Göbekli Tepe Comparisons to Other Sites

About Göbekli Tepe Comparisons to Other Sites

When archaeologists set Göbekli Tepe beside any other monument in the world's published corpus, the comparison nearly always breaks the corpus first. Klaus Schmidt, who recognized the site's Neolithic age in 1994 and directed excavations from 1996 until his death in 2014, made the case in Sie bauten die ersten Tempel (C.H. Beck, 2006) and again in the English edition Göbekli Tepe: A Stone Age Sanctuary in South-Eastern Anatolia (ex oriente, 2012, translated by Mirko Wittwar from a revised German edition). The five-meter T-pillars are pre-pottery, pre-agriculture, and roughly seven thousand years older than Stonehenge's first sarsen ring. Every comparison below has to carry that gap as a constant.

Geophysical surveys have since identified at least twenty enclosures inside the mound, with around two hundred T-pillars across the site in total, the majority still unexcavated. Only Enclosures A, B, C, D, and a smaller set in Layer II have been excavated in detail, with additional enclosures (Enclosure H announced in 2017, others identified by geophysics) under investigation by the German Archaeological Institute, the Şanlıurfa Museum, and Istanbul University, in coordination with Necmi Karul's Tas Tepeler / Stone Hills Project. Schmidt's "first temples" framing has been the field's working hypothesis since 2006, but it has been contested in print and the contest sharpens what comparisons are honest. This page walks through five comparison axes — chronological siblings, megalithic-precision peers, skull and ancestor cult parallels, urbanism timeline anchors, and the anti-comparison with every post-agricultural monument in the corpus — and notes where the comparison holds and where it collapses.

The closest peer Göbekli Tepe has, in time and toolkit, is Karahan Tepe, roughly 40–60 km east in the same Sanliurfa Province (sources vary on the exact figure). Both belong to the cluster of Pre-Pottery Neolithic A and B sites that Necmi Karul of Istanbul University has labeled the Tas Tepeler ("Stone Hills"). Karahan Tepe was discovered in 1997 by Bahattin Çelik of Harran University, with limited initial work in 2000 and 2011; sustained, full-scale excavations began in 2019 under Karul. The chamber that has drawn the most attention contains eleven phallus-shaped pillars carved directly out of bedrock, around a carved human head set into the wall. That is a different iconographic register from Göbekli Tepe's animal-coded T-pillars, but the structural grammar — sunken round chamber, carved bedrock features, central anthropomorphic focus — is unmistakably the same vocabulary.

The chronological overlap is also tight. Dietrich, Köksal-Schmidt, Notroff, and Schmidt's 2013 paper "Establishing a Radiocarbon Sequence for Göbekli Tepe" (Neo-Lithics 1/13: 36–47) places the wall plaster of Enclosure D at roughly 9984 ± 42 BP, which calibrates into the late tenth millennium BCE — the early PPNA. Karahan Tepe's earliest phases sit in the ninth millennium BCE, overlapping Göbekli Tepe's later Layer III and Layer II. Both sites were ultimately backfilled rather than abandoned to erosion. The 2023 announcement by the Turkish Ministry of Culture and Tourism added a 2.3-meter seated human statue at Karahan Tepe (gripping the phallus with both hands) and a painted limestone boar at Göbekli Tepe — described by Karul as the first painted Neolithic statue known from the wider Tas Tepeler / Anatolian Pre-Pottery Neolithic record. The painted boar sat on a stone bench beside an H-symbol, a crescent, and snake reliefs: a cluster that recurs on Pillar 18 at Göbekli Tepe, where the same H and crescent motifs appear on the central figure's belt. The shared vocabulary across that PPN landscape suggests a single symbolic system, not parallel invention.

However, the two sites are not interchangeable. Göbekli Tepe is on a hilltop with no springs or year-round water; Karahan Tepe sits closer to a usable water source and shows more evidence of dwellings cut into the bedrock. If Banning's 2011 critique (discussed below) is taken seriously, Karahan Tepe may be the better candidate for "ritually charged housing," and Göbekli Tepe the better candidate for purely periodic ceremonial gathering — though Karul has resisted the binary, treating both as ritual-residential complexes on a sliding scale.

