It would probably come as a surprise to many of you to know that the evidence that the Great Library of Alexandria ever actually existed is surprisingly thin. There is no physical evidence of such a structure, and the earliest reference to it is now known to be a forgery. Later Roman references to the supposed destruction of the library by Julius Caesar date from over century later. If nothing else, many scholars agree that such writers ludicrously exaggerated the size of the library’s supposed collection.
Still, it is true that Alexandria was, in the third and second centuries BC, an important centre of learning and intellectual activity. If nothing else, scholars agree, the very legend of the Great Library seems to have acted as an inspiration for the ancient world. The notion of a library – a collection of stored wisdom, preserved for the ages – remains an important notion.
But when did libraries begin? What even is a ‘library’?
It’s one of those words that we think we all understand until we have to think a little deeper about it, and then it’s not so clear anymore.
Surely a library is a collection of texts? But what kind of collection of texts? The codex or book as we know it didn’t really exist until the early common era, so it’s not books as we normally think about it. Prior to that texts were inscribed onto stone, they were pressed into clay tablets, they were written on papyrus sheets which formed long rolls or scrolls. They were scribbled onto stone, onto bone, and broken potsherds called ‘ostracon’. Can such a collection be said to constitute a library?
Further we think of libraries as somehow accessible to the public but our most ancient collections are the purview of the one per cent. Yeah, about one per cent of the population that could read and write. Those were overwhelmingly men, and overwhelmingly elite, rich men to some degree, though there were some women who are noted in the literary and cultural history of libraries.
Just as the development of writing itself apparently stemmed from the need to keep administrative records, especially commercial transactions, many ancient collections were purely administrative repositories. Tax records, military rolls, inventories and mountains of receipts. Should these collections be more properly regarded as ‘archives’, rather than ‘libraries’? Any more than your shoebox of receipts for tax time are a ‘library’?
It seems more correct, then, to consider a library as a repository dedicated to the preservation and production of knowledge itself, rather than mere records.
The earliest texts known include Sumerian lexical lists: lists of signs for concepts and objects, much like a dictionary. Like Johnson’s first English dictionary, these are profoundly important innovations. They represent the standardisation of the symbolic forms of language. If you’ve ever read early English texts with their idiosyncratic and often bizarrely phonetic spelling, you’ll appreciate how fundamental it was to set out agreed-upon ways of recording word symbols, so that every literate person could understand them.
Thus lexical lists, while seemingly boring at first, I think in fact they are the first pieces of philosophy.
However by the second half of the 3rd millennium B.C.E truly non-administrative documents had begun to appear in both Mesopotamia and Egypt. It marked the appearance of literature in world history, written largely on clay tablets in the cuneiform script in Mesopotamia, and upon papyrus sheets and potsherds in Egypt.
Such physical media set innate constraints on the size of collections. The Epic of Gilgamesh, for instance, takes up about 12 large, heavy, clay tablets. Some medical texts exceed 100 tablets. On the other hand, papyrus was lightweight and thin, allowing rolls over 40 metres long to be produced.
On the other hand, papyrus is fragile and quickly broke down in the humid Nile Delta, as well as being vulnerable to fire. It survives better in the dry deserts, but not many scholars and scribes lived in the desert. Clay tablets, on the other hand, are incredibly durable. This leads to what is called ‘survivorship bias’: certain types of texts are far more abundant in the archaeological record, simply by dint of how they were made and stored.
Tomb and temple complexes, with their extensively inscribed stele and other objects, could also be considered libraries. But tombs were often looted or destroyed.
Taking all that into account, the best candidates for the earliest libraries we know of are the temple complexes of Egypt and Sumeria.
The earliest surviving collections of recognisable libraries may be from the Sumerian cities of Shuruppak, Abu Salabikh and Ebla, from around 2600 BC.
The site of Ebla is a bit more clear as something we might recognize as a library, located about 60 km south of Aleppo in modern Syria. Ebla was populated by a Semitic-speaking people […] until their capital was of course destroyed, perhaps by the king of Mari, Ishki'mari – just before he himself was conquered by Sargon of Akkad around 2330 BCE. Or about 250 years after the sites of Shuruppak and Abu Salabikh.
Within the administrative quarter of the palace complex several rooms have been found, sporting mostly administrative finds numbering in the hundreds of tablets, no less. However the most spectacular find was an apparent archive somewhat preserved since the conflagration that destroyed that palace itself.
‘Room L.2769’ measures about five by three and a half meters and originally contained an L-like series of three wooden shelves stocked with tablets, with further tablets and baskets tucked below. The ashes of those baskets are still there, totaling about 2100 or so tablets.
Along with the usual administrative stuff, there are a collection of 12 incantation tablets, including an Akkadian hymn to the sun god Shamash. More importantly, it is clear that texts were grouped by category. It was a long way from the Dewey Decimal System, but it was clearly a library as we’d recognise it today.
