Great Sphinx · Egypt · 29.97527, 31.13768 · PYR record in the living atlas
The Sphinx was not built. It was subtracted. A ridge of Mokattam Formation limestone, 73 metres from paw to tail and 20 metres from base to the top of the head, was left standing while the rock around it was quarried away into a horseshoe-shaped enclosure. Everything contentious about the monument is contentious because of that enclosure: its walls are cut faces of natural rock, and cut faces of natural rock weather in ways people argue about.
Two details of the making get skipped and both matter. The paws are not part of the subtraction: they were extended forward in masonry blocks rather than cut from the living rock, so the monument is not the single piece of stone it is usually described as. And it was painted. Traces of red pigment survive on the face, and there is evidence the whole monument once carried several colours, though the scheme cannot be recovered. The face is about four metres across.
It faces the equinox, and that is where the argument starts
The Sphinx faces due east, towards sunrise at the equinoxes. That is an uncontested survey fact, and for an atlas built around alignment it is the first thing that ought to be said about the monument. What it is not is an established intention. The statue was cut from a ridge of bedrock that already ran in a particular direction, and whether the due-east orientation is deliberate solar symbolism or a consequence of the lie of that ridge is an open question, examined among others by Juan Antonio Belmonte. The alignment is measured. The intent behind it is not. Holding those two sentences apart is most of what this site has to teach.
Nobody knows what the builders called it
The standard dating puts the carving somewhere in the range 2600 to 2500 BC, most often in the reign of Khafre, whose valley temple sits beside it and whose pyramid rises directly behind on the same axis. Pinning it to that single 26-year reign is firmer than the evidence supports, because no contemporary inscription names a builder at all. Here is the part the textbooks tend to skip. No Old Kingdom name for the statue survives. The name we have, Hor-em-akhet, Horus of the Horizon, is New Kingdom, and its best-known appearance is on the Dream Stele that Thutmose IV set between the paws around 1400 BC, recording a dream in which the buried statue asked him to clear the sand off it. That stele is the only ancient text that has ever been read as naming Khafre in connection with the Sphinx, and the reading rested on a partial syllable in a damaged line. The relevant lines had flaked away by 1925. The one inscription everybody cites was carved eleven centuries after the fact and no longer says what it was said to say.
That does not make the conventional date wrong, and it is important to be clear about why. The case for Khafre is archaeological, not textual: the sequence of the quarry, the temples and the causeway. But a reader is entitled to know that the famous documentary link is a great deal thinner than its reputation. Whose face it is remains an active question inside Egyptology rather than outside it. Khafre is the leading candidate. The Egyptologist Vassil Dobrev has argued for Khafre's half-brother Djedefre instead, which is a live minority position within the discipline and not a fringe one.
The nose is gone, and so is the ceremonial beard, fragments of which are held in the Egyptian Museum in Cairo and in the British Museum. Whether that beard was an original Old Kingdom feature or a New Kingdom addition is not settled either. Specialists are divided, and accounts that call it flatly a later addition are reporting one side of a live disagreement as though it were the verdict.
The enclosure has been dug out repeatedly, and the excavation history is part of the evidence. Giovanni Battista Caviglia cleared sand in 1817. Auguste Mariette began systematic clearance in the 1850s. Gaston Maspero surveyed in 1886 and Émile Baraize worked the site in 1931 and 1932. Selim Hassan's campaign of 1936 to 1938 exposed the full body for the first time in modern history and documented the enclosure walls, the associated temples and the Dream Stele; it is the single most important excavation the enclosure has had. Mark Lehner's 1991 Yale dissertation, "Archaeology of an Image: The Great Sphinx of Giza", is the stratigraphic and construction-sequence study that everything since is measured against.
The claim, at full strength
In 1990 the geologist Robert Schoch visited Giza with John Anthony West, and in 1991 he presented his reading of the enclosure to the Geological Society of America, pointedly without using the word Atlantis. His argument is about the shape of the weathering, not its depth. Wind and sand, he argues, cut horizontally and pick out soft beds preferentially, producing a fluted, layer-following profile. Water falling from above cuts vertically and rounds, producing the undulating vertical profile he identifies on the enclosure walls. Egypt has been hyper-arid for a very long time, so sustained rain pushes the carving back before roughly 5000 BC, and in some versions further.
