Ancient History & Civilisation


The Question of Shape

And when the Sieve turned round and round,

And every one cried, ‘You’ll all be drowned!’

They called aloud, ‘Our Sieve ain’t big,

But we don’t care a button! we don’t care a fig!

In a Sieve we’ll go to sea!’

Edward Lear

The most remarkable feature provided by the Ark Tablet is that Atra-hasīs’s lifeboat was definitely, unambiguously round.

No one had ever thought of that possibility. Confronting the fact comes, initially, as a shock. For everyone knows what Noah’s Ark, the real Ark, looks like. A squat wooden affair with prow and stern and a little house in the middle, not to mention a gangplank and several windows. No respectable child’s nursery at one time was ever without one, with its chewed pairs of lead or wooden animals.

A classic example of a toy Noah’s Ark and animals in painted wood; from about 1825 and probably German.

(picture acknowledgement 7.1)

Sunday entertainment.

(picture acknowledgement 7.2)

The tenacity of the conventional Western vision of the Ark is remarkable, and remains, at least to me, inexplicable, for where did it come from in the first place? The only ‘evidence’ that artists or toymakers had before them was the description in the Old Testament where, as we will see, Noah’s Ark is altogether a different proposition.

Whatever the pattern was before, we can now see that the Mesopotamian ark from Old Babylonian times was unquestionably round. We learn this fact from the new Ark Tablet, the remarkable and unexpected contents of which will now hold our attention for many pages to come. For this tablet, with its sixty lines, has more to offer than any other cuneiform tablet I have ever encountered, and it is the duty of any self-respecting Assyriologist to give such a document the full squeeze treatment and ensure that no possible item of information inside it is left unextracted.

We have seen that the tablet begins with a classic ancient speech advocating a boat of recycled reeds. Without pause Enki lays out unambiguously for Atra-hasīs what he is to do, which is to build something altogether different:

Draw out the boat that you will make

On a circular plan;

Let her length and breadth be equal,

Let her floor area be one field, let her sides be one nindan (high).

10.       You saw kannu ropes and ašlu ropes/rushes for [a coracle before!]

Let someone (else) twist the fronds and palm-fibre for you!

It will surely need 14,430 (sūtu)!

Reading lines 6–7 for the first time was certainly an adrenalin-stirring moment, and my first reaction – as anybody’s would have been – was can this be right? A circular plan … ?

But then, thinking it over, staring into space with the tablet precariously poised over the desk, the idea began to make sense. A truly round boat would be a coracle, and they certainly had coracles in ancient Mesopotamia and when you thought about it a coracle is exceptionally buoyant and would never sink and if it happened to be difficult to steer or stop from going round and round that would not matter, because all it had to do was keep its precious contents safe and dry until the waters receded. So, no need to gasp and stretch one’s eyes. On the contrary, it made a lot of sense, and what was going on here was something serious and valid and highly interesting …

The Akkadian word for the Ark is, here too, eleppu, ‘boat’. The phrase ‘circular plan’ in Akkadian is eṣerti kippati, in which eṣertu means ‘plan’, and kippatu ‘circle’. The Ark Tablet does not use a special word for coracle, although there was one in Akkadian,quppu, as we will see.

Enki tells Atra-hasīs in a very practical way how to get his boat started; he is to draw out a field-sized plan of the round boat on the ground. The simplest way to do this would have been with a peg and a long string; the peg is stuck in what becomes the middle of the circle, the boat-builder walks the taut string round to mark the circumference, much as described later in this chapter by Colonel Chesney in laying out a differently shaped boat. The stage is thus set for building the world’s largest coracle, with a base area of 3,600 m2, with a diameter of, near enough, 70m. Atra-hasīs actually probably did not need to be told such elementary stuff. There is good background from other cuneiform texts where the word uṣurtu, the more common form of eṣertu, is used of the plan of a building detectable on the ground.

Then comes Enki’s remark, ‘let her length and breadth be equal’, at first sight disconcerting because everyone knows what a circle looks like and therefore what a circular boat would look like. This is a god speaking, however, who is not concerned with the theoretical nature of circles but with reinforcing the image of a round boat; unlike any other boat, it has neither prow nor stern but is the same width – or as we would say, diameter – in all directions. Enki’s instructions to be build a coracle were very specific, given the plan he had in mind, and his servant Atra-hasīs had to be clear on this.

A circle within a square forming part of an exercise in Sumerian geometry; this large tablet is the teacher’s reference copy with all the answers.

(picture acknowledgement 7.3)

Atra-hasīs in the Ark Tablet, one senses, knew as much about boats as the next man, although Enki did have to encourage him about details, suggesting that he could get help (lines 10–12) as he began to contemplate just what lay ahead of him in building the world’s first Super Coracle.

It was obviously a sound idea to tackle the first reading of this new inscription with the familiar Flood Story texts close at hand, and there were further surprises to come. I discovered before long that two of the tablets, both conveniently in the collection of the British Museum and easily consulted, also proved on reinvestigation to feature an ark that was round. The crucial cuneiform signs were in one case damaged and in the other without good context, but in both the key word kippatu, ‘circle’, was there in the clay.

