Making headlines: Was Göbekli Tepe built by Aboriginal Australians?

Most astonishing news are spreading these days again. An article published November 14th in Epoch Times [external link] was claiming a truly amazing discovery already in it’s headline: “Australian Aboriginal symbols found on mysterious 12,000-year-old-pillar in Turkey”. Since then these news were repeatedly picked up and reproduced, and meanwhile we received many requests by interested readers here, asking if this was true.

This unusual hypothesis was based on superficial similarities between selected images and symbols occurring in both cultural contexts (Anatolian Pre-Pottery Neolithic Göbekli Tepe and Aboriginal Australia). Aboriginal signs and symbols apparently painted onto a medicine man’s chest or carved into stones are paralleled with reliefs on Göbekli Tepe’s pillars or other objects known from this and contemporary sites of the region.

Before having a closer look into these assumed analogies, a few general remarks should be noted. According to latest research, the Australian continent most likely was settled about 65,000 years ago (cf. Clarkson et al. 2017), probably via the Malay archipelago. So, both cultures were contemporary at some point? Yes, but in different parts of the world – with an ocean between them. And with an incredibly long period of isolation in case of the Australians (Bergström 2016). This obvious chronological gap and the huge geographical distance of about 15,000 km would make a direct relation and interaction of both cultural phenomena rather difficult.

On the other hand the key evidence, the Aboriginal elder’s chest symbol, is of a much younger date. The referenced depiction seems to be the reproduction of a photograph from the late 19th / early 20th century Spencer and Gillen anthropological expeditions through the continent (further information and resources on the website of the “Reconstruction the Spencer and Gillen Collection Project”: [external link]). The original photograph was first published in 1904 (Spencer & Gillen 1904, 58 Fig. 33) – significantly later, about circa 12,000 years later to be precise, than the Göbekli Tepe pillar was carved. Again quite a distance.


Fig. 1: A shaman or medicine man with extensive body painting, Worgaia, Central Australia. Process print. (Wellcome Collection:, CC BY 4.0)

Pillar 28

Fig. 2: Pillar 28 from Enclosure C with a symbol resembling the Aboriginal body painting only at first glance. However, upon closer inspection it becomes clear that there are significant differences. (Photo: K. Schmidt, emphasis: J. Notroff)

Yet the rather recent study of Aboriginal symbolism at least allows us to understand the icon in question here, describing two persons communicating. However, the supposed analogy at Göbekli Tepe is (at least in this combination) a much rarer exception in early Neolithic iconography – one of which we cannot say what it actually meant to the prehistoric people carving it, due to a lack of any related sources. Besides, and that might be the strongest point to highlight here: the Göbekli Tepe symbol does not at all look exactly the same. In fact, upon closer inspection it appears quite different (cf. Fig. 2): what appears to be a straight horizontal line in the Aboriginal sign is more of an “H”-shaped icon on the T-Pillar – with two vertical ‘arms’ at each side (which are obviously missing in the Australian example).


Fig. 3: Carved graffito of a woman. Most likely a later addition to a stone slab in the ‘Lion Pillars Building’, Göbekli Tepe (Layer II, PPNB). (Photo: K. Schmidt, DAI)

The same has to be said about the other symbols mentioned and in particular the female figure from Göbekli Tepe’s so-called Lion’s Pillar Building which is linked to Aboriginal depictions of the Rainbow Serpent, going by the name of Yingarna – sometimes depicted as a women in some Aboriginal cultures. Again, we know context and meaning of one of these (the Australian example), but do not have any further sources for the other one at Göbekli Tepe – which, by the way, was not part of the original decoration, but is a later added (and again exceptional) graffito.

These already dubious analogies are further complicated by assumptions aiming to underline the argument, but lacking a clear factual basis. Substantiating that the Göbekli Tepe pillar symbols must be “similarly sacred” since the “pillar depicts a deity” is of course highly interesting, but not without problems as already discussed earlier. And the assertion that the enclosures at Göbekli Tepe were “… created by a society that was wiped out by a cataclysmic event”, referring to a recently published study causing a similar  news output, did not go without criticism too.

