Keros Project - 2025

Keros island, Cyclades. The BSA’s Keros Project returned to the field in 2025, now in the form of a synergasia with the Hellenic Ministry of Culture, directed by Demetris Athanasoulis (Ephorate of Antiquities of Cyclades), with Michael Boyd (BSA) and Evi Margaritis (Cyprus).

The new, five-year collaborative project asks questions around the environmental economy of Keros through the ages, and the effects of past climate change, as well as the periods of earliest habitation. The aim for 2025 on Dhaskalio was to excavate the 7th century CE chapel on its summit, and to investigate the third millennium BCE terracing system underlying the site’s dense architecture, while also beginning some larger trenches. On Keros the aim was to investigate the site’s wider environmental aspects and to understand its earliest period.

On (now islet of) Dhaskalio, seven trenches were opened (Trenches P, Q, R, T, U, V and W), placing a total of 608 m2 under excavation. Trenches P, Q, R and T were excavated to significant depth (Fig. 1), while Trenches U, V and W underwent surface clearance in preparation for significant excavation in 2026. The aim of the long and narrow Trenches R and T was to furnish information on the system of terrace walling used to create flat spaces for building. This is of particular interest as the entire site seems to have been built in a planned fashion, so the terracing and drainage systems underpinning it attest to incipient urban planning. In Trench R a pair of braziers, and unusual pyrotechnological ceramic form whose exact use is still unclear, was found almost intact and fully restorable. Located in a room with a hearth, it should be possible finally to determine how these enigmatic vessels were used. Trench P is a large open area intended to understand the detailed utilisation of space over a large area. Trench Q, located on the summit, primarily investigates the 7th century CE Byzantine chapel, where a plaster floor was located, but has also provided much information on the use of the summit in the final phase of Early Bronze Age occupation (Fig. 2). A storeroom here, filled with ceramic vessels, was excavated toward the end of the season, and will provide many restorable vessels. This storeroom is from the end of the Early Bronze Age occupation of the site around 2200 BCE. A small marble figurine of the so-called Dhaskalio sub variety was also found here.

On Keros itself, excavations took place in the coastal zone to the west of the low hill of Gerani (Polygon 2 of the Keros survey), with the principal aim of uncovering evidence for occupation of the island in the centuries before Dhaskalio, the Final Neolithic to Early Bronze I periods. Initially small (1m x 1m) test trenches were excavated. These uncovered much material from the fourth, third and second millennia BCE, as well as some structures, initially thought to be likely to be terrace walls. Neolithic finds were located in a geomorphