The Ark Was Round: Irving Finkel on Cuneiform, Floods, and a Seal From 9000 BC

Irving Finkel tells Lex Fridman that the first attempts we can honestly call writing appear in the middle of the fourth millennium, around 3500 BC. People living between the Euphrates and the Tigris already used clay for buildings and objects; eventually they used it as a writing surface. The essence of the invention, he says, is an agreed system of symbols: one person makes a sign, another sees it and knows what sound it engenders, “a bit like a gramophone record.” All scholars can actually claim is that the oldest surviving evidence sits around that date. Whether writing itself began anywhere near 3500 BC seems to him “very, very unlikely.”

The pictures came first in the archaeological record: a foot meant a foot, barley meant barley. Then someone realized that a picture of a foot could also stand for the *sound* of the word for foot. That, Finkel insists, is the genius step. Once signs could carry sound, scribes were free of pictographs and could record language, grammar, proverbs, and literature. A trained writer could take down Sumerian or Babylonian, and even a language he did not speak, purely by ear. Long before the alphabet was a dream, this “bewildering-looking” system lasted well over three millennia, closer to four, and it took a long time for the more sensible alphabet to displace it.

Received Assyriology says the order was pictures first, sounds second. Finkel finds that hard to believe. If a community urgently needed marks that everyone could read, why not start with signs for sounds? Speakers already shared a language; they could have isolated vowels and consonants, made fifty signs, and written anything. Pictographs are the least flexible system, almost useless for grammar. His hunch is that the Sumerian pictures sit at the end of a very long tradition of picture-messages among traders who did not share a tongue — three little sheep for three sheep, like a cigarette with an X through it. “That’s what I think happened.” Many Assyriologists, he adds, would leave the room.

Finkel’s controversial exhibit is a small object from Göbekli Tepe in southeastern Turkey, a site dated to about 9000 BC — some six thousand years before Stonehenge and seven thousand before the Giza pyramids. Buried in a color plate that archaeologists unwisely put online, among heaps of stone, is a round green piece shaped like an Egyptian scarab: arched back, flat bottom, hieroglyphic signs cut into the face. Nobody had written a word about it. To Finkel it is obviously a stamp. Pressed into clay or another sealing material, it would leave an official impression. That is writing, he says, thousands of years before southern Iraq.

His old professor used to argue that writing evolved in Mesopotamia because cities, ziggurats, and crowds forced people to organize. If that logic holds for Sumer around 3000 BC, Finkel asks, how did Göbekli Tepe’s planners finish buildings of astonishing quality without it? He does not believe seasonal nomads threw up that architecture with a megaphone. A finished seal with a flat face is not a doodle on a pot. It implies contracts, brick payments, an official who had to ratify. He quotes Sherlock Holmes — the line is not original to Conan Doyle, he notes — that it is theoretically possible to infer Niagara Falls from a raindrop. The Göbekli seal is his raindrop.

He is willing to imagine that people wrote on leaves that later vanished. In the Indus Valley, almost everything that survives is seals of three to five signs, probably “me, son of my dad.” Scholars write “abject nonsense” about that script, he says; the real writing was likely on perishable palm leaf, as people in parts of India still write today. Why would it have been different then? Inertia, he argues, is nearly as strong as evolution, and people who talk about progress have no idea how powerful it is. Lex remarks that Finkel’s whole trade is Holmesian: a few pieces, a reconstructed world. Even the cuneiform hoard in modern museums is a raindrop beside millennia of human talk.

Assyriologists can now read Akkadian, Babylonian, and Sumerian more or less fluently. The problem is an unspoken assumption: that what sits in the world’s collections is something like all there ever was. Finkel calls that absurd. In the Ur III period, around 2100 to 2000 BC, city-states recorded tithes, offerings, and taxes on tablets the size of a matchbox. He calls the scribes “the ancestors of the Inland Revenue.” Thousands of those tablets exist, scattered through museums after nineteenth-century excavators and local diggers emptied the find-spots. They probably came out of two rooms. Two dump rooms are not a country, a religion, or a history.

Once in a while a real library surfaces. In the 1850s, Layard and others dug at Nineveh, the Assyrian capital, and found the collection of Ashurbanipal, a king who styled himself ruler of the world. He was a trained scholar; he impounded tablets and tried to house prevailing knowledge under one roof, an Assyrian cousin of Alexandria. The tablets lay higgledy-piggledy across a great hall and its corridors, many broken, many burnt. For more than a century, specialists joined fragments: a goddess trying to seduce Gilgamesh in a garden, a missing line, another shard. The working assumption was that if you fitted every piece, you would have the whole library.

