A Sanctuary … or so fair a House?

Göbekli Tepe is situated on the highest point of the Germuş mountain range in southeastern Turkey. The spot is hostile to settlement; the next accessible springs are located in a distance of about 5 km northeast (Edene) and to the southeast (Germuş). A number of pits at Göbekli Tepe’s western slope could represent cisterns to collect rain water; although their exact date could not have been determined yet. With a total capacity of 153,12 cubic metres (cf. Herrmann-Schmidt 2012) they may have accumulated enough water for people to stay there for a longer periods of time, but probably not during the whole rainless summer. The next Neolithic settlements so far known were found in the plain in immediate vicinity of nearby springs, like for example Urfa-Yeni Yol.

From its discovery onwards, the interpretation of Göbekli Tepe’s suprising architecture has centered around the terms ‘special purpose buildings’ (Sondergebäude), ‘sanctuaries’, or even ‘temples’. Naturally, this line of interpretation has been called into question. As already discussed here, it is indeed quite challenging to use a rather strictly defined historical terminology and complex spiritual concepts to describe the material remains of prehistoric phenomena. Even more while cult, ritual and ultimately religion are concepts often cited but rarely clearly defined by archaeologists.

Just recently a colleague challenged the existence of pure domestic or ritual structures for the Neolithic, arguing that archaeologists tend to impose modern western distinctions of sacred vs. profane on prehistory, while anthropology in most cases shows these two spheres to be inseparably interwoven (Banning 2011, 624-627). In his eyes, Göbekli Tepe rather was a settlement with buildings rich in symbolism, but nevertheless domestic in nature. Undisputedly, this boundary is perceived much stricter today after centuries of secularization in the western hemisphere, although it should be noted that this differentiation indeed also is known from non-western societies, too. Banning’s arguments that in-house inhumations, caches and wall paintings are demonstrating that ‘the sacred’ clearly is leaking into everyday live in the Near Eastern Neolithic (Banning 2011, 627-629) and that therefore a clear distinction is impossible to define, is valid, too, of course. In fact the idea of manifestations of the sacred in houses or parts of houses is neither new, nor surprising as already M. Eliade pointed out in his seminal work on the entanglement of the sacred and profane. Yet Eliade also emphasized that belief and faith of course could focus within special places and structures particularly dedicated to give ‘the sacred’ a room: “… the sanctuary – the center par excellence was there, close to [man], in the city, and he could be sure of communicating with the world of the gods by entering the temple.” (Eliade 1959, 43). All this is essentially theoretical thinking, based on historical sources and ethnologic observation. But going back to prehistoric periods which are denying such direct access, we are thrown back again at a selection of what is left physically and intentionally – exclusively. In case of the enclosures unearthed at Göbekli Tepe this means to focus on the material culture found in this context and the structures themselves.

Göbekli Tepe_Fig. 3

Pillar 31, one of the central pillars of Enclosure D, illustrates the anthropomorphic appearance of the T-shaped pillars due to the depiction of arms, hands, and a loincloth. (Photo: N. Becker, DAI)

Among these, still the monumental T-shaped pillars can be regarded as the site’s most prominent and most defining moment. While they remain faceless, the depiction of arms, hands, and clothing clearly identifies these up to 5.5 m high pillars as anthropomorphic, but distinctively also larger than life at the same time. Their highly abstracted character must be considered intentional, in particular since we know of the existence of more naturalistic and life-sized sculptures like for example the contemporaneous ‘Urfa man’ and numerous heads of similar sculptures discovered at Göbekli Tepe. So, even though we cannot know if these buildings actually were really meant to house gods or deities, the peculiar role of these larger-than-life anthropomorphic images forming the centre and main element of the enclosures at Göbekli Tepe remain conspiciously disctinctive to the life-sized sculpture heads which were apparently carefully deposited in the backfill.

