A poem recovered from the ground
No complete ancient manuscript of the Epic of Gilgamesh exists. The poem as we read it is a reconstruction, put together from hundreds of broken clay tablets and fragments found at many sites, written over a span of roughly two thousand years. Modern editions mark gaps, restorations and uncertain readings. That the story can be read at all is the result of nineteenth-century excavation, cuneiform decipherment and the patient work of scholars matching fragments.
The twelve tablets
The best-preserved form is known as the Standard Babylonian version. It is written in Akkadian on twelve tablets, each holding roughly one major stage of the story: Gilgamesh’s rule, Enkidu’s creation and arrival, the expedition to the Cedar Forest, the Bull of Heaven, Enkidu’s death, the search for Utnapishtim and the flood story on the eleventh tablet. The twelfth is generally regarded as an addition: it translates part of an older Sumerian poem and does not follow naturally from tablet eleven. A scribe named Sin-leqi-unninni is credited in ancient tradition with this edition, usually dated to the later second millennium BCE, though how much he composed or edited is debated.
Earlier versions exist. The Old Babylonian period, roughly 2000 to 1600 BCE, has tablets with earlier Akkadian forms of the story, some of which are shorter and differ from the Standard Babylonian text. Before them stand the separate Sumerian poems. Copies of the epic in various forms have also come from Hattusa, the Hittite capital, and from Megiddo, a sign that the story was known well beyond southern Mesopotamia.
Ashurbanipal’s library at Nineveh
Most of the Standard Babylonian text comes from one place: Nineveh, a capital of the Assyrian Empire on the Tigris, near modern Mosul. Ashurbanipal, who ruled Assyria in the seventh century BCE, assembled a large collection of tablets in his palaces, with works on omens, medicine, rituals, hymns, letters and literature. His scribes collected and copied texts from older sources in Babylonia, and some sources describe an active search for tablets in temple libraries.
When Nineveh fell to a coalition of Babylonians and Medes in 612 BCE, the palaces were burned. Fire is usually destructive, but for clay it can be preserving: tablets that had been only sun-dried were often partly baked, and many survived under collapsed buildings, though often shattered into pieces. This is why the Assyrian archaeology of the site has produced tens of thousands of fragments.
Excavation and the British Museum
The ruins of Nineveh lie in the mounds of Kuyunjik and Nebi Yunus. In the mid-nineteenth century, British and French excavators, including Austen Henry Layard and, later, his assistant Hormuzd Rassam, worked there. The tablets from Kuyunjik were shipped to the British Museum, where they accumulated in large numbers, many unread. At that time cuneiform had only recently been deciphered, through the work of several scholars in the 1840s and 1850s, and there were few people able to read the tablets at all. The Rosetta Stone had unlocked Egyptian hieroglyphs decades earlier, but Akkadian had no equally convenient key and was decoded largely from multilingual royal inscriptions.
George Smith and the flood tablet
George Smith, a former engraver who had taught himself Assyrian, worked at the British Museum sorting and joining fragments. In 1872 he came across a tablet that described a flood, a boat, birds sent out to find land, and a mountain on which the boat came to rest. He recognised the story as comparable to the account in the Book of Genesis. He announced his find at a meeting of the Society of Biblical Archaeology in London in December 1872, an event reported widely in the press.
The sensation was considerable. A newspaper, the Daily Telegraph, then funded Smith to travel to Nineveh in 1873 to look for the missing portions of the tablet, and he reported finding a fragment that filled a gap. He made further trips and died of illness in Aleppo in 1876, aged only thirty-six. Smith’s translations, published as The Chaldean Account of Genesis, introduced the epic to the public, although at first it was understood mainly through its flood story and the poem’s name was not yet settled; the name of the hero was at first read as “Izdubar” because the cuneiform signs were not yet interpreted correctly.
Piecing fragments together
Since Smith’s day, the work has been mostly a matter of joining. Scholars match tablet fragments by shape, handwriting, line spacing and content, sometimes decades apart and across collections. Fragments at the British Museum in London have been joined with others that were excavated or acquired elsewhere. This is called making a “join”, and a new join can add a line, a name or a missing section.
Further evidence has appeared from other sites in Iraq and Turkey. Tablets from Sippar, Babylon, Uruk and Assur preserve first-millennium copies, some different from those in Ashurbanipal’s collection, and the archives of Hattusa have contributed versions in Hittite and Hurrian. In 2015, a previously unknown fragment of the fifth tablet, describing events in the Cedar Forest, was reported at the Sulaymaniyah Museum in Iraqi Kurdistan, having been acquired by the museum rather than excavated by it. Finds that are not recorded in an excavation are a concern for scholars, because they lose their archaeological context.
A modern critical edition by Andrew George, published in 2003, collected all the known manuscripts and remains the standard scholarly reference, with new texts added since.
What remains uncertain
Even with the abundance of sources, much is unknown. Some lines remain broken. The relationship between the Old Babylonian and Standard versions is debated, and so is how much Sin-leqi-unninni changed. Dating is based on handwriting, language, find spots and the scribal colophons that sometimes name the copyist and the owner. Methods such as these are part of how we date tablets generally, and they offer date ranges, not exact years.
Why the tablets matter
The Gilgamesh tablets are valuable in several ways: as literature, as evidence for religious belief, as a source of the oldest surviving flood narrative and as a record of how scribes preserved and revised their traditions. They also show how knowledge survives by accident. A poem was composed, copied by scholars, stored in a royal library, buried in a burning palace and forgotten for over two thousand years, before being read again by a man who had left school as a teenager and learned the script on his own. Its recovery is a story as striking as the poem’s own.