In 1906, a German-American expedition working through the ruins of ancient Nippur — a religious center of Mesopotamia, south of modern Baghdad — unearthed a small clay prism. It was cracked, incomplete, about the size of a hand. On it, in neat cuneiform script, someone had inscribed a list of kings stretching back tens of thousands of years before the Great Flood. Not hundreds. Not thousands. Tens of thousands. The earliest rulers on this list reigned for periods of 28,800 years. 36,000 years. 43,200 years. The total time covered before the Flood: 241,200 years.

This was the Sumerian King List, and more than a century later, nobody has fully explained it.

The Document Itself

The most complete surviving copy is the Weld-Blundell Prism (WB 444), now housed at the Ashmolean Museum in Oxford. It dates to roughly 1827 BCE, during the reign of King Sîn-māgir of Isin, and it is a four-sided clay prism inscribed on all faces, meant to be rotated on a spindle and read column by column. Several other partial copies exist — the Nippur fragment mentioned above, tablets from Larsa and Ur, additional prisms and clay tablets scattered across collections in Philadelphia, Berlin, and Istanbul. The scholar Thorkild Jacobsen published the definitive critical edition in 1939, collating over a dozen sources into a composite text. His work remains the standard reference.

The structure is repetitive and deliberate. The list follows a strict formula: a city is named as the seat of kingship, a sequence of rulers is given with their reign lengths, and then the phrase “City X was defeated; its kingship was carried to City Y.” This pattern continues from before the Flood through historical dynasties that we can independently verify — the kings of Ur, Kish, Uruk, and others who left their own inscriptions, who appear in other records, whose buildings archaeologists have dug up.

Here is where it gets strange: the text draws no distinction between its mythological and historical sections. It does not say “and now we transition from legend to fact.” The antediluvian kings and the historical kings are presented in the same grammatical structure, the same formulaic language, the same scribal hand. Whatever the Sumerian scribes believed they were doing, they were not writing two different kinds of document.

The Numbers

Let’s look at the antediluvian section directly. The King List names eight kings (in some versions, ten) who ruled in five cities before the Flood. Here are the first few entries, using Jacobsen’s composite:

  • Alulim ruled in Eridu for 28,800 years.
  • Alalgar ruled in Eridu for 36,000 years.
  • En-men-lu-ana ruled in Bad-tibira for 43,200 years.
  • En-men-gal-ana ruled in Bad-tibira for 28,800 years.
  • Dumuzid, the shepherd, ruled in Bad-tibira for 36,000 years.

And so on. Then: “The Flood swept over. After the Flood had swept over, and kingship had descended from heaven, kingship was in Kish.”

From Kish forward, the reign lengths begin to shrink — but slowly. The first dynasty of Kish includes rulers credited with reigns of 1,200 years, 960 years, 900 years. These numbers gradually descend toward plausible human lifespans as the list approaches the historically verifiable period, roughly the mid-third millennium BCE.

The standard scholarly explanation, articulated by Jacobsen and others, is that these numbers are symbolic or schematic — perhaps reflecting a base-60 (sexagesimal) mathematical system, perhaps serving a theological purpose about the declining vitality of humanity after the Flood. This is reasonable. The Sumerians did use base-60 mathematics (we still have 60 seconds in a minute because of them). And the numbers do cluster around values that are neat multiples of 3,600 — a sar, or “great circle,” in Sumerian counting.

But “symbolic” is not the same as “meaningless,” and the explanation has gaps that are rarely discussed honestly.

Why “It’s Just Symbolic” Doesn’t Fully Satisfy

The first problem is specificity. If the numbers are purely schematic — round figures plugged in to convey “a really long time” — why do the reign lengths vary from king to king? Why does Alulim get 28,800 years while En-men-lu-ana gets 43,200? If you are inventing numbers to fill a theological template, uniformity is simpler. The variation implies that the scribes believed there was a meaningful difference between these reigns, that one king ruled longer than another and this fact mattered.

The second problem is consistency across sources. The King List exists in multiple copies from different cities and different centuries, and while the antediluvian numbers vary somewhat between versions, they are not wildly different. A purely symbolic system should be more fluid — if the numbers don’t mean anything specific, why would scribes in Larsa bother to roughly preserve the values used in Nippur? The degree of conservation suggests the numbers were being transmitted as data, not as decoration.

The third problem is the transition. The reign lengths do not jump from 36,000 years to 30 years overnight. They decline in stages — from five figures to four, four to three, three to two — across many dynasties and centuries of the text. This gradual curve has a structure to it that is hard to explain as pure literary convention. The scholar Jean-Jacques Glassner, in his Mesopotamian Chronicles (2004), noted that this tapering follows something closer to a mathematical function than a narrative choice. Whether that function reflects an actual belief about human longevity or a sophisticated literary device, it has more internal logic than the dismissive “just symbolic” reading grants it.

And there is a fourth, often overlooked point: the Sumerians themselves appear to have taken these numbers seriously. Other Mesopotamian texts — the Dynastic Chronicle, portions of the Epic of Gilgamesh, Berossus’s third-century-BCE Babyloniaca (which transmits the Mesopotamian king tradition into Greek) — all reference these extreme lifespans without irony or qualification. When Berossus, a Babylonian priest writing for a Greek audience, presented the antediluvian kings with their enormous reigns, he was not telling a fairy tale. He was presenting what his tradition considered history. We may disagree with him, but we should at least register that the people who created and maintained this list over centuries treated it as a record, not a poem.