Early Neolithic domestic architecture is well known in the upper Euphrates region due to the long and secure stratigraphy of rectangular buildings at Çayönü Tepesi (Schirmer 1988; 1990; Özdoğan 1999) and extensive excavations at Nevalı Çori (Hauptmann 1988) for instance, both stiuated in Turkey. Contemporaneous with Göbekli Tepe in this sequence would be Çayönü’s so-called grillplan-phase (PPNA), the ‘channeled’ ground plans (early PPNB; attested also in Nevalı Çori), and the ‘cobble paved buildings’ (middle PPNB). Research of the last 20 years in the region has revealed that almost every settlement site of the 10th and 9th millennium BC, which was excavated more extensively, shows a spatial distinction into living quarters and workshop areas and furthermore produced special buildings or free spaces for apparently communal or ritual activity. Characteristic traits of these so-called special purpose buildings are benches at the inner walls, rich and elaborate inner fittings as well as outstanding installations and finds like (stone) sculptures and sometimes human burials – as the examples of Nevalı Çori’s ‘Terrazzo Building’, Çayönü’s ‘Skull’, ‘Terrazzo’ and ‘Flagstone Buildings’ or the communal buildings at Jerf el Ahmar and Mureybet (northern Syria) demonstrate, to just name some.


‘Special purpose buildings’ of the PPN: 1. Çayönü, ‘Flagstone Building’ (after Schirmer 1983, fig. 11c), 2. Çayönü, ‘Skull Building’ (after Schirmer 1983, fig. 11b), 3. Çayönü, ‘Terrazzo Building’ (after Schirmer 1983, fig. 11a), 4. Nevalı Çori (after Hauptmann 1993, fig. 9), 5. Jerf el Ahmar (after Stordeur et al. 2000, fig. 9), 6. Mureybet (after Stordeur et al. 2000, fig. 2), 7. Jerf el Ahmar (after Stordeur et al. 2000, fig. 5).


Reconstruction of the ‘Terrazzo Building’ at Nevalı Çori where T-Pillars were found for the first time. (Photo: H. Hauptmann, reconstruction: N. Becker, DAI.)

At Göbekli Tepe no traces of this well-documented typical domestic PPN architecture could have been proven as of yet. But the existing structures at the site clearly mirror features and layout of those outstanding communal ‘special purpose’ buildings which usually are the exception within settlements. At Göbekli Tepe, however, this building type is not an exception, but the general rule – almost overrepresented compared to other settlement sites, while whole object classes (like clay figurines for instance) known from these settlements are almost completely absent.

Summing up, from our point of view there seems to be ample evidence to interpret Göbekli Tepe as a peculiar place formed of special purpose structures related to cult and ritual with distinct and fixed life-cycles of building, use, deconstruction and burial. All of these stages seem to be marked by specific ritual acts, of which the last, i.e. those related to burial and deposition of symbolic objects are naturally best visible in the archaeological record. What remains is largely a problem of adequate terminology to address these buildings and the site as a whole. If ‘temple’ is understood as a technical term for specialized cult architecture, one could indeed consider this lable for Göbekli Tepe. If the term is defined in our western perception as a place where a god is present, maybe ‘sanctuary’’ would be a more neutral description; alternatively the auxiliary construction of ‘special purpose buildings’ (Sondergebäude) may be used to escape any trap of culturally bound denominations. But in any case one thing is sure: the idea that Göbekli Tepe’s buildings are “so fair a house” seems not the most convincing interpretation of the available evidence so far.

A more detailed discussion of this question can be found in:

O. Dietrich and J. Notroff, A sanctuary, or so fair a house? In defense of an archaeology of cult at Pre-Pottery Neolithic Göbekli Tepe. In: N. Lanerie (ed.), Defining the Sacred. Approaches to the Archaeology of Religion in the Near East. Oxford & Philadelphia: Oxbow 2015, 75-89.


E. E. Banning, So Fair a House: Göbekli Tepe and the Identification of Temples in the Pre-Pottery Neolithic of the Near East, Current Anthropology 52/5, 2011, 619-660.

