As we arrive at the site in the mountians of southeastern Turkey, a pale moon still hangs in a sky shifting from black to blue. Groups of local workers have arrived minutes before by tractor from a village down the hill. Still dressed in coats and cardigans against the morning chill, they wait for the day’s excavation to start while the archaeologists collect their tools and instruments, equipment, and journals.
Finally, first light sounds the bell for the workday to begin, the sun rising just above the eastern horizon. Workers and archaeologists head to the excavation trenches, a caravan of shovels and buckets, of head scarves and hats.
Everyone knows his or her place and assignment; soon the air is filled with the sound of pickaxes and of chanting and laughing. Soil is shifted, rocks are moved. Basket after basket of debris is brought out of the trenches.
What has been brought to light here would have been thought of as impossible by archaeologists just 20 years ago. Four-meter-high monolithic limestone pillars—in a distinct shape resembling the letter T—are arranged in 10- to 30-meter-wide circles around sets of two, even larger, central pillars. The pillars are richly decorated. Many of them show animals: jumping foxes, snarling predators, bulls with heads lowered to attack. Others, especially the central pillars, have depictions of arms and hands, showing that the pillars are themselves abstract images of human beings.
What makes this site, Göbekli Tepe, so special is its early date. The structures described here belong to the 10th and early 9th millennia BCE, the so-called Pre-Pottery Neolithic (PPN), a period before the development of pottery when subsistence was still based on hunting and gathering and stone was the material of choice for tools. In this period, a development started that lies at the foundations of our modern world: a process labeled “Neolithization”—the gradual transition from hunting and gathering to food production, from small-scale highly mobile groups to large-scale permanently sedentary communities. Plants and animals became domesticated, the landscapes were altered to allow settlement and agriculture, and social inequalities developed.
The Transition to Permanent Settlements
Semipermanent settlements existed during the Epipalaeolithic (roughly 12,500 to 9600 BCE), but this process gained momentum in the Old World during the PPN, which is generally subdivided into an earlier PPNA phase (ca. 9600–8800 BCE) and a later PPNB (ca. 8800–7000 BCE). The transition from hunting and gathering to agriculture and the raising of livestock is set relatively late within the PPN, between 8300–8000 BCE. The reasons for that crucial change were long sought in environmental catalysts. When V. Gordon Childe coined the term “Neolithic Revolution” in 1936, he proposed climate change as the main driving force leading to permanent agricultural settlements. Increasing aridity was believed to have driven people to concentrate in oases; population pressure forced them to adopt innovative ways of food production.
Diametrically opposed are approaches that emphasize the importance of cultural-cognitive triggers. French archaeologist Jacque Cauvin´s book The Birth of the Gods and the Origins of Agriculture (1994) represents a programmatic text for this line of thought. Starting from the observation that a sheer explosion of imagery—a “revolution of symbols”—took place in the Epipalaeolithic and, thus, predated the adoption of agriculture by millennia, he reached the conclusion that cognitive changes, a new “psycho-cultural” mindset, allowed people to interact with their world in a completely new way—equally or even more important in the process of cultural change than ecological reasons. Cauvin developed his theory before the significance of Göbekli Tepe was understood. This site and other discoveries made during the last three decades in Upper Mesopotamia, a region seen for a long time at the periphery of the Neolithization processes, offer new evidence in support of the cultural-cognitive model.
It could be argued, however, that there is no single explanation as to why various hunter-gatherer communities moved toward permanent settlement. Perhaps some sites on the margins of the Neolithization process, those with no evidence of early agriculture like Göbekli Tepe in Upper Mesopotamia, were established to mark a changing worldview. This shift, tied to the beginning of cult or religion, then led to the domestication of plants and animals.
The Rediscovery of Göbekli Tepe
Göbekli Tepe, situated about 15 km northeast of the modern town of Şanlıurfa, was recognized as an archaeological site in 1963 during a large-scale survey conducted by the Universities of Istanbul and Chicago. Peter Benedict, the surveyor, described the site as a cluster of several mounds of reddish soil separated by depressions. He identified Göbekli Tepe as Neolithic, but did not understand the importance of the site.
Between 1983 and 1991, large-scale rescue excavations in advance of the construction of the Atatürk Dam were under way at another important PPNB Neolithic site: Nevalı Çori, 50 km to the north of Şanlıurfa. Under the direction of Harald Hauptmann, a settlement with large rectangular domestic buildings was excavated. However, excavations also revealed one building that was different from anything known from the Neolithic Near East. Not only were several monumental stone sculptures discovered, but the rectangular building had T- or Gamma (Γ)-shaped pillars running along the walls, interconnected by a bench, and a pair of T-shaped pillars in the center. These pillars could be understood as highly abstracted depictions of the human body, with representations of arms and hands. Differing so completely from the rest of this settlement’s architecture, the excavators concluded they had found a communal building, maybe used for ritual gatherings.