Old Babylonian Atrahasis

In Old Babylonian Atrahasis the section which describes the Ark is closely related to the wording of the Ark Tablet but is incomplete. In line 28 we can now recognise the partly preserved word kippatu:

“The boat which you are to build

[Let its … ] be equal [(…)] […]

28         […] circle … […]

Roof it over like the Apsû.”

The cuneiform signs readable in line 28 are: […] ki-ip-pa-ti x x [x (x)].

Assyrian Smith

Lines 1–2 of Assyrian Smith, close enough in date to the first-millennium Gilgamesh XI tablets, contain the same important matter, but although the word has long been correctly read its significance could never be appreciated, and even now it is still not quite clear how this passage should be understood because it is incomplete.

“[…] … let [its … be …]

2           […] … like a circle … […]”

The cuneiform signs in line 2 are: […] x ki-ma kip-pa-tim x […]

There is a crucial difference in the second case, one thousand years on, in that the boat, or some characteristic of it, is now ‘like a circle’, which of course is not the same thing as being a circle, but it would be a stern sceptic who insisted that this was unconnected with the shape of the vessel itself, in view of the other two accounts. It is evident that Enki’s description befuddled Atra-hasīs, who in this later Assyrian version of the story emerges as much more self-effacing than his Old Babylonian counterpart and asks for a guide drawing; one imagines a hand reaching down with Rembrandt’s pointed finger to trace the explicit shape on the ground:

Atra-hasīs opened his mouth to speak,

And said to Ea, [his] master,

“I have never built a boat …

Draw the design on the ground

That I may see [the design] and [build] the boat.”

Ea drew [the design] on the ground.

Assyrian Smith: 11–16

Here, in a flash of cross-millennial understanding, we encounter a recognisable human being. Atra-hasīs, going about his daily life and far from thinking about saving the planet, has been charged all of a sudden – by Enki himself – with an impossible responsibility for which he is perhaps Mesopotamia’s least suitable candidate. He has never built a boat, and for him verbal descriptions are not enough: if he is going to have to do this he wants a clear plan. This professed reluctance or lack of skill to undertake an enormous task suddenly thrust upon him has parallels with Moses in the Book of Exodus, who cries ‘Who am I that I should go … ?’ or with the prophet Jeremiah who, taken aback when called by God to be a prophet, initially protests that he is too young and inexperienced to speak in public.

We now have three cuneiform flood tablets in which the Mesopotamian Ark’s shape is given as (or in one case, likened to) a circle.

Could a round ark, therefore, be the Mesopotamian norm? Emboldened by this giddy progress – and it must be stressed that such an undertaking was courageous in the extreme – I decided to have another look at Gilgamesh XI: 48–80, which promotes that hugely famous – but very strange – cuboid ark. I say emboldened because this particular passage is one of the most celebrated in cuneiform with a classical status verging on that of Homer. To tamper with the text of Gilgamesh XI is probably to invite arrows and hot pitch.

Assyriologists have long known that Old Babylonian manuscripts like the Ark Tablet or Old Babylonian Atrahasis lie behind the Assyrian version of the whole Gilgamesh story that we know today from the Nineveh library; Jeffrey Tigay gave an enlightening examination of this matter in 1982. Such ancestor tablets were by then already a millennium or more old. Their texts, as we can see from what survives today, were not always identical; words could change their meanings or become obscure, cuneiform signs tend to get damaged, and the finished literature that the ancient editor-scribes who produced Assurbanipal’s beautiful library manuscripts finally bequeathed us had run through many hands. Deliberate changes and interpolations were also made along the way, and signs of editorial work – sometimes over hasty – are occasionally still perceptible. With the help of the newly arrived Ark Tablet the parallel description of the boat and its building in Gilgamesh XI turns out to be a fertile and revealing case study. We can see that an Old Babylonian account of building a round ark, closely related to that of the Ark Tablet, lies right under the surface in Gilgamesh XI, and we can understand how in the interim its message has become heavily disguised. No one who pored over this story in Assurbanipal’s reading room would ever have guessed that Utnapishti’s gargantuan Ark was also once a giant coracle made of bituminised rope.

This is a big and bold claim which must be substantiated forthwith. To undertake tilting at this windmill requires another sprinkling of cuneiform philology – which will, I hope, suffice to prove the point.

Information about the shape of Utnapishti’s Ark as we receive it in Gilgamesh XI is split into two sections; first as instructions from Ea; second in Utnapishti’s account of the construction.

The instructions from Ea:

The boat that you are going to build,

29         Her dimensions should all correspond:

30         Her breadth and length should be the same.

Cover her with a roof, like the Apsū.

Gilgamesh XI: 28–31

Next come twenty-six lines of quite separate narrative explaining what Utnapishti was to say to the elders and giving ominous warnings as to what he was to look out for, with no ark information. Then Utnapishti records:

On the fifth day I set in place her (outer) surface:

58         One “acre” was her area, ten rods each her sides stood high,

Ten rods each, the edges of her top were equal.

I set in place her body, I drew up her design.

I gave her six decks,

I divided her into seven parts.

