Karahan Tepe: Gobekli's Eerie Twin

Structure AB is a semi-subterranean oval chamber, carved partially into the bedrock, measuring approximately 7 meters by 6 with a trapezoidal plan and rounded corners. Our own research file calls it the most iconic space at Karahan Tepe and states that it contains at least eleven carved limestone pillars set in a roughly oval configuration. That eleven does not rest on our file alone: three independent sources reach the same total. Two of them give eleven directly; a third counts ten pillars cut from the bedrock, with the eleventh standing free. A specific number corroborated four ways gets stated without hedging, and this one is.

Several of those pillars are carved as naturalistic phallic forms: tapered, rounded shafts of varying heights rising from the bedrock floor, with anatomically explicit sculpted heads. The independent sources describe the bedrock-cut set collectively in the same terms, and one of them gives each pillar a height of roughly 1.6 to 1.7 meters. This is an iconographic choice unlike anything found at Gobekli Tepe.

A carved human head emerges from the bedrock wall of Structure AB: a naturalistic face with defined eyes, nose, mouth, and chin, sculpted in three dimensions as though a figure were pushing through the stone from behind, on an elongated neck that may itself represent a serpent. One detail in that sentence is corrected rather than repeated. Our own research file places the head on the chamber's north wall. Two independent sources place it on the western wall, and this article uses the western wall. Our file additionally states that the head sits directly above a carved serpent; that spatial relation was not independently re-checked, so it travels here as our file's own statement rather than as a verified one.

A continuous rock-cut bench runs along the chamber's circumference, which suggests an audience or a congregation space.

Karahan Tepe was identified during a 1997 surface survey by the archaeologist Bahattin Celik, who noted concentrations of T-shaped pillar fragments, stone tools, and carved stone fragments lying on the surface of a low hilltop. A correction belongs here rather than in a footnote. Our own research file gives Celik's affiliation as Istanbul University. Independent checking puts him at Harran University, and Istanbul University is in fact the correct affiliation for Necmi Karul, who directs the later excavation, which suggests our file may have conflated the two men rather than getting each of them wrong separately. This article states Harran University for Celik.

Celik published the surface finds in 2000 and again in 2011, classifying Karahan Tepe as a Gobekli Tepe-type Pre-Pottery Neolithic ritual site. Limited test trenches followed between 2000 and 2014. For more than two decades after the survey, that was the extent of it: a known site, surveyed and published, largely undug.

Full-scale systematic excavation began in September 2019 under the Tas Tepeler project, Stone Hills, directed by Prof. Necmi Karul of Istanbul University and sponsored by the Turkish Ministry of Culture and Tourism. Karul's Istanbul University affiliation is independently confirmed. Tas Tepeler takes in at least twelve Pre-Pottery Neolithic sites in the Sanliurfa region: Gobekli Tepe, Karahan Tepe, Sayburc, Harbetsuvan Tepesi, Gurcu Tepe, Ayanlar Hoyuk, Yeni Mahalle, Cakmak Tepe, Kurt Tepesi, Tasli Tepe, Sefertepe, and Hamzan Tepe. Karahan Tepe is the second-most intensively excavated of them, after Gobekli Tepe.

Excavation is active and ongoing, and almost all of this hill is still underground. Our research file puts roughly 5 percent of the estimated site area uncovered as of early 2026. A 2025-dated secondary source says only 1 percent. The precise fraction evidently depends on which season and which estimate of total area is being described, so the honest statement is a range rather than a figure: somewhere between about 1 and 5 percent of Karahan Tepe has been dug. Everything below is a report on that fraction.

Radiocarbon dates from the 2019 to 2024 excavation campaigns place Karahan Tepe at approximately 9400 to 8200 BCE, running from the Pre-Pottery Neolithic A into the early Pre-Pottery Neolithic B. That makes it broadly contemporary with Gobekli Tepe's earliest phases, at about 9600 to 8000 BCE, and possibly slightly later. The opening date is the soft end of this. One independent source gives the same 9400 to 8200 BCE window; another gives between 10,000 and 9500 BCE with construction around 9750 BCE. They cluster in the same broad window, and this article treats the exact starting date as approximate rather than settled.