Competing Theories

So if the numbers are not purely symbolic, what are the alternatives? Several have been proposed, none fully convincing.

Misunderstood units of time. Perhaps the “years” are actually months, or seasons, or some other unit. Dividing by 12 brings some of the later antediluvian figures closer to plausible lifespans, but it makes the earlier ones (28,800 ÷ 12 = 2,400 years) still absurd by biological standards. The reduction also breaks the post-Flood sections, where the numbers already approximate real lifespans and dividing them produces reigns of a few months. This theory creates as many problems as it solves.

Dynastic rather than individual reigns. Maybe “Alulim” is not a single person but a dynasty, a lineage of kings using the same throne name over thousands of years. This is plausible in principle — some Egyptian dynasty names span centuries — but the King List explicitly says “one king” for each entry, and the names change with each listing. The text does not read as describing dynasties. It reads as describing individuals.

Astronomical or calendrical encoding. Some researchers, including the independent scholar Jöran Friberg, have argued that the antediluvian numbers encode astronomical data — orbital periods, precession cycles, or relationships between celestial bodies. The numbers do show tantalizing mathematical relationships. For example, 28,800, 36,000, and 43,200 are all multiples of 3,600, and their ratios (4:5:6) are musically harmonious. Whether this reflects deliberate astronomical encoding or simply the properties of base-60 arithmetic is unresolved. The hypothesis is clever but unproven.

Literal record of non-human rulers. This is the ancient astronaut reading, popularized by Zecharia Sitchin and others, which takes the text at face value: beings with enormous lifespans actually ruled Sumerian cities before the Flood. Mainstream scholarship rejects this completely, and for good reason — there is no physical evidence of non-human rulers, and Sitchin’s translations have been thoroughly criticized by actual Assyriologists (most notably Michael Heiser). But it should be acknowledged that this reading is, at minimum, what the text itself claims to describe. The question is whether the text is reliable, not whether it says what it says.

The Flood as Historical Pivot

One element of the King List that has gained unexpected support from archaeology is the Flood itself. The text treats the Flood as a real event — a historical boundary between two eras of kingship. And excavations at Ur, Kish, Shuruppak, and other Mesopotamian sites have revealed flood deposits: layers of water-laid sediment interrupting urban strata, datable to the late fourth or early third millennium BCE.

These were not global floods. They were massive, catastrophic river floods — the kind that would have seemed world-ending to a civilization built entirely along the Tigris and Euphrates. Sir Leonard Woolley, excavating Ur in the 1920s, found an eleven-foot deposit of clean clay separating earlier and later occupation layers. He announced he had found evidence of Noah’s Flood. He was probably wrong about the universality, but he may not have been wrong about the cultural memory. The Sumerians experienced devastating floods, and they built their historical framework around at least one of them.

This matters for interpreting the King List because it means the text’s historical framework is not entirely fictional. It contains a real event — a catastrophic flood — embedded within an otherwise extraordinary narrative. The question, then, is how much else in the framework might reflect distorted memory rather than pure invention.

What the Scribes Were Doing

The most sophisticated recent scholarship treats the King List as a political document with theological underpinnings. The idea of kingship “descending from heaven” and passing from city to city served a specific ideological purpose: it legitimized whatever dynasty currently held power by placing it in an unbroken chain reaching back to the gods. The scholar Piotr Steinkeller of Harvard has argued that the List was compiled during the Third Dynasty of Ur (circa 2112–2004 BCE) precisely to justify Ur’s dominance by constructing a linear succession of kingship that in reality was nothing of the sort — multiple dynasties ruled simultaneously, not sequentially, and the List’s one-at-a-time structure is a deliberate fiction.

This is almost certainly correct as a political explanation. But it does not explain the numbers. If you are fabricating a legitimizing genealogy, you do not need your founding kings to reign for 43,200 years. You need them to be prestigious, divine, ancient — but the specific lengths are overkill for propaganda. They suggest the scribes were working with received tradition, numbers they inherited rather than invented, and felt obligated to preserve even as they shaped the list’s structure for political ends.

This is the crux of the problem: the King List is clearly a politically motivated compilation, but it incorporates material that does not serve its political purpose and may predate it by centuries or millennia. The antediluvian section feels bolted on — ancient, inherited, strange — and the compilers included it because it was part of the tradition, not because it helped their argument.

The Uncomfortable Space

We are left, then, in an uncomfortable space. The numbers are too structured to be random, too consistent across sources to be casual, too varied to be purely formulaic, and too enormous to be literal. No single explanation accounts for all the features of the text. The symbolic reading explains the most but explains it incompletely. The alternative readings each illuminate one aspect while failing on others.

What the Sumerian King List ultimately reveals is the limit of our categories. We want ancient texts to be either history or myth, either factual record or literary creation. The King List refuses this division. It presents its mythological and historical content as a single continuous narrative, and the transition between them is seamless — which either means the scribes did not recognize the distinction, or they recognized it and rejected it.

What kind of knowledge were the Sumerians preserving in those impossible numbers? What did 28,800 years mean to a scribe in Isin who carefully inscribed it on wet clay, knowing he himself might live to fifty? Was he recording what he believed to be true? Encoding something we no longer have the key to decode? Or transmitting, faithfully and without understanding, something that was already ancient and opaque when it reached his hands — a signal from a past so deep that even the Sumerians, the people we call the beginning of civilization, considered it unimaginably old?