Checked · The Architect Question

Is Graham Hancock right? A checkable answer, claim by claim

The claim

An advanced civilisation existed before the end of the last ice age, was destroyed around 12,900 years ago, and its survivors passed knowledge to the cultures that came after.

Our verdict

Speculative

The right answer is not one answer. Break the position into its separate claims and they score differently: one is a live scientific controversy, one is solidly established, several are unsupported, and one is a genuinely good argument that mainstream archaeology has answered badly. Treating it as a single yes or no is what makes the conversation useless.

This page exists because the usual answer to this question is a tone rather than an argument, and the tone loses.

Graham Hancock has been writing on this since Fingerprints of the Gods in 1995, and the audience is enormous. That audience is not stupid. It is, in large part, made up of people who noticed that professional archaeology sometimes states things more confidently than its evidence supports, and who then went looking for someone who would say so.

They were right about the first part. We spend a whole book demonstrating it, and this site publishes our own recall figures because most surveys do not publish theirs. Where they went next is a different question, and it is worth answering carefully rather than loudly.

The scorecard

The position is not one claim. Split it, and the parts do not score alike.

ClaimWhere it stands
A cosmic impact contributed to the Younger Dryas*Under review.* A live, peer-reviewed controversy, argued across four papers in Earth-Science Reviews between 2021 and 2024. Not fringe, and not established.
Sea level rose about 134 metres and drowned habitable coastline*Published.* Lambeck and colleagues, 2014, from roughly a thousand observations worldwide. Larger than most retellings say.
Archaeology has been wrong about how early complex societies appear*Published, and it keeps happening.* Göbekli Tepe from about 9600 BCE, and Upano dated by excavation to around 500 BCE, both moved the model.
Remote sensing keeps finding more than expected under forest*Published.* The Maya lidar survey and the Upano survey both did, and both are peer-reviewed in Science.
Mainstream archaeology suppresses inconvenient findings*Refuted as stated*, and it has a real version underneath it. That has its own page.
An advanced pre-ice-age civilisation existed*Speculative.* No settlement evidence, no material culture, no burials, no subsistence remains. The absence is not proof, and it is also not nothing.
Survivors of it seeded later cultures*Speculative*, and dependent on the previous row, which is not established.
The impact and the lost civilisation are one claim*Refuted.* Two independent propositions, welded by the retelling and by some critics. This is the most useful distinction on the site.

Why the mixed scorecard is the honest answer

A reader who has been told the whole thing is nonsense, and who then discovers that Göbekli Tepe really is that old, or that sea level really did rise that far, has been given a reason to distrust everything else they were told.

Published

The model has been wrong about the timing of monumental construction, and it was corrected by excavation and radiocarbon, in public, by the discipline itself. Anyone claiming archaeology never revises is contradicted by Göbekli Tepe. Anyone claiming archaeology is reliably right about origins is contradicted by the same site.

Göbekli Tepe: earliest monumental phase about 9600 BCE, Pre-Pottery Neolithic A, dated by accelerator mass spectrometry, built by hunter-gatherer groups before the full development of agriculture. Upano: Rostain and colleagues, Science, 11 January 2024, dated by excavation.

Both of those sentences are true at once, and a page that only prints one of them is doing advocacy.

Where the case actually fails

It fails at the last two rows, and it fails for a specific and boring reason rather than a rhetorical one.

A civilisation is not an event. It is quarries, mines, spoil, middens, trade goods moving between places, agriculture, burials, and the accumulated rubbish of people doing the same things for centuries. That record is dense, continuous and mostly not underwater. The submerged coastline explains a gap in one zone; it does not explain the absence of inland traces of a society that would have had inland traces.

That is the argument. It is not "no serious scholar believes this" and it is not a comment on anybody's motives. It is a statement about what kind of evidence the claim predicts and where that evidence would be.

What would change our mind

  • A stratified site with material culture, subsistence remains and a secure date before 12,900 years ago, showing complexity beyond what is currently attested for that period.
  • A worked artefact of a technology that has no local developmental sequence, dated securely, in a context nobody can explain by later intrusion.
  • Agricultural or metallurgical residues at a pre-Younger-Dryas horizon, replicated by a laboratory with no stake in the argument.

None of that requires the impact hypothesis to be true, which is the point of keeping the two propositions apart.

We rank in this cluster deliberately, and we know what that looks like. A book whose thesis is "take this seriously and check it" that then declines to appear where those readers are is not being careful. It is being squeamish.

Where this leaves things

The five labels, applied to this question. Nothing is left unlabelled.

Published

Sea level rose about 134 metres. Göbekli Tepe is about 11,600 years old. Remote sensing keeps finding more than expected. All three, without qualification.

Under review

The Younger Dryas impact hypothesis. Genuinely open, in a mainstream journal, in 2024.

Candidate

Underwater prehistoric archaeology on drowned coastlines. Worth funding, and it may find things.

Refuted

The suppression claim as usually stated, and the welding of the impact to the civilisation.

Speculative

The advanced pre-ice-age civilisation itself, and everything downstream of it.

Sources

Related

Where this is argued in full

Answering the Architect Question

Both cases at the same length, measured and printed at the back. The argument runs across four papers in one journal and you can read all of it.

Both cases at the same length, measured and printed at the back. It is written and not on sale, and advance readers can read it first.