There is no pottery at Karahan Tepe. The subsistence economy runs on wild plant gathering and hunting, primarily gazelle, aurochs, and equids, with no evidence of domesticated cereals and no evidence of domesticated animals. The monumental construction on this hill firmly predates agriculture in this region.

Zooarchaeological analysis identifies gazelle, Gazella sp., as the dominant faunal taxon, followed by aurochs, Bos primigenius, then equids, Equus sp., wild boar, and smaller mammals. All are wild species and none show morphological indicators of domestication. Bird remains include large raptors, eagles and vultures, consistent with the prominent vulture iconography at Gobekli Tepe.

Charred wild plant remains include einkorn wheat, Triticum boeoticum, wild barley, Hordeum spontaneum, and various pulses, all of them in their wild morphological forms, with non-shattering rachises absent. That is the signature of intensive wild plant exploitation, and it is specifically not the signature of cultivation.

Limestone and basalt ground-stone tools, mortars, pestles, and querns, processed those wild plant foods. The sheer energy invested in stone grinding equipment suggests intensified wild cereal processing, which is a plausible precursor to domestication. Large quantities of limestone flaking debris around the pillars and inside the structures document that the monumental sculptures were carved on site rather than imported finished.

The chipped-stone assemblage is dominated by flint and obsidian: pressure-flaked projectile points, bladelets, scrapers, and micro-drills. The obsidian was imported from sources in eastern Anatolia, Bingol and Nemrut Dag, at distances of 250 to 400 km. Long-distance exchange networks existed here among people who were not farming.

Concentrations of animal bone, grinding equipment, and fire-cracked stone associated with the monumental structures are consistent with communal feasting: ritual meals involving large quantities of prepared food. Dietrich and colleagues documented similar feasting assemblages at Gobekli Tepe in 2012 and read them as evidence that large-scale ritual gatherings motivated surplus food production and ultimately fed into the domestication process. That reading is picked up again in section 07, where it belongs to a live argument rather than to the finds themselves.

At Gobekli Tepe the structures sit on the natural hilltop surface. At Karahan Tepe much of the monumental architecture has been carved directly into the bedrock: floors, benches, and even some of the pillars emerge from the living rock, which blurs the boundary between architecture and sculpture. The technique is a hybrid, combining free-standing pillars erected in sockets with rock-cut excavation, and it is not prominently seen at Gobekli Tepe.

Karahan Tepe also displays a wider range of architectural forms than its neighbor's predominantly circular enclosures: oval structures, sub-circular structures, and rectangular structures stand on the same hill. Our file reads that variety as experimentation with spatial organization, or as different functional requirements. Which of those it is, the file does not say.

Geophysical survey, georadar and magnetic, has revealed at least 250 T-shaped pillars across the site, most of them still in situ or visible only in survey data. Independent sources agree: one states that over 250 T-shaped pillars have been identified across the hillside, another that more than 266 were observed as of the year 2000. That is potentially more pillars in total than Gobekli Tepe, which has about 200 across roughly 20 enclosures, most of them still unexcavated. Karahan Tepe's are smaller, though. Its largest stand approximately 3 to 4 meters, where Gobekli Tepe's largest reach 5.5 meters and 16 tonnes.

The exposed and excavated remains cover approximately 1 hectare, with ground-penetrating radar indicating substantial unexcavated structures beyond them. Our file puts the total site extent across the limestone hilltop at approximately 11 hectares. An independent source gives almost 10 hectares, rising by another five if the quarries for the T-shaped columns are counted in. Those are broadly consistent rather than identical, and this article carries both rather than picking the tidier one.

A partially quarried, unfinished T-shaped pillar has been found in situ in the bedrock quarry immediately next to the site. Channel-cutting and wedge-splitting marks in that quarry document the extraction process step by step. The pillars were made on this hill, out of this hill. No pillar was transported in from a distance, and there is no transport mystery here to solve for the stonework itself, whatever the obsidian's own supply chain looked like.

