PUBLIC RESEARCH PREVIEW Research in progress · uncertainties and source access are recorded throughout.

SITE-042 · 2026-10-30

Göbekli Tepe

About 9,600–8,000 BCE · a settlement with monumental communal buildings

GöbeklitepeGobekli TepeTürkiyeAnatoliaNeolithicMonumentsRadiocarbonSerpent imagery
Working research. Assessments and source-access limitations are retained from the research record. This entry has not completed its public-release review.

Why this discovery matters

Göbekli Tepe, near Şanlıurfa in southeastern Türkiye, demonstrates impressive communal construction and symbolic expression early in the Neolithic. UNESCO places its monumental buildings around 9,600–8,200 BCE: the earliest part is roughly 11,600 years before today. Its importance concerns the capabilities and organization of these communities, rather than the first appearance of humans. [S118 · UNESCO World Heritage Centre. Göbekli Tepe, World Heritage List 1572.]

The evidence challenges a simple sequence in which fully developed farming must come first and large communal buildings follow. However, plant exploitation, cultivation, domestication and permanent settlement are different processes. Göbekli Tepe does not establish that religion alone caused agriculture. [S119 · German Archaeological Institute. Göbekli Tepe research project 5746.] [S121 · Dietrich, L., Meister, J., Dietrich, O., Notroff, J., Kiep, J., Heeb, J., Beuger, A. & Schütt, B. (2019)]

Discovery and recognition

The Istanbul–Chicago survey identified the site as Neolithic in 1963. Klaus Schmidt recognized the significance of its worked stone and T-shaped pillars during renewed investigation in 1994; excavations began in 1995, in cooperation with the Şanlıurfa Museum and German Archaeological Institute. The excavation team’s historical account also describes local villagers guiding the visit. “Discovery” here means archaeological recognition and investigation of a locally known landscape. [S124 · Dietrich, O. (2016-06-02)]

What has been found

Round, oval and rectangular buildings contain limestone T-shaped pillars, some reaching approximately 5.5 metres. Arms, hands and clothing details identify certain pillars as stylized human forms. Wild-animal reliefs and abstract designs provide evidence for a sophisticated visual vocabulary; their precise narratives are not recoverable simply by naming the animals. [S118 · UNESCO World Heritage Centre. Göbekli Tepe, World Heritage List 1572.] [S121 · Dietrich, L., Meister, J., Dietrich, O., Notroff, J., Kiep, J., Heeb, J., Beuger, A. & Schütt, B. (2019)]

A 2019 study examined more than 7,000 food-processing artefacts using several complementary approaches. Grinding-tool wear, experiments and phytoliths support substantial cereal processing. The authors interpreted aspects of food production as provision for work feasts. Grinding cereals is evidence for plant use; by itself, it does not establish that those cereals were domesticated. [S121 · Dietrich, L., Meister, J., Dietrich, O., Notroff, J., Kiep, J., Heeb, J., Beuger, A. & Schütt, B. (2019)]

The chronology is not one date

Source and scopeReported calendar spanHow we use it
UNESCO monumental-building overviewc.9,600–8,200 BCEBroad monument record on the atlas and globe. [S118 · UNESCO World Heritage Centre. Göbekli Tepe, World Heritage List 1572.]
DAI project overviewc.9,500/9,250–8,000/7,750 cal BCEPreserve the source’s alternative approximate endpoints. [S119 · German Archaeological Institute. Göbekli Tepe research project 5746.]
Gresky & Clare’s 2026 settlement overviewc.9,600–8,000 BCEDossier overview, spanning PPNA and early/middle PPNB. [S120 · Gresky, J. & Clare, L. (2026)]

These summaries differ in scope and chronological framing. None is a single laboratory result or a confidence interval for every structure. We have not resolved their endpoint differences through a complete sample audit. The calendar BP plotting axis is referenced to 1950; the approximate “years ago” phrase above is not a conventional radiocarbon measurement.

Building D: the sample-to-building ledger

Open the Building D evidence ledger for 13 identifiable laboratory determinations, their materials, grids/loci, architectural relationships, published calibrations and unresolved phase assignments. Seven original calibration plots have been checked; six older apatite/carbonate outputs remain unverified. This is not yet a complete archive of all later excavation measurements. [S125 · Dietrich, O., Köksal-Schmidt, Ç., Notroff, J. & Schmidt, K. (2013)] [S126 · Platform for Neolithic Radiocarbon Dates (PPND), ex oriente. Göbekli Tepe, site 25; introduction by Oliver Dietrich; legacy entries cite Dietrich (2011)]