Two things are worth stating for accuracy. Schoch's own range is far more conservative than the 10,500 BC date that circulates under his name; that figure belongs to a separate argument about the constellation Leo, made by other people. And Schoch is not an indiscriminate believer: he dived the Yonaguni formation off Japan, which the alternative-history circuit treats as a sunken city, and concluded it was most likely natural. When someone with that record makes a claim, the correct response is to check it rather than to sneer. The rainfall argument did not begin with West either. It follows earlier geological observations by R. A. Schwaller de Lubicz, and West's contribution was to press them.
What the counter-evidence actually says
The rock itself does a lot of work. The Sphinx is cut through nummulitic limestone beds of markedly different hardness, so any weathering agent whatsoever will produce a deeply uneven result. On top of that, salt crystallisation inside the pores, driven by moisture moving through the stone, splits it from within; the Getty Conservation Institute's work at the site between 1990 and 1992 concluded that continual salt crystallisation would account for at least part of the deterioration. K. Lal Gauri, writing in Geoarchaeology in 1995, argued for that mechanism driven by atmospheric moisture such as dew rather than rainfall. James Harrell proposed wet sand banked against the walls. Peter Lacovara has pointed out that some of the enclosure wall profile is quarrying damage rather than weathering at all.
Then the context. The blocks cut out from around the Sphinx's body went into the temple standing directly in front of it, which ties the carving to a construction programme. Zahi Hawass's objection is arithmetical: roughly 1,100 years separate Khafre from the first major recorded restoration in the Eighteenth Dynasty, and nobody has shown why that span, in a climate that was not uniformly dry, is insufficient. Kenneth Feder's objection is about everything the theory needs and Giza has never produced: a civilisation capable of this work leaves settlements, burials, storage, surplus and hierarchy, and none of it is there before the Fourth Dynasty.
The climate literature is genuinely split, and we will not pretend otherwise. Rudolph Kuper and Stefan Kröpelin's work places the onset of arid conditions in the Egyptian Sahara by about 5300 BC, which does not reach Schoch's window. Other geoarchaeological work reports significant rainfall events continuing until the end of the Old Kingdom around 2200 BC, which cuts the opposite way and gives water-driven erosion centuries to work inside the conventional chronology. Both cannot be fully right, and the disagreement is between mainstream specialists, not between mainstream and fringe.
The version that gets less airtime
The most interesting challenge to the standard date is not Schoch's. It is Colin Reader's, and it moves the Sphinx by centuries rather than millennia, to the First or Second Dynasty, roughly 3000 to 2800 BC. Reader's argument is about catchment. The plateau immediately west and north-west of the Sphinx was solid bedrock before Khufu and Khafre quarried it for pyramid stone. Intact rock sheds rainwater as surface runoff; a backfilled quarry soaks it up. So heavy runoff into the enclosure was possible before the quarries existed and largely impossible afterwards, which implies the enclosure was already open when the plateau was intact.
He supports it with the distribution of the damage. The western wall and the western end of the southern wall, precisely where runoff would concentrate, are the worst affected, while the body of the Sphinx weathers more evenly. He also identifies an earlier phase in the eastern part of the Khafre temple complex, and reads the causeway's line as respecting a route that already existed. His proposal needs no unknown civilisation and no missing millennia, only an earlier Egyptian dynasty on ground that Egyptians demonstrably occupied, and it has had a correspondingly more serious hearing, including published exchanges with Mark Lehner and Hawass.
A sphinx before the Sphinx
In 2023 Leif Ristroph's group at New York University's Applied Mathematics Laboratory published an experiment in Physical Review Fluids that nobody in this argument had thought to run. They embedded a harder inclusion in soft clay and ran fast flows over it to model abrasion by wind-borne sand, then watched what came out. The forms that emerged repeatedly had a raised head, an undercut neck, forward-projecting paws and an arched back. They had made yardangs, the wind-carved ridges that litter deserts, and the yardangs looked like couchant lions.
Ristroph is careful, and we will be too: he calls it a possible origin story, not a finding about this monument. It dates nothing. What it does is make a third option cheap. If the ridge at Giza was already a lion-shaped yardang, then it carries pre-carving weathering, which is one of Schoch's observations, and it was finished by Khafre's masons, which is the archaeologists' sequence, and the two are no longer in conflict. Nobody has yet tested that specific proposition against the actual rock, and somebody should.
What is under it
The hidden-chamber question gets asked more than any other, and it has a straight answer: there is one real measurement, and it is inconclusive. Ground-penetrating radar was run near the left front paw by Michael Oristaglio, Steinar Haldorsen and colleagues, on fieldwork in 2001, and it detected subsurface anomalies. Those anomalies are consistent with voids. They are equally consistent with natural fractures in the limestone, which this plateau has in quantity. No follow-up excavation has been officially authorised, so nothing has been opened, and nothing has been ruled out either.
