collection, the path of the copying, and the several reports of loss do not collapse into one fire or one number of rolls. Each report keeps its own text and its own open edge.

What supports this

Strabo describes a staffed house of scholars inside the royal palaces, and a satirist of the third century BCE already mocks that house.
Galen, the Letter of Aristeas, and the Suda, writing at very different dates, all describe collecting, copying, or cataloguing. They do not describe the same moment, and they do describe an institution.
The four reports of loss name different centuries, different writers, and different objects, from dockyard fire to a ruined quarter to a temple to a thirteenth-century story.
Suetonius has scribes sent to Alexandria to copy books in the late first century CE, and Ammianus still finds teachers and sky study in the city before 391.

What pushes back

Strabo does not say library, and no dug room has been identified as one, so the building is still a text.
The counts of rolls contradict one another, and some contradict themselves across manuscripts, which leaves the scale of the house unfixed.
A reading that one or more fires removed unique books, and a reading that the Arabic report records a real burning, are not answered by keeping the reports side by side.

What would settle it

It holds while the institution is attested and no single destruction count is forced. It moves if a dug room, or a catalogue in its own hand, shows one building and one end.

Emerging narratives

Other published readings. Each may be right. None is adopted here unless the evidence carries it.

An emerging narrative

The fires took books that were only here

One or more of the reported fires destroyed copies that had not been made elsewhere. The plays, medical books, and treatises we know only by title or by quotation are missing because those copies burned.

What supports this

Several ancient writers, Plutarch among them, describe a fire in 48 BCE that reached a great library or a great mass of books, and they give large figures even though the figures disagree.
Ammianus describes the loss of the Bruchion, the quarter of the palaces, under Aurelian, and he places libraries of great value in the Serapeum before 391.
The Pinakes, Eratosthenes' own measurement book, and Herophilus's anatomical writings do not survive as books. Quotation is not the book.

What pushes back

Caesar does not mention books. Dio can be read as waterfront storehouses. The numbers for 48 BCE run from 40,000 to nearly 700,000, and some of those numbers split by manuscript.
Strabo's Mouseion is still a working house after 48 BCE, and Suetonius still has books in Alexandria worth sending scribes to copy.
The plays we have are the ones school copying kept. A fire in one city is not the only way a book stops being copied. Euclid and the Almagest show that books from this world did leave.

What would settle it

It stays possible for any title we cannot find in a second city. It weakens where a second copy, or a writer after the fire still using the book, is on the table.

An emerging narrative

Most of it had already been copied out

The library's work was copying. Books moved to Pergamon, to Rome, and into the later manuscript tradition. What we lack is largely what copyists stopped copying, not a single pile of unique rolls lost on one day.

What supports this

Galen describes a house whose daily work was making copies, keeping originals, and returning the new rolls. A library run that way is a place books leave, as well as a place they gather.
Suetonius has Roman scribes copying Alexandrian books in the late first century CE. The medieval text of Homer follows the Alexandrian editors. The Almagest survives in a fifteenth-century Greek manuscript, and in Arabic translation.
Plutarch's report of 200,000 volumes from Pergamon, whatever one makes of the charge, is a story about books moving between royal libraries, not about one sealed room.

What pushes back

No inventory shows the palace collection transferred as a collection. Survival of some books does not measure the books that were never copied.
Herophilus, the Pinakes, and Eratosthenes' own treatise are names without the books. If the only copies were here, copying-out does not save them.
The ancient writers who give huge figures for a fire are describing a loss they thought was large, not a loss they thought was only duplicates.

What would settle it

It stays strong for every work we can still read, and for every work a later writer quotes from a copy. It stays weak for a title that never appears again after a reported fire.

An emerging narrative

A collection was still burned about 642

When Amr ibn al-As took Alexandria, a body of books was still in the city, and it was destroyed on Umar's order, as the thirteenth-century Arabic writers relate.

What supports this

Abd al-Latif, Ibn al-Qifti, and Bar Hebraeus, independently enough to be three texts rather than one sentence, report a burning by Amr with Umar's approval.
The longer story names a scholar, a letter, and a method, the baths for six months, and it gives the collection a history back to Ptolemy II.
A report can be written down late and still carry an older memory. Books could have remained in the city after the earlier wars. Silence in a nearer narrative is not, by itself, a proof that nothing happened.

What pushes back

The texts are from the early thirteenth century, about five and a half centuries after the conquest. Conquest narratives closer to the event, in the sources used here, do not tell it.
The scholar in the long story is the one later readers identify with John Philoponus, whose other dated work sits in the sixth century, so the meeting in 642 is part of what is open.
If the palace library and a Serapeum collection had already been lost, the story still has to say which books were left to heat the baths. The figure 54,120 is a number inside the speech, not a catalogue.

What would settle it

It stays open. A nearer text that tells the same story, or a ground of burned books tied to that year, would move it. The absence of that text does not delete the report.

Should the current view be reevaluated?

Open to revision

The house of scholars is well enough attested to stand. The size of the collection, the number of destructions, and the share of books that were unique are not. Keeping the reports side by side is the careful reading, and it is not a final count.

What drives this

Strabo's Mouseion, and no identified library room.
Counts of rolls that disagree, including inside single authors.
Four reports of loss, each with a different object and a different date of writing.
Books that survive by later copying, beside titles that do not survive at all.
A noon-shadow measurement whose stadion has no single length.

This is the museum’s own weighing of the evidence. It is not a visitor vote, and it is separate from the slider.