In the end, an all too simple identification of similarly appearing symbols and equation of the complex individual cultural phenomena behind them, would be highly speculative – and is doing injustice to both cultures and their rich iconography. Furthermore, the focus on only a few selected images and symbols in an otherwise rich (and diverse) iconographic repertoire seems arbitrary. There are many more differences to be noted between these two cultures and places than actual similarities.

Additionally, Aboriginal Australians are not at all forming one single homogenous culture complex. More than 400 distinct Australian Aboriginal peoples could have been identified meanwhile, distinct groups with an own language (including a specific symbolic language) and peculiar culture (Horton 1994).

So, to answer the question posed in this headline: No. No, Göbekli Tepe was not built by Aboriginal Australians. The superficial similarities in iconography and art are exceptional coincidences in the best and misinterpretations in the most unfavourable case. With the same line of argument one could think, the early Neolithic hunter-gatherers of Göbekli Tepe were already inventing the letter “T” due to the characteristic shape of these omnipresent pillars …


Bergström, A., Nagle, N., Chen, Y., McCarthy, Sh., Pollard, M. O., Ayub, Q., Wilcox, St., Wilcox, L., van Oorschot, R. A. H., McAllister, P., Williams, L., Xue, Y., Mitchell, R. J., Tyler-Smith, Ch., Deep Roots for Aboriginal Australian Y Chromosomes, Current Biology 26(6) (21 March 2016) 809–813. [external link]

Clarkson, Ch., Jacobs, Z., Marwick, B., Fullagar, R., Wallis, L., Smith, M., Roberts, R. G., Hayes, E., Lowe, K., Carah, X., Florin, S. A., McNeil, J., Cox, D., Arnold, L. J., Hua, Qu., Huntley, J., Brand, H. E. A., Manne, T., Fairbairn, A., Shulmeister, J., Lyle, L., Salinas, M., Page, M., Connel, K., Park, G., Norman, K., Murphy, T., Pardoe, C., Human occupation of northern Australia by 65,000 years ago, Nature 547 (20 July 2017), 306–310. [external link]

Horton, D., The Encyclopedia of Aboriginal Australia: Aboriginal and Torres Strait Islander History, Society, and Culture, Canberra 1994.

Spencer, B. S. & Gillen, F. J., The Northern Tribes of Central Australia, London / New York 1904. [external link]


#FactCheck: It’s on the internet, so it must be true. Right?

Recently we got a couple of notes and comments regarding a number of finds supposedly attributed to the early Neolithic site of Göbekli Tepe. Finds we surprisingly never addressed or mentioned or published anyway. Of course this raised questions. Not only among our readers, but also in the research team. So, I picked three of these mysterious objects and tried to follow their path all the way back. A lesson in ‘alternative facts’, their easy creation – and quick multiplication on the world wide web.

The first of these items we’re having a look into is this detailed and rather ‘life-like’ sphinx-sculpture – linked to a caption saying “The oldest known sphinx was found in Göbekli Tepe …”.

C3g5gN8XUAMB9b7.jpg large

Stylistically, this sculpture already looks quite different from any other prehistoric piece of art we know of, yet those coming from Pre-Pottery Neolithic Göbekli Tepe in particular. The carefully sculpted face and body are more reminiscent of ‘classic’ ancient Greek or Roman sculptures – or those of the 18th and 19th century ‘romanticising’ sculptures. Also, the material does not really look like the yellowish-worn limestone so characteristic for Göbekli Tepe’s sculptures.

The statement connected to this depiction is even more surprising – since actually there is no single sphinx known from Göbekli Tepe (or any other related PPN-site of the region) reported. In fact, composite – hybrid – creatures are generelly very rare in prehistoric art. With a few excepetions, the concept of chimaeras seems to have become a more common (or at least more often depicted) mythological trope at a later time in human history.

Finally, with a little bit of research the sculpture in question here can be identified as being part of the garden decoration of the Royal Palace of La Granja [external link] in the Spanish town of San Ildefonso – dating to the 18th century. AD.