I divided her interior into nine …

Gilgamesh XI: 57–63

This is some boat! Square in cross-section, six decks, multiple rooms …

However, in Gilgamesh XI line 58 the highly significant ark word kippatu, = ‘circle’, is also found. Here, let us beware, it is not spelled in simple signs, but is written with the Sumerian ideogram GÚR. In his great Gilgamesh publication Andrew George took this word as ‘area’ (George 2003, Vol. 1: 707 fn. 5) and translated the first part of the line as ‘one “acre” was her area’. With the benefit of the Ark Tablet we can retain the real meaning and take the word to refer to the Ark’s shape, thus translating kippatu here as ‘circle’.

Taking this step establishes that Utnapishti’s Ark in the Gilgamesh story was actually circular with a base area of one acre (ikû), exactly like the giant coracle of Atra-hasīs!

Ark Tablet 9:

Let her floor area be one ‘acre’, let her sides be one rod (high).

Gilgamesh XI 58:

One ‘acre’ was her circle, ten rods each her sides stood high …

In Gilgamesh XI the statement in lines 29–30 that the boat’s dimensions should all correspond and her length and breadth should be the same have become divorced from the crucial issue of her roundness, for this is only referred to further on (and non-explicitly) in line 58. This separation within the text of features that belonged together imposed the unfounded idea of a ‘square’ boat, far from the original meaning. This had the effect of displacing the original circular ground plan idea, enabling the very improbable cube to come into existence.

Where does this leave us? Another round ark, but this time submerged and almost lost to view. Given that some Old Babylonian text of the same ‘family’ as the Ark Tablet underlies the classical text of Gilgamesh XI: 28–31 and 58–60, we can assume that originally there was one instruction speech by Ea, and that development of the text disrupted the original simple format. This simple ‘proto-Gilgamesh’ instruction speech probably originally read as follows:


The boat that you are going to build


Draw up her design;


Her dimensions should all correspond,


Let her breadth and length be equal;


Let one ‘acre’ be her circle, let her sides stand one rod high;


The edge(s) of her top must be equal.


Cover her with a roof, like the Apsû!

The Ark as Coracle

Enki, looking down, knew all about coracles, and the reasons for his upgraded choice of ark model are, as already indicated, clear and intelligible. Atra-hasīs’s Ark did not have to go anywhere; it just had to float and bob around, settling, when the waters subsided, wherever it had drifted or been carried. The coracle in question was to be traditionally built of coiled rope basketry coated with bitumen; it would be unimaginably huge, but a lot of room was going to be needed.

Coracles, in their unassuming way, have played a crucial and long-running role in man’s relationship with rivers. They belong, like dugout canoes and rafts, to the most practical stratum of invention: natural resources giving rise to simple solutions that can hardly be improved upon. The reed coracle is effectively a large basket transferred to water, sealed with bitumen to prevent waterlogging, and its construction is somehow natural to riverine communities, so that coracles from India and Iraq, Tibet and Wales, are close cousins, if not easy-to-confuse twins.

Up until now no one seems to have afforded the ancient Mesopotamian coracle much attention, but with the arrival of the Ark Tablet on the Flood Story scene it suddenly becomes a very interesting creature indeed. There is hardly a mention of the coracle instandard works on ancient Mesopotamian boats, nor even the distinction of a specific word for coracle identified in the Akkadian language.

Or is there?

There is a cuneiform story known as the Legend of Sargon which is of huge significance within the pages of this book, and we will come back to it later in conjunction with the biblical story of Moses in the bulrushes. In the cuneiform version King Sargon of Akkad (2270–2215 BC) explains how his mother had deposited him, a new baby, on the River Euphrates in what is always translated as a ‘basket’, to go wherever the waters might take him:

I am Sargon, the great king, king of Akkad,

My mother was a high priestess but I do not know who my father was,

My uncle lives in the mountains.

My city is Azupirānu, which lies on the bank of the Euphrates.

My mother, a high priestess, conceived me, and bore me in secret;

She placed me in a reed quppu and made its [lit. my] opening watertight with bitumen.

She abandoned me to the river, from which I could not come up;

The river swept me along, and brought me to Aqqi, drawer of water.

Aqqi, drawer of water, lifted me up when he dipped his bucket,

Aqqi, water drawer, brought me up as his adopted son.

Aqqi, water drawer, set me to do his orchard work;

During my orchard work Goddess Ishtar loved me;

For fifty-four(?) years did I rule as king …

The Akkadian word quppu in line 6 of this composition has, so far, only three meanings according to modern Assyriological dictionaries: ‘wicker basket’, ‘wooden chest’ and ‘box’. In modern Arabic the word for ‘coracle’ is quffa, which also primarily means ‘basket’, since a coracle is nothing more than a large basket, manufactured like a basket and waterproofed, and this is the local word that has been heard up and down the bank of the Euphrates in Iraq wherever coracles were in use. Akkadian and Arabic are fellow members of the Semitic language family and share many historical words in common. We can say, therefore, that quppu and quffa are cognate words (for ‘p’ in Akkadian comes out as ‘f’ in Arabic), and can see that the two words share the same range of meanings, from basket to coracle. Given this I think we can conclude therefore that Babylonian quppu also had the specific meaning ‘coracle’, most especially with regard to the experience of the baby Sargon.

