Catalhoyuk · Turkey · 37.6664, 32.8286 · PYR record in the living atlas
You did not walk to your front door at Catalhoyuk. You crossed the roofs of your neighbours, found the opening above your own hearth, and climbed down a ladder through the smoke hole. There were no streets. Mudbrick houses stood wall against wall in one continuous block, and the roofscape was the public space: the place people worked, cooked in summer and moved through the town. The East Mound carries around eighteen building levels, dated by Bayesian modelling of the radiocarbon sequence to roughly 7100 to 5950 BC, and stands up to 20 metres above the Konya plain. Older summaries, UNESCO's included, give a wider span of about 7500 to 5600 BC; the modelled sequence is what the current excavators work from, and the two figures should not be read as one correcting the other so much as one being tighter than the other. There is a second and smaller West Mound beside it, carrying a later Chalcolithic occupation whose extent, and whose relationship to the people of the East Mound, are not established.
People rebuilt on the same footprints for more than a thousand years, and the mechanism was deliberate rather than accidental. When a house became dilapidated it was demolished, its walls cut down to a low platform, and a new house built directly on top. That is what produced the stack of superimposed floors the mound is made of, until the wreckage of their own houses had lifted the town above the fields that fed it.
The interiors were standardised to a degree that is hard to put down to convenience. Hearth and oven sit at the south end of the main room, sleeping platforms to the north, repeated with striking regularity from house to house and from level to level. Large deposits of animal bone point to communal feasting, and in 2024 researchers described what may be a fragment of bread from the site, roughly 8,600 years old and potentially among the oldest known. How a community organised building, feasting and burial on this scale with no identified civic architecture and no evident hierarchy is an open research question, not a solved one.
The famous population figure is an assumption, not a count
Almost every popular account gives Catalhoyuk several thousand residents at its peak. In June 2024 Ian Kuijt of Notre Dame and Arkadiusz Marciniak of Adam Mickiewicz University published a reassessment in the Journal of Anthropological Archaeology that put the figure at 600 to 800 people. Their objection is not to the excavation data but to the arithmetic laid over it. Earlier estimates counted the houses visible in a level and assumed they all stood, occupied, at the same time. Kuijt's own analogy is that this is like assuming every airport hotel built in the past fifty years is open simultaneously and permanently full. Allow that houses were abandoned, rebuilt and left empty on a rolling cycle, and the town shrinks by close to an order of magnitude.
This is a proposal rather than a settled correction, and the published range it is arguing against runs from about 2,800 to 10,000, with much of the recent literature sitting between roughly 3,500 and 8,000. The number turns entirely on how many houses stood occupied at the same moment, and that cannot be established directly from the stratigraphy, which is why it is an open question rather than a figure awaiting refinement. But the direction matters. If Catalhoyuk held hundreds rather than thousands, the thing it demonstrates is not that farming produced cities quickly. It is that people chose to pack themselves into extraordinary density long before they had any of the reasons cities usually give for it.
Living above the ancestors
The houses were also tombs. Adults and children were buried beneath the sleeping platforms of the rooms their households used, sometimes many to a building, and the floors were replastered over them. Walls carry some of the oldest paintings on plaster anywhere: geometric panels, hunting scenes, vultures stooping over headless human figures. Bull skulls were set into walls and benches with the horns projecting into the room. Some human skulls were lifted back out of the ground, plastered and painted into portrait-like heads, and kept in circulation among the living before being reburied.
The sequence was more elaborate still. A 2022 study led from the University of Bern found that some remains had been partly painted with red ochre, and that some had been exhumed and reburied, so a single interment beneath a floor is not the whole of the practice. In August 2025 Arkadiusz Marciniak's team reported a structure they have called a House of the Dead, holding the remains of twenty individuals.
James Mellaart read the heavy seated female figurines, including the well known one flanked by two leopards, as the earliest evidence of a Mother Goddess. Marija Gimbutas took that reading up and built it into a general Old European framework in The Goddesses and Gods of Old Europe in 1982, which is how a claim about one Anatolian mound acquired reach far beyond Anatolia. Ian Hodder's project, which ran from 1993 to 2017, recovered large quantities of male and animal figurines alongside them and read the seated figures instead as elders and ancestors. Neither reading is proven. What the excavation shows solidly is the absence of the usual markers of rank: no palace, no rich quarter, no building set apart, and house sizes and burial goods running remarkably even. Hodder's conclusion from that, that religion and ritual were embedded in ordinary domestic life rather than housed in temples, is an interpretive position rather than a settled finding. Whether any building here served as a shrine is unresolved, and so is what the wall paintings and the circulated skulls meant to the people who made them.
The DNA says the houses were not simply families
In June 2025 a team led by Eren Yuncu and Mehmet Somel, with Eline Schotsmans, published 131 ancient genomes from the East Mound in Science, recovered from around 400 skeletons and covering 23 buildings that held more than one burial. The result overturned the obvious assumption. Where people buried under the same floor were related at all, the connection ran through the mother's line. Simulations of residence patterns fitted the data best at roughly half mixed and half matrilocal, with essentially no contribution from strictly patrilocal residence. Husbands moved into their wives' households rather than the reverse.