The second object on this list is a bit trickier. Indeed it seems to be a rather large stele showing the relief of some upright standing person. Stone and setting somehow look similar to what we would expect from Göbekli Tepe – although the style of depiction and the building visible to the right suspiciously do not really look unfamiliar from an Anatolian early Neolithic point of view.

DC16JMzXkAEExUbAgain, a closer look at the actual depiction makes it clear that we are dealing with something different here. The person shows characteristical features we would expect from an ape. In particular the head, face, and tail give it away: this is a depiction of Hanuman, a Hindu god in the shape of a monkey. With this information at hadn it is only a question of further reading to identify the stele as one of the monuments at the Virupakasha Temple in Hampi [external link], a World Heritage Site in southern India – dating back as much as the 7th century AD, but definitely not the Neolithic period.

Finally, the last find examined here is this standard-like object of which the caption claims it is a “G[ö]bekli Tepe god of the constellation of Orion.” Besides the challenges to project modern (or at least historical) star constellations into prehistory and even more prehsitoric art (as we already discussed here) and the terminological issues surrounding lables like ‘god’, ‘gods’, and even ‘temple’ (discussed here earlier), this piece, again, does look like nothing we know from the site of Göbekli Tepe or any other Neolithic find spot in the region.

DBubx1SXkAAeSld.jpg large

In fact, a stylistic comparison eventually offers a lead to modern Iran. A couple of elaborate objects dating to the Bronze Age period (i.e. the 3rd millennium BC) from this region show quite some parallels. They are linked to the so-called Jiroft culture which, unfortunately, still remians raher obscure since the majority of finds associated with Jiroft are coming from illegal excavations and we do not know much about them. But again, they’re neither related to the Anatolian Neolithic nor the site of Göbekli Tepe.

These three examples should be sufficient and representative for the point we would like to make with this blog post: it’s easy to come up with an idea, it’s easy to form hypotheses. Honestly, we as scientists do it too, all the time. But every hypothesis needs to be checked, material needs to be compared, sources reviewed. We’re living in times where facts can be made up as ‘alternatives’. Stay curious, but stay suspicious. Check the facts, check the sources. And when in doubt … ask an archaeologist (or any other expert).

What is the connection between Göbekli Tepe and…

Göbekli Tepe is often compared with other megalithic architecture. Stonehenge is an example here, others include the temples of Malta, the Taulas of Menorca, or the Moai of the Easter Islands. And fairly often, people also construct or believe in direct relations between these sites.

I believe this partly happens  because people tend to categorize things in relation to other things they already know.  Especially Stonehenge – for many people the iconic example for megaliths par se – can be found in every popular history book, making such comparisons with other sites with large standing stones, some of them decorated with reliefs, easy. But there is a little more to it. I remember that my  schoolbooks used to invoke the idea of a somehow interrelated Neolithic „Megalithic Culture“ that spread throughout Europe by migration. This was in the later 1980ies. Of course by this time the diffusionist view on the spread of megaliths had long been discredited by Colin Renfrew, i.a.  based on radiocarbon dates (you can see him talk about this here – external link). But textbooks just hadn’t noticed what was going on in academia. This is, by the way, a problem that archaeologists should address in some way also today.

But back to Göbekli Tepe. As we really get a lot of inquiries regarding possible interrelations of important megalithic sites, I thought I should post a short checklist here to show how different these sites really are. So here they are, in chronological order; please note that I am just writing down the main points from memory, if you have further questions please post them in the comments.

Göbekli Tepe


Göbekli Tepe, Enclosure C, illustrating the characteristic layout of the older buildings (copyright DAI, photo K. Schmidt).

Location: southeastern Turkey, on the highest point of the Germus mountain range.

Built / used between: ca. 9500-8000 cal BC, Pre-Pottery Neolithic.

By: Hunter-gatherer groups from a catchment area of about 200km around the site using stone tools.

