What Was the Library of Alexandria? The Myth and What Actually Happened

It was founded in the third century BC under the first two Ptolemies, the Macedonian dynasty that took Egypt after Alexander's death, probably on the advice of Demetrius of Phalerum, and it was part of a larger institution called the Mouseion, a temple of the Muses that functioned as a research establishment. Scholars were given stipends, lodging, meals and freedom from taxes, and they lived and worked in the palace complex, which makes it a reasonable ancestor of the modern research university and of the word museum. The collecting policy was aggressive. Ships arriving at the port were searched for books, which were copied with the copy returned and the original kept, and Ptolemy III is said to have borrowed the official Athenian texts of the great tragedians against an enormous deposit and then kept them and forfeited the money. Estimates of the holdings vary from forty thousand to seven hundred thousand scrolls, and all of them come from later sources with reasons to exaggerate.

The scholarship was substantial and a good deal of it still stands:

• •Eratosthenes, the chief librarian, calculated the circumference of the Earth from the difference in shadow angles between two Egyptian cities and got within a few percent of the correct figure
• •Euclid worked in Alexandria and produced the Elements, the most successful textbook ever written, in use for over two thousand years
• •Aristarchus of Samos proposed that the Earth orbits the sun, seventeen centuries before Copernicus, and was not believed
• •Herophilus and Erasistratus performed human dissection, distinguished nerves from tendons and arteries from veins, and identified the brain as the seat of intelligence
• •The librarians developed textual criticism as a discipline, comparing manuscripts of Homer, marking suspect lines and establishing the editions on which all later texts depend
• •Callimachus compiled the Pinakes, a catalogue of the collection in 120 scrolls, which is the ancestor of every library catalogue since
• •The Hebrew scriptures were translated into Greek as the Septuagint, according to tradition under royal commission

There is no single destruction. Four events are usually cited and each is partial or doubtful. Julius Caesar, besieged in Alexandria in 48 BC, set fire to ships in the harbour and the fire spread to warehouses near the docks, destroying a quantity of scrolls; the library itself appears to have survived, since it is referred to afterwards. Aurelian's suppression of a revolt in the 270s AD damaged the palace quarter where the library stood, and this is probably the most destructive single episode. The Christian mob that destroyed the Serapeum in 391 destroyed a daughter library rather than the main collection, which by then may no longer have existed. And the story of the Arab conquest in 641, in which a caliph orders the books burned as either contradicting or duplicating the Quran, first appears in a source written six centuries after the event and is rejected by essentially all historians. The most likely account is unremarkable: royal funding declined, scholars were expelled during a political purge in 145 BC, papyrus decayed in a humid climate, and an institution that depended on continuous patronage simply ran down over centuries.

The losses from antiquity are real and enormous and were not caused by one fire. Of the roughly ninety plays Aeschylus wrote, seven survive; of Sophocles's hundred and twenty, seven; the works of Sappho exist as fragments, many recovered from scraps used to wrap mummies. But these losses happened through the ordinary mechanism by which ancient texts disappeared, which is failure to be copied. A papyrus scroll lasts perhaps two or three centuries, so any text that survived did so because someone chose to reproduce it, repeatedly, across two millennia, and what survived is largely what was taught in schools, admired by later critics or useful to the church. A work that stopped being copied vanished regardless of how many libraries held it. The idea that one building's destruction cost us the knowledge of the ancients gets the mechanism wrong, and it also obscures how much did survive, including through Byzantine copying and through Arabic translation and commentary.

The story is attractive because it offers a single dramatic cause for a diffuse and depressing process, and because it can be pointed at whoever the teller dislikes: the version blaming Christians was popularised by Gibbon and by later anticlerical writers, the version blaming Muslims by Crusade-era Christian sources, and the version blaming Caesar by Roman-sceptical ones. It also flatters a particular view of history in which progress is linear and can be interrupted, whereas the actual record of ancient science shows work that was ignored rather than destroyed, including Aristarchus's heliocentrism and Hero's steam device. A large modern library and cultural centre, the Bibliotheca Alexandrina, opened on a site near the original in 2002 with support from Unesco, which is a fitting response to a building whose real achievement was institutional rather than architectural: paying scholars to work together in one place and cataloguing what they had.

The library was part of a royally funded research institution in Ptolemaic Egypt that paid scholars, collected books aggressively by copying everything arriving in the port, and produced Euclid's Elements, Eratosthenes's measurement of the Earth and the first systematic library catalogue. It was not destroyed in a single fire: Caesar's fire damaged dockside stores, Aurelian's campaign wrecked the palace quarter, the 391 destruction was of a daughter library, and the Arab burning story appears six centuries later and is rejected. Ancient texts were lost mainly by ceasing to be copied.