KIA-44149, charcoal in wall plaster between Pillars 41 and 42 (L9-68, locus 782.3), measured 9,984 ± 42 conventional radiocarbon BP. Its original plot preserves two intervals: 9,745–9,727 cal BCE (1.5%) and 9,675–9,314 cal BCE (93.9%), totaling 95.4%, using OxCal 4.1.7 and IntCal09. The small interval was omitted from the original caption, but is visible in the plot. No recalibration has been performed. [S130 · Dietrich, O. & Schmidt, K. (2010)] [S125 · Dietrich, O., Köksal-Schmidt, Ç., Notroff, J. & Schmidt, K. (2013)]

The crucial distinction is the age of carbon versus the event of applying plaster. The first report treated the date as construction evidence. The 2022 architectural account associates this plaster with a late repair. The sample cannot yet be assigned securely to first construction without reconciling its field position, wall sequence and possible inherited carbon. [S130 · Dietrich, O. & Schmidt, K. (2010)] [S128 · Breuers, J. & Kinzel, M. (2022)]

Three charcoal samples beside the ring walls need stratigraphic correlation. A tooth in basal fill dates animal carbon, not necessarily filling. Five bone-apatite measurements are retained but quarantined from phase dating because post-burial carbon exchange can bias them. Soil and carbonate formation describe still different events. [S125 · Dietrich, O., Köksal-Schmidt, Ç., Notroff, J. & Schmidt, K. (2013)] [S126 · Platform for Neolithic Radiocarbon Dates (PPND), ex oriente. Göbekli Tepe, site 25; introduction by Oliver Dietrich; legacy entries cite Dietrich (2011)] [S130 · Dietrich, O. & Schmidt, K. (2010)]

Rebuilding and filling were not one event

The 2019 section separates probable lower intentional fill from later slope deposits. The 2022 architectural analysis distinguishes repeated rebuilding; its phase estimates must remain separate from the laboratory records until the relevant sample links are recovered. [S121 · Dietrich, L., Meister, J., Dietrich, O., Notroff, J., Kiep, J., Heeb, J., Beuger, A. & Schütt, B. (2019)] [S128 · Breuers, J. & Kinzel, M. (2022)]

Excavations reported in 2024 add structural damage, roof remnants, voids in rubble and walking surfaces within the fill. These support repeated slope inundation, with people moving over surfaces in the partly buried building. An object in the fill could have been present beforehand or moved with debris; its position alone does not demonstrate a ritual burial deposit. [S127 · Clare, L. (2024)]

The ledger therefore distinguishes original construction, rebuilding, use, possible deliberate disposal, slope movement and later soil formation. A proposed earthquake explanation requires its own evidence and dating links; it is not adopted as an established event. [S129 · Kinzel, M. (2024)]

The later register: recovered, but not yet phase-linked

Explore the 2026 laboratory register for all 141 site-wide determinations in Clare’s Table S1. The 13 existing Building D / adjacent / overlying IDs match exactly in conventional age and error. These are repeated records, not 13 additional samples. The workbook supplies materials, Figure 2 exclusions and bibliographic references, but no loci or architectural phases. [S131 · Clare, L. (2026)]

Cross-checking the cited publications identifies 14 Building B records and 10 K10-13/23 sounding records. Those are not Building D dates; the remaining 104 rows have no recovered field-context link in this audit. A retained regional plotting date is not automatically a trustworthy construction date. [S132 · Breuers, J. (2022)] [S133 · Kinzel, M. & Clare, L. (2020)]

The register page compares the 2020 southeast-area phases with construction, modification, damage, repair and later occupation. It keeps that numbering separate from Building D’s own building phases. No new secure D sample-to-numbered-phase assignment has been established. [S133 · Kinzel, M. & Clare, L. (2020)]

Breuers’ D sediment section stops roughly 1.15 m above the floor. Its proposed final-fill chronology is based on lithic comparisons; it is not a directly dated floor-to-fill sequence. [S132 · Breuers, J. (2022)]

Field-context key: search outcome

The field-key search audit records no additional secure joins. Kinzel’s 2024 Table 1 is a damage inventory, not a laboratory register. Its repair-date discussion does not identify the underlying assays in the examined passage. [S129 · Kinzel, M. (2024)]

Schönicke’s 2025 dissertation is the named source for Poz-130450, Poz-130451, Poz-131050, Poz-131108 and Poz-131109. The repository lists the PDF, but full-text access failed; these five contexts remain unresolved. A 2023 microarchaeological study identifies sampling areas in DR-2 and Room 16, including floor locus L9-80-122 and pit fills 130, 132 and 134. These are geochemical sampling contexts, not an explicit key to those radiocarbon IDs. [S131 · Clare, L. (2026)] [S134 · Schönicke, J. (2025)] [S135 · Uzdurum, M. et al. (2023)]

The Pöllath study remains a bibliographic lead under two title variants. The final text was not located; this does not demonstrate that no field register exists. No excavation-team contact was made. [S132 · Breuers, J. (2022)] [S129 · Kinzel, M. (2024)] [S136 · Kinzel, M., Clare, L. & Sönmez, D. Built on Rock—Towards a Reconstruction of the Neolithic Topography of Göbekli Tepe. Istanbuler Mitteilungen 70 (2020)]