Nevalı Çori was flooded in 1991. But one of the members of the excavation team, Klaus Schmidt (1953–2014), wanted to find out whether similar settlements existed in the Urfa region. In 1994, he visited Göbekli Tepe. The moment of (re)discovery is best described in his own words:
“October 1994, the land colored by the evening sun. We walked through slopy, rather difficult and confusing terrain, littered with large basalt blocks. No traces of prehistoric people visible, no walls, pottery sherds, stone tools. Doubts regarding the sense of this trip, like many before with the aim to survey prehistoric, in particular Stone Age sites, were growing slowly but inexorably. Back in the village, an old man had answered our questions whether there was a hill with çakmaktaşı, flint, in vicinity, with a surprisingly clear ‘Yes!’ And he had sent a boy to guide us to that place….We could drive only a small part of the way, at the edge of the basalt field we had to start walking…. Our small group was made up of a taxi driver from the town, our young guide, Michael Morsch, a colleague from Heidelberg, and me. Finally, we reached a small hill at the border of the basalt field, offering a panoramic view of a wide horizon. Still no archaeological traces, just those of sheep and goat flocks brought here to graze. But we had finally reached the end of the basalt field; now the barren limestone plateau lay in front of us.…On the opposed hill a large mound towered above the flat plateau, divided by depressions into several hilltops.… Was that the mound we were looking for? The ‘knocks’ of red soil Peter Benedict had described in his survey report, Göbekli Tepe, or to be more precise, Göbekli Tepe ziyaret?…When we approached the flanks of the mound, the so far gray and bare limestone plateau suddenly began to glitter. A carpet of flint covered the bedrock, and sparkled in the afternoon sun….We assured ourselves several times: these were not flint nodules fragmented by the forces of nature, but flakes, blades and fragments of cores, in short, artifacts.…Other finds, in particular pottery, were absent. On the flanks of the mound the density of flint became lower. We reached the first long-stretched stone heaps, obviously accumulated here over decades by farmers clearing their fields….One of those heaps held a particularly large boulder. It was clearly worked and had a form that was easily recognizable: it was the T-shaped head of a pillar of the Nevalı Çori type.”
At that moment, Göbekli Tepe was nearly untouched and could be reached only by foot or horse. The tepe or tell (mound), with its height of up to 15 m and a diameter of 300 m, is the only colorful spot and a widely visible landmark on the otherwise barren Germuş mountain range at the edge of the Harran plain near the modern border to Syria. A first test trench was opened at the base of the southeastern slope, revealing rectangular buildings characteristic of what was later termed Layer II, dating to the early and middle PPNB (ca. 8800 BCE).
Excavating Monumental Enclosures
During this first field season in 1995, one of the local landowners, whose holdings extended into the mound’s southeastern depression, started to clear his fields of stones that obstructed ploughing. He uncovered the heads of two huge T-shaped pillars and started to smash one of these with a large sledgehammer. Fortunately, he was persuaded to stop, and, in 1996, work resumed in this area. What was brought to light here was the first of the monumental enclosures Göbekli Tepe is now so well known for, belonging to the site´s older Layer III, dating to the PPNA (9600–8800 BCE) and perhaps into the very early PPNB.
These PPNA enclosures are the most impressive part of Göbekli Tepe´s archaeology. Monolithic T-shaped pillars were arranged in 10- to 30-m-wide circular enclosures. The pillars in the circle, standing up to 4 m high, are interconnected by walls and benches and are always oriented towards a central pair of even larger pillars. All enclosures were backfilled, buried intentionally at the end of their use-lives.
It is still unclear whether they originally had roofs, however, much hints at subterranean structures. A geophysical survey confirmed that these monuments were not restricted to a specific part of the mound. More than ten enclosures have been located in addition to the nine already under excavation— the latter designated A to I in order of discovery.
Enclosure D is the largest and best preserved so far. Two huge monolithic T-central pillars are surrounded by a circle formed of—at the current state of excavation— 11 pillars. Most of these pillars show depictions of animals in flat relief: foxes, birds—such as cranes, storks, and ducks— and snakes are the most common species. The two pillars in the center, measuring about 5.5 m tall and weighing some 10 metric tons, were found on pedestals only 20 cm deep, which are, like the rest of the floor, carved out of the carefully smoothed bedrock. The central pillars demonstrate very well the human-like appearance of the T-shaped pillars. The oblong T-heads can be regarded as abstract depictions of the human head, the narrow side representing the face. Clearly visible are arms on the shafts with hands brought together above the abdomen (see page 10).
The depiction of belts and loincloths in the shape of animal skins underlines the impression that these pillars indeed have an anthropomorphic meaning. Since the loincloth reliefs cover the genital region of the pillarstatues, we cannot be sure about the gender of the two individuals depicted in the center. But some help may come from the clay figurines from the PPNB site of Nevalı Çori. Among these figurines, which depict both male and female individuals, only the males are wearing belts. Thus, it is highly probable that the pair of pillars in Enclosure D represents two male individuals, too. Indeed, it is striking that the iconographic and symbolic world present at Göbekli Tepe is one dominated by masculinity. Whenever the gender of one of the animals depicted is indicated, it is a male specimen. Of course, females may be represented when gender is unclear.