We can say more. Sargon’s autobiographical fragment undoubtedly alludes directly to the national Mesopotamian Flood Story, exactly as the story of Moses refers back to Noah’s Ark in the Book of Genesis. The baby was to be one of the greatest kings of Mesopotamia, his life saved at the outset against all odds by a bitumen-sealed, basket-like vessel launched on water into the unknown. The description of sealing the opening with bitumen is a direct textual parallel to the traditional Flood Story account.

There is an additional dimension to this. In the Gilgamesh account there is a striking poetic image at the end of the great storm on the seventh day:

The sea grew calm, that had fought like a woman in labour.

Gilgamesh XI: 131

It is easy to take this as a simple metaphor, but it would carry deeper meaning for a Mesopotamian. There is a cycle of magical spells to aid a woman in travail which share the image that the unborn child within the amniotic fluid is a boat in a stormy sea, moored in the darkness to the ‘quay of death’ by the umbilical cord and unable to break free to be washed out into the world. The round, nutshell Ark containing the whole seed of life, tossed on the waters before reaching anchorage, is undoubtedly likened to a storm-battered foetus, albeit obliquely; the voyage to eventual safety is re-enacted each time a baby is born.

According to F. R. Chesney, writing in the late nineteenth century, the smallest Iraqi coracle recorded was ‘3 feet 8 inches in diameter’. The chances are, then, that wee Sargon’s coracle, woven of reeds and waterproofed, was the smallest specimen ever made. If so, we have the unique privilege here of simultaneously documenting at one blow the world’s smallest and largest Iraqi coracles!

Now that we have the ancient name and two extremes in size we are entitled to look a little further into the question of normal coracles in ancient Mesopotamia. Where in fact are all the others? Since the Ark Tablet uses the general word eleppu for the round craft, it is natural to wonder whether other eleppus in cuneiform texts might not sometimes refer to a coracle, but only the odd example can be quoted as we proceed.

Although this humble riverine vessel has largely slipped by unnoticed under the radar, I maintain that skin-covered or bitumen-coated coracles must have crossed the Euphrates and Tigris waters, this way and that, more or less since the beginning of time. Pictorial evidence supports this. From the middle of the third millennium BC some of the hard stone cylinder seals that were used to ratify clay documents by rolling over the surface and leaving a customised imprint depict boats in their carved scenes. Most are evidently classic Mesopotamian reed boats with high prow and stern of the school that we have branded (from the Ark point of view) ‘prototype’, but we can distinguish at least one with the characteristic rounded profile, or rather cross-section, of a coracle. This seal is from the Iraqi site of Khafajeh on the Diyala River, seven miles west of Baghdad, and it appears to depict a genuine coracle in about 2500 BC.


Nearly two thousand years later we see the Assyrian army, nothing if not practical, making excellent use of campaign coracles, and fortunately for us these were depicted in accurate detail by the court sculptors within the daily-life and military scenes of the famous palace wall-sculptures.

The Assyrian King Shalmaneser III (859–824 BC) left us a graphic account of a military campaign in Mazamua (an Assyrian province on the northwestern slopes of the Zagros Mountains, modern Suleimaniyah), during which he was forced to use ‘reed boats’ and ‘skin-covered boats’ to pursue the enemy:

They became frightened in the face of the flash of my mighty weapons and my tumultuous onslaught and they swarmed into reed boats on the sea. I went after them in skin-covered coracles (and) waged a mighty battle in the midst of the sea. I defeated them (and) dyed the sea red like red wool with their blood.

The earliest coracle from the Khafajeh seal.

(picture acknowledgement 7.4)

King Sennacherib’s ancient four-man, heavy-duty coracle at work.

(picture acknowledgement 7.5)

Ship-to-shore: a heavily laden 20th century coracle approaches the bank.

(picture acknowledgement 7.6)

In a sculpture from the palace of King Sennacherib (705–681 BC) at Nineveh (see previous page), two sturdy pairs of Assyrian oarsmen negotiate the fast river currents in a heavy-duty coracle laden with bricks. Their long steering poles end in a curved hook and are apparently weighted at the lower end, perhaps with lead ingots. A fellow Assyrian astride an inflated animal-skin lilo on either side is spearing fish for their lunch. The men are seated on top of the coracle, which is loaded to the full and more, and seems to have some kind of bench running around the top. The oars are secured in a rowlock device. The coracle sides are marked with horizontal and vertical lines, which do not represent the lower layers of bricks inside the vessel but rather some external characteristic of its finish, probably panels of skin stitched together. The top rim or gunwale is clearly shown as a tightly bound and distinct reinforcing element although the binding is not shown at the right-hand edge.

These sepia snapshots in stone of ancient coracles in use are invaluable to us in demonstrating the existence and practical utility of the vessel in the ninth to eighth centuries BC. No doubt, as was certainly the case later, coracles were made in a range of sizes, from the two-person ‘water-taxi’ to a substantial craft capable of transporting, à la Noah, serious numbers of livestock.

