Taş Tepeler

Taş Tepeler / News / The people who dig Göbekli Tepe think its enclosures had roofs

What the excavators say · Göbekli Tepe · 24 July 2026

The people who dig Göbekli Tepe think its enclosures had roofs

The postcard image of Göbekli Tepe — open rings of freestanding stones under the sky — is probably wrong. Lee Clare, who runs the excavation, argues the great buildings were walled and roofed, closer to sunken rooms than to a stone henge.

The people who dig Göbekli Tepe think its enclosures had roofs
Enclosure D at Göbekli Tepe. The smaller pillars stand within the surrounding wall rather than free of it — a detail the excavation reads as evidence the buildings were enclosed and roofed, not open rings.

German Archaeological Institute, Göbekli Tepe Project, Enclosure D, Göbekli Tepe, Copyright: DAI — used with permission, via .

The picture in your head

Almost every article, documentary and travel photo shows Göbekli Tepe the same way: circles of tall stones standing in the open air, like a Neolithic Stonehenge you could walk between under the sky. It is a striking image. According to the excavation's own director, it is also misleading.

The reason the open-ring picture took hold is simple — that is roughly how the enclosures look once they have been dug out and the fill removed. What survives is the stone. What did not survive is anything that sat on top of it.

What Lee Clare actually argues

In lecture and interview material, Dr Lee Clare — head of the Göbekli Tepe research project for the German Archaeological Institute (DAI) — makes a plain case that these were not freestanding stone circles. The smaller pillars were set into the surrounding walls rather than standing apart from them, and the two great central pillars would have needed structural support. Read that way, the buildings were enclosed and very probably roofed.

This is not a lone hunch. The question of roofs has been discussed on the project for years — the architect Dietmar Kurapkat modelled roofed reconstructions long ago, and the team has often described the older enclosures as semi-subterranean, dug down into earlier deposits. What is notable is that the current director now states the roofed reading as the more likely one, against the open-air image that dominates popular coverage.

If he is right, standing inside Enclosure D in the Neolithic would have felt less like walking through an open ring and more like entering a dim, enclosed room — walls close around you, the carved central pillars rising in a covered space built to hold perhaps twenty or thirty people.

Why it changes the story

A roof is not a small detail. An open ring reads as a monument meant to be seen from outside and from above — a display. A walled, roofed room reads as an interior: a space you go into, for something that happens among a limited group of people, out of sight.

That shift lines up with the rest of how the excavation now talks about these buildings — as gathering places for a community rather than open-air sanctuaries. The architecture and the interpretation point the same way.

What is uncertain

No roof survives. If timber-and-thatch or beam roofs existed, they were dismantled or rotted away, and firm structural proof is still lacking — this remains an argued position, not a settled fact, and other reconstructions are possible. The point Clare makes is that the evidence of the walls and pillars favours enclosed, roofed buildings over open rings.

This position comes mainly from spoken lecture and interview material, and the project's formal architectural publications may refine or qualify it. Treat it as the director's considered reading, likely to be spelled out more carefully in print.

Sources

  1. Tepe Telegrams (DAI) — Ask an archaeologist (about Göbekli Tepe): the roof question
  2. Tepe Telegrams (DAI) — Frequently Asked Questions

Follow the digs

New finds from the Taş Tepeler sites, a few times a year — join the list.

Explore the sites →

Last reviewed 9 September 2026. Something wrong?