Four Of Six Control Quarries Show The Same Fissures. One Is 26th Dynasty.
The enclosure's rounded vertical fissures are real and water cut them. So are the fissures on a quarry face inscribed in 591 BC. Meanwhile the west wall is losing 1.4 millimetres a year to a rising water table.
NAZLET ES-SAMMAN, Egypt — Quarry face 4 lies eleven kilometres south of the Sphinx, cut into the same limestone bed, and it carries a hieratic dedication dated to the fourth regnal year of Psamtik II. That is 591 BC.
Two geomorphologists who were not told what they were looking at scored its weathering profile at 8.0 on a ten-point scale.
They scored the west wall of the Sphinx enclosure at 8.4.
The gap is smaller than the two scorers' average disagreement with each other.
What was never in dispute
The enclosure walls are rounded at the top, deeply fissured vertically, and undercut at the base in a way that wind abrasion does not produce. This newspaper accepts that as measured fact and has never printed otherwise.
Water made those profiles. Nobody serious argues the point.
The argument has only ever been about which water, and when.
The control nobody had run
The profile has been read as a clock since the early 1990s: this much rounding needs this much rainfall, therefore this many years. A clock needs calibrating against faces of known age in the same rock, and in thirty-five years of argument nobody had surveyed one.
So we commissioned surveys at six quarry faces within eleven kilometres, all cut into the same member, three of them securely dated — one by inscription, two by mason's marks.
One operator, one scanner, 0.1mm resolution. The cross-sections were printed at identical scale, stripped of site names, shuffled, and posted to two geomorphologists who were not told that the Sphinx was in the set.
Four of the six
Four controls scored above 6.8. Two scored below 2.5 — both faces standing more than three metres above the local water table, both draining freely.
And at quarry 4 the fissures do not stop at the inscription. They run through it, truncating three sign-forms mid-stroke.
Whatever cut those fissures was working on that face after the text was cut. After 591 BC.
What the profile does track
Not age. Against any estimate of how long a face has been exposed, the blind scores correlate at r = 0.11, which is nothing.
Against chloride in the surface twenty millimetres, r = 0.88. Against height above the local water table, r = −0.79.
That is salt weathering, and it is not exotic. Groundwater wicks up through the pore space, evaporates at the face, and leaves halite crystallising in the voids. The crystals grow along bedding planes and joints, which is why the damage comes out vertical, rounded and undercut. It is the commonest destructive process in dry-country masonry on earth.
One point four millimetres a year
We registered three surveys of the west wall to the same eleven control points: stereo plates shot in 1978, a terrestrial laser scan from 2011, and a scan we flew ourselves in March 2025.
Between 1978 and 2011 the wall lost a mean 0.6mm of surface a year.
Between 2011 and 2025 it lost 1.4mm a year.
Over the same period Nazlet es-Samman roughly trebled in built area without a mains sewer. Local monitoring wells put the water table about four metres higher than in the early 1980s. The water arriving at the base of the Sphinx enclosure is, in large part, septic seepage.
The scheme that was funded, and the wall that came out of it
A dewatering scheme for the plateau was funded in 2008 at $8.4m: well points, a collector main, and a boundary polygon drawn around the enclosure and the causeway.
In 2011 the package was value-engineered. The revised polygon saved $1.9m. The west wall fell outside it.
Ing. Hoda Mansour-Delacroix, who was the project engineer, kept the minute and gave us a copy.
"The west wall was the cheapest thing to remove, because there is no visitor route on it and nobody counts it in the footfall model," she said. "That is the whole reason. It is not a secret. I have said it in three meetings and it has been minuted twice."
What this desk takes from it
The profile is not a clock. It is closer to a hygrometer, and what it is reading is now.
Professor Ellingsen-Attia's reply runs below, unedited, and he is right that we overreached: what we have shown is not that the clock is wrong but that it was never calibrated, which is a duller and more durable finding.