Finkel thinks the opposite. Southern Babylonians, in his view working with Elamites from Iran, took Assyria in a pincer, ran through the capital, and set fire to it. They took women, jewelry, and gold. People say they destroyed the library in a fit of pique. He does not believe it. The library was the giant brain of a world empire, written in a language and a script the conquerors shared. They would have taken the tablets home, cart after cart. What remained on the floor were duplicates, broken pieces, and things dropped in the rush. Lex jokes that Finkel is starting trouble without a pause. “It’s common sense,” Finkel says.

Cuneiform, named in the nineteenth century from Latin *cuneus*, “wedge,” does not write isolated consonants. Scribes placed a vowel before or after: *ab* and *ba*, *ib* and *bi*, *ub* and *bu*, *eb* and *be*, across four vowels and the full set of consonants. To write *museum*, a Babylonian would put down *mu*, then a sign for the middle, then *um*, and squash the syllables together in the head. Signs also carried extra values because they came from Sumerian words that were probably once distinguished by tone. After the tones vanished, *ba* number one, *ba* number two, and *ba* number three all still meant *ba*. Students must learn the list and pick the right reading from context. The scribes left no spaces between words.

A university course therefore means two languages at once. Babylonian is Semitic, kin to Hebrew, Aramaic, Arabic, Ethiopic, and Syriac; nineteenth-century decipherers could fall back on those dictionaries. Sumerian is related to no known language. Languages do not exist in a vacuum, Finkel says, so Sumerian must once have had a family — perhaps in China, Russia, or somewhere else in Asia. Writing arrived just in time to catch this one representative. The relatives are gone forever unless something amazing happens. That isolation still bewitches him.

He will not accept that Neanderthals, after 800,000 years of European life, ice ages, and child-rearing, had no speech. They had the same apparatus and a brain that answers to stimulus. Picture two hunters after rhinoceros; one keeps shouting “legs” until the other realizes the word is a warning about *his* legs. Once is enough: I am me and you are you. People who claim early humans could not distinguish an ego from the world strike him as ridiculous. Cut your hand and the blood is yours, the pain is yours, you might bleed to death. That is not someone else’s hand. Consciousness of being a separate body, he thinks, did not wait for philosophy.

When Finkel was a student the grammars and dictionaries already existed; the heroic work was done. The Rosetta of cuneiform was a trilingual cut high on a mountain at Bisutun in Iran, where Darius recorded his career in Elamite, Babylonian, and Old Persian. Persian was still alive in the nineteenth century, and the Old Persian column used a simple style of wedges. Scholars recognized the language, read the name *Darayawush*, and noticed that the other two columns were about the same length. The Persian text repeated a formula: I am Darius, great king, king of the world, son of, grandson of. Those reiterated phrases became a chisel into the other two tongues.

The golden key was the discovery that Babylonian was Semitic. One of the first words identified was “river,” Akkadian *naru*, matching Arabic and Aramaic *nahr*. Dictionaries of Arabic and Hebrew came off the shelf, and the rest of the inscription yielded, with many remaining difficulties. Without the trilingual, Finkel is not sure the script would have been cracked when it was. Henry Rawlinson, a young British officer, climbed the face with “some miserable kid,” made squeezes overlooking the plain, and is still called the Father of Assyriology. Finkel prefers stepfather. Rawlinson’s first thick book on the inscriptions was almost entirely wrong.

The person who saw how the system actually worked, Finkel says, was Edward Hincks, a clergyman in Killyleagh, Northern Ireland, with five daughters and a church — “possibly a card-carrying genius, if not jolly, jolly close.” Hincks had been trying to beat the French to Egyptian hieroglyphs. He ran into a wall, glanced at cuneiform for help, and cracked it instead. His decisive insight was that one sign can have more than one sound and more than one meaning. “I tried to shelter you from the horrible news,” Finkel tells Lex. Learning it is not a walk in the park: about five years, or four for someone like Lex. On top of the multiple values, there are no word spaces. You start at the first sign, run its readings, test whether *ba* plus *ab* could begin a word, and walk forward. The variables were restricted on purpose, so the system is learnable. Once mastered, he says, it is “very beautiful.”

If the alphabet is obviously more useful, why did cuneiform last three thousand years, approaching four — in Finkel’s phrase, the most successful writing system ever? Inertia is the first answer. The second is power. People who could read and write ran archives, served as temple clerks, and wrote for the king. Most of the public could not. They reserved knowledge, interpretation, and philosophical inquiry. Scribal schools were cliquey in the way Oxford and Cambridge are rivals. Literacy for all would never have been proposed; it would have been anathema. A man on a Saturday soapbox shouting that children must be taught would, Finkel thinks, have been taken away.