Further south, a little later, we get hard information on Babylonian coracles in Greek, from the redoubtable Herodotus, writing his Histories in the second half of the fifth century BC when cuneiform writers were very alive and fertile; his book is one of the world’s ultimate bestsellers. An ongoing dispute persists about whether or not Herodotus actually went to Babylon himself, or about how reliable his statements are, and so forth, but when it came to facts about coracles he knew which way was up:

They have boats plying the river down to Babylon which are completely round and are made of leather. In Armenia, which is upstream from Assyria, they cut branches of willow and make them up into a frame, around the outside of which they stretch watertight skins to act as a hull; they do not broaden the sides of the boat to form a stern or narrow them into a prow, but they make it round, like a shield. Then they line the whole boat with straw and send it off down the river laden with goods. Their cargo is most commonly palm-wood casks filled with wine. The boats are steered by two men, who stand upright and wield a paddle each; one of them pulls the paddle towards his body and the other pushes the paddle away from his body. These boats vary in size from very large downwards; the largest of them can manage cargo weighing five thousand talents. Each boat carries a live donkey – or, in the case of larger boats, several donkeys. At the end of their voyage to Babylon, when they have sold their cargo, they sell off the frame of the boat and all the straw, load up the donkeys with the skins, and drive them back to Armenia. They do this because the current of the river is too strong for boats to sail up it, and that is why they make these boats out of skin rather than wood. Once they have got back to Armenia with their donkeys, they make themselves more boats in the usual way.

Herodotus, Histories Bk 1

Tigris coracles in the hands of professionals later caught the fancy of the Romans in the fourth century AD, who, with an eye to stowage and manoeuvrability, brought Tigris barcarii all the way from Arbela on the Tigris to South Shields in Tyneside to build coracles and run their river transports there, perhaps thereby introducing the first coracles to the British Isles. The Latin barca is a small boat carried on a ship and convenient for shipping cargo to shore, a common use of the coracle. Interestingly, an existing Latin term was applied instead of adapting the contemporary local Tigris word, which at that time was surely a form of quppu/guffa.

Early evidence for the British coracle.

(picture acknowledgement 7.7)

It is this practical background that makes sense of the Ark Tablet coracle. Some remote poet once asked himself or was enquired of by a listener – given that the Flood had really happened, and the Ark had really been built – what did the thing actually look like? What kind of vessel would be spacious enough, unsinkable yet buildable? Not a pointed magurgurru, by any means. Looking out over the river, rapt in a daydream, one can readily imagine that the solution would present itself in a lightning bolt of understanding: acoracle, a round coracle, on a – how you say? – cosmic scale …

We are entitled to focus in on an ancient river scene thronged with coracles because these traditional craft remained in use unchanged on the rivers of Mesopotamia right down into the first half of the last century, although in today’s Iraq they are, sadly, extinct. Coracles in general are a much studied and understood phenomenon, and the coracles of Iraq hold a more than respectable position among them. Many nineteenth- and early twentieth-century photographs taken there show coracles, portrayed either as specific studies or as part of the inevitable river background to daily life. E. S. Stevens, whose useful 1920s coracle-construction photographs are reproduced here, wrote evocatively:

we rattled over devious ways, splashing through the flood when we came to it, until the four lean horses came to a stop where a gufa was drawn up onto the bank. A gufa is a large bowl-shaped basket, made water-tight by a coating of bitumen. Some of these round craft are huge; ours would have held thirty people easily. We got in, and the gufachi slung a towing rope over his body, and waded upstream … When we had reached the actual river-bed, he jumped in with his helpers and began to paddle the boat across at an angle; for Samarra, on the high opposite bank, was by this time a good distance down-stream. The current was so swift and strong that it took only a few minutes before he landed us at the landing-stage below the city.

Stevens 1923: 50

Then there is the enigmatic E. A. Wallis Budge, later Keeper in the British Museum, an old coracle hand himself who knew them to be useful even in battle. At Baghdad in 1878 (he confesses) there was a little trouble over a tin of important clay tablets which had been mistaken by customs for a case of whisky and which needed to be deftly manoeuvred onto a British gunboat:

This procedure did not please the Customs’ officials, several of whom leaped into kuffahs and followed us as fast as their men could row. They overtook us at the gangway ladder, and tried to cut me off from the ship by thrusting their kuffahs in the way; and as some of them jumped on to the rounded edge of my kuffah, and tried to drag out of it my trunks and the box of Tall Al-’Amarnah Tablets, I became anxious lest the box of tablets be lost in the Tigris.

The “kuffah” [Budge added] … is a large basket made of willows and coated with bitumen inside and out. It is perfectly circular, and resembles a large bowl floating on the stream; it is made in all sizes, and some are large enough to hold three horses and several men. The small ones are uncomfortable, but I have journeyed for days in large ones, over the flood waters of the Euphrates around about Babylon, and on the Hindiyah Canal, and slept in them at nights.

Budge 1920: 183

Three stages in building a coracle as recorded by E. S. Drower (née Stevens).

(picture acknowledgement 7.8)

Walking the plank coracle-style.

(picture acknowledgement 7.9)

I am only sorry he didn’t bring one back for the British Museum.

