In the mid-tenth century, probably in Aleppo, Ishaq ibn Shahram reported having visited a magnificent ancient temple in Byzantium. The temple was a three-days’ journey from Constantinople. It had a huge iron gate controlling its entrance. The ancient Greeks had used the temple for prayer and apparently reading. The Byzantines had locked the temple gate in the fourth century when they converted to Christianity. Ishaq ibn Shahram, who was engaged in important diplomatic missions, persuaded the Byzantines to unlock the temple for him. Here’s what Ishaq ibn Shahram reportedly saw:
a structure composed of various kinds of marble and huge rocks, furnished with inscriptions and drawings in an abundance and beauty such as I had neither seen nor heard of before. That sanctuary contained so many ancient books that a goodly number of camels would have been needed to carry them (indeed, Ishaq put the number at as many as one thousand camels). Some of the books were dilapidated, others were well-preserved and some were worm-eaten. In that sanctuary I also saw many interesting sacrificial implements of gold and other materials. 
Given that a reasonable load for a camel is 300 pounds, a thousand camels could carry 300,000 pounds of books. A large personal library about this time and place took 12 camels to transport. Similar large personal libraries in the Islamic world contained about 20,000 books. These facts suggest that the ancient Greek library that Ishaq ibn Shahram saw alleged contained about 1.7 million books. That figure almost surely is largely exaggerated.
Other accounts of large ancient libraries exist. The Library of Alexandria was reported to have held about 500,000 books. The Cordoba library of Al-Hakim II (reigned 961-976) reportedly had a catalog of 44 volumes of 50 folios each and contained 400,000 books. Like Ishaq ibn Shahram’s account, these other accounts of large ancient libraries are not well-attested. The best evidence for the size of an ancient library are the charred remains of 1,800 manuscripts from Herculaneum’s Villa of the Papyri. The Ulpian Library in Rome is estimated to have held roughly 20,000 scrolls. The libraries at Alexandria and Pergamum were likely larger than the Ulpian Library. Ishaq ibn Shahram’s story is some additional evidence for large ancient libraries.
* * * * *
- reason for skepticism about huge ancient libraries
- how the ancient Library of Alexandria was destroyed
- libraries as a focus for political competition
 Ishaq ibn Shahram probably traveled by sea. A three-day’s journey could have taken him to Ephesus, Miletus, or Pergamum. Dodge (1970) p. 585, n. 52, states:
By the tenth century, the great temple of Apollo Didymaeus at Branchidae near Miletus and the famous libary at Pergamum were almost certainly in ruins. It is likely, therefore that this library was a second-century building at Ephesus with the famous temple of Diana nearby.
The Epheus library is known as the Library of Celsus. An additional complication is that the story states that the temple’s “immediate vicinity was inhabited by Sabians and Chaldeans, whom the Byzantines allowed to profess their religions, exacting a poll-tax from them.” The Sabians and Chaldeans tend to be associated with southeastern Anatolia.
 Ishaq ibn Shahram says his travel to the temple occurred in the reign of Sayf al-Dawlah ibn Hamdan (945-67). Other sources mention Ishaq ibn Shahram visiting Byzantine Emperor Basil II in 981 as part of a diplomatic mission from Adud al-Dawlah. See Kraemer (1986) p. 78. In Kopf’s translation, Ishaq ibn Shahram’s story suggests an ongoing relationship with the Emperor: “I never ceased to entreat him, both in writing and verbally when at his court ….” Perhaps Ishaq ibn Shahram traveled to Constantinople multiple times over many years. An alternative explanation is that Ishaq ibn Shahram’s story is mistaken or falsified. Papyrus books surely wouldn’t have lasted six centuries. Parchment books might have.
 HP pp. 358-9. Skepticism about the size of ancient libraries seems to have colored translations of this passage. Ibn Abi Usaibia explicitly took the passage from ibn al-Nadim’s Fihrist. Dodge (1970) v. 2, p. 585, translates al-Nadim’s sentences on the library thus:
“… In this temple there were numerous camel loads of ancient books.” He exaggerated to the extent of a thousand camel [loads].
Montgomery (2000), p. 117, corrected Dodge’s translation thus:
In this temple there were numerous camel loads of ancient books. [There must have been] a thousand camel loads.
A translation of the passage via ibn Abi Usaibia into German and then into English has:
“…several camel-loads of old books were found in this temple.” He went on exaggerating until eventually he spoke of a thousand camel-loads.
Rosenthal (1975) p. 50. Kopf’s translation, which is consistent with Mongomery’s, does not indicate exaggeration.
 While the largest personal libraries in the Islamic world in the eleventh and twelfth centuries contained 20,000 or more books, the largest personal libraries in Europe after the fall of Rome reached 3,000 to 4,500 volumes in the late sixteenth century and about 50,000 volumes in the mid-eighteenth century. Blair (2010) p. 54.
Blair, Ann. 2010. Too much to know: managing scholarly information before the modern age. New Haven [Conn.]: Yale University Press.
Dodge, Bayard. 1970. The Fihrist of al-Nadim: a tenth-century survey of Muslim culture. New York: Columbia University Press (relevant text).
HP: Ibn Abi Usaybi’ah, Ahmad ibn al-Qasim. English translation of History of Physicians (4 v.) Translated by Lothar Kopf. 1971. Located in: Modern Manuscripts Collection, History of Medicine Division, National Library of Medicine, Bethesda, MD; MS C 294.
Kraemer, Joel L. 1986. Philosophy in the renaissance of Islam: Abū Sulaymān Al-Sijistānī and his circle. Leiden: E.J. Brill.
Montgomery, Scott L. 2000. Science in translation: movements of knowledge through cultures and time. Chicago: University of Chicago Press.
Rosenthal, Franz. 1975. The classical heritage in Islam. Berkeley: University of California Press.