Main characteristics: The oldest layer III (10th millenium BC) is characterized by monolithic T-shaped pillars weighing tons, which were positioned in circle-like structures. The pillars were interconnected by limestone walls and benches leaning at the inner side of the walls. In the center of these enclosures there are always two bigger pillars, with a height of over 5m. The circles measure 10-20m. The T-shape of the pillars is clearly an abstract depiction of the human body seen from the side. Evidence for this interpretation are the low relief depictions of arms, hands and items of clothing like belts and loinclothes on some of the pillars. Often the pillars bear further reliefs, mostly depictions of animals, but also of numerous abstract symbols. Layer III is supraposed by layer II, dating to the 9th millenium BC. This layer is not characterised by big round enclosures, but by smaller, rectangular buildings. The number and the height of the pillars are also reduced. In most cases only the two central pillars remain, the biggest measuring around 1,5m.



Miniature madel of a Maltese temple from Mġarr, Museum of Valletta (Photo: O. Dietrich).

Temples of Malta

Location: Malta and Gozo, islands in the Mediterranean Sea, temples are spread widely, sometimes forming clusters.

Built / used between: The Neolithic and the Bronze Age. However, the actual  ‘Temple Period’ falls within the 4th millennium BC and the 3rd millennium BC. Temples were constructed using stone tools.

By: The local population of these islands, evidence for external contact is rare.

Main characteristics: The temples are made of limestone orthostats forming walls. They usually have an oval forecourt and a facade with an entrance made up of three megaliths, of which two are supporting the third, forming a trilithon. Inside is a passageway of similar construction leading to an open paved space flanked by apses. Decorations inside the temples include spiral motifs, animals and surfaces covered entirely with drilled holes.

Further reading: For an easily accessible and well written overview: Trump, D.H. 2002. Malta: Prehistory and Temples. Midsea Books: Malta. Also, check out the Website of the UNESCO World Heritage List entry [external link].



Stonehenge, features of all construction phases (Drawn by en:User:Adamsan, CC BY-SA 3.0, Wikimedia Commons).


Location: Wiltshire, England.

Built / used between: several building phases between 3100 and 1600 BC.

By: People from a wider catchment area, some of the raw material was transported over vast distances, e.g. the so-called bluestones from nowadays Wales, metal tools available during the later phases.

Main characteristics: The iconic view of Stonehenge shows a ring of  standing stones around 4 m high, partly still forming trilithons. But Stonehenge has a highly complex building history that includes many changes to the layout of the site, accumulating to two megalithic stone rings and two orthostat arrangements surrounded by wooden posts and earthworks. Further, Stonehenge is part of a Neolithic/Bronze Age cultural landscape marked by earthworks and burial mounds.

Further reading: Mike Parker Pearson is the person to ask about Stonehenge and here is a great overview article that´s also freely accessible: Parker Pearson, M. 2013. Researching Stonehenge: Theories Past and Present. Archaeology International. 16, 72–83. DOI:



Taula of Trepuco (Juan Costa Archiv, CC BY-SA 3.0, Wikimedia Commons).


Location: On the Balearic island of Menorca.

Built / used between: roughly between 1000 and 300 BC.

By: The local, so-called Talaiotic Culture, which is restricted to Menorca.

Main charateristics: Taulas (meaning tables) are formed of a vertical pillar (sometimes made up of several stones) on which another stone rests horizontically. They are around 4 m high and usually stand within u-shaped buildings.

Further reading: There is not so much published about the Taulas in English and available to access freely online, if you are able to read Spanish, this artcle may be a good start: Daniel Albero Santacreu, D.A. 2009-2010. Análisis arquitectónico de los recintos de taula de la isla de Menorca: significación técnica y simbólica de los parámetros constructivos. Mayurqa 33, 2009-2010: 77-94 [external link].



Moai at Ahu Tongarik (Rivi, CC BY-SA 3.0, Wikimedia Commons).


Location: Easter Island, Polynesia.

Built / used between: 1250-1500 AD.

By: the Polynesian colonizers of the Easter island.

