Ancient History & Civilisation

Appendix 2

Investigating the Text of Gilgamesh XI

1. The Shape

The argument that Utnapishti’s Ark was originally a round coracle like that described in the Ark Tablet raises three problems that need to be addressed:

Problem (a):

How does a round boat turn into a square one?


The Ark Tablet gives us explicitly the height of the sides of the boat, and this makes perfect sense of the boat’s proportions:


1 NINDAN igārātuša
And let be one nindan (high) her sides.

Atrahasis Vocabulary Box:

Akkadian , ‘let be’.

Sumerian NINDAN = Akkadian nindānu, ‘nindan measure’.

Akkadian igāru, pl. igārātu, ‘wall’.

Utnapishti’s description of the finished sides of his boat in Gilgamesh XI runs over two lines and repeats the measurement:


10 NINDAN.TA.ÀM šaqqā igārātuša
Ten nindan each stood high her sides,


Ten nindan each.

Gilgamesh Vocabulary Box:

Akkadian šaqû, ‘to be high’.

Sumerian NINDAN = Akkadian nindānu, ‘nindan measure’.

Akkadian igāru, pl. igārātu, ‘wall’.

TA.ÀM just means ‘each’.

This repetition of ‘ten nindan each’, acceptable enough in a Romantic poem, reads very awkwardly in Akkadian. It could just be an error, for this can easily happen when an editor is amalgamating separate written sources to produce one text. More probably, though, the repetition of ‘ten nindan each’ was introduced by some Gilgamesh redactor to make sense of the text in his hands, reasoning that if length and breadth were identical, as it says in line 30, he should give each its height. Losing sight of the circle at this point gave the old description – ‘her breadth and length should be the same’ – which originally reinforced the idea of the circular plan – a wholly different meaning, which led to the permanent misunderstanding in Gilgamesh XI that Utnapishti built himself asquare ark. The original, simple round vessel, subjected to subsequent textual elaboration, thus jelled into an implausible cube, and the Assyrian text, vivid and meaningful enough in itself, left Utnapishti with a waterborne life capsule that was utterly impractical. The Ark whose vital statistics are quoted in Gilgamesh XI: 61–3 has provoked a good deal of discussion ever afterwards but it is, from a historical point of view, a phantom.

Problem (b):

How can the wall height in Gilgamesh XI be ten times higher (ten nindan = sixty metres) than those in the parent text tradition (one nindan = 6 metres)?


The crucial point is that the measure of one nindan for the wall-height in the Ark Tablet results, as we shall see, in a coracle of normal proportions, so this has to be taken very seriously. In cuneiform writing, 10 is represented by a single diagonal cuneiform wedge and 1 by a single upright. The 10 now found in the Gilgamesh XI tablets could either be an ancient misreading of the original number ‘1’ or could reflect a deliberate ‘upgrading’ of the numeral because of the idea that everything about the Ark was going to be big.

Problem (c):

Why, in the Gilgamesh story, does Utnapishti only draw up a work-plan after five days of hard labour when the basic shell of the craft is already finished?


The explanation again comes from comparing the received text with the Ark Tablet version. The out-of-place verbal form in Gilgamesh line 60, ‘I drew up her design’, which has always been interpreted as a verb in the past tense, should really be understood as the imperative, ‘draw up her design!’ as Ea commands Atra-hasīs in Ark Tablet line 6. (The cuneiform spelling makes these two similar-sounding verbal forms confusable.) Originally this line belonged right after the contents of Gilgamesh line 31, when the hero was receiving his instructions, in parallel with the Ark Tablet.

2. The Interior

These are the Akkadian verbs for constructing the five-star floating hotel as the late poet described it:

urtaggibši ana 6-šu, ‘I gave her six decks’ (verb: ruggubu)

aptarassu ana 7-šu, ‘I divided her into seven parts’ (verb: parāsu)

qerbīssu aptaras ana 9-šu, ‘I divided her interior into nine’ (verb: parāsu)

Gilgamesh XI: 61–3

Proposition: It seems to me that these three lines derive ultimately from the very ‘fingers of bitumen’ passage that is missing in Gilgamesh, signifying a gross misinterpretation of the underlying text.

