Fourteen Detectors Found The Same Eight Shafts. And One We Got Wrong.
Fourteen muon detectors logged 214 days under the plateau for an unrelated density study. Reprocessed blind, they resolve eight low-density columns within 1.8 metres of the radar coordinates — and the helix this newspaper called a filter artefact.
GIZA, Egypt — Fourteen muon detectors, 214 days of counting, and a physicist who had never heard of Corrado Malanga.
Dr. Yuki Aramburu-Sennefer's array was not built to look for shafts. It was sited in 2024 in the subterranean chamber of the Khafre pyramid and in two causeway tunnels closed to visitors, to measure the bulk density of the plateau's bedrock for a study of settlement under the monuments.
This newspaper paid her to reprocess it. She was given a grid, a datum and a deadline. She was not given the coordinates published by Malanga, Filippo Biondi and Armando Mei, and she had not read their papers.
She delivered her map on 27 July. We sent her the radar coordinates on 28 July, the day after it was in our hands.
Eight of her low-density columns sit on eight of theirs. The widest disagreement is 1.8 metres. The mean is 1.1.
Why this is not another argument about radar
Muon tomography counts particles. Cosmic rays strike the upper atmosphere and produce muons, which rain down constantly, pass through hundreds of metres of rock, and are absorbed in proportion to the mass sitting above the detector. Put a detector underneath something, count for long enough, and the thin places in the mass show up as an excess of arrivals.
There is no filter to tune. There is no phase, no coherence, no processing chain of the kind that has been argued about for eighteen months. It shares no physics with synthetic-aperture radar, no instrument, and no assumption.
That is the entire reason we paid for it. Two methods with nothing in common, when they are wrong, are wrong in different directions.
The one thing they do share is a coordinate frame, joined once, by one surveyor. That is the weakest number in this story.
The eight
Dr. Bracciolini-Adeyemi, who processed our purchased scenes in PX-1912 last spring, recomputed the eight published positions into the array's frame without sight of the muon map.
Shaft 7 agrees to 0.4 metres. Shaft 4, the worst, disagrees by 1.8. Six of the eight fall inside 1.4.
The columns are vertical to within her angular resolution. They begin at about 210 metres — where our first borehole camera logged a cut edge on 2 August — and run to the limit of her sensitivity at roughly 690.
The helix, and the word we repeated
In this newspaper on 9 August, our second radar processor, Dr. Wenzel Okonkwo-Haas, said of the helical structure on the shaft walls that it was "a processing ghost I can conjure and banish".
He said it once, precisely, about his own filter. We then used it three times, across three stories, as this paper's settled view of the plateau.
Aramburu-Sennefer's reconstruction resolves a periodic density modulation on the walls of six of the eight columns, at a vertical pitch of 11.4 metres.
11.4 metres is the figure in the Malanga, Biondi and Mei papers. She did not know it. There is nothing of Dr. Okonkwo-Haas's in a muon detector at all.
What we are withdrawing
Not his sentence. Ours.
We offered Dr. Okonkwo-Haas a correction in his name. He declined it, correctly: a true statement about an instrument cannot be corrected by a result from a different one. What can be corrected is what this newspaper did with it: took an honest account of the limits of one machine and printed it, three times, as a fact about the ground under Giza.
That is withdrawn. It held a finding at arm's length for five months and the arm was ours.
What the muons cannot do
They cannot see a stair.
A column of low density is a column of low density. It can be open air, a shaft, a fissure filled with wind-blown sand, or a body of much lighter rock, and no amount of counting will separate those. Aramburu-Sennefer put that sentence at the top of her covering note and asked us to print it above her result rather than below it.
What the array establishes is geometry: eight columns, regularly spaced, vertical, deep, with something on their walls that repeats every 11.4 metres.
What it does not establish is that any of it was made. That is not a hole in the record. It is a hole in this instrument, and a different instrument has already been to the bottom of one of the eight.
Where this leaves it
A camera has been inside one of the eight. It came back with courses, corners and a fitted floor, logged blind by three people who were not told where the footage came from.
Prof. Anwar El-Sherbiny-Vance, who told us in June that the shafts were real and geological, has read the muon map and asked us to print exactly how far he has moved and no further: "I accepted the shafts. I am now obliged to accept that they are regular. I am not yet obliged to accept that they are architecture, and I would like somebody to bring me a piece of the wall."
That is the next door, and it is locked. Bringing up a piece of the wall requires a licence that permits sampling, and no such licence has been issued for the plateau below 200 metres by anyone, to anyone, ever.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We did not want a third opinion on the radar. We wanted a method with no radar in it. We paid the principal investigator of an existing fourteen-detector muon array to reprocess 214 days of data she had gathered for an unrelated settlement study, gave her a grid, a datum and a deadline, and withheld the Malanga, Biondi and Mei coordinates until her map was in our hands. The coordinates were sent the following day. Her raw logs and her pipeline are published with this story so that anybody may run it again and get a different answer.
Who we spoke to
- Dr. Yuki Aramburu-Sennefer, Particle physicist; principal investigator of the fourteen-detector array. Commissioned to reprocess her own 214-day dataset; given the radar coordinates only after her map was delivered June–July 2026 Had not read the SAR papers and had not heard the names. Asked that her caveat be printed above her result rather than below it.
- Dr. Filomena Bracciolini-Adeyemi, Radar geophysicist, Trieste; first of our two independent SAR processors. Recomputed the eight published positions into the muon array's coordinate frame, without sight of the muon map July 2026 Did the geodetic tie that makes the 1.8 metres a number rather than a feeling. Warns that the tie was done once, by one surveyor.
