First came the temple, then the city

What was the purpose of 11,000 year old monument at Göbekli Tepe?

Located on a limestone ridge 15 km (9 miles) from the town of Şanlıurfa in southeastern Turkey is a site unlike any other known in the early Neolithic world of Southwest Asia. Göbekli Tepe is thought to be the world’s oldest temple. It comprises a series of stone circles that draw superficial comparison to Stonehenge, but it predates the well-known Salisbury Plain monument by seven millennia.

Göbekli Tepe was noted as far back as the early 1960s, but was largely ignored for thirty years. Not until 1994 was it visited by German archaeologist Klaus Schmidt, who believed that the site was Neolithic. He began excavating there the following year, and work has been ongoing ever since.

At the lowest level of the site, Layer III, Schmidt discovered series of semi-submerged circular or oval enclosures. Each comprises a dry-stone wall, into which up to twelve T-shaped limestone pillars are set, often joined to one another by stone benches. At the centre of each enclosure are two more pillars, which tend to be larger than the surrounding ones. The pillars range in height from 3 to 5 m (10 to 16 ft.) and weigh up to 10 tonnes. They were quarried from limestone plateaus close to the site, where a number of incomplete pillars remain in situ. One weighs over 50 tonnes, larger than any of the finished pillars so far excavated. Currently, four enclosures, designated A to D, are undergoing excavation, but geomagnetic surveys suggest that least twenty exist. Many of the pillars are carved with bas-reliefs of animals, including snakes, wild boar, foxes, lions, aurochs, wild sheep, gazelle, onager, birds, various insects, spiders, and scorpions. Where sexual characteristics are present, they are always male. The images are large, often life-size, and semi-naturalistic in style. Some pillars exhibit pairs of human arms and hands, suggesting that they represent stylised anthropomorphic beings. However, it is unclear as to whether they represent gods, shamans, ancestors, or even demons. There are also a number of mysterious abstract symbols that have been interpreted as pictograms (Schmidt, 1995; Schmidt, 1998; Schmidt, 2000; Schmidt, 2003; Peters & Schmidt, 2004).

Pictograms are graphic symbols used to convey meaning, often by pictorial resemblance to a physical object. They are widely used in present-day road and other public signage to denote traffic lights, pedestrian crossings, speed cameras, etc. If the Göbekli Tepe symbols were indeed pictograms, then the origins of writing may extend back into the early Neolithic, thousands of years before the appearance of writing systems such as cuneiform and hieroglyphic script.

No traces of houses have been found and there is little doubt that Göbekli Tepe was a ritual centre, possibly the first of its kind anywhere in the world (Schmidt, 1998). Unlike Stonehenge, the people who built Göbekli Tepe lacked a mixed farming economy. This overturned the conventional wisdom that such major projects could only be realised by fully-established farming communities. “First came the temple, then the city”, as Schmidt put it. How are we to interpret this temple?

One possibility is that the animals depicted in the various enclosures are totemic. It could be that the site was frequented by a number of groups, each of which identified itself with a different animal or animals and travelled to the site to perform rituals in its own particular enclosure (Peters & Schmidt, 2004). Another possibility is that Göbekli Tepe was associated with shamanistic practices (Lewis-Williams & Pearce, 2005).

A project on the scale of Göbekli Tepe would have required a large number of labourers and craftsmen. Coordinating the activities of all these people, to say nothing of providing them all with food and shelter, would have been a major undertaking. It should also be remembered that unlike the builders of Stonehenge, the Göbekli Tepe people were still not yet full agriculturalists. Such an undertaking was almost certainly beyond the capabilities of a few shamans and their communities. Instead, it seems likely that the monument was constructed by a hierarchical, stratified society, with powerful rulers. The shamans might have had more in common with priests (Peters & Schmidt, 2004). The link between rulers and religion, so prevalent in later times, might have already started to take shape.

The totemic and shamanistic explanations are not necessarily mutually exclusive, and if the totemic view is correct, then it possible that animals depicted in each enclosure could provide clues as to the origins of particular groups. For example, wild boar depictions predominate in Enclosure C. This suggests a group originating from the north, where pigs account for up to 40 percent of the animal remains found at PPNA sites. Combinations of wild boar with aurochs and cranes, as seen in Enclosure D, suggest an ecotone of steppe and river valley, such as along most water courses in the Euphrates and Tigris drainage regions (Peters & Schmidt, 2004).