The peer most often paired with Göbekli Tepe in popular accounts is Çatalhöyük, the Neolithic mound on the Konya Plain in central Anatolia. The pairing is geographically tidy and chronologically suggestive — Çatalhöyük's eighteen Neolithic levels run from about 7400 to 6200 BCE on the AMS-refined sequence, and the broader occupation extends to roughly 5600 BCE. That leaves a gap of six hundred to a thousand years between Göbekli Tepe's deliberate burial around 8000 BCE and Çatalhöyük's first mudbrick houses. The two sites are not contemporaries, but they sit on the same Anatolian arc of post-glacial transformation.

Where the comparison gets sharper is the symbolic vocabulary. Both sites are dominated by wild and predatory animals rather than the species their inhabitants ate most. At Göbekli Tepe, gazelle accounts for over sixty percent of the faunal remains while the carved iconography is dominated by foxes, vultures, snakes, scorpions, and aurochs — the dangerous, the scavenging, the venomous. At Çatalhöyük, Ian Hodder's team has documented the same disjunction: bull bucrania, leopard reliefs, and vulture wall paintings dominate the symbolic register while the diet centered on sheep, goat, and cereals. Hodder argues in Religion in the Emergence of Civilization: Çatalhöyük as a Case Study (Cambridge, 2010) that the wild-animal symbolism encodes a relational worldview in which danger and the hunt remained ritually charged long after they ceased to be subsistence-critical.

The skull-cult dimension links the sites more tightly. Çatalhöyük's plastered skulls — crania removed from primary burials, replastered, painted with ochre, and curated within "history houses" before being reburied — were interpreted by Hodder and Peter Pels in their 2010 chapter "History Houses" as material vehicles of social memory rather than relics in a Western "sacred-relic" sense. Julia Gresky and colleagues' 2017 paper "Modified human crania from Göbekli Tepe provide evidence for a new form of Neolithic skull cult" (Science Advances 3:e1700564) reported three partially preserved crania from Göbekli Tepe with deep flint-tool incisions along the sagittal axis and a drilled perforation in one parietal bone, with red ochre traces. Gresky's team interpreted the modifications as a previously undocumented variation of Anatolian-Levantine skull cult — possibly suspended display rather than the buried-and-recurated treatment at Çatalhöyük or the plaster-faced skulls of Jericho and 'Ain Ghazal. The same broad practice; different operational mode. The supported reading is that Göbekli Tepe sits at the upstream end of a thousand-year tradition that Çatalhöyük and the wider Levantine PPNB inherited and modified.

Most public descriptions of Göbekli Tepe reach for Stonehenge as the comparison. The chronological fact that has to lead any such comparison is that Stonehenge's central sarsen settings were erected around 2500 BCE, with the average sarsen weighing about twenty-five tons and the heaviest stones (such as the Heel Stone) closer to thirty tons. Geochemical analysis published by David Nash and colleagues in Science Advances in 2020 traced fifty of the fifty-two sarsens to West Woods, Wiltshire, twenty-five kilometers north of Stonehenge. Göbekli Tepe predates Stonehenge's sarsen ring by approximately seven thousand years. The standing T-pillars at Göbekli Tepe are estimated at five to ten tons each in the published Schmidt-era figures, with the two 5.5-meter central pillars in Enclosure D estimated at roughly 8–10 tons each in Schmidt-team figures (per the Tepe Telegrams "How did they do it?" post on monolith making and moving). One unfinished pillar still in the bedrock quarry weighs roughly fifty tons and would have been the largest had it been completed. The numbers are smaller than Stonehenge's sarsens, but the technological context — flint chisels, no metal, no wheels, no draft animals — is far more austere.

