In response to:

Lost History of the Lost Library from the June 14, 1990 issue

To the Editors:

From Professor Hugh Lloyd-Jones’s review of Luciano Canfora’s book on the library of Alexandria [NYR, June 14], one learns, with astonishment, that the author, and perhaps even to some degree the reviewer, are still disposed to lend credence to the story of how the great library of Alexandria was destroyed by the Arabs after their conquest of the city in 641 AD, by order of the Caliph ‘Umar.

This story first became known to Western scholarship in 1663, when Edward Pococke, the Laudian Professor of Arabic at Oxford, published an edition of the Arabic text, with Latin translation, of part of the History of the Dynasties of the Syrian-Christian author Barhebraeus, otherwise known as Ibn al-‘Ibri. According to this story, ‘Amr ibn al-‘As, the commander of the Arab conquerors, was inclined to accept the pleas of John the Grammarian and spare the library, but the Caliph decreed otherwise: “If these writings of the Greeks agree with the book of God, they are useless and need not be preserved; if they disagree, they are pernicious and ought to be destroyed.” The books in the library, the story continues, were accordingly distributed among the four thousand bathhouses of the city, and used to heat the furnaces, which they kept going for almost six months.

As early as 1713, Father Eusèbe Renaudot, the distinguished French Orientalist, cast doubt on this story, remarking, in his History of the Patriarchs of Alexandria published in that year, that it “had something untrustworthy about it.” Edward Gibbon, never one to miss a good story, relates it with gusto, and then proceeds: “For my own part, I am strongly tempted to deny both the fact and the consequences.” To explain this denial, Gibbon gives the two principal arguments against authenticity—that the story first appears some six hundred years after the action which it purports to describe, and that such action is in any case contrary to what we know of the teachings and practice of the Muslims.

Since then, a succession of other Western scholars have analyzed and demolished the story—Alfred J. Butler in 1902, Victor Chauvin in 1911, Paul Casanova and Eugenio Griffini, independently, in 1923. Some have attacked the internal improbabilities of the story. A large proportion of books of that time would have been written on vellum, which does not burn. To keep that many bathhouse furnaces going for that length of time, a library of at least 14 million books would have been required. John the Grammarian who, according to the Barhebraeus story, pleaded with ‘Amr for his library, is believed to have lived and died in the previous century. There is good evidence that the library itself was destroyed long before the Arabs arrived in Egypt. The 14th century historian Ibn Khaldun tells an almost identical story concerning the destruction of a library in Persia, also by order of the Caliph ‘Umar, thus demonstrating its folkloric character. By far the strongest argument against the story, however, is the slight and late evidence on which it rests. Barhebraeus, the principal source used by Western historians, lived from 1226 to 1289. He had only two predecessors, from one of whom he simply copied the story and both preceded him by no more than a few decades. The earliest source is a Baghdadi physician called ‘Abd al-Latif, who was in Egypt in 1203, and in a brief account of his journey refers in passing to “the library which ‘Amr ibn al-‘As burnt with the permisison of ‘Umar.” An Egyptian scholar, Ibn al-Qifti, wrote a history of learned men in about 1227, and includes a biography of John the Grammarian in the course of which he tells the story on which the legend is based. His narrative ends: “I was told the number of bathhouses that existed at that time, but I have forgotten it. It is said that they were heated for six months. Listen to this story and wonder!” Barhebraeus merely followed the text of Ibn al-Qifti, omitting his final observation on the number of baths. This number is provided by other Arabic sources, in quite different contexts.

To accept the story of the Arab destruction of the library of Alexandria, one must explain how it is that so dramatic an event was unmentioned and unnoticed not only in the rich historical literature of medieval Islam, but even in the literatures of the Coptic and other Christian churches, of the Byzantines, of the Jews, or anyone else who might have thought the destruction of a great library worthy of comment. That the story still survives, and is repeated, despite all these objections, is testimony to the enduring power of a myth.

Myths come into existence to answer a question or to serve a purpose, and one may wonder what purpose was served by this myth. An answer sometimes given, and certainly in accord with a currently popular school of epistemology, would see the story as anti-Islamic propaganda, designed by hostile elements to blacken the good name of Islam by showing the revered Caliph ‘Umar as a destroyer of libraries. But this explanation is as absurd as the myth itself. The original sources of the story are Muslim, the only exception being Barhebraeus, who copied it from a Muslim author. Not the creation, but the demolition of the myth was the achievement of European scholarship, which from the 18th century to the present day has rejected the story as false and absurd, and thus exonerated the Caliph ‘Umar and the early Muslims from this libel.

But if the myth was created and disseminated by Muslims and not by their enemies, what could possibly have been their motive? The answer is almost certainly provided in a comment of Paul Casanova. Since the earliest occurrence of the story is in an allusion at the beginning of the 13th century, it must have become current in the late 12th century—that is to say, in the time of the great Muslim hero Saladin, famous not only for his victories over the Crusaders, but also—and in a Muslim context perhaps more importantly—for having extinguished the heretical Fatimid caliphate in Cairo, which, with its Isma’ili doctrines, had for centuries threatened the unity of Islam. ‘Abd al-Latif was an admirer of Saladin, whom he went to visit in Jerusalem. Ibn al-Qifti’s father was a follower of Saladin, who appointed him Qadi in the newly conquered city.

One of Saladin’s first tasks after the restoration of Sunnism in Cairo was to break up the Fatimid collections and treasures and sell their contents at public auction. These included a very considerable library, presumably full of heretical Isma’ili books. The break-up of a library, even one containing heretical books, might well have evoked disapproval in a civilized, literate society. The myth provided an obvious justification. According to this interpretation, the message of the myth was not that the Caliph ‘Umar was a barbarian because he destroyed a library, but that destroying a library could be justified, because the revered Caliph ‘Umar had approved of it. Thus once again, as on so many occasions, the early heroes of Islam were mobilized by later Muslim tradition to give posthumous sanction to actions and policies of which they had never heard and which they would probably not have condoned.

It is surely time that the Caliph ‘Umar and ‘Amr ibn al-‘As were finally acquitted of this charge which their admirers and later their detractors conspired to bring against them.

Bernard Lewis
Princeton, New Jersey

Hugh Lloyd-Jones replies:

I am delighted that my review has elicited Professor Lewis’s learned and interesting letter. But if he had looked a little more closely at what I wrote, he would have seen that I do not believe that when the Arabs conquered Egypt there was very much left in the library for them to destroy. Whether the Arabs of that time would have destroyed a great library belonging to an alien culture is an interesting question which Professor Lewis is far better qualified to answer than I am, so that I am disappointed that nothing in his letter throws light upon this problem.

This Issue

September 27, 1990