Main charateristics: Monolithic human figures with facial features, arms/hands, up to 10 m high and integrated into ceremonial sites   The letter consist of a levelled plaza, from which a ramp led up to a rectangular platform, where the moai stood.

Further reading: A classic is Routledge, K. 1919. The mystery of Easter Island. The story of an Expedition. London: Hazel, Watson & Winey [external link]. There isa vast amount of literature though, and also an ongoing research project by the German Archaeological Institute [external link].


The sites discussed here may have had similar social functions as centers for gatherings, expressions of belief systems etc. for the societies that built them. But I believe that this short comparison also shows clearly that we are dealing with very different sites indeed, evolving in different timeframes and regions, and rooted in a very specific local cultural background each. Their architecture is hardly comparable. Superficial similarities like the T-shape of GT´s pillars and the Taulas can be explained much better by a similar function, e.g. as roof supports, than by direct interconnections between the builders over large chronological and spatial distances.





More than a vulture: A response to Sweatman and Tsikritsis.

We already expressed a couple of thoughts and remarks on a paper published in Mediterranean Archaeology and Archaeometry in which Martin B. Sweatman and Dimitrios Tsikritsis have suggested (original article accessible here: [external link]) that the early Neolithic monumental enclosures at Göbekli Tepe were space observatories and the site’s complex iconography the commemoration of a catastrophic astronomical event (‘Younger Dryas Comet Impact’).

Meanwhile we were putting together a more elaborate reply with further arguments and references which, in our opinion, challenge the interpretation and add more context to the paper’s discussion of Göbekli Tepe’s iconography in the light of the early Neolithic in Upper Mesopotamia. The editors of Mediterranean Archaeology and Archaeometry kindly agreed to publish our response in the same journal as the original article by Sweatman and Tsikritsis.

The paper (J. Notroff, O. Dietrich, L. Clare, L. Dietrich, J. Schlindwein, M. Kinzel, C. Lelek-Tvetmarken, D. Sönmez: More than a vulture: A response to Sweatman and Tsikritsis. Mediterranean Archaeology and Archaeometry 17(2), 2017, 57-63.) can be accessed online: [external link].

Fig. 1

Aerial view of the mound of Göbekli Tepe with excavation areas. (Photo: E. Kücük, DAI)

Our reservations, which are not meant to silence any further archaeoastronomic discussion for Göbekli Tepe at all but rather comment on a number of discrepances we see in the interpretation, are summed up here:

1. The original layout of Göbekli Tepe’s monumental round-oval buildings is still subject of ongoing research (none of these structures are completely excavated as of yet). One should be aware that many of the T-pillars incorporated into the enclosures at Göbekli Tepe are not standing in their original positions and the buildings underwent significant modification during their life-cycles. Building archaeology studies have revealed that in many cases pillars were ‘recycled’, i.e. pulled out and used elsewhere. The monuments as we see them today are the culmination of multi-phase building and rebuilding events. Additionally, there is the significant possibility that we are dealing with roofed structures; this fact alone would pose limitations to a function as sky observatories.

2. The chronological frame Sweatman and Tsikritsis suggest for Pillar 43 (10950 BC +/- 250 years) is still 700-1000 years older than the oldest radiocarbon date so far available for Enclosure D (which stems from organic material retrieved from a wall plaster matrix). While there is evidence for later re-use of pillars (see above), assuming such a long tradition of knowledge relating to an unconfirmed (ancient) cosmic event appears extremely far-fetched. So far, any available date for Göbekli Tepe rather marks the end than the beginning of the Younger Dryas.

3. The assumption that asterisms are stable across time and cultures is not convincing. It is highly unlikely that early Neolithic hunters in Upper Mesopotamia recognized the exact same celestial constellations as described by ancient Egyptian, Arabian, and Greek scholars, which still populate our imagination today.