M. Eliade, The Sacred and the Profane. New York: Brace & World 1959.

H. Hauptmann, Nevalı Cori: Architektur, Anatolica XV, 1988, 99-110.

H. Hauptmann, Ein Kultgebäude in Nevalı Cori. In: M. Frangipane, H. Hauptmann, M. Liverani, P. Matthiae and M. Mellink (eds.), Between the Rivers and Over the Mountains. Archaeologica Anatolica et Mesopotamica Alba Palmieri dedicata. Rome: Università di Roma “La Sapienza”, 37-69.

R. A. Herrmann and K. Schmidt, Göbekli Tepe – Untersuchungen zur Gewinnung und Nutzung von Wasser im Bereich des steinzeitlichen Bergheiligtums. In: F. Klimscha, R. Eichmann, C. Schuler and H. Fahlbusch (eds.), Wasserwirtschaftliche Innovationen im archäologischen Kontext. Rahden/Westf.: Verlag Marie Leidorf GmbH, 2012, 57-67.

A. Özdoğan,  Çayönü. In: M. Özdoğan and N. Başgelen (eds.), Neolithic in Turkey. Istanbul: Arkeoloji ve Sanat Yayınları, 1999, 35-63.

W. Schirmer, Zu den Bauten des Çayönü Tepesi, Anatolica XV, 1988, 139-159.

W. Schirmer, Some Aspects of Building at the ‘Aceramic Neolithic’ Settlement of Çayönü Tepesi, Wolrd Archeology 21/3, 1990, 363-378.

D. Stordeur, M. Brenet, G. der Aprahamian and J. C. Roux, Les bâtiments communautaires de Jerf el Ahmar et Mureybet horizon PPN A (Syrie), Paleórient 26/1, 2000, 29-44.

Could we really call it a ‘temple’?

Of course, magazines have to sell stories – and superlatives always are a good argument in this case. People just love to hear about the biggest, oldest, and most spectacular. And what could be more spectecular than a headline like “The Oldest Temples in the World”? That’s how you sell a find, don’t you? Yet, as scientists we need to show some healthy reservation – in particular when dealing with such phrases and terms which obviously have developed a certain history on their own. It’s all too easy to make up a good story or ‘hypothesis’, but substantiating such proposition is where real research actually starts.

Against the background of the historical definition of ancient Roman or Greek or Near Eastern temples for instance, this peculiar type of building implicitly forms places to worship a deity or deities in our language use – the existence of this concept of ‘divinity’ is crucial to the temple as home of a god or goddesses in antiquity. It is a futile task trying to answer this complex question based on the archaeological record exclusively. We know to identify the temples of ancient Rome and Greece and the Near East and to name the gods these were housing due to the written record those cultures have left to us. Delving deep into the prehistory of the Anatolian Neolithic, however, confronts us with a sudden lack of any sources other than the material record. The challenge in relying to physically tangible sources solely to grasp rather spiritual concepts is obvious. Yet, the material culture of Göbekli Tepe and related sites and the elements of monumental architecture in particular may offer a lead worth following.

Göbekli Tepe_Fig. 2

Main excavation area with monumental PPN A enclosures (Photo: N. Becker, DAI).

The T-shaped pillars forming the major and most prominent feature of Göbekli Tepe’s architecture need to play a crucial role in our observations here. While large and highly abstracted, they also clearly own human characteristics: some of these pillars show arms on their sides and hands brought together above the abdomen. There are elements of clothing depicted in relief as well: stola-like garments draped around pillars’ shoulders and fox-skin loincloths depicted dangling from belts. This emphasizes quite impressively that the T-pillars apparently have to be understood as monumental anthropomorphic sculptures. Most interestingly, however, is that they are always depicted faceless. There are no eyes, no nose or mouth present, these pillar-statues remain bereft of individuality on first glance – only to be distinguished, at least in the case of the central pillars of Enclosure D for example, by peculiar symbols below their heads – not unlike where one would wear necklaces. So, while still nameless to us, the Neolithic people may well have recognized who it was depicted here towering above them.

