What Are The Main Taş Tepeler Sites?

The twelve active excavations

Taş Tepeler — 'Stone Hills' — is a Turkish-led research programme covering a cluster of Pre-Pottery Neolithic sites in Şanlıurfa province. In September 2025 it reached twelve active excavations with the addition of Ayanlar Höyük.

Göbekli Tepe — the UNESCO World Heritage site, the earliest and most fully published of the monumental sites, with T-pillar enclosures and, more recently, extensive domestic evidence on the same mound.

Karahan Tepe — the second great monumental site, known for the rock-cut chamber with pillars rising from the bedrock floor and the human head carved into its wall.

Sayburç — a site inside a living village, holding a narrative relief on a bench face that is among the earliest known scenes rather than isolated figures.

Sefertepe — T-pillars and quadrangular buildings, and a chamber where human skulls were gathered together.

Çakmaktepe — early-horizon material, mortars, and deposits of burnt animal heads inside and outside special buildings.

Harbetsuvan Tepesi — T-pillars but, per 2025 geophysics, no large circular buildings; excavated with a Japanese–Turkish team.

Gürcütepe — four shallow mounds excavated by Klaus Schmidt in the late 1990s, later than Göbekli Tepe, showing the village world that followed.

Ayanlar Höyük (Gre Hut) — roughly 14 hectares, essentially unexcavated until September 2025.

Mendik Tepe, Yoğunburç, Yenimahalle and Söğüt Tarlası — the less-published members of the group, with excavation and reporting still developing.

A note on which names count

Published lists of Taş Tepeler sites do not always agree, and it is worth knowing why.

There is a difference between sites under active excavation in the project and sites known from survey that belong to the same regional tradition. Places such as Kurt Tepe, Hamzan Tepe and Taşlı Tepe were identified through survey work as T-pillar sites and are often named alongside the others, but they are not among the twelve active digs.

Some lists include them; some substitute them for Yenimahalle or Söğüt Tarlası. Neither list is wrong — they are counting different things. We list the twelve active excavations and name the survey sites separately.

The scale of the programme

At its fifth anniversary in 2025 the project reported 36 academic institutions involved — 15 Turkish and 21 international — and 219 researchers working across all twelve sites simultaneously.

That is an unusual concentration of effort on a single region and period, and it is the reason the picture keeps changing. Findings that were current three years ago have been superseded; several of the sites had no excavation data at all before this programme began.

It also means published summaries date quickly. Anything written about Taş Tepeler should be read with an eye to when it was written.

Why they are treated as a group

Because they share a vocabulary. T-shaped pillars, special or communal buildings distinct from houses, animal relief carving, human imagery, and a concern with heads and the dead recur across the group.

And because they are close together. These sites sit within a landscape people could walk across. Harbetsuvan is about 7 km from Karahan Tepe; Karahan Tepe is around 35 km from Göbekli Tepe. Whatever was happening was happening between neighbours.

But grouping them should not flatten them. Harbetsuvan lacks the large circular buildings its neighbours have. Gürcütepe is substantially later. Sayburç has a narrative relief nobody else has produced. The group is a network of variations, not a set of copies.

Which ones you can actually visit

Göbekli Tepe is fully developed for visitors, with a protective shelter, walkways and a visitor centre.

Karahan Tepe is open with visitor infrastructure in place and is the natural second stop.

The remainder are working excavations rather than visitor sites. Access varies by season and by permission, some have no facilities at all, and several are on land where arriving unannounced is not appropriate.

If you want to see more than the two flagship sites, going with someone who knows the current access situation is the practical answer — it changes year to year.