The Sumerian form of his name is Bilgames, and it is the older one; Gilgamesh is the Akkadian development, and it is the form the great epic uses. He is listed in the Sumerian King List as the fifth ruler of Uruk's first dynasty, credited with a hundred and twenty-six years and described as the son of a lillu-spirit and lord of Kullab, one of Uruk's two quarters. An inscription of a king of Umma from around the twenty-fourth century BCE mentions the wall of Uruk, and a votive object naming a contemporary of his has been read as independent confirmation that a ruler of that name existed somewhere around 2700 BCE. None of this is proof. It is the same order of evidence that supports a historical Arthur, and it points to a real Early Dynastic king around whom a literature accumulated for two thousand years.
Five Sumerian poems, and then a book
The epic was not written; it was assembled. Five separate Sumerian narrative poems about Bilgames survive from the Old Babylonian scribal schools: the expedition against Huwawa in the cedar mountain, the killing of the Bull of Heaven, the fight with Aga of Kish, his death, and Gilgamesh, Enkidu and the Netherworld. They are independent, they do not share an arc, and Enkidu in them is a servant rather than an equal.
Around 1800 BCE an Akkadian poet made a unified epic from some of this material, opening with the line šūtur eli šarrī, "surpassing all other kings". Several centuries later that was reworked into the version usually read today: eleven tablets, opening instead with ša naqba īmuru, "he who saw the deep", and attributed by a Neo-Assyrian scribal catalogue to Sîn-lēqi-unninni, an exorcist. The attribution is the closest thing Akkadian literature has to a named author for a major work, and what he did was editorial as much as authorial — a frame added at both ends, the flood story imported from a different poem entirely, and the whole turned into an argument about mortality.
Then there is Tablet XII, which is a literal Akkadian translation of the second half of one of the Sumerian poems, appended to a story it flatly contradicts. In it Enkidu has never died at all, and descends below on an errand — two wooden playthings of Gilgamesh's have fallen through the floor of the world and someone has to fetch them. Assyriologists read it as a piece of scribal housekeeping rather than a continuation of the plot, and printed editions usually quarantine it behind a note explaining why it is there.
Two-thirds god, which is not a fraction anyone should be able to be
His mother is the goddess Ninsun, "lady wild cow", a genuine deity with her own cult; his father Lugalbanda is an earlier king of Uruk who was himself deified and has his own Sumerian epic. The text says Gilgamesh is two-thirds divine and one-third human, which is genealogically impossible and is usually explained by the sexagesimal arithmetic Babylonians used, in which two-thirds was a standard, comfortable fraction rather than an awkward one. Whatever the reasoning, the epic wants him more than half a god and definitively not immortal, and the entire plot lives in that gap.
The shape of the story, and where it turns
The opening is an abuse of power — he exhausts the young men of Uruk and takes what he wants from its brides — and the gods' answer is not punishment but a counterweight, a wild man made from clay to match him. They fight in the street, the door-frame shakes, and afterwards they are inseparable. The two of them go to the cedar forest and kill its guardian Humbaba against explicit advice; they refuse Inanna and kill the Bull of Heaven she sends. The gods rule that one of them must die for it, and it is not the king.
Everything after that is grief. He will not permit the burial until a worm falls from his friend's nose. He dresses in skins and walks east, past the scorpion-people at the mountain of the sunrise, through twelve leagues of darkness, to the ale-wife Siduri at the shore and across the waters of death to Utnapishtim, who survived the flood and was granted the thing Gilgamesh wants. The answer he gets is that the grant was unique and unrepeatable, that he cannot stay awake for seven nights let alone defeat death, and that a plant of rejuvenation on the seabed is the best consolation available — which a snake eats while he bathes. He comes home and shows the boatman the city walls, in the same words the poem used to describe them in its first lines. That is the ending: the work outlasts the man, and it is offered without triumph. The full reading is on Gilgamesh and the search for immortality.