Pfeiler 18 mit Podest 2

In particular the central pillars of Enclosure D illustrate the anthropomorphic character of Göbekli Tepe’s T-shaped pillars (3D-model: HS Karlsruhe; Photos: N. Becker, DAI).

With a height of about 5.5 m it is particularly the T-pillars’ larger-than-life appearance which seems so remarkable – especially given that their highly abstract character is intentional and not to the result of deficient craftsmanship. Apart from the numerous animal sculptures uncovered at Göbekli Tepe, the so called ‘Urfa Man’ gives witness to Neolithic sculptors’ ability to portray the human body naturalistically. This oldest known statue of a man, about life-size, was found during construction work in the area of the Pre-Pottery Neolithic site of Urfa-Yeni Yol.  In contrast to the cubic and faceless T-pillars, whose identity and meaning apparently seems to a different one, ‘Urfa Man’ has a face, his eyes depicted by segments of black obsidian sunk into deep holes (a mouth, however, is missing). From Göbekli Tepe there are known several limestone-heads, too. They have a breaking edge in the neck area indicating that they originally were part of larger statues much like ‘Urfa Man’ himself.


So-called Urfa Man is considered the oldest known life-sized sculpture of a man (Photo: J. Notroff, DAI).

Göbekli_ZOrA_Abb. 17

Collection of limestone heads, supposedly parts of sculptures similar to ‘Urfa Man’, from Göbekli Tepe (Photos: N. Becker, DAI).

As already noted in the beginning, we know little of the beliefs these people might have followed, so it would seem rather bold to denote these monumental pillar-statues as personifications of ‘deities’. But faceless, larger than life and highly abstract, they clearly seem to be set on a quite different level than the naturalistic life-sized sculptures like ‘Urfa Man’ and the Göbekli Tepe stone heads. They seem to represent something more, supposedly something beyond the self-referential depiction of human beings. Together with the obviously narrative character of other depcitions on these T-pillars which clearly exceed simple decorative purposes, this perception feeds the impression that we are confronted here with a complex iconography – with mythological narrations probably even.

Göbekli Tepe_Fig. 11

P43 emphasises the narrative character of the T-pillars’ relief (Photo: B. Steinhilber).

It is these T-pillars in particular which form the centre and most important element of the site of Göbekli Tepe, so they naturally become a strong argument in the interpretation of these enclosures as well. If we after all would like to call them ‘temples’ or still hesitate to use this term finally comes down to the definition one applies. But differing so noticeably from the well-known general types of contemporary settlement patterns (and also apparently lacking most of the material culture which is so typically for clearly domestic contexts), we confidently name these structures ‘communal’ or ‘special purpose buildings’ with all due scientifical propriety. This is even more compelling since apparently almost every settlement site of the period and region seems to have produced at least one comparable communal structure of similar design and layout. Only at Göbekli Tepe there is a noticeable cumulation of this peculiar building type – but this should be topic of another contribution.

Further reading
N. Becker, O. Dietrich, Th. Götzelt, Ç. Köksal-Schmidt, J. Notroff, K. Schmidt, Materialien zur Deutung der zentralen Pfeilerpaare des Göbekli Tepe und weiterer Orte des obermesopotamischen Frühneolithikums, Zeitschrift für Orient-Archäologie 5, 2012, 14-43.

O. Dietrich, J. Notroff, A sanctuary, or so fair a house? In defense of an archaeology of cult at Pre-Pottery Neolithic Göbekli Tepe, in: N. Lanerie (ed.), Defining the Sacred. Approaches to the Archaeology of Religion in the Near East. Oxford & Philadelphia 2015, 75-89.

J. Notroff, O. Dietrich, K. Schmidt, Gathering of the Dead? The Early Neolithic sanctuaries of Göbekli Tepe, Southeastern Turkey, in: C. Renfrew, M. J. Boyd and Iain Morley (eds.), Death Rituals, Social Order and the Archaeology of Immortality in the Ancient World. “Death Shall Have no Dominion”, Cambridge 2016, 65-81.