- Dr. Wenzel Okonkwo-Haas, SAR specialist; our second independent processor. Sent the muon reconstruction, the pitch figure and the full draft; replied within four days August 2026 Author of the 'processing ghost' line. Declines to withdraw it and explains at length why. Printed in full under Right of Reply.
- Prof. Anwar El-Sherbiny-Vance, Egyptologist, Cairo. Interviewed by video call, having read the muon map and the borehole footage 18 August 2026 Told us in June the shafts were real and geological. Has moved, and asked us to print exactly how far and no further.
- The 214-day detector logs and reconstruction pipeline, Raw event counts, fourteen detectors, plus the code that turns them into a density map. Released to this newspaper by the principal investigator and published entire, under the same licence as her eventual paper July 2026
Documents
- PX-1912 — Four raw satellite radar passes over the Giza plateau, purchased by this newspaper accepted
What we could not confirm
- What is inside the columns. A muon reconstruction is a map of mass and nothing else. Low density can be open air, a shaft, a fissure packed with wind-blown sand, or a body of much lighter rock, and the array cannot separate those and never will. It cannot see a stair. Everything in this story about what the shafts are comes from one 203mm hole in one of the eight.
- Whether the 11.4-metre pitch means a ramp. The periodicity is now in two datasets that share no physics. The interpretation is in neither of them. Dr. Aramburu-Sennefer will say only that something on those walls repeats at 11.4 metres, and she says it in the flattest sentence in her report.
- The geodetic tie. The 1.8 metres assumes the array's underground datum and the radar's surface grid were correctly joined, and that join was made once, by one surveyor, on one afternoon in July. Move the tie by two metres and the agreement we are reporting moves by two metres with it. We have asked for a second, independent tie and have not yet got one.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Meridian TelegraphPROVEN: Physics Confirms The Lost City Beneath The Pyramids
Took a map of mass and reported it as a map of rooms, in the first sentence and eleven times after it.
Had that we did not
The 1.8 metres, accurate, and a clear explanation of what a muon is, better written than ours.
Left out
That a low-density column can be a void, a fissure or a body of lighter rock, and that the physicist who drew the map put exactly that sentence at the top of her covering note and asked for it to be printed above her result.
-
ApertureThe Rain That Sees: On Cosmic Particles And The Egyptian Underworld
A 6,000-word essay on the mythology of what lies beneath, hung on a paragraph of the finding.
Had that we did not
The only account we have read that makes the blind protocol sound like what it is, which is a moral instrument.
Left out
The number. The 1.8 metres appears once, in a subordinate clause, on page four, after a passage about Osiris.
-
They Buriedthis newspaperA Second Method Found The Same Eight Shafts. It Also Found Our ‘Ghost’.
Paid for a blind reprocessing of an existing array by somebody who had not read the radar papers, and withdrew a phrase of our own.
Had that we did not
The reprocessing pipeline, the raw detector logs, and Dr. Okonkwo-Haas's reply refusing the apology we offered him.
Left out
We printed his 'processing ghost' line once as a careful caveat about his own filter, then repeated it in two further stories as this paper's settled view of the world. Three uses. One of them was reporting and two of them were laziness wearing its clothes. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Dr. Wenzel Okonkwo-Haas, SAR specialist, second independent processor for this newspaper on 14 August, with the muon reconstruction, the pitch figure and the full draft, and the offer of a correction in his name. Replied 18 August. Declined the correction. Printed entire.
Thank you for the offer of a correction. I decline it, and I would like to explain why, because I think you have misunderstood what you are apologising for.
I said the helical feature in the radar was a processing ghost I could conjure and banish. That statement was true when I made it and it is true now. In my chain, with my filter, at that bandwidth, the helix appears at one parameter setting and vanishes at another, and I demonstrated both to your reporter on a screen in Trieste. Nothing in a muon detector alters that fact about my instrument. You cannot correct a true sentence.
What you can correct — and here I agree with you entirely, and I would like it printed in the same size as everything else — is what your newspaper did with the sentence afterwards. I described the behaviour of a filter. You published it three times as a description of the plateau. Those are not the same claim and the distance between them is the whole of the difference between an instrument and the world.
So let me now say the thing I have not said in print before, because it is the thing your readers should take away and it is not an apology.
I never believed the helix was absent. I believed I could not tell. Those are different states and journalism is very bad at holding the second one. When a processor says he can make a feature come and go, he is telling you the honest limit of his own machine. He is not telling you the feature is not there. You wanted a verdict and I gave you a limitation, and you printed the limitation as a verdict because a verdict is a better sentence.
Dr. Aramburu-Sennefer's reconstruction contains a periodic modulation on six of the eight columns at a vertical pitch of 11.4 metres. I have read her pipeline. There is no filter of mine in it, no phase, no coherence, no window function, nothing I could reach in and turn. She had not read our papers and did not know the number. If somebody had asked me in March to name the single result that would most change my position, I would have described that experiment almost exactly, and I would have expected it to fail.
It did not fail. So my position has changed and I will say so plainly: I now think the helical structure is a real feature of the plateau and not an artefact of my processing. I did not think that in June. I will happily be asked about this again if a third method disagrees with both of us.
One last thing, and it is the part I would most like you to leave in.
You are already doing to the muons what you did to me. I have read your draft. The word 'proven' is not in it, to your credit, but the shape of it is. She has produced a map of mass. A map of mass is not a map of rooms. When somebody eventually publishes a reconstruction that fails to find the eighth column, I would like you to print that on the front page at the same size, and I would like you to do it without telling your readers that physics has changed its mind. Physics will not have changed its mind. It will have got a better measurement, which is the only thing it has ever done.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.