Weigh it yourself

Anonymous. Nothing here is stored except on this browser. Your lean does not change the assessment beside it.

Weigh each report on its own Unsure The fires took books that were only here

If the contested reading is right

What would follow if a reading that is not the consensus turned out to be true, and what could settle it. This explains the stakes. It does not adopt the reading.

The fires destroyed copies that existed nowhere else

Large parts of Greek drama, medicine, and Eratosthenes' own book would be missing because of those days. A new papyrus would be a recovery, not only a duplicate of a text we already have.

Why it is still worth a test

Titles known from quotation, searched for in papyri and in later libraries, with the date of the last writer who still had the book.

The books had mostly already been copied elsewhere

The missing plays and treatises would be what copyists stopped copying, over many centuries. A fire in Alexandria would be one path of loss among others, and the books we still read would be the evidence that the path was not closed.

Why it is still worth a test

A map of which Alexandrian books are quoted after 48 BCE, after the 270s, and after 391, and in which city the quotation was made.

The report of about 642 records a real burning

A collection would still have been in the city at the conquest, and the earlier fires would not have been the end. The daughter library, the palace library, or some other store would have to be named.

Why it is still worth a test

A text nearer to 642 that tells the story, and a context for the books that is not only the thirteenth-century speech.

Eratosthenes' 250,000 stadia is the circumference in his own unit

A librarian would have measured the Earth with a shadow and a surveyed distance, inside the life of this house. The kilometer value would still wait on the stadion, and the sky work would be part of what the institution did, not a decoration.

Why it is still worth a test

The length of the stadion he used, from a source that states it, and a check of the meridian and the tropic against the 5,000 stadia. Not a mile drawn in afterwards.

Sources

Commentary on Hippocrates' Epidemics, the ships and the fifteen talents Galen, in a translation at Attalus.org. He is repeating Zeuxis.
The Letter of Aristeas Translation of R. H. Charles. More than 200,000 rolls, and a hope of 500,000.
Ammianus Marcellinus, Book 22, chapter 16 Perseus. The Serapeum, 700,000 volumes, and the Bruchion under Aurelian.
Plutarch, Life of Caesar Loeb translation at LacusCurtius. The fire from the dockyards and the great library.
Plutarch, Life of Antony 58 Loeb translation. Calvisius's charge of 200,000 volumes from Pergamon.
Cassius Dio, Roman History 42.38 English version in which the docks, the grain stores, and the library burn. The warehouse reading of the same sentence is weighed in the room.
Suetonius, Life of Domitian 20 Loeb translation at LacusCurtius. Scribes sent to Alexandria to copy and correct books.
Alexandria, Library of Dreams Roger S. Bagnall, 2002. The counts, the manuscript splits in Gellius and Orosius, and the several reports of loss.
Eratosthenes and the circumference of the Earth Nature, 1943. Cleomedes' 250,000 stadia, and Strabo's 252,000.
The House of Ptolemy, on the Museum and the Pinakes E. R. Bevan, at LacusCurtius. The Suda's 120 books, and the librarians.
Ibn al-Qifti on the library Roger Pearse, 2010, with a translation. The scholar, the 54,120, and the baths.
Bar Hebraeus, Abd al-Latif, and the library Roger Pearse, 2010. The short remark and the longer story, both thirteenth century.
The auditoria on Kom el-Dikka Grzegorz Majcherek. Twenty lecture halls, late fifth to early sixth century, not the Ptolemaic library.

Image credits

The Bibliotheca Alexandrina, opened in 2002, beside the harbor. A new building. It is not the Ptolemaic library, and no ancient wall of that library has been identified under it. . Vyacheslav Argenberg, CC BY 4.0. Shown unedited.
Cleomedes' report, drawn for this room. One fiftieth of a circle, 5,000 stadia, 250,000 stadia. Strabo's 252,000 is not drawn. No mile figure. Not a direction of a library wall. . Diagram: Our Deep History.
A nineteenth-century drawing of the library's hall, by O. Von Corven. An artist's reconstruction. It is not a plan of a dug room, and no ancient wall has been matched to it. . O. Von Corven, Public domain. Shown unedited.
The column called Pompey's Pillar, on the hill of the Serapeum of Alexandria. The inscription honors Diocletian. This is the temple later readers give a daughter library. It is not the royal library, and it is not the bull galleries at Saqqara. . Daniel Mayer, CC BY-SA 4.0. Shown unedited.
A Roman marble bust of Serapis in the Vatican, with a grain measure on his head. A copy of a Greek type. It was not lifted from the Serapeum of Alexandria. The god is the one that temple housed. . Jastrow, Public domain. Shown unedited.
A gold coin in the Pergamon Museum, Berlin. The file's description reads the portraits as Ptolemy II and Arsinoe II, and the inscription as "of the siblings." A portrait of that court, not a coin struck for the library. . MatthiasKabel, CC BY-SA 3.0. Shown unedited.
PSI 1092, in the Biblioteca Medicea Laurenziana, Florence. A papyrus of the first century BCE with Callimachus's poem on Berenice's lock. A later copy of a court poem about a constellation, not a roll lifted from the royal shelves. . Unknown photographer, Public domain. Shown unedited.
Folio 143 of BnF Grec 2392, a fifteenth-century Greek copy of Ptolemy's Almagest. Not Ptolemy's own hand. He worked in Alexandria in the second century CE. This page is how a book from that city reached a later reader, by being copied. . Bibliothèque nationale de France, Public domain. Shown unedited.

← Egypt World map