4. Sweatman and Tsikritsis’ contribution appears incredibly arbitrary, considering images adorning just a few selected pillars. Meanwhile more than 60 monumental limestone T-pillars are known from Göbekli Tepe – among these many feature similar carved low reliefs of animals and abstract symbols, a few even as complex as Pillar 43 (e.g. Pillar 56 in Enclosure H). Furthermore, the iconographic programme is not restricted to the limestone pillars; it is known from other find groups (including stone vessels, shaft straighteners, and plaquettes) not only from Göbekli Tepe but also from numerous contemporary sites in the wider region.

Fig. 3

Pillar 56 from Enclosure H is another example for the rich and often complex iconography of Göbekli Tepe. (Photos & drawing: N. Becker, DAI)

5. Göbekli Tepe’s iconography is actually even more complex than the paper suggests. The animals depicted on the pillars seem to follow an intentional pattern, whereby each building has a different emphasis, i.e. with one animal or more being especially prominent. If we interpret these differences as an expression of community and belonging, this could hint at different groups having been responsible for the construction of particular enclosures. In other words, specific enclosures may have served the needs of different social entities. For this reason, it is extremely problematic to pick out any one pillar and draw far-reaching but isolated interpretations while leaving out its context. A purely substitutional interpretation ignores these subtler but significant details. Details like the headless man on the shaft of Pillar 43, interpreted as a symbol of death, catastrophe and extinction by Sweatman and Tsikritsis, silently omits the clearly emphasised phallus which must contradict the lifeless notion; rather, this image implies a more versatile narrative behind these depictions. It should also be noted that there are even more reliefs on both narrow sides of Pillar 43 which apparently went unnoticed in the study at hand.

Fig. 4

Distribution of the appearance of figurative representations in the enclosures of Göbekli Tepe. Note: The different state of excavation as well as chronological depth of construction periods have to be considered; later added graffiti as well as symbolically reduced icons were not included. (Graphic: J. Notroff & N. Becker, DAI)

Fig. 2

Pillar 43 from Enclosure D and its particularly rich relief-decoration – actually extending not only on the pillar’s western broadside (left), but also the southern (middle) and northern (right) narrow sides. (Photos: K. Schmidt, N. Becker, DAI)

Pre-Pottery Neolithic iconography, by far exceeding the realms of Göbekli Tepe, is often especially concerned with articulation and disarticulation of the human body. Particularly the depiction of severed human heads or headless bodies in combination with necrophagous animals (preferably but not exclusively vultures) is a well-known theme and may be rooted in a complex multiphase Pre-Pottery Neolithic mortuary ritual. Similar depictions of a bird grasping a human head are known from Göbekli Tepe as well as life-sized human sculpture heads which were deposited within the buildings.

Fig. 5

Fragmented sculpture from Göbekli Tepe showing a bird of prey crouched on a human head. (Photo: N. Becker, DAI)


Meanwhile both authors of the orginal study replied to our response (same issue of MAA, see link above), stressing that “… given the statistical basis o[f] [their] interpretation, any interpretation inconsistent with [theirs] is very likely to be incorrect.” (Sweatman and Tsikritsis, Comment, MAA 17(2), 66). Admittedly though, we still would like to express our doubt that human creativity really can be treated as a statistical case solely.

What you get is … what you want to see: For example Göbekli Tepe on a 4th millennium seal print from Susa.

Recently, I stumbled upon a blogpost by Graham Hancock [external link]. I was looking for something completely different, i.e. the “fallout” of the rather unfortunate meteor theory proposed by two researchers from Edinburgh in April. What I found however sent me off in a completely different direction. As it is a prime example how false interpretations of images arise, and how they could have been prevented right from the start, I thought I should write a few words about that blog post here.

In his short text, Hancock explains that an independent researcher, while browsing the images in the online database of the ‘Cuneiform Digital Library’ [external link], found a depiction of the enclosures of Göbekli Tepe with their iconic T-shaped pillars. On a seal impression from Susa, dating to the Uruk V period. The settlement phase Uruk V constitutes together with Uruk IV the Late Uruk Period. The details of the absolute chronology of this period, which sees the invention of writing (i.e. proto-cuneiform script starting from Uruk IVa) and the cylinder seal, is still under debate, but a general date between 3500-3100 BC seems to be safe. Göbekli Tepe is currently dated between c. 9500-8000 BC. So, there is some chronological and regional distance between the sites (Susa lies in nowadays Iran). “Nice mystery here”, to cite Hancock. But let’s have a critical look at the evidence, which is always a good idea when doing science.