The pattern also decayed over time. Among children and adolescents buried together in one building, 63 per cent were genetically related in the earliest periods, falling to 30 per cent in the middle periods and 22 per cent in the late. Sharing a house floor slowly stopped meaning sharing ancestry. Among those same younger burials, graves containing objects were five times more frequent for females than for males, a difference the authors report as statistically significant. That is not a matriarchy and it is not the goddess Mellaart wanted. It is something narrower and much better evidenced: the female line held the house. What that implies for authority and for social organisation more broadly is a separate question, and researchers disagree about it. No consensus has been reached.
The excavator forged his own evidence
Mellaart identified the mound in 1958 and dug it in a short run of seasons in the early 1960s, and he is the reason the world knows the place exists. He was also, on the evidence found in his own flat, a forger. In February 2018 the geoarchaeologist Eberhard Zangger, working with Mellaart's son Alan, went through the London study after Mellaart's death and found the workshop: draft texts, and pieces of slate carrying pictorial carvings that were plainly the sketches behind murals Mellaart had published as reconstructions of paintings at the site.
The scale is what makes it serious. Mellaart had recorded one genuine landscape painting at Catalhoyuk. Over the following decades he produced roughly seventy more that had no original behind them. The same flat yielded about 500 typed pages presented as translations of Bronze Age documents, the so-called Beykoy text, and eyewitnesses tied the Dorak treasure affair of 1959 to the same pattern, although the Dorak business is a separate matter from the murals and was the reason he was barred from excavating in Turkey in the 1960s. Excavated objects from Catalhoyuk are still what they always were. Mellaart's later reconstructions are not, and any drawing of a Catalhoyuk mural is worth checking against which of those two categories it belongs to. Work to reconcile his surviving original field notes against the physical stratigraphy is still running, and results are still coming. This correction came from inside the discipline, not from outside it, and that is worth saying plainly.
The digging did not stop with him, or with Hodder. Since 2018 the work has been led largely by the Adam Mickiewicz University team from Poznan, alongside Turkish institutions, in the TP, TPC and East Areas.
The oldest map, or a leopard
One painting from level VII shows a block of rectangles below a two-peaked shape shedding dots. Before anything else about it: the version almost everyone has seen is a Mellaart-era reconstruction, and after the 2018 findings its authenticity is itself in some doubt. The doubt attaches to the reconstruction rather than to the original painting, but it belongs at the front of any discussion of what the image shows. Mellaart read it as a town plan with an erupting volcano behind it, which would make it both the oldest known map and the oldest known depiction of an eruption. Stephanie Meece has argued instead that it is a leopard skin laid over a geometric panel, pointing out that no other prehistoric map is known and that leopards run through the site's imagery everywhere else.
The geology has since been tested independently. In January 2014 Axel Schmitt and colleagues published zircon dating of pumice from Hasan Dagi, the twin-peaked volcano visible from the mound, in PLOS ONE. They obtained an explosive eruption at 8,970 years before present, give or take 640 years, which overlaps the radiocarbon range of the level holding the painting. That does not show the painting is a map. It shows the painters could have watched the mountain erupt. The authors say as much themselves and name the leopard reading as the live alternative. The question is genuinely open, and the useful part is that the eruption was real whether or not anybody painted it.
One claim this site does not carry
No peer-reviewed study has identified a deliberate cardinal or solstitial orientation at Catalhoyuk. The plan is organic and modular, houses added against houses, and individual buildings face varied directions. That is worth stating rather than passing over, because it makes the site a useful control: a large, long-lived, plainly deliberate settlement with no alignment claim attached to it at all.
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. Catalhoyuk returned 7 golden-ratio pairs where chance predicts 7.4, giving p = 0.5812.
That is almost exactly what chance would predict, and it is not significant. The control points the same way: the arbitrary ratio sqrt(2) scored 1.57 times its expected count against phi's 0.95, so nothing here singles out the golden ratio.
Worth flagging: the engine could reach only 9 catalogued sites here, so this result carries little weight in either direction. That is a limit on our data, not a verdict on the region.
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 Catalhoyuk record in the living atlas to see the mound in 3D, place it among the other catalogued sites of the Konya plain, and run the alignment and ley line engines on the real coordinates. The Catalhoyuk site-emblem tee carries the twin-leopard throne, the site name and the exact coordinates in the Pyramason Chronicle style.
Sources worth reading
Bayliss, A. et al., "Bootstrapping Bayesian Chronology Building in the Neolithic of the Eastern Mediterranean", Radiocarbon (2015). Marciniak, A. et al., "Fragmenting Times: Interpreting a Neolithic Sequence through Bayesian Modelling from Catalhoyuk East, Turkey", Antiquity (2015). Haddow, S. et al., "Burial and bodily practices at Catalhoyuk", Journal of World Prehistory (2023). Hodder, I., The Leopard's Tale: Revealing the Mysteries of Catalhoyuk, Thames and Hudson (2006).