The deeper comparison is not "Göbekli Tepe is the world's first Stonehenge" but the opposite: Stonehenge sits inside a megalithic tradition that, on Bettina Schulz Paulsson's Bayesian-modeled radiocarbon analysis published in PNAS 116(9), 3460–3465 (2019, "Radiocarbon dates and Bayesian modeling support maritime diffusion model for megaliths in Europe"), originates in northwest France around 4500 BCE and spreads by sea routes through Iberia, the western Mediterranean, and eventually Britain and Ireland. The Carnac Stones in Brittany — the Ménec, Kermario, and Kerlescan alignments, with several thousand standing menhirs across roughly six and a half kilometers — are the upstream end of that Atlantic megalithic horizon. Göbekli Tepe is not in Schulz Paulsson's tradition. It is older than Carnac by about five thousand years, and the diffusion her data describes does not extend east into Anatolia. The supported reading is that monumental stone setting is a recurrent, not a singular, human response: Atlantic Europe found it independently of southeastern Anatolia, and Anatolia found it five thousand years before northwest France. Avebury — the larger but later British henge, c. 2850–2200 BCE — sits inside the same Atlantic horizon, well downstream of Stonehenge's earliest stages but part of the same insular megalithic tradition that Schulz Paulsson's diffusion model accommodates.

The closest stone-by-stone parallel may be the Megalithic Temples of Malta — Ġgantija on Gozo, Hagar Qim, Mnajdra, Tarxien — built between roughly 3600 and 2500 BCE. Until Schmidt published the Göbekli Tepe radiocarbon data, Ġgantija was widely cited as the world's oldest free-standing stone architecture; the southerly Ġgantija temple dates to about 3600 BCE, and UNESCO continues to call the Maltese temples among the earliest free-standing stone structures on Earth. Schmidt's results pushed that title back by roughly six thousand years. The two traditions share corbelled niches, large standing slabs, and a clear ceremonial focus, but the Maltese temples were built by a fully agricultural society with pottery, domesticated grains, and sheep-goat herding — exactly the cultural package that did not exist at Göbekli Tepe.

The other comparison Göbekli Tepe forces is timeline. Eridu in southern Mesopotamia is the city the Sumerian King List names as the first place where kingship descended from heaven. Founded around 5400 BCE in the early Ubaid period and rebuilt across eighteen superimposed mudbrick temples beneath the eventual Ur III ziggurat of Amar-Sin (c. 2047–2039 BCE), Eridu represents the Old World's clearest case of a city growing out of a temple. Fuad Safar and Seton Lloyd's 1946–1949 excavations, finally published in 1981 as Eridu by the Iraq Ministry of Information, traced the unbroken stratigraphy from the small Ubaid shrine to the late ziggurat platform.

The comparison with Göbekli Tepe is not "they are similar." It is that Eridu, founded forty-six centuries after Göbekli Tepe was buried, is the earliest Old World case where ceremonial monumentality, permanent residence, surplus storage, and recognizable proto-urbanism all coincide. Göbekli Tepe has the first two — ceremonial monumentality and possibly periodic residence — but not the rest. Schmidt's inversion of V. Gordon Childe's Neolithic Revolution model, articulated across the 2006 and 2012 books and developed in Oliver Dietrich, Manfred Heun, Jens Notroff, Klaus Schmidt, and Martin Zarnkow's 2012 paper "The role of cult and feasting in the emergence of Neolithic communities" (Antiquity 86, 674–695), proposed that the labor pull of feasting at sites like Göbekli Tepe — gazelle and aurochs bones in the fill suggest enormous meat consumption, possibly accompanied by fermented cereal beverages — may have been the catalyst that drove the domestication of wild einkorn and the transition to settled life. Eridu, on this reading, is not Göbekli Tepe's descendant; it is what eventually happens once the temple-pull thesis has worked its way through five thousand years of Mesopotamian aggregation.