School still had a lower as well as an upper path. Teachers farmed out boys who would rather have been playing football, but who could read and write, into low-level work: contracts, letters, the four-day business of marrying off a daughter with witnesses present. Talented students went into medicine, law, the palace, the priesthood, architecture — someone had to know load-bearing and brick measures — or army logistics, moving stores and animals. At the top sat a thin intellectual layer writing astronomy, theoretical grammar, bilingual treatises, and commentaries on what words meant. “There was an intellectual high-level top, and then there were lots of professional scribes, and then the kids who left school as soon as possible… like today.”

Lex quotes Wittgenstein: the limits of language are the limits of the world. Finkel thinks Babylonian, in vocabulary and verbal subtlety, rivals Arabic and English; you can say almost anything. What the grammar cannot mark are modal verbs — could, might, should, ought. That matters because a huge share of scholarly literature takes the form of omens. A typical line, translated the usual way, runs: if a lizard runs across the breakfast table, the queen will die. Thousands of such sentences fill modern books as if A simply equals B.

Finkel says that equation is impossible. You are the king’s chief diviner, a sheep’s liver in your hands, the king standing there. If you announce that the queen will die and she does not, you look like an idiot. If she does die, you are responsible. All you can ever say is that there is a sign the queen *could* die — and the proper ritual or magic must swing into action to defer the danger. A equals B was never true. With A, B could be, might be, ought to be. The court diviner was a philosopher who knew what the king wanted to hear and how to wrap bad news so it would not wreck the room. “It’s the most beautiful thing. It’s so subtle. It’s like a violin concerto.”

Medical texts get the same mechanical “he will get better.” Finkel asks whether Lex has ever met a doctor who says that. Doctors say all being well you will be back on your feet, or this condition usually goes fine. Never “you will,” because what if you die? Colleagues still translate the grammar-book future automatically, with “no self-appraisal of the folly of it.” Change “will” to “might” and the whole text is different. The Chicago Assyrian Dictionary, begun in the 1920s, funded for decades by the National Endowment for the Humanities, and staffed by most of the world’s best Assyriologists, is his model of how translation should work: not A means B, but a proverb’s sense against a letter’s. He ranks it with the electric guitar as America’s greatest cultural achievement. It is as long as a table. He will show it to Lex at the museum.

Mesopotamia had a large pantheon. Three gods sat at the top — Anu, Enlil, and Ea — with hundreds of gods and goddesses beneath them. Village deities were worked into a theological phone book: lesser gods amalgamated, then given jobs in the households of greater ones, with temples and specialties, including illness and the dead. A baby was placed under one god’s tutelage. Because gods were humans on a larger scale, they could be forgetful, uninterested, or on holiday, so offerings were prompts and small bribes. Gods were “great when they were there but not always there.” Finkel thinks that still describes a lot of modern life. People who did not grow up in a stifling religion get interested when they are seriously ill or hit by disaster. Few walk around in religious awe, and a good proportion of clergymen he has met do not either.

In southern Iraq the difference was immediacy. There are no skyscrapers, no night lights; people slept on the roof under an unsmogged sky. The gods were not an artifice you had to talk yourself into. People did not “believe” in gods or ghosts. They took them for granted. Nobody in the inscriptions Finkel has found asked whether ghosts existed or whether he had really seen one. A person who died peacefully went down through the gates of the netherworld and stayed there. Families buried the dead under the courtyard and poured fluid and food down a hole, a symbolic offering close in spirit, he says, to practices in Islam and Judaism. Mortality itself does not strike him as a tragedy — the tragedy would be if we were not supposed to die. Whining after a death is, in his word, infantile, because every relationship has a finite clause.

Mankind’s great mistake, he argues, was the creation of monotheistic religions, “because they brought evil into the world.” If you believe in one god, I am right and you are wrong; the footing is already dogmatic. Inquisitions and the rest follow from that claim of superiority. Polytheism, with specialties, favorite gods, temples, and offerings, did not affront individuality in the same way. In antiquity, he says, there was no religious prejudice and no racial prejudice; those are modern disadvantages. What has been done in the name of religion is “absolutely staggering.” A temple in Kolkata that he visited with his wife Joanna felt to him much like the ones in Mesopotamia must have felt.