This is as far as I think we can go in investigating Mesopotamian ark shapes on the basis of the known cuneiform Flood Story tablets. We know that tradition varied between the long and pointed ˚makurru (antiquated, unsuitable and unseaworthy) or the round and hospitable quppu (modern, practical and preferred). Later processes of textual accretion ‘developed’ the latter model into a tall, multi-floored tower of a cruise ship that was apparently endorsed by Gilgamesh himself (utterly unusable).

The next old photograph shows a cluster of traditional Tigris riverboats at the end of the nineteenth century. Side by side with plentiful round coracles are boats called taradas, whose characteristic outline, viewed from above, corresponds closely to the biconvexmakurru shape in the Old Babylonian diagram. The tarada is made of wood, with mast and sails, but in shape such boats are descendants of the ancient makurru. Looking at the two possibilities I think we can agree that Enki chose his round coracle Ark wisely.

J. P. Peters described his photograph of 1899 as ‘A Scene on the Tigris at Baghdad, showing characteristic native boats, the long taradas, and the round, pitch-smeared kufas, with bridge of boats beyond.’

(picture acknowledgement 7.10)

Noah’s Ark in Genesis

From here, as good investigators, we must follow the Ark trail where it naturally leads, which is to the Hebrew Bible and beyond.

Make yourself an ark (tēvāh) of gopher wood [came the instruction]; make rooms (qinnīm) in the ark, and cover it (kāpar) inside and out with pitch (kopher). This is how you are to make it: the length of the ark three hundred cubits, its width fifty cubits, and its height thirty cubits. Make a roof for the ark, and finish it to a cubit above; and put the door of the ark in its side; make it with lower, second, and third decks.

Genesis 6:14–16

Noah’s Ark as illustrated in Martin Luther’s bible, reflecting the Hebrew description.

(picture acknowledgement 7.11)

Such was the order to Noah, facing in his turn the awful task of saving the world more or less single-handedly with the help of a custom-order boat. This is the breakdown of the specs:


tēvāh (unknown word for rectangular boat)


gopher-wood (unknown species)


qinnīm (cells; the basic word means ‘bird’s nest’)


pitch or bitumen (kopher), smeared on (kāphar), inside and out


300 cubits (ammah) = 450 ft = 137.2 m


50 cubits = 75 ft = 22.8 m


30 cubits = 45 ft = 13.7 m


1 cubit high(?)





Compare the sparser data for Moses’ ‘arklet’ in Exodus 2:2–6:


tēvāh (unknown word for rectangular boat)


gomeh, bulrushes; rush/reed/papyrus; wicker


hamār, slime; bitumen/asphalt; bitumen; zefeth, pitch.

The biblical word tēvāh, which is used for the arks of Noah and Moses, occurs nowhere else in the Hebrew Bible. The flood and baby episodes are thus deliberately associated and linked in Hebrew just as the Atrahasis and Sargon Arks are linked associatively in Babylonia.

Now for something extraordinary: no one knows what language tēvāh is or what it means. The word for the wood, gopher, is likewise used nowhere else in the Hebrew Bible and no one knows what language or what kind of wood it is. This is a peculiar state of affairs for one of the most famous and influential paragraphs in all of the world’s writing!

The associated words kopher, ‘bitumen’, and kāphar, ‘to smear on’, are also to be found nowhere else in the Hebrew Bible, but, significantly, they came from Babylonia with the narrative itself, deriving from Akkadian kupru, ‘bitumen’, and kapāru, ‘to smear on’. In view of this it is logical to expect that tēvāh and gopher are similarly loanwords from Babylonian Akkadian into Hebrew, but there has been no convincing candidate for either word. Suggestions have been made for gopher-wood, but the identification, or the non-Hebrew word that lies behind it, remains open. Ideas have also been put forward over the centuries concerning the word tēvāh, some linking it – because Moses was in Egypt – with the ancient Egyptian word thebet, meaning ‘box’ or ‘coffin’, but these have ended nowhere. The most likely explanation is that tēvāh, like other ark words, reflects a Babylonian word.

I have a new suggestion.

A cuneiform tablet dealing with boats from around 500 BC, now in the British Museum, mentions a kind of boat called a ṭubbû which is found at a river crossing, apparently as part of a vessel swap among boatmen:

… a boat (eleppu) which is six cubits wide at the beam, a ṭubbû which is at the crossing, and a boat (eleppu) five and a half (cubits) wide at the beam which is at the bridge, they exchanged for (?) one boat which is five cubits wide at the beam.

BM 32873: 2

The Babylonian ṭubbû tablet, front.

(picture acknowledgement 7.12)

The Babylonian ṭubbû tablet, back.

(picture acknowledgement 7.13)

The consonants t (in tēvāh) and ṭ (in ṭubbû) are distinct from one another, so it is impossible that ṭubbû, a masculine noun of unknown etymology, and tēvāh, a feminine noun of unknown etymology, represent the same word etymologically. I think that the Judaeans encountered the Akkadian boat word ṭubbû used for the Ark in the story along with the other Akkadian ark words and Hebraised it as tēvāh. In this case the original consonants are less important; the idea was to render the foreign word, for it was only to be used twice in the whole Bible, once for Noah, once for Moses. The relationship between the words is thus that they are neither cognate nor loaned: the Babylonian was given a Hebrew ‘shape’. It is much the same as the way in which Nebuchadnezzar’s eunuch Nabu-sharrussu-ukin became Nebu-sarsekim in the Book of Jeremiah. This would perforce mean that the word ṭubbû must have occurred in place of eleppu, ‘boat’, for Utnapishti’s Ark, in some first-millennium BC Babylonian source for the Flood Story that we do not have now.