The smaller pillars in the circle walls seem to look towards the central, larger pair. The benches along the walls underline the impression of a gathering. Whatever gathering took place here, it does not seem to be one of equals, given the difference in size of the pillars. Another distinction exists between the clearly anthropomorphic, but abstract, pillars and more naturalistic human sculptures recovered in larger numbers from the site. In particular, several life-sized human heads were unearthed. They are made of limestone and must once have been part of large sculptures, as illustrated by their broken edges. The heads seem to have been intentionally removed and were in many cases deposited next to the pillars during refilling of the enclosures. While their exact relation to each other remains unknown, it seems possible to assume that both represent different hierarchical levels. Whether the pillars show important ancestors or deities is yet unclear.
Depictions of Animals
The elaborate reliefs on most of the T-pillars rarely depict single animals. More often they show complex, potentially narrative, scenes. One striking example is Pillar 43 in Enclosure D, whose whole broad western side is covered by a variety of motifs (see page 22). A big vulture is dominant, lifting its left wing while the right wing points towards a sphere or disc nearby. Although the meaning of depictions like this remains unknown, it is clear that they exceed simple decorative purposes. The complex iconography hints at important mythological content.
A distinctive variation in the animal spectrum is depicted in each circle. While in Enclosure A, the snake prevails, in Enclosure B, foxes are dominant. In Enclosure C, boars predominate, while Enclosure D is more diverse, but strongly emphasizes birds. All depicted animal species are wild and no domesticates are attested in the archaeofaunal remains at the site. Göbekli Tepe clearly is a place of hunter-gatherers. But how could small hunter-gatherer groups with little hierarchy build such a monumental site? Any answer needs to examine possible incentives for cooperative action. And a prime candidate is cult or religious belief.
Belief Moves Stones…and People
The Neolithic quarry areas that provided building material for the monuments are situated on the limestone plateaus surrounding the site. The maximum distances that had to be covered to bring workpieces to the enclosures were 600 to 700 meters. The terrain is uneven and sloping, and the megaliths are of impressive size. There are signs of ongoing construction, deconstruction, and reconstruction activity in the enclosures. Most likely the act of working at the site was central to the builders, and repeated periodically. It seems as if constant construction activity was a way to bring people together. And there is significant evidence for congregations at the site. Most important are the iconographic differences between the enclosures. It is possible that they reflect the emblematic or totemic animals of the different groups constructing these buildings.
Naturalistic depictions and abstract symbols are not only present on the pillars, but also on functional objects like shaft straighteners (grooved stone tools used for fabricating arrows) or bowls, as well as on small stone tablets which likely served no other function than to bear these signs. There is a strong possibility that this imagery was thoroughly readable; many elements and whole scenes repeat on several pillars or objects, and they appear to have a narrative character as shown above, likely codifying core elements of mythological/ritual knowledge. This typical iconography appears in concentrated form at Göbekli Tepe, but elements of it reappear in sites throughout Upper Mesopotamia (southeastern Turkey, northeastern Syria, and northern Iraq), outlining a community of people with a common symbolic background. Göbekli Tepe, and other sites (perhaps Nevalı Çori), may have served as central places where people gathered for construction work and to perform cultic acts inside the buildings. The massive amount of backfilling inside the enclosures of Göbekli Tepe provides an answer for a possible mode to gather larger groups of people. The material consists of limestone rubble from the quarries, flint artifacts, and animal bones smashed to get to the marrow, clearly the remains of large meals. Enclosure D alone, the largest of the four circles, comprised nearly 500 cubic meters of debris. With traces of settlement absent, this certainly fuels the idea of large, ritualized feasts.
Regular meetings and collective activities are crucial to hunter-gatherer societies in many ways, serving purposes such as the exchange of goods, marriage partners, and information; the transfer of innovations; and the strengthening of social cohesion in times of hardship. Cultic activities seem to have been a way to render regular meetings possible, not only by establishing fixed times, but also by providing a framework in which peaceful congregations of groups from different territories were possible.
OLIVER DIETRICH is a Research Assistant at the German Archaeological Institute (Deutsches Archäologisches Institut, DAI) in Berlin and has worked on the Göbekli Tepe Project since 2005. LAURA DIETRICH is a Research Assistant at DAI, with a focus on the archaeology of food. JENS NOTROFF is a Research Assistant at DAI and has worked on the Göbekli Tepe Project since 2006.
For Further Reading
Dietrich, O., Heun, M., Notroff, J., Schmidt, K., and M. Zarnkow. “The Role of Cult and Feasting in the Emergence of Neolithic Communities. New Evidence from Göbekli Tepe, South-eastern Turkey.” Antiquity 86 (2012): 674–695.
Schmidt, K., Göbekli Tepe. A Stone Age Sanctuary in South-Eastern Anatolia.</Berlin: ex oriente e.V., 2012.