The New World's parallel is Caral in the Supe Valley of north-central Peru, where Ruth Shady's excavations beginning in 1994 documented monumental platform mounds, sunken circular plazas, and quipu-precursors in a society that had no pottery, no metallurgy, no defensive walls, and no clear evidence of warfare. Caral's main occupation runs from roughly 3000 to 1800 BCE — still five thousand years younger than Göbekli Tepe. The comparison that holds is that ceremonial monumentality preceded both pottery and warfare in two unrelated traditions: Caral built sunken circular plazas without pottery in the New World; Göbekli Tepe built T-pillar enclosures without pottery in the Old. The comparison that does not hold is direct lineage. The Atlantic between them is a vacuum the evidence does not cross.

The comparison axis that does the most analytical work is the negative one. Of the dozens of peer sites in the Satyori ancient-sites corpus, every single one was built by a society that had crossed the Neolithic threshold — pottery, domesticated cereals or maize, herded animals, semi-permanent or fully sedentary villages. Göbekli Tepe is the only site in the published corpus built by mobile foragers without those technologies. That fact is what makes the conventional Neolithic Revolution narrative untenable in its 1930s form, and it is also what makes any "Göbekli Tepe is just an older Stonehenge" comparison category-mismatched at the foundation.

The pushback against Schmidt's "temple first" framing has been most clearly stated by E. B. Banning of the University of Toronto in "So Fair a House: Göbekli Tepe and the Identification of Temples in the Pre-Pottery Neolithic of the Near East" (Current Anthropology 52(5), 619–660, 2011). Banning argued that the sacred-versus-domestic distinction Schmidt used to read the enclosures is a modern Western imposition; that ethnographically, in many house-based societies the most elaborately decorated buildings are also residences; and that Göbekli Tepe's enclosures may be ritually charged houses rather than purpose-built temples without permanent occupation. Banning specifically pointed to dwelling-scale features in some areas of the mound, fill containing domestic refuse, and the lack of definitive evidence ruling out residence. Hodder's "history houses" interpretation at Çatalhöyük, also developed across 2006 and 2010, offers a coherent middle position: the buildings can be domestic and ritual at the same time, with elaboration tracking social memory rather than sacred-profane separation.

Schmidt and his successors at the Deutsches Archäologisches Institut have replied through the Tepe Telegrams blog and follow-up papers, conceding that small Layer II dwellings exist but maintaining that the central Layer III enclosures show no in-situ hearths, food-processing residues, or sleeping platforms, and that their fill is dominated by feasting debris rather than household waste. The argument is unresolved. The site was buried by a combination of deliberate fill and slope-collapse processes around 8000 BCE — the relative weights of those factors are now contested within the DAI team that succeeded Schmidt, where current Tepe Telegrams writers describe the filling as much more complex than the original purely-deliberate model proposed. The 5% problem — only a fraction of the mound has been excavated — keeps the question open. What the comparison axis demonstrates is that Göbekli Tepe's outlier status is not rhetorical: there is no other site in the corpus where the residence/ritual question even has to be asked at this scale, because every other site has obvious houses.

The provocative astronomical readings — Sweatman and Tsikritsis's 2017 paper in Mediterranean Archaeology and Archaeometry 17(1), 233–250, arguing that Pillar 43's "Vulture Stone" encodes the date of the Younger Dryas comet impact around 10,950 BCE; Andrew Collins's Cygnus alignment hypothesis; Giulio Magli's 2013 Sirius-rising proposal — sit inside this same anti-comparison frame. They are interesting precisely because Göbekli Tepe is anomalous. Notroff, Dietrich, Dietrich, and colleagues' reply, "More than a vulture: A response to Sweatman and Tsikritsis" (Mediterranean Archaeology and Archaeometry 17(2), 57–63, 2017, with eight authors including Lelek Tvetmarken, Kinzel, Schlindwein, Sönmez, and Clare), made three concrete objections: the Sweatman date for Pillar 43 (10,950 ± 250 BCE) is roughly 1,400–1,600 years earlier than the calibrated radiocarbon date for the Enclosure D wall plaster (Dietrich et al. 2013); Pillar 43 has additional reliefs on its narrow sides that the constellation reading does not account for; and projecting Greek-era constellation boundaries back onto Pre-Pottery Neolithic carvings assumes asterism-stability that is not supported by any independent evidence. The Sweatman hypothesis has not been ruled out — it has been challenged on specific grounds, which is the right way for a contested claim to live in the literature. The 2020 geometric analysis by Gil Haklay and Avi Gopher in Cambridge Archaeological Journal 30(2), 343–357, which found that the centers of Enclosures B, C, and D form a near-equilateral triangle to within centimeter-scale tolerance, is a more cautious archaeoastronomical claim that has not drawn the same contestation.