An alternative possibility is that the Hebrew word tēvāh is a so-called Wanderwort, one of those basic words that spread across numerous languages and cultures, sometimes as a consequence of trade, whose original etymology or language becomes obscured (a good example is chai and tea), lasting for ever. We would have then an old, non-Semitic word for a very simple kind of river boat – conceivably even ultimately ancestral to the English tub – which appears as ṭubbû in Babylonian, tēvāh in Hebrew. One could imagine readily enough that such a simple word for a simple boat might survive along the waterways of the world for endless centuries. Turned upside down these boats produce a dull ‘dub’ sort of thumpy thud. It is curious that tub, like ark, can mean box, chest and boat. Ironically this Babylonian word ṭubbû, like tēvāh, is rare too: it occurs twice in the tablet just quoted and nowhere else!

Either proposal would account for the biblical name for the Ark: either the Judaeans encountered the ark word ṭubbû and Hebraised it to tēvāh, or they called the Ark tēvāh because it corresponded to the shape characteristic of that kind of old boat which was known to them as a tēvāh and to the Babylonians as a ṭubbû.

But, again, what about the shape?

The traditional river craft of Iraq once included a type of boat which in shape and proportions closely resembles the Ark as described in Genesis. Lieut.-Col. Chesney, compiling a government survey, himself witnessed such boats being made and used in the 1850s:

A remarkable kind of boat is constructed at Tekrít and in the marshes of Lamlúm, but more commonly near the bituminous fountains of Hít. At these places the operation of boat building is an every-day occurrence, and extremely simple. The self-taught shipwrights have not, it is true, the advantage of docks, basins or even slips; yet they can construct a vessel in a very short time, and without employing any other tools than a few axes and saws, with the addition of a large metallic ladle to pour out the melted pitch, and a wooden roller to assist in smoothing it. The first step in this primitive mode of ship-building is to choose a level piece of ground of suitable size, and sufficiently near the edge of the water; on this the builders trace out the size of the vessel’s bottom, not with mathematical precision it is true, still a line is used, and a certain system followed, the floor or bottom of the boat being the first object.

This procedure is exactly similar to that in the Ark Tablet when Enki instructs Atra-hasīs on how to lay out the plan for the boat described above. Chesney continues:

In the space marked out a number of rough branches are placed in parallel lines, at about a foot distance; other branches are places across them at similar distances, and interlaced. These, with the addition of a sort of basket-work of reeds and straw, to fill up the interstices, form a kind of rough platform, across which, to give the necessary stability, stronger branches are laid transversely from side to side, at distances of about eight or twelve inches. The bottom being in this state, the work proceeds to the second stage, by building up the sides. This is done by driving through the edge of the former, upright posts, about a foot apart, of the requisite height; these are filled up in the same way, and the whole is, as it were, consolidated by means of rough pieces of timber, which are placed at intervals of about four feet from gunwale to gunwale.

Having completed detailing the structural aspects of the boat, Chesney goes on to describe the next stage of waterproofing, again parallel to the Ark Tablet:

All parts are then coated with hot bitumen, which is melted in a hole close to the work, and reduced to a proper consistency by a mixture of sand or earth. This bituminous cement being spread over the frame-work, the application of a wooden roller gives the whole a smooth surface, both within and without, which after a brief space becomes not only quite hard and durable, but impervious to water, and well suited for navigation. The usual shape of the boats thus constructed is much like that of a coffin, the broadest end representing the bow; but others are of a neater shape. Such a boat, 44 feet long, 11 feet 6 inches broad, and 4 feet deep, drawing 1 foot 10 inches of water when laden, and only 6 inches when empty, can be constructed at Hít in the course of one day …

Chesney saw at once that the shape and proportions of such vessels strongly recalled the biblical Ark, arguing rather plausibly that Noah could have produced a boat of this type without much trouble:

The ark, as we are all aware, was three hundred cubits in length, fifty cubits in breadth, and thirty cubits in height, finished in a cubit, or sloping roof. These dimensions, presuming the smallest cubit to have been in use, would give 450 feet for the length, 75feet for the breadth, and 45 feet for the depth of this enormous structure, whose burthen, making an allowance for the cross-beams with which it was braced and the supports, would be upwards of 40,000 tons. From the description just given of the Hít boats, it will be seen that there is not anything to prevent the people of that town, or of the neighbouring country, from constructing such a vessel, a larger scantling only being necessary for the frame-work. The lower story being intended for quadrupeds, must necessarily have been divided into compartments; and these divisions, as a matter of course, would support the second floor, which was appropriated for the people, whose apartments, again, supported the upper story, or that allotted for the birds. As this arrangement required three floors and a roof, the divisions and the necessary supports would have given sufficient stability to the whole structure; therefore the objections raised on account of the supposed difficulty of the work, may be considered as obviated, more particularly as the ark was destined to remain and be floated on the same spot …

Thanks to the archaeologist John Punnett Peters we have a photograph of several boats of this kind, in construction or finished, taken in 1888. Judging by his caption he, too, was irresistibly reminded of Noah’s Ark.