Here is the one this desk cannot put down. The oldest monumental sculpture on the plateau is dissolving at one and a half millimetres a year because a village grew faster than its drains, the fix was costed, funded and then trimmed, and the argument about whether the thing is four thousand years old or twelve thousand has absorbed most of the attention anyone has paid it since 1991.
The west wall was not in the polygon. Nobody hid that. It is in an appendix, and it has been public for fifteen years.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We did not argue about the profiles. We had seven faces measured by one operator with one instrument, printed the cross-sections at identical scale with no site names and in random order, and sent them to two geomorphologists who were told nothing except how to score. Both sheets are published entire, including the face on which they differ by 1.3 points and the note in which the second scorer identifies the wrong profile as the Sphinx.
Who we spoke to
- Dr. Marta Szilágyi-Bouderbala, Geomorphologist, Ljubljana. Scored seven anonymised profile sets; her sheet is published entire March–May 2026 Was not told the Sphinx was in the set and says she did not guess which face it was.
- Dr. Théo Nwachukwu-Palmar, Geomorphologist, second scorer, no contact with the first. Scored the same seven sets from the same printed cross-sections April–May 2026 Guessed which face was the Sphinx. Guessed wrong. Says so in his own sheet.
- Ing. Hoda Mansour-Delacroix, Project engineer, Giza dewatering scheme, 2008–2011. Interviewed in person twice, and by telephone once after seeing the 2025 figures June 2026 Supplied the revised boundary polygon and the value-engineering minute from her own file.
- Prof. Ruaridh Ellingsen-Attia, Geologist; has argued since 1991 that the enclosure profile requires prolonged running water. Sent the full draft, the scoring sheets and the point cloud; 24 days to reply July 2026 Reply printed in full and unedited. It is the strongest thing on this page.
- 1978 stereo plates, 2011 terrestrial scan, 2025 aerial scan, Three surveys registered to the same eleven control points. Plates borrowed from the surveying school that shot them; 2025 scan flown and paid for by this newspaper 1978, 2011, 2025
What we could not confirm
- Whether quarry 4's face was opened in the 26th Dynasty or is an older face inscribed later. Our epigrapher reads the text as a quarry-opening dedication; two others read it as commemorative. The fissures truncate three sign-forms either way, which dates the fissuring but not the face.
- Where the chloride comes from. We measured salt; we did not trace it. Septic seepage from the village is the likeliest source and likeliest is not a source. We were permitted twelve samples of loose spall and wanted forty from drilled powder.
- That the 2011 descope caused the acceleration. The descope and the change in loss rate both fall in the same year and we cannot separate either from a water-table rise that began in the 1980s.
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 TelegraphTHE SPHINX WEPT FOR TEN THOUSAND YEARS
Ran the enclosure profile as settled proof of a pre-dynastic date, sourced to one conference paper from 1992 and to nobody who has been to the site since.
Had that we did not
The profiles, which are real, and the observation that wind does not make them, which is correct.
Left out
That no control face had ever been surveyed. It does not occur to the piece that one could be.
-
Signal & LedgerGiza Dewatering Package Retendered At Reduced Scope
Covered the 2011 descope as a procurement story, accurately, and five years before anybody else had the figures.
Had that we did not
The revised contract value, the value-engineering rationale, and the revised boundary polygon as an appendix.
Left out
Which wall came out of the polygon. The word Sphinx does not appear anywhere in the piece.
-
They Buriedthis newspaperFour Of Six Control Quarries Show The Same Fissures. One Is 26th Dynasty.
Ran the control nobody had run, blind, and then measured what the wall is actually losing.
Had that we did not
Both scorers' full sheets, the seven anonymised cross-sections, the 2025 point cloud, and a reply that says our headline overreaches.
Left out
We commissioned this survey to settle a dating argument and put the erosion rate in the fifth section. A wall losing 1.4mm a year was the news. We made it the aftermath. — V. Ashcombe-Doyle, standards editor
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.