A T-shaped pillar bearing a realistically carved human face, with eyes, nose, and jaw, is the most explicit anthropomorphic rendering of the T-pillar concept found so far at Karahan Tepe. It confirms and then exceeds the standard reading of T-pillars as stylized human figures. Smithsonian Magazine reported the find in October 2025: a pillar about 4.5 feet tall, which converts to roughly 1.4 meters, with a prominent brow bone and a straight nose, dated to approximately 11,000 years old, the 10th millennium BCE, and found in a structure that also contained domestic elements such as grinding stones. Necmi Karul is quoted directly on both points. “The human-faced pillar we encountered in a structure containing domestic elements, such as grinding stones, fills this gap,” and “This example provides clear evidence that T-shaped pillars symbolized humans.”

Multiple carved three-dimensional human heads have been recovered at Karahan Tepe, some sculpted from bedrock and some as portable objects, with naturalistic detailed eyes, noses, mouths, and jaw lines. One head from Structure AB appears to depict an older individual, with defined wrinkles.

T-shaped pillars and walls at the site carry carved reliefs of serpents, a prominent and recurring motif across multiple structures, along with a leopard or large cat carved in high relief on a pillar, and geometric patterns.

Structure AD, a smaller chamber separate from Structure AB, contains phallus-shaped pillars carved directly from the bedrock rather than free-standing. One of those sculptures stands approximately 2.3 meters tall, integrated with the surrounding rock floor. It is part of the hill and has never moved.

In September 2023, excavators uncovered a 2.3-meter human statue at Karahan Tepe. It was found broken in two, originally fixed in a seated position on a stone bench, holding its own phallus with both hands, with a strong V-neck collar motif, carved ribs, and a square raised protrusion between those ribs. It came out of a newly opened enclosure alongside a separate vulture or bird statue positioned beneath a circular holed stone on the northern wall. Five independent outlets covered the find in October 2023, and it has been widely described as one of the earliest known naturalistic, true-to-life-scale human sculptures. Now the housekeeping, because two things at this site are described at about 2.3 meters and they are not the same thing. Structure AD's phallic pillar is bedrock, fixed, and part of the floor it rises from. This seated statue is free-standing, was found broken in two, and comes from a different enclosure entirely. The shared height is a coincidence of measurement, nothing more.

Ground-stone objects recovered from the site include stone bowls, pestles, and carved miniature replicas of the T-shaped pillars themselves, which points at both utilitarian and symbolic activity in the same assemblage. The bowls show sophisticated manufacturing, including rotary grinding and internal polishing, which indicates specialized craft production rather than expedient tool-making.

Microscopic analysis of some pillar surfaces has revealed faint traces of red ochre. That raises the possibility that the carved reliefs and sculptures were originally painted, which would dramatically alter the visual experience of these ritual spaces. Note how thin that evidence is: faint traces, found under a microscope. It is enough to raise the possibility and not enough to establish it, and what these surfaces looked like when they were new is not necessarily what stands on the hill today.

Where Gobekli Tepe's relief program emphasizes animals, foxes, wild boar, snakes, vultures, Karahan Tepe's emphasis on human heads, phallic forms, and serpents may indicate a distinct ritual program, potentially oriented toward human ancestry, fertility, and chthonic symbolism rather than, or in addition to, totemic animal associations. Karul has made the same contrast in public, saying that human symbolism dominates Karahan Tepe while animal imagery is more prevalent at Gobekli Tepe, and offering a reason for it: that “the growing number of human sculptures can be read as a direct outcome of settled life,” with people gradually distancing themselves from nature and placing “the human figure and the human experience at the center of the universe.” The inventory of motifs is Tier 1. The explanation for the inventory is Tier 2, and it is an interpretation offered by the excavator rather than a result the excavation produced.

Karul, in publications of 2020 and 2021, reads the combination of phallic pillars, the human head, serpent iconography, and bench seating as a ritual space connected to themes of fertility, ancestor veneration, death-and-rebirth symbolism, or clan initiation. Note the shape of that sentence. It is four candidate functions joined by or, from the excavator himself. No scholarly consensus exists on the specific function of Structure AB, and the excavation is still running. This article does not pick one of the four, because our file does not, and because there is nothing yet out of the ground that would justify picking.