The fullest Gilgamesh we have is a twelve-tablet work from the Nineveh library, long enough to feel Homeric. Gilgamesh was a real king of Uruk who, like Alexander, went on living in story after death. Tales circulated orally while he was alive and after; around 1800 BC people began writing them in Sumerian or Babylonian; eventually they were woven into the long adventure about humanity, immortality, and man in the hands of the gods. Even the polished Nineveh version still shows its oral roots. Every speech is introduced with “Gilgamesh opened his mouth to speak and addressed his friend Enkidu,” and the reply is framed the same way. A reader does not need that scaffolding. It is what a narrator says when a scene is being enacted, frozen into the clay like a stage direction.

Finkel thinks fireside narrative is as old as people gathering after dark. Cave paintings of running antelope, done in mineral color at unsurpassable quality, force him to allow for strong creative talent very early in *Homo sapiens*. Populations contain imbeciles and Einsteins, and somewhere along the line Rembrandts. He imagines that half the cave paintings in Europe were done by one traveling person, because a genius painter in every family is impossible. Stories told the young what happened: famous battles, the flood, how fire was made, how the wheel was invented. Even a tale told without a moral usually has one if you think about it.

A visitor to the British Museum once brought in a tablet about eight inches by three. Finkel saw almost at once that it was part of a flood story, dated around 1700 BC. The gods had decided to wipe out the population of Mesopotamia because the people were so noisy the gods could not sleep after lunch. The third god in the triumvirate thought this deplorable and warned a man named Atra-Hasis to build a boat, specifying shape, materials, and quantities — a sixty-line blueprint. The shocking detail is that the boat was round. Everyone who knows the Bible pictures a coffin-shaped ark. Mesopotamian rivers actually used round coracles, cargo boats that almost never sank. The story simply scales one up until male and female animals, Atra-Hasis, his wife, and his three sons can float until the waters fall and the world starts over.

In 1872 George Smith, a gifted reader at the Museum, found a Nineveh tablet that ran in parallel with Genesis: waters, boat, landing on a mountain, the hero Utnapishtim releasing a bird three times, the third not returning. That was not two independent memories of the same event, Finkel says. It was literary dependence. The Nineveh Gilgamesh version and the Hebrew Bible sit in roughly the same centuries, the seventh and sixth. The 1700 BC tablet is a thousand years older; nobody thinks the Bible was written then, so Mesopotamian primacy is settled. Jerusalem does not flood. Mesopotamia does — too little water, too much, sometimes far too much. A catastrophic regional inundation, perhaps a tsunami sweeping toward the Persian Gulf, would have drowned houses and animals and left survivors in boats. He does not think anyone ever forgot it. Graham Hancock’s Younger Dryas asteroid-and-global-flood idea he rates as negligible. The Mesopotamian story, he believes, does rest on a real disaster, but nothing that touched America, China, or Birmingham.

After the destruction of the Temple, Judeans lived in Babylon under Nebuchadnezzar for three generations. They spoke Hebrew and Aramaic; Babylonians spoke Babylonian and Aramaic; clothes and skin matched. Without a charter they would have intermarried and disappeared. Finkel thinks authorities compiled a history from the beginning of time to explain the mess — chosen people, no temple, no country — and recycled Babylonian material they had never had in their own philosophical tradition, including creation. The Babylonian gods punished noise; Jewish philosophers, facing a vengeful Almighty, recast the crime as sin. “They took one narrative and they recycled it for their own purposes.” The plot itself is irresistible: you have five days or two weeks, the clock ticks, one person might save the world. Utnapishtim, then Noah, then Hollywood.

Noise, he later adds, was a euphemism. Before the flood the gods had not created death, so numbers of people and animals exploded. Afterward they institute barren women, men who cannot have children, princesses without offspring — a sophisticated Malthusian arrangement. He does not think the giant ark was ever built; people who survived were on ordinary boats, and the tablet is a literary construction. A one-third replica was built on a lagoon in Kerala by three specialists in medieval Arab boats, using computer-limited wooden ribs, woven covering, and bitumen. Indian bitumen was poor; Iraqi bitumen could not be exported. The hull leaked. Finkel asked the producer whether he had ever been in a rowing boat with a dry bottom. Seepage is a feature. The proportions, he thinks, come from oral performance: a storyteller facing river people could not say “the biggest coracle you have ever seen” and sit down. He had to fetch real quantities, scale them, and hypnotize the room with numbers. The tablet sits on the cusp of oral and written literature.