Hancock´s post refers to a fragment of a cylinder seal impression, for which the ‘Cuneiform Digital Library’ database gives a scanned black and white photo and some background information, like the material (clay), the collection (Louvre, Paris) and the primary publication (MDP 43, 676). It is also clear that the image is rotated – most likely accidentally – by 180° compared to the original publication (the number is upside down). And there they are, the two T-shaped pillars encircled by an oval, shown two times. A perfect abstract depiction of a round building from Göbekli Tepe´s older layer, as it seems. Alright, the pillars inside the perimeter wall are missing. But who cares? It could be an abstracted depiction of something a few millennia older but apparently still very well known.

The seal impression is fragmentary and highly damaged. It is obvious that the original image was more complex. If we turn the image correctly and look a little closer, in front of the left Ts, which now do not resemble Ts anymore, there is an indication of some more depictions that are hard to identify on the b&w photograph. That is why finds were and mostly still are drawn in archaeology, and in any case described extensively. And where to find a drawing and description of the find? In MDP 43 of course.


I perfectly understand that this is the point at which those with a general interest in archaeology and browsing through an online database might be lost. MDP refers to the series “Mémoires de la Délégation Archéologique en Iran”. Why the “P”? Because the series was first called “Mémoires de la Délegation en Perse” and the abbreviation never changed. If we look the find up in volume 43 of this series, written by Pierre Amiet and  dealing with “Glyptique Susienne”, the scene is described as “two figures sitting on the left, on curved seats, in front of apparatuses made up of two supports with square bases and an elongated oval element”. And the drawing of the sealing shows just that. The persons are touching this “oval element” with their hands. The publication has some more depictions of this kind on sealings, and at least some, such as MDP 43, nr. 673 or 674 are less fragmentary. It becomes immediately clear that we are not dealing with a depiction of T-shaped pillars, but of two supports with square feet at the bottom and a knob at the top, connected by an oval.

The depictions of people interacting with this “apparatus” are part of a group of sealings that shows people at work, and some of the images with the supports strongly hint in the direction of weaving (esp. nr. 673), the “oval” most probably being the depiction of a thread.


So, absolutely no “nice mystery here”. Just a misinterpretation of a highly fragmentary depiction. While dealing with prehistoric imagery things like that can happen quickly. Because the human brain interprets things in relation to former experiences and knowledge. In the case at hand, I have seen images of Göbekli Tepe´s round to oval enclosures with their iconic pair of monumental T-shaped pillars. Then I see two T-shapes on a scan of a b&w image of a highly-weathered fragment of a clay seal impression. And immediately make a connection between the two. Science starts when I challenge that superficial connection in the way described above with some simple questions that work not only in the case at hand:

  1. General chronological and cultural-historic reasoning: What is the cultural background of the artefact I am looking at and how old is it? How likely might it be that the people making it depict an object or a site millennia older and not something well-known to them?
  1. Iconographical reasoning: How was the way of depicting things in that particular period, may the shape I am looking at fit the way of representing certain devices / objects / things? What else might be depicted that perfectly fits the cultural background / everyday activities of the people making the artefact?
  1. Challenging the evidence / its documentation: Is the depiction fragmentary or hard to evaluate for other reasons? What kind of documentation is available to me? Does it allow me to fully comprehend what is depicted? Or do I need further information before I can make up my mind?
  1. Go to the sources. Archaeological artefacts, or artefacts similar to the one at hand are usually published somewhere, and these publications may hold further information and better images. It may be tricky to identify the available sources.

So, finally: Why not ask an archaeologist, some of us are nice people willing to help!