Defence: The verb ruggubu (from the root RGB), ‘to roof’, in the form urtaggibši occurs only in one passage, Gilgamesh XI according to the Chicago Assyrian Dictionary. There are admittedly not many contexts in life where deck-providing might be a central issue, and it is specious to argue that ‘to roof’ is not the same as ‘to provide with decks’, since the effect is the same. However, there is the very similar-sounding verb rakābu, rukkubu, šurkubu (root RKB) in the Old Babylonian account:

I caused the kilns to be loaded (uštarkib) with 28,800 (sūtu) of bitumen into my kilns …

I ordered the kilns to be loaded (uštarkib) with fresh bitumen … in equal measure …

Ark Tablet: 21 and 25

Perhaps the later tablet-editors found the Old Babylonian verb uštarkib, ‘I caused (the kilns) to be loaded’, confusing when not applied to vehicles or boats, as it almost always is, and as I certainly first interpreted it when struggling to understand those lines in theArk Tablet. Perhaps, too, in an effort to make sense of an unclear passage, they associated the underlying root rkb with the noun rugbu, ‘loft’ or ‘upper room’, and came up with a derivative verb – as one can in Semitic languages – ruggubu, meaning ‘to fit withrugbus’. It is as if, in English, one were to say ‘deckify’ or ‘loftisise’; terms not in the dictionary but transparent in meaning.

Let us continue a little further in this vein. In Gilgamesh, the verb parāsu appears twice in the form aptaras, ‘I divided’, once with reference to the Utnapishti boat’s interior. This is an echo of the verb aprus, ‘I divided’, in the Ark Tablet, where the distinction between exterior and interior was the main point:

I apportioned (aprus) one finger of bitumen for her outsides.

I apportioned (aprus) one finger of bitumen for her interior.

Ark Tablet: 18–19

Careless recycling of an Old Babylonian text like the Ark Tablet could also explain the oddity that feminine suffixes for the boat are not given in Gilgamesh: 61–3, although they do appear correctly from line 64 onwards.

I suspect, too, that the Old Babylonian signs ŠU.ŠI standing for ‘finger’ in Ark Tablet 18–20, which include the cabins, were later interpreted as šūši, 60, and that the three 60s of the original became disassociated from bitumen-thicknesses. Instead they were thought to have something to do with the decks and chambers and evolved by distinct numerological activity and cosmological speculation into the sequence 6, 7 and 9, undoubtedly compounded by the conviction that the vessel itself was a giant, straight-sided cube. A sort of Babylonian midrashic development, subtle and full of allusion, then played long on Utnapishti’s over-inflated time-capsule, the simple 1,000-year old text subjected to theological-cum-philosophical interpretation and symbolical elaboration, as has been discussed at length in George 2003, Vol. 1: 512–13. (The theoretical Assyriological idea advocated by several scholars that Utnapishti’s Ark was connected with the multi-layered ziggurat temple at Babylon is rendered less innovative by the fact thatGilgamesh XI: 158 actually refers to the Ark, once landed on the mountain, as the ziggurat!)

The growing number of floors and subdivisions is also a practical response, for not all species were equally compatible, and humans might want separate quarters. For these reasons, one can understand how the Ark blossomed into a five-star skyscraper hotel with certain cosmic resonances. My suspicion, however, is that the Gilgamesh Ark as it came to be launched from Nineveh is primarily the consequence of textual misunderstanding compounded by the insertion of narrative without care as to overall meaning coupled with interventive editing. I doubt that many people who knew or heard the story ever believed for a moment that the Ark was really a perfect cube.

If you find an error please notify us in the comments. Thank you!