The Royal Game of Ur is a twenty-square board of idiosyncratic shape. In the 1920s Leonard Woolley, digging royal graves at Ur, found four or five boards with dice and pieces, dated around 2600 BC. The game mattered enough to take into the next world. Later boards turn up across Mesopotamia, Egypt, Syria, Lebanon, Jordan, Turkey, Greece, and Crete, down to the end of the first millennium. Tutankhamun had two or three sets in his tomb. For nearly three thousand years it was a world-conqueror on the scale of chess or backgammon, and it spread without written rules and without players necessarily sharing a language. A merchant in a bar would have a go, go home, try to remember the shape, and reconstruct the play. After about 2000 BC the far squares migrated onto a twelve-square central avenue with four squares on each flank — the form that then stuck.

It is a race game: pieces enter at a corner, turn, run the middle, and bear off, with rosette squares as safety or extra throws and room to block. Chance is real, but Finkel’s reconstruction builds in a measurable quota of strategy. Snakes and Ladders is almost pure luck; chess is pure strategy; backgammon is the grown-up mix, and Ur belongs to that family. A British Museum tablet from the second century BC, 2,300 years after the first boards, names pieces and discusses throws for a descendant of the early game. Working backward, Finkel produced rules that people now play worldwide, including in Iraqi cafés. Games should not last four days; a short match, best of three, is enough. After a Getty lecture, twenty commercial sets went onto small tables. Three minutes of rules, then silence, then someone shouting “I hate you! I’m never playing this game with you again.” Happily married couples, he claims, were phoning lawyers by afternoon. All good board games provoke that salutary fury.

He thinks the appetite is human. Board games sublimate rivalry onto a safe terrain where you can still get angry. More important is what India calls *time pass*, not quite pastime: what you do when it is too hot to do anything, grandmother and grandchildren under a tree, stories and another round, winning not quite the point. Gambling arrives later, because coinage is late, but a school tablet quotes a scrap Finkel reads as a gambler’s lament: “Oh my astragal, oh my astragal, woe is me.” Knucklebones served as dice; he is sure people bet a sack of grain or a roomful of goods on a throw. The late rules tablet itself is set in a bar, with moves tallied in food, drink, and women.

Finkel will not call the British Museum an art museum. The Louvre is basically that; the Met is named for art and prioritizes design, color, and shape. The British Museum, as he sees it, is the narrative of the human species in objects: achievements and the horrible things too, with no favored religion or country. It serves two horizons. One is the whole world. The other is the unborn — babies not yet here, and their children. He calls it a lighthouse in a universe of darkness, ignorance, stupidity, and skepticism, one of the places where you can talk about truth, beauty, elegance, and intelligence without affronting people who have none of those qualities. If Martians burst into the Great Court and demanded a fast briefing on the world, a walk through the essential objects would not give them an insane picture. The only missing piece would be Chuck Berry’s “Johnny B. Goode,” and that recording, he notes, has already been sent into space.

The usual complaint — that most of the collection is not on display — is not, for him, a valid criticism. The point is stockpiling for future examination. Distance shifts what an object means; the further back you go, the sharper the view of its precedents and its contemporaries. Storage and contemplation are inestimable. The Museum holds about 130,000 cuneiform tablets. Asked for something that makes him smile, he offers a Babylonian joke: a mosquito lands on an elephant and asks whether it is too heavy. You would not try it in a pub. Lex asks what wisdom the ancients had that we have lost. Finkel will not type out precepts. They lied and esteemed truth, fell in love, committed adultery, committed murder. Put them on a bus in the same clothes and you would not know. They were spared, he thinks, the electronic universe, which he calls disastrous because it reduces the vitality of the human component, like a drug that seems fine until you cannot stop.

He is not a Luddite about railway engines and kettles. What offends him is the short clip on a phone, “utterly, utterly wicked,” leaving children who cannot articulate, spell, or make meaning clear in English, “the best, most literary, and most beneficial language that’s ever been created.” His exhibit is the phrase “like I went,” used to mean “said” — a construction so hard to define grammatically that, if Sumerian *to go* actually meant *to speak*, Assyriologists would be lost. Lex notes that slang and sailors’ talk existed then too, and some of it did reach clay. Finkel’s larger point is that English is unimaginably rich: run a finger down a full Oxford English Dictionary and you may turn several pages before a familiar word. Vocabulary comes from reading, not from “Do you want more tea?” Lex suggests the Museum’s own long view: emojis will not be what two hundred years remember; Einstein, ideologies, rockets, the Moon will. Finkel does not disagree, but he will not let go of the claim that language, properly used, is a crucial human tool. He invites Lex to the Museum, to see “these confounded things,” the Chicago dictionary, and a grammar book.