That is a published measurement that has not been resolved, which puts it on the open list and not on the confirmed one or the refused one. Anybody citing it as evidence of a chamber is overstating it. Anybody waving it away is understating it.
The real threat under the monument is more boring and more urgent. An integrated geophysical survey of the plateau published in 2019 put the groundwater on average around 15 metres below the Sphinx. That is currently safe, and it is monitored. Rising groundwater and pollution from the neighbouring village of Nazlet El-Samman are the main long-term conservation risks to the statue, which is a duller sentence than any of the theories above and a more consequential one.
A 2026 claim, recorded and not repeated
In March 2026 the Italian researcher Filippo Biondi announced that satellite radar scans appeared to show a second, buried Sphinx-like structure on the Giza plateau. Zahi Hawass dismissed the claim as unsubstantiated. Independent archaeologists noted that radar anomalies frequently have ordinary geological causes. Nothing has been excavated.
We record the claim and its status and take no position beyond that, because as things stand there is nothing to take a position on. The distinction this site forces is the one worth carrying away from it. The 2001 radar anomalies under the paw are a published measurement nobody has resolved, so they are an open question and we name them as one. An announcement with no published survey anyone can check behind it, and no excavation, is not a finding, and we will not pass it on as one. The difference is not how strange a claim sounds. It is whether there is a measurement to argue about.
Where we come down
The weight of evidence still favours a Fourth Dynasty Sphinx, and the strongest live challenge is Reader's rather than Schoch's. Attempts to settle it with a number have not helped: a 2014 surface luminescence study by Ioannis Liritzis and Asimina Vafiadou in the Journal of Cultural Heritage returned Sphinx Temple dates spread from about 2220 BC to about 3100 BC, with errors of several hundred years and the authors themselves flagging small samples and incomplete bleaching. Thirty-odd years into this dispute, what has not arrived is a decisive measurement, and that is worth saying plainly rather than dressing either side up as settled.
There is a structural reason no decisive measurement has arrived, and it deserves to be better known than it is. The Sphinx cannot be radiocarbon dated. It is carved bedrock, not organic material, and there is nothing in the statue to date. The mortar samples that have been dated come from later repair phases, and what they date is the repairs. Any dating of the monument itself has to be argued from context, stratigraphy and style. That is why this argument has lasted thirty years, and why it will not end with a laboratory result.
What our engines found
We ran the same test here as everywhere else, with the settings fixed in advance and identical at every one of the fifty sites we cover: a 200 km search radius, at least 12 km between sampled sites, 15 sites, a 2 per cent tolerance, and 2,000 Monte Carlo runs against a null model resampled from real catalogued sites in the same region. The settings are frozen because the answer moves if you tune them, and tuning them after seeing the answer is exactly the trick we object to elsewhere. Great Sphinx returned 43 golden-ratio pairs where chance predicts 30.1, giving p = 0.1049.
That is a little more than chance would predict, and it is not significant. The control points the same way: the arbitrary ratio pi/2 scored 1.88 times its expected count against phi's 1.43, so nothing here singles out the golden ratio.
We publish the misses because a tool that only reports its hits is not measuring anything. Every figure above is reproducible from the same coordinates, the same settings and the same seed.
One limit worth stating plainly, because it is the one most often glossed over. This test compares positions between monuments. It says nothing about any building's own proportions, its height, base, slope or perimeter. That is a separate question, and at several sites a genuinely open one.
Explore it yourself
Open the Great Sphinx record in the living atlas to see the plateau in 3D, place the Sphinx among its neighbours and run the geometry engines on the real coordinates. The Great Sphinx Site-Emblem T-Shirt carries the erosion-contrast artwork, the site geometry and the exact coordinates.
Sources worth reading
Lehner, "Archaeology of an Image: The Great Sphinx of Giza" (Yale doctoral dissertation, 1991), the construction-sequence reference. Boury, Weady and Ristroph, "Sculpting the Sphinx", Physical Review Fluids (2023), for the yardang experiments. Gauri, Geoarchaeology (1995), on salt crystallisation and atmospheric moisture. Liritzis and Vafiadou, Journal of Cultural Heritage (2014), for the surface luminescence dates and their error bars. Reader's published exchanges with Lehner and Hawass, for the strongest alternative chronology. Oristaglio, Haldorsen and Englund's ground-penetrating radar work near the left paw, reported to the Society of Exploration Geophysicists (2005). The 2019 integrated geophysical survey of Giza plateau groundwater in Geoscientific Instrumentation, Methods and Data Systems.