What flood myths actually preserve, and why the lost civilization was never lost

I visited Göbekli Tepe two years ago. This is what stuck with me, and where it led.

A new excavation area, a hundred meters from the main enclosures of Göbekli Tepe, August 2024. The roof is up; the digging is only beginning.

The Hill That Shouldn’t Be One

On August 12th, 2024, I travelled to Göbekli Tepe. With temperatures above 40°C (104°F) and a boiling sun, one could ask why. My curiosity could not be tamed; I had to see and feel this place. As I stood on this hill in southeastern Anatolia that, strictly speaking, shouldn’t be one, I wondered: what was this place? Now an ochre swelling in a dried-out landscape, fifteen kilometers from Şanlıurfa, the city where local tradition places the birth of Abraham. Göbekli Tepe means “potbelly hill”, some hear “hill of the navel”. A name so harmless it borders on comedy, for what lies underneath.

At school in Røros, Norway, I was taught that Stonehenge marked the dawn of monumental civilization, five thousand years old, a marvel. Then you stand here, looking down into circular enclosures of carved T-shaped pillars, foxes and serpents and vultures worked into limestone, and the archaeologists tell you what the dating says: more than twice as old. Around 9600 BCE. Before writing, before pottery, before the wheel, before, in all probability, the first farm. One of them, a man from Istanbul I met on site, put it this way: at Göbekli Tepe, wherever you put the spade in the ground, you find something that shakes your world.

But the fact that has never let go of me is not the age. It is this: the place was not lost. It was interred. The enclosures were filled in with tons of rubble, bone, and flint packed around the pillars; whether by the hands of the builders themselves or by the slow shrug of the hillside is still argued over among the excavators. Either way, the effect was the same. Burial became preservation. The mound is nine hectares: twelve soccer fields of archive, entirely man-made, fifteen meters deep at its highest point. After three decades of excavation, at most a tenth of it has been opened. Ground-penetrating radar suggests sixteen more enclosures and nearly two hundred carved pillars still standing in the dark. For eleven and a half thousand years they have held their carvings, and then, within living memory, someone opened the ground and began to read.

The so-called totem pole of Göbekli Tepe: faces upon faces, carved nearly ten thousand years ago. Şanlıurfa Archaeology Museum.

An hour’s drive east lies a sister site, Karahan Tepe, younger in excavation, older, perhaps, in its promise. No tourist buses. Wind, stones, a chamber of pillars rising from the bedrock, serpents everywhere in the carvings, and a handful of archaeologists who look into the millennia the way one looks into an unfinished manuscript. My guide that day, Tahe, walked me through it, and at some point he began to tell me a story that was old when Babylon was young: of Oannes, the being that rose from the sea with the body of a fish and the head of a man, and taught human beings writing, agriculture, and law. He told it the way you pass something on, not the way you cite something.

I flew home from Türkiye with so many questions, and they have followed me ever since. While it is interesting to understand who built this, it might be even more interesting to understand who buried it, and for whom.

I started writing this piece after a long conversation with a good friend from India some weeks ago, where the stories are similar and the patterns are, to say it mildly, interesting. And this is what I wanted to share with you this morning.

Everyone Remembers the Water

Tahe’s story did not only follow me home; it led me to its relatives. One of them was the story my friend and I discussed. Somewhere around 800 BCE, in a ritual composition called the Shatapatha Brahmana, Indian priests fixed in verse a story that was already old. A man named Manu finds a small fish in his washing water. The fish asks for protection and promises, in return, to save him. Manu raises it; it outgrows every vessel; finally he releases it into the sea, and the fish, Vishnu himself, in his first descent, tells Manu the year the flood will come, and instructs him to build a ship.

Change the names of this story and all of a sudden you are in Mesopotamia, a thousand years earlier, where the god Ea whispers through a reed wall to Utnapishtim, in the very landscape where Oannes and his seven fish-cloaked sages, the Apkallu, were said to have brought the arts of civilization out of the sea. Change them again and you are in Greece, where Prometheus warns Deucalion; in the Andes, where Viracocha drowns the giants of Titicaca; in the Aztec codices, where the fourth sun dies in water and a couple rides out the deluge inside a hollowed log. Another fascination of mine over the years has been the Zoroastrian tellings out of Persia, and also here the same story arrives frozen: Yima is warned not of rain but of a killing winter, and builds an underground vault stocked with pairs of every living seed, an ark with the water taken out and the structure left standing.