Pierre Amiet, Glyptique Susienne. Des origins à l´ époque des Perses Achéménides. Cachets, sceaux-cylindres et empreintes antiques découverts à Suse de 1913 à 1967. Mémoires de la Délégation Archéologique en Iran XLIII (Paris 1972).

Archaeoastronomy, meteor showers, mass extinction: What does the fox say? (And what the crane? The aurochs?)

Recently a (peer-reviewed) paper published by M. Sweatman and D. Tsikritsis, two researchers of the University of Edinburgh’s School of Engineering, has made headlines, suggesting that the Göbekli Tepe enclosures actually were space observatories and that some of the reliefs depict a catastrophic cosmic event (the original publication in Mediterranean Archaeology and Archaeometry 17(1), 2017 is accessible online here [external link]).

A selection of the carved reliefs found on many of Göbekli Tepe’s T-shaped pillars is linked to and interpreted as depiction of actual stellar constellations. In particular Pillar 43, which is indeed an outstanding (but actually not exceptional) example of the site’s  rich and complex iconography, is interpreted as record of a meteor shower and collision – with quite serious consequences for life on earth 13,000 – 12,000 years ago (this whole ‘Younger Dryas Impact’ hypothesis [external link] actually is disputed itself [external link], so making Göbekli Tepe a ‘smoking gun’ in this argument should absolutely ask for a closer look).

GT06_P43_N09.32_ 600_A4.jpg

Pillar 43 in Göbekli Tepe’s Enclosure D. (Photo: K. Schmidt, DAI)

Debate regarding a possible astronomic link and interpretation of the architecture and the characteristic pillars in particular are as old as the history of research regarding Göbekli Tepe, but as of yet no convincing proof for an actual celestial orientation or observation of such phenomena could have been put forward. We always were and still are open to consider these discussions. So, of course we were looking into the new study with quite some interest, too. After all it is a new and fascinating interpretation. However, upon closer inspection we as excavators of this important site would like to raise a few points which may challenge this interpretation in our point of view:

1. There still is quite a significant probability that the older circular enclosures of Göbekli Tepe’s Layer III actually were subterranean buildings – possibly even covered by roof constructions. This then somehow would limit their usability as actual observatories a bit.

2. Even if we assume that the night sky 12,000 years ago looked exactly like today’s, the question at hand would be whether a prehistoric hunter really would have put together the very same asterisms and constellations we recognise today (most of them going back to ancient Egyptian, Babylonian, and Greek scholars and descriptions)?

3. Contrary to the article’s premise the unearthed features at Göbekli Tepe are not  shrouded in mystery. Published over the last years and decades, there is ample scientific literature available which unfortunately did not find its way into the study. The  specific animals depicted in each enclosure’s iconography for instance seems to follow a certain intention, emphasizing different species in different enclosures. A purely  substitutional interpretation ignores these more subtle but significant details. This also can be demonstrated for instance with the headless man on the shaft of Pillar 43, interpreted as symbol of death and mass extinction in the paper – however silently omitting the emphasised phallus in the same depiction which somehow contradicts the lifeless notion and implies a much more complex narrative behind these reliefs. There are even more reliefs on both narrow sides of P43 which went conpletely uncommented here.

4. It also seems a bit arbitrary to base this interpretation (and all its consequences as described in the paper) on what seems to be some randomly selected pillars and their iconography (the interpretation thus not covering “much of the symbolism of Göbekli Tepe” as stated in the paper, but merely the tip of that iceberg). In the meantime more than 60 monumental T-pillars could have been unearthed in the older Layer III – many of these showing similar reliefs of animals and abstract symbols, a few even as complex as Pillar 43 (like Pillar 56 or Pillar 66 in enclosure H, for example). And it does not end there: the same iconography is prominently known also from other find groups like stone vessels, shaft straighteners, and plaquettes – not only from Göbekli Tepe, but a variety of contemporary sites in the wider vicinity.

So, with all due respect for the work and effort the Edinburgh colleagues obviously put into their research and this publication, there still are – at least from our perspective as excavators of this important site – some points worth a detailed discussion. A more thorough exchange with the excavation team could have clarified many of these concerns.