The second of J. P. Peters’ photographs, which he described as ‘A Noachian Boatyard at Hit on the Euphrates.’

(picture acknowledgement 7.14)

So now we have a real, functional boat-style candidate that is neither long and thin (Sumerian-type), round (Atra-hasīs-type) or square (Utnapishti-type), but which matches the oblong Genesis ark description to a disconcerting degree. It is reasonable, I suggest, to assume that the Hebrew description in the Bible reflects an oblong boat of this pattern, which, like the coracle, was surely commonly seen on the rivers of Mesopotamia in antiquity, and was encountered there by the Hebrew poets. Unfortunately neither Chesney nor Peters records the nineteenth-century Arabic name, but all things considered it seems not unlikely that this type of boat was called ṭubbû in Akkadian, tēvāh in Hebrew.

The existence of such boats contributes an important element to our assessment of the Hebrew encounter with the Babylonian story. If the oblong shape of the Hebrew ark reflects an existing type of Babylonian boat easily seen ‘out of the window’, this has direct implications for the transmission of the story.

It is conceivable that, while Utnapishti at Nineveh ended up tweaking a square ark out of a circular one, another and unknown cuneiform edition tweaked this a little further into an oblong, convinced that a cubic boat would never work and swayed by the existence of the oblong barge-type called a ṭubbû. While retaining virtually the same base area (15,000 cubits2 as against 14,400 cubits2), the length and width of the Ark were adjusted to round numbers reflecting the relative proportions of such a barge.

The importance and brevity of the biblical description of Noah’s Ark meant that successions of scholars, religious and otherwise, have pored over these lines of Noah text. The rabbis have left us many details to amplify the simple narrative.

Noah, for example, is supposed to have planted cedar trees one hundred and twenty years in advance with the double advantage that the population would have time to turn away from sin, and the trees could grow tall enough. The ark is variously attributed three hundred and sixty cells, or chambers, ten by ten yards, and nine hundred cells, six by six yards. Some authorities saw the top floor for the unclean beasts, the middle for the humans and clean beasts, and the bottom for refuse, while others favoured the reverse, while there was a trapdoor to allow waste disposal into the sea. Atra-hasīs, emptying pans, must have often mused rancorously over this supposedly humorous Akkadian fable:

An elephant spoke to himself and said, ‘Among the wild creatures of the god Shakkan there is no one who can defecate like me.’ The sipidiqar-bird answered, ‘And yet, I, in my own proportion, I can defecate like you.’

Since the sky was sealed off from the Ark’s inhabitants day and night there would have been darkness but the Rabbis explained that Noah hung up precious stones which shone like the noonday sun. The rounding up of all the animals, with their fodder, had been handled by a team of angels, while the hand-picked animals behaved in an exemplary manner and did not go in for reproduction while on board. Noah never slept for he was up the whole time feeding the inmates. Another thing: while the loading was going on, imposing lions guarded the gangplank to prevent the wicked from sneaking on board, which reminds me of the lions at the back door of the British Museum, which, however, are there to discourage visitors from leaving.

The Berossus Ark

Berossus, as we have seen in Chapter 5, gives no description of the boat beyond its dimensions:

He (Xisuthros) did not disobey, but got a boat built, five stades long and two stades wide

Patai writes that its length was ‘five stadia or furlongs – about 1,000 yards – and its breadth was two stadia – about 400 yards’. In the Armenian version of Eusebius’s Chronicles, which is based on Berossus, the length of the ship is given as fifteen furlongs, that is, nearly two miles.

The Ark in the Koran

Nuh’s [Noah] lifeboat Ark had no special name, but is referred to as safina, the common word for boat, Sura 54:3 describing it as ‘a thing of boards and nails’. There is no Koranic counterpart to the details of building the Ark or its appearance, although Abd Allah ibn Abbas, a contemporary of Muhammad, wrote that when Noah was in doubt as to what shape to make the Ark Allah revealed to him that it was to be shaped like a bird’s belly and fashioned of teak wood. In Islam, too, there was much later discussion and analysis of the story and its implications by the religious authorities. Abdallah ibn Umar al-Baidawi, writing in the thirteenth century, explains that in the first of its three levels wild and domesticated animals were lodged, in the second the human beings, and in the third the birds. On every plank was the name of a prophet. Three missing planks, symbolising three prophets, were brought from Egypt by Og, son of Anak, the only one of the giants permitted to survive the Flood, and the body of Adam was carried in the middle to divide the men from the women. There was a tradition that Noah had to say, In the Name of Allah! when he wished the Ark to move, and the same when he wished it to stand still.

An abundance of shapes, then. But we must return to the primary model. First, we must build our coracle.

English ladies on tour by coracle in the 1880s, but not entirely relaxed.

(picture acknowledgement 7.15)

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