The details vary the way dialects vary. The grammar does not. A warning arrives from outside the human world. One person listens. Something small is carried across the destruction: seed, pairs, pattern, name. And on the far side, the world is repopulated not from what was powerful in the old one, but from what was portable.

The temptation is to draw hard conclusions. So what I want to explore is where a coincidence this size stops being a coincidence and becomes a finding.

The Seduction

It is worth being honest about why Graham Hancock persuades millions, because the reasons are not stupid, and because I know the pull from the inside.

Two years ago, standing in that heat among the pillars, I wrote a question into my notebook that I would later find, almost word for word, in his books: What if these stories are not fantasy, but memory? Tahe had just told me about Oannes. I knew the flood was everywhere: Noah, Manu, Deucalion, Quetzalcoatl. I knew about the cataclysm hypothesis, the comet, the sudden cold snap around 10,800 BCE. And in front of me stood a monument that, by everything I had been taught, should not exist. For an afternoon, the lost civilization was the most natural idea in the world. I did not resist it because I was rigorous. I resisted it later, at a desk, and it took work.

That is the first thing to concede: the pattern is real. Hancock did not invent the universality of the flood story; comparative mythologists have been cataloguing it since the nineteenth century. When he stands in front of the pillars and says this should not exist, monumental architecture millennia before agriculture was supposed to allow it, he is pointing at a genuine anomaly, one that forced archaeology to rewrite its own textbooks. His demand for humility has a factual basis: the site was surveyed in the 1960s and dismissed as a medieval cemetery. It sat misidentified for three decades. Anyone who claims the archaeological record is complete has not read its history.

And he tells the story with wonder, at a moment when his opponents too often answer wonder with a raised eyebrow. The average viewer of Ancient Apocalypse is not choosing between Hancock and the peer-reviewed literature. They are choosing between someone who says the deep past is astonishing and concerns you and someone who says please stop asking. That is not a fair fight, and the institutions keep losing it.

He is right about the pattern. He is right that the past is deeper and stranger than the popular imagination allows. The problem lies elsewhere, in what he does with a pattern once he has found one. It took me some time to see it, because I had briefly done it myself.

What Graham Hancock Gets Wrong About the Flood

His method has a shape, and once you see it, you see it everywhere in his books: an anomaly is presented, the conventional explanation is declared insufficient, and the gap is filled with the hypothesis: a lost, advanced, seafaring civilization, destroyed around the end of the Ice Age, whose survivors sailed the world seeding agriculture, astronomy, and the memory of the deluge.

Notice what kind of claim this is. It predicts nothing that would count against it. Every absence of evidence is explained by the cataclysm itself, the proof drowned with the proof-bearers. Every anomaly, anywhere, of any kind, is compatible with it. A theory that can absorb all possible findings is not a strong theory; it is a story wearing the costume of one.

Even the cataclysm at its center illustrates the method. There is a live scientific controversy about whether a cosmic impact contributed to the sudden cold spell of the Younger Dryas, around 10,800 BCE, a genuine, unresolved debate among geologists about climate. In Hancock’s hands it becomes something else: a debate about civilizations, which it is not, settled in his favor, which it is not either. The same happens on the hill itself. In 2024, headlines announced that a pillar at Göbekli Tepe bore “the world’s oldest solar calendar”, 365 V-shaped marks, counted and interpreted by a researcher who is himself a proponent of the impact-and-lost-civilization thesis. The excavation team rejects the reading. Online, the headline long ago hardened into fact. This is how the edifice is built: contested claims poured, layer by layer, into the shape of certainty.