Eventually, the enclosures at Göbekli Tepe complex were filled in and buried with debris. Animal remains and stone artefacts mixed in with the soil suggest that the filling material came from a typical late PPNA settlement refuse dump. The settlement has not been found, but the amount of debris involved suggests that it was not far away. Subsequently, a far less impressive complex was constructed over the first, comprising rectangular pits with smaller pillars, averaging about 1.5 m (5 ft.) (Peters & Schmidt, 2004).

Establishing a chronology for the site is difficult. Plant remains from the settlement debris have been dated to around 9000 BC (Kromer & Schmidt, 1998), but this does not tell us when the site was first occupied. Assuming that the debris accumulated while the Layer III site was in use, the first occupation of the site would be no later than this date. Based on dates for soil overlaying the filling debris, the Layer III complex was probably buried around 8000 BC (Peters & Schmidt, 2004). Dates for carbonates formed on the stone walls as a result of their burial suggest it could have been no later than 7700 BC (Pustovoytov, 2002).

Just why the Layer III complex was buried and the Layer II complex built over it is not known. A possible clue comes from the site of Nevali Çori, 30 km (18 miles) away. Unfortunately, this site was submerged following the construction of the Atatürk Dam. Prior to flooding, the site was excavated between 1983 and 1991 by Harold Hauptmann from the University of Heidelberg. The site was first occupied around 8500 BC, at the start of the PPNB, and occupation spanned three phases before final abandonment around 7600 BC ( Ex Oriente eV Scientific Society, 2011). It comprised some 29 rectangular multi-roomed houses and a ‘cult building’ – marking a shift from circular houses to the rectangular constructions that have largely characterised human dwellings ever since. The cult building dates to the site’s second and third phases. It was approximately square, measuring 13.9 by 13.5 m (45 ft. 7 in. by 45 ft. 4 in.), and was cut about 3 m (10 ft.) into the slope behind it. Access was via two downward steps. A stone bench ran all the way around the interior, broken by pillars similar to those at Göbekli Tepe and again surrounding a central pair, although they resembled the Hebrew letter ד (daleth) rather than the letter T (Peters & Schmidt, 2004; Lewis-Williams & Pearce, 2005). There is clearly a connection between the two sites, and possibly the shift to rectangular architecture is why Göbekli Tepe was filled in and rebuilt along rectangular lines.


1. Schmidt, K., Investigations in the early Meospotamian Neolithic: Göbekli Tepe and Gürcütepe. Neo-Lithics (2/95), 9-10 (1995).

2. Schmidt, K., Beyond Daily Bread: Evidence of Early Neolithic Ritual from Göbekli Tepe. Neo-Lithics (2/98), 1-5 (1998).

3. Schmidt, K., Göbekli Tepe, Southeastern Turkey A Preliminary Report on the 1995-1999 Excavations. Paléorient 26 (1), 45-54 (2000).

4. Schmidt, K., The 2003 Campaign at Göbekli Tepe (Southeastern Turkey). Neo-Lithics (2/03), 3-8 (2003).

5. Peters, J. & Schmidt, K., Animals in the symbolic world of Pre-Pottery Neolithic Göbekli Tepe, south-eastern Turkey: a preliminary assessment. Anthropozoologica 39 (1), 179-218 (2004).

6. Lewis-Williams, D. & Pearce, D., Inside the Neolithic Mind (Thames & Hudson, London, 2005).

7. Kromer, B. & Schmidt, K., Two Radiocarbon Dates from Göbekli Tepe, South Eastern Turkey. Neo-Lithics (3/98), 8-9 (1998).

8. Pustovoytov, K., 14 C Dating of Pedogenic Carbonate Coatings on Wall Stones at Göbekli Tepe (Southeastern Turkey). Neo-Lithics (2/02), 3-4 (2002).

9. Ex Oriente eV Scientific Society, PPND – the platform for Neolithic Radiocarbon Dates, Available at (2011).