What this network of comparisons reveals about Göbekli Tepe is more limited than the popular framing usually allows. The site is not the first stone circle, the first temple, or the first calendar — those titles depend on prior definitional work that the comparisons above complicate. What Göbekli Tepe demonstrably is, in the current literature, is the oldest known site in the world where mobile hunter-gatherers cooperated to quarry, transport, and erect multi-ton carved stone monuments at landscape scale, repeated across multiple enclosures, over centuries, and then deliberately buried. No other entry in the Satyori library does all of those things in that combination. Every comparison either narrows on one axis (Karahan Tepe, the closest sibling, lacks the T-pillar density and the Vulture Stone iconography) or opens an immense chronological gap (Stonehenge, Malta, Eridu, Caral, Çatalhöyük all sit one to seven millennia downstream). The site's place in the ancient world is not "ancestor of everything that came after." It is "the existing baseline against which every other early monument has to be re-explained."

Significance

The Göbekli Tepe comparison work matters because it forces a definitional reckoning the popular framing usually sidesteps. Klaus Schmidt's 2006 inversion of V. Gordon Childe's Neolithic Revolution model — that the temple may have preceded the farm, not followed it — only carries weight if Göbekli Tepe is genuinely chronologically and culturally distinct from every other monument. The five comparison axes above (Karahan Tepe siblings, Çatalhöyük post-pottery parallels, the Stonehenge-Carnac-Malta megalithic horizon, the Eridu and Caral urbanism timelines, and the anti-comparison with every other post-agricultural peer) confirm that distinctness in the current literature. E. B. Banning's 2011 challenge in Current Anthropology sharpens rather than dissolves the case: even the strongest critique concedes that no peer site in the corpus poses the temple-versus-house question at Göbekli Tepe's scale, because every other site has unmistakable houses. The site's position is not "first of a familiar type" but "outside the type the corpus was built to compare."

Connections

Göbekli Tepe — the parent entity. This sub-page focuses on cross-site comparisons; the parent covers Göbekli Tepe's stratigraphy, iconography, deliberate burial, and the Schmidt-era excavation history in standalone depth.

Karahan Tepe — the closest sibling site, sixty kilometers east in the same Tas Tepeler cluster, with phallic-stelae chambers and overlapping ninth-millennium-BCE radiocarbon dates.

Çatalhöyük — the Anatolian Neolithic mound (c. 7400–5600 BCE) that inherits the wild-animal symbolic register and skull-cult practices in modified form, six hundred to a thousand years downstream.

Stonehenge — the canonical megalithic peer, c. 2500 BCE for the central sarsens, roughly seven thousand years younger than Göbekli Tepe.

Carnac Stones — Brittany's PPN-era megalithic horizon (c. 4500–2300 BCE), the upstream end of Schulz Paulsson's 2019 Atlantic-megalithic diffusion model that Göbekli Tepe sits outside.

The Megalithic Temples of Malta — Ġgantija and the Tarxien complex (c. 3600–2500 BCE), the title-holder for "oldest free-standing stone architecture" until Göbekli Tepe pushed the threshold back by six millennia.

Eridu — the Sumerian "first city" founded c. 5400 BCE; relevant as the timeline anchor for when ceremonial monumentality and urbanism finally coincide in Mesopotamia.

Caral — the Norte Chico ceremonial complex in Peru (c. 3000–1800 BCE), the New World parallel for pre-ceramic monumentality without warfare or metallurgy.

Avebury — the larger but later British henge (c. 2850–2200 BCE) included for completeness in the European megalithic-precision comparison.

Further Reading