The deeper flaw is quieter, and it is the one that should bother us. Hidden inside the lost-civilization hypothesis is an assumption about who is capable of what. The hunter-gatherers of Anatolia could not have conceived those enclosures on their own; the peoples of the Andes and the Yucatán could not have raised their stones unaided; someone older, wiser, whiter in the tradition’s ugliest versions, must have taught them. Archaeologists did not sign an open letter against Ancient Apocalypse in 2022 because they fear new ideas. They signed it because the old idea underneath, indigenous achievement always requires a vanished tutor, has a long and documented history, and they recognize it on sight.

And here is the irony that undoes the whole edifice: the boring, consensus, peer-reviewed account of Göbekli Tepe is more radical than his. No lost civilization built it. Hunter-gatherers did: people without pottery, without writing, without agriculture, organizing labor at monumental scale around something sacred. The evidence now suggests the temple may have come before the farm; that it was religion, not surplus grain, that first gathered humans into permanence. Hancock offers a mystery solved by invention. Archaeology offers a fact that revises what we thought a human being was.

The boar of Building D, Göbekli Tepe, carved in limestone by people without pottery, writing, or agriculture. Şanlıurfa Archaeology Museum.

His failure is not credulity; it is a lack of imagination about what a story can actually be.

The Myth Is the Artifact

Read the flood stories again, not as testimony about an event, but as engineering. Consider what a myth must survive. Not water: time. Centuries of it, in most cases with no writing at all, passing mouth to mouth through famines, wars, migrations, the death of every individual carrier. Any message crossing that channel faces the same enemies an engineer would list today (noise, loss, corruption), and the myths carry countermeasures an engineer would recognize. Compression: an entire catastrophe reduced to one man, one warning, one vessel, small enough to hold in a single memory. Redundancy: the Vedic schools recited their texts forward, backward, and in interlocking word-pairs, so that any error announced itself: checksums, performed aloud for three thousand years. Priority encoding: notice what the ark never carries. No gold, no thrones, no armies. Seed, pairs, pattern, name. The stories agree with total unanimity about what deserves to cross the water, and it is never wealth. It is always information sufficient to restart the world.

The vocabulary is anachronistic, and I use it deliberately. No Vedic reciter thought in checksums; no Aboriginal elder spoke of transmission channels. This is a reading, not a claim about intent, the way we may say a bird’s wing solves an aerodynamic problem the bird has never formulated. What matters is that the solution works, whether or not its makers could name it.

And we now have proof that it works. In 2016, the geographer Patrick Nunn and the linguist Nicholas Reid published a study of twenty-one coastal traditions of Aboriginal Australian peoples describing lands now under the sea. Checked against the postglacial record, the stories preserve accurate accounts of coastlines drowned between seven and ten thousand years ago, geography transmitted orally, generation to generation, across roughly four hundred generations, arriving intact. No writing. No lost civilization. Just the technology of the sacred story, performing exactly as designed.

This is the reading Hancock’s method makes impossible. If the myths are corrupted memories of one event, they are evidence, interesting only for what lies behind them, dig sites in narrative form. But the flood story was not invented once and scattered. It was invented again and again, wherever rising water met a species that had learned to speak, because every culture that survives catastrophe discovers the same truth and encodes it the same way: what cannot be told cannot be saved.

The lost civilization Hancock is searching for was never lost. Its archive has been in continuous transmission for ten thousand years, hiding in the one place no one thought to look for an artifact: the telling itself.

The War That Never Needed Fighting

For three centuries the modern West has treated myth as a rival to be defeated. The Enlightenment’s founding gesture was subtraction: strip away the gods, the floods, the fish-men, and what remains is the truth. Myth was error with a good memory. On this account, the flood stories are what humanity believed before it learned to know, and the only respectable thing to do with them is to explain them away.

The project succeeded so completely that it created its own counter-movement. Max Weber named the price: disenchantment. Two philosophers of this city where I write, Horkheimer and Adorno, saw the deeper irony from their Californian exile in 1944: an enlightenment that merely destroys myth without understanding it does not escape myth; it becomes one. Their institute had fled Frankfurt ahead of the catastrophe and would return to it after: thought itself, carried across the water and brought home. A culture that answers every “what does it mean?” with “actually, here is what really happened” leaves an entire register of human need unaddressed. That vacancy does not stay empty. Something always moves in.