Some researchers have suggested that the seated statue from the 2023 enclosure represents a deceased ancestor associated with the building it was found in. That is a suggestion attached to a find, in the same family as the ancestor-veneration reading above, and it carries the same weight: plausible, unfalsified, and unproven.

The Structure AB assemblage, phallic pillars plus serpent imagery plus a head coming out of stone, has been compared to cross-cultural shamanic cosmologies, with the head-through-wall motif read as a portal between worlds; the neuropsychological model of cave art developed by Lewis-Williams, built on entoptic phenomena and trance states, has been speculatively applied to it. There is no direct evidence for this at Karahan Tepe: no psychoactive plant residues, and nothing standing in for them. There is also a specific reason to be more careful here than usual. The head-emerging-from-stone motif has no direct parallel at any other known Pre-Pottery Neolithic site. Cross-cultural analogy works by matching a thing to comparable things, and this particular thing has no comparables in its own period and region, which makes analogies drawn from other continents and other millennia weaker here rather than stronger.

As at Gobekli Tepe, there is evidence that some structures at Karahan Tepe were deliberately buried after a period of use, filled in with rubble, soil, and refuse. At Gobekli Tepe that practice is what preserved the enclosures for 11,000 years, and it is interpreted there as intentional decommissioning. Karahan Tepe's own evidence for the practice is still being assessed, which is why this sits at Tier 2 and not higher.

Why anything was buried at all has two readings and neither has won. Some researchers interpret the deliberate backfilling of pillar enclosures at both sites as a form of ritual killing or burial of the structure, analogous to the intentional breaking of objects documented in Neolithic contexts elsewhere, which would imply buildings with finite ritual lifespans: built, used, ceremonially decommissioned, replaced. Others interpret backfilling pragmatically, as site management, preparing a stable surface for new construction above, with no ceremonial content whatsoever. Both readings explain the same fill.

Whatever the buildings meant, making them took coordination. Quarrying heavy stones, moving them even short distances, carving elaborate reliefs, and maintaining the structures afterward is not work a single family band does casually. Estimates for Gobekli Tepe run to more than 500 workers; Karahan Tepe's smaller scale may need fewer, but it still demands organizational capacity beyond one band. That implies nascent social hierarchy, or at least temporary leadership structures, among Pre-Pottery Neolithic populations, which is a revision of the standard model of hunter-gatherer social organization.

The identification of twelve or more Pre-Pottery Neolithic sites with T-shaped pillar architecture across the Sanliurfa plateau demonstrates something that took decades to become visible: Gobekli Tepe was not unique. It was the largest node in a regional network of ritual and ceremonial sites. Karul's Tas Tepeler framework proposes a coordinated or interrelated social system connecting them, possibly involving seasonal aggregation events at which dispersed hunter-gatherer bands converged for ritual, feasting, exchange, and social reproduction. The geographical distribution spans approximately 200 km east to west across the northern Harran Plain.

Sayburc, a Pre-Pottery Neolithic site excavated since 2021 and located approximately 50 km northwest of Karahan Tepe, contains a relief panel depicting two human figures flanking wild animals, bulls and leopards, which has been described as the first known Neolithic narrative scene. Its discovery confirms that the regional pattern extends beyond Karahan Tepe and Gobekli Tepe.

Both Karahan Tepe and Gobekli Tepe show minimal evidence of residential occupation, with no securely identified hearths, cooking areas, or sleeping quarters inside the monumental structures, though recent work at Gobekli Tepe has challenged the purely temple interpretation of that site. The two were likely occupied at least partially at the same time. That raises questions our file states and does not answer: whether they served different communities, different functions within one social network, or different seasons of one mobile group's annual cycle.

The traditional model, Childe's Neolithic Revolution, holds that surplus agriculture was a prerequisite for monumental architecture, social stratification, and organized religion. Gobekli Tepe challenged that model, and Karahan Tepe and the wider Tas Tepeler network reinforce the same challenge: monumental construction, elaborate ritual, long-distance exchange, and probable social coordination, all among communities with no evidence of food production. Revised models propose that ritual and social aggregation drove sedentism and agriculture rather than the reverse, making the monuments a catalyst for farming rather than its product. The debate continues, and it is largely Gobekli Tepe's argument to carry; what Karahan Tepe adds is that the pattern is not one anomalous hill.