The god of the dead nobody expects
Outside the literature, Gilgamesh had a working cult, and its content is surprising. Incantations and funerary texts address him as a judge in the underworld — a magistrate of the dead who hears cases alongside the Anunnaki, receives offerings in the month of Abu, and is invoked in rituals against ghosts. Ur-Nammu's coronation hymn and the Sumerian poem on his death both put him in that office. So the man who fails to escape death ends up administering it, which is a piece of theology the epic itself never mentions and which its readers in Babylon would all have known.
The text is still growing
More of this poem exists now than at any point since antiquity, and the number rises. Fragments continue to be identified in museum drawers, in the Sippar library, and among tablets excavated a century ago and never fully read. The best-publicised recent case is a manuscript that surfaced in Iraqi Kurdistan and was edited in the Journal of Cuneiform Studies in 2014, restoring roughly twenty lines to the fifth tablet. Its effect on the reading is out of all proportion to its length, and what it does to the cedar forest episode is set out on the Humbaba page.
The trade in such tablets has its own record. A fragment carrying the dream sequence, bought by a major American retail chain in 2014 with a false provenance and displayed in a Washington museum, was seized by United States authorities, ordered forfeited in 2021 and returned to Iraq. Any account of this epic's recovery that stops at the Victorian excavations is missing the part that is still happening.
Where he went after cuneiform
Because the epic was unread for two millennia, it has no medieval reception and no folkloric descent; its entire modern life dates from the 1870s onwards, and that lateness is itself unusual. Rilke, reading the first German translation, called it the epic of the fear of death and wrote about it with something close to shock. Bohuslav Martinů set it as an oratorio in 1955. It became a standard text in world-literature curricula, a fixture in Iraqi national self-presentation under several very different governments, and a name that turns up in fantasy and games with almost no connection to the tablets — most visibly as a recurring character in the Fate franchise, where he is a golden-armoured king with a treasury of prototype weapons.
One television episode is worth naming because it is doing something more careful: Star Trek: The Next Generation's "Darmok" builds an entire hour around a species that speaks only in mythological allusion, and the captain's demonstration of what that would mean in his own language is a retelling of Gilgamesh and Enkidu at Uruk.
Points that keep being argued
Were Gilgamesh and Enkidu lovers?
The text supports the question without settling it. Ninsun's dreams tell Gilgamesh he will love Enkidu kīma aššate, like a wife, and he caresses the dream-object of the coming friendship as one would a wife; the lament compares his grief to that of a widow, or of a bride's loss depending on the restoration. Against that, the Akkadian words used are the ordinary vocabulary of intense kinship and friendship, ancient Mesopotamia had no category corresponding to modern sexual identity, and the poem never depicts sex between them. Serious translators — Andrew George among them — describe the language as erotically coloured while declining to convert that into a modern label. Both the flat denial and the confident assertion overstate the evidence.
How much of the epic is actually missing?
Roughly a fifth of the standard version remains unrecovered, and the gaps are not evenly spread — Tablet VII and the ending of Tablet X are especially damaged. This is why translations differ so much more than translations of Homer do: a translator has to decide whether to restore a broken line from a parallel manuscript, to leave a gap, or to paraphrase across it. George's edition marks every restoration; older popular versions silently smooth them, which is why a reader moving between two English Gilgameshes can find whole incidents present in one and absent in the other. The tradition hub sets out how the tablets survived at all.
Is the flood chapter the original flood story?
No — it is an insertion. Tablet XI takes a flood narrative that already existed independently in Atrahasis and in a Sumerian version, and puts it in the mouth of its survivor as a speech. That is why the flood's causes are barely explained in Gilgamesh while Atrahasis devotes tablets to them. What the transplant achieves is dramatic rather than theological: the story is told to a man who has come a very long way to hear that it cannot help him. The relationship between this account and the biblical one is set out on the flood myths compared.