Graham Hancock moved in. This, more than Netflix budgets or Joe Rogan, explains his reach: he is often the only voice in the room granting that the deep past means something, that the flood stories are addressed to us. His answer is wrong, but he is answering a question the debunkers refuse to hear. Every fact-check that responds to wonder with a raised eyebrow recruits for him. You do not defeat a bad answer by suppressing the question.

There is a third posture, and it is neither belief nor debunking. It is reading. The flood myths are not failed history and not sacred revelation; they are compressed truth: catastrophe survived, encoded in the only storage medium available, transmitted with an accuracy we can now, in at least one documented case, measure across seven thousand years. To read them this way requires exactly the two capacities the old quarrel kept apart: the rigor to check the transmission, and the reverence to receive the message. Rigor without reverence produces the raised eyebrow. Reverence without rigor produces Atlantis.

We are now confronted with the need for a re-enlightenment. And with the advancements of technology, we can move beyond our own self-evident truths, from an information society drowning in its own data, through what we have referred to as a knowledge society, toward what we now need: a society of understanding.

The war between myth and enlightenment never needed fighting. The myths were never trying to be science. They were trying to be survivable, and at that, they are the most successful texts our species has produced. The possibilistic future lies in the unification of science and philosophy, mythology and enlightenment.

The Question I Brought Home

There is one more turn, and I offer it as a question, because that is all it currently is.

In recent years a framework from theoretical physics, the Quantum Memory Matrix, developed by Florian Neukart and colleagues, has proposed that spacetime itself may function as a storage medium: that every quantum interaction leaves an imprint in the cells of space, and that information, once written into the world, is never destroyed but retained in its fabric. Parts of the formalism have been tested on quantum hardware, imprinted states written into qubits and retrieved with high fidelity. A proof of principle, no more; the framework remains a hypothesis among hypotheses. But it is a hypothesis with a startling shape: it describes a universe whose most basic operation is remembering.

I want to be precise about what this does and does not license. The physics, if it holds, earns one sentence: the universe retains a record. It does not earn a second sentence, that the myths knew this, that Manu’s fish swam up from the Planck scale, that Göbekli Tepe encodes cosmology. The gap between those two sentences is real, and I intend to leave it open. Hancock’s entire method consists of closing such gaps with confidence; the honest move is to stand in them and look both ways.

But standing in the gap, one is allowed to notice a symmetry. For ten thousand years, wherever catastrophe met consciousness, human beings responded by doing one thing: compressing what mattered into a form that could survive its carriers, and writing it into the most durable medium they had: the sacred story, the buried sanctuary, the recited verse. Imprint, preserve, retrieve. If the physicists are right, this is also what reality does, not once per apocalypse, but at every point of spacetime, at every instant, automatically and without loss.

Which leaves the question I have not been able to put down since that hill in Anatolia: when we began to tell stories against forgetting, were we inventing something, or noticing, for the first time, what the world had been doing all along?

Whatever happened at Göbekli Tepe, it took eleven and a half thousand years to find out it was being saved for someone; it was us.

Looking out over the Harran plain from the hill, August 2024.

Some referenced sources in this piece and parts of a forthcoming book on the topic:

Shatapatha Brahmana 1.8.1 (Manu) – Andrew George, The Epic of Gilgamesh, Tablet XI – Verbrugghe & Wickersham, Berossos and Manetho (Oannes/Apkallu) – Avesta, Vendidad Fargard 2 (Yima) – Nunn & Reid, Australian Geographer 47(1), 2016 – Klaus Schmidt, Göbekli Tepe: A Stone Age Sanctuary; German Archaeological Institute, Tepe Telegrams (site data: ~9 ha, ≤10% excavated, ≥20 enclosures) – Society for American Archaeology, open letter on Ancient Apocalypse, Nov 2022 – Horkheimer & Adorno, Dialectic of Enlightenment – Neukart, “The Quantum Memory Matrix,” Entropy 26, 2024.