The Tas Tepeler survey has identified twelve or more sites so far. Systematic satellite imagery analysis and ground-penetrating radar survey suggest potentially dozens more Pre-Pottery Neolithic sites across the Harran Plain and the surrounding uplands. The full extent of the network remains unknown, which is a statement about survey coverage rather than a discovery.

The Pre-Pottery Neolithic sites of Upper Mesopotamia emerge in the immediate aftermath of the Younger Dryas cold event, at roughly 10,900 to 9700 BCE. Researchers have hypothesized a causal link, with climatic stress forcing social reorganization, ritual innovation, and intensified wild cereal processing that eventually contributed to agriculture. The chronological coincidence is suggestive and the causation is undemonstrated, and an alternative model proposes the opposite driver: that post-Younger-Dryas climatic improvement, not stress, gave the ecological opportunity for population growth and social complexity. The Younger Dryas has its own file and its own wing on this site, and this article takes it only as a date lying on one side of a site.

The 2025 excavation season uncovered Structure BH, a rectangular building measuring approximately 9 meters by 7. It has two-tiered benches, four T-shaped pillars, walls preserved to 3.5 meters in height, a large stone basin, and a channel that appears to have carried liquid through the building, which suggests organized communal activity of some kind. Large flat stone slabs seal the chamber's upper section. Archaeologists plan to remove them individually and enter from above during the 2026 excavation season, which is expected to run until October 2026 under Necmi Karul as excavation coordinator of the Tas Tepeler project. The deliberate sealing is consistent with the pattern section 06 records, credible but still under assessment at this site, in which special-purpose buildings may have been intentionally filled and sealed rather than simply abandoned.

Reporting dated November 2025 records roughly 30 newly unearthed artifacts from the Karahan Tepe excavations, described in that reporting as illustrating the transition from hunting and gathering to more settled ways of life over 11,000 years ago. The haul includes statues of humans and animals, various vessels, necklaces, and beads. Three items are named specifically: a stone figurine with stitched lips, further carved stone faces, and a black serpentinite bead with expressive faces carved on both of its sides.

Astronomical alignment proposals, stellar, solar, or precessional encoding in T-shaped pillar enclosures, have been made for Gobekli Tepe by Sweatman and Tsikritsis in 2017, where they are contested, and some have extended them to Karahan Tepe. The extension is the problem. There is no published archaeoastronomical study specific to Karahan Tepe at all. That is not a claim that has been tested and failed. It is a claim that has not been tested.

Two cautions from our file's own counter-arguments section apply with unusual force here. Many Tas Tepeler sites remain incompletely excavated, so extraordinary claims about them should be treated as preliminary; and interpretation of any site is vulnerable to confirmation bias, since natural formations, weathering, and coincidental alignments can look intentional to someone already expecting intention. At a site where between about 1 and 5 percent of the ground has been opened, both of those cut in every direction, including against the readings this article has just called credible.

Claims that Karahan Tepe represents technology from a lost advanced civilization, or extraterrestrial intervention, have no support. The lithic technology is stone-tool-based, the construction methods are bedrock quarrying with stone and antler tools, and the iconographic repertoire is fully consistent with documented Pre-Pottery Neolithic material culture. The unfinished pillar still lying in the quarry next door, with the channel-cutting and wedge-splitting marks around it, is a direct physical answer to how these were made, found at the place where they were made.

No, the site is not 50,000 or more years old; proposals that date it that way misrepresent the radiometric evidence. No, it is not the oldest temple in the world superseding Gobekli Tepe; the same radiocarbon record that dates Karahan Tepe puts it broadly contemporary with its neighbor, possibly slightly later. No, the carved human head in Structure AB is not a portal or a stargate; that is a modern fantasy projected onto Neolithic iconography with no archaeological basis of any kind. And no, there is no documented advanced mathematics encoded in the pillar arrangements; those claims lack any published measurement or peer-reviewed analysis behind them.