A settlement as well as a ritual place

The DAI now describes domestic activity and evidence for permanent settlement alongside the special buildings. The familiar image of an exclusively ceremonial hill, visited by people who lived elsewhere, is therefore too restrictive. Ritual importance remains compatible with everyday residence. [S119 · German Archaeological Institute. Göbekli Tepe research project 5746.] The 2024 report cautions that permanent occupation during the earliest PPNA phases remains inconclusive; evidence for settlement should not be projected uniformly across the entire sequence. [S127 · Clare, L. (2024)]

A paper published on 22 September 2026 reports two burial contexts: a collective burial found in 2017 in trench DR1, and an individual burial recognized in 2012 and excavated in 2022 in L09-65. The authors place them in PPNB contexts, broadly 8,700–8,000 BCE. This is a contextual period assignment, not a direct radiocarbon age for each person. [S120 · Gresky, J. & Clare, L. (2026)]

The study also supports erosion and slope movement as explanations for at least some dispersed human remains in building fills; it does not exclude other mortuary practices. A single deliberate burial event for the whole site should not be assumed. [S120 · Gresky, J. & Clare, L. (2026)]

Serpent imagery, with its context preserved

The project identifies a net-like arrangement of snakes on Pillar 1 in Building A. It is indexed as TR-005 in the traditions and imagery register. The original name and intended narrative are unknown. The source does not identify this motif as feathered or winged; we retain its snake identification without treating the depiction as evidence for an otherwise unidentified animal. [S123 · Dietrich, O. (2016-04-23)]

A second object, the decorated bench beneath the wild boar in Building D, carries three snakes among other motifs. It is indexed as TR-006. The monolith was reused; its carving is not directly dated by the suggested late placement of the boar. [S127 · Clare, L. (2024)]

Connections across the collection

Ohalo II offers an earlier comparison for plant processing, discussed by the cereal-study authors. Similar food-processing evidence does not by itself demonstrate direct transmission between communities. [S121 · Dietrich, L., Meister, J., Dietrich, O., Notroff, J., Kiep, J., Heeb, J., Beuger, A. & Schütt, B. (2019)]

For comparison with White Sands and Latnija, carry over the same question: what exactly was dated, and how does it relate to the human activity? A footprint surface, a hearth and a wall’s mortar have different formation histories. Comparing those relationships is more useful than ranking all three by a headline age.

Karahantepe and the regional comparison

Karahantepe now has a dossier and comparison table, with its own atlas/globe record. Shared forms support a regional tradition, while precedence and direct transmission remain unresolved. [S198 · German Archaeological Institute (12 February 2026)] Its building-closure interpretation must not override the evidence for repeated slope deposition at Building D. [S127 · Clare, L. (2024)]

The post–Younger Dryas comparison connects both settlements to the earlier daily-life collection without treating either as a pre–Younger Dryas profile. Nevalı Çori and Sayburç remain further research leads. [S124 · Dietrich, O. (2016-06-02)] [S120 · Gresky, J. & Clare, L. (2026)]

Questions to investigate next

  • Recover the field-context key for the 2026 register, especially the 70 rows labelled “This study”; continue tracing the phase-linked Pöllath study. The site-wide list is recovered, but its relationship to that earlier promised study is unverified.
  • Reconcile locus 782.3 and the plaster sample with the rebuilt walls and repair sequence.
  • Recover the six unverified legacy calibration outputs and clarify the 8880 ± 60 versus unnamed 8860 ± 60 BP soil reports; do not silently merge them.
  • Compare domestic and special buildings at Göbekli Tepe, Nevalı Çori and Karahantepe before inferring particular leaders, institutions or labor systems.
  • Track plant-use and domestication evidence separately from interpretations about feasting and belief.

Atlas and globe coverage

Three shared records are plotted at a generalized site location. The two broad records remain: AT-GT-MONUMENTS for the UNESCO monument overview and AT-GT-BURIALS for the PPNB burial contexts. Their ranges can overlap; this does not establish contemporaneity between a particular burial and a particular building phase. AT-GT-D-PLASTER separately plots the wall-carbon uncertainty envelope, including a clearly disclosed gap between calibration segments. It is not a construction date or occupation range. Unresolved phase assignments and unsuitable legacy measurements have an explicit unplotted record. No sample age has been substituted for the whole site. Atlas/globe coverage checked against the 2026 register: plotted bounds remain unchanged because no new defensible phase date was established; the unplotted explanation now records this result.

Explore Göbekli Tepe on the globe · Map and timeline · Later burial contexts

Building D carbon-age marker · Full sample ledger

Atlas/globe coverage checked for the Karahantepe comparison (2026-10-30): existing Göbekli Tepe records and limits remain unchanged. No new date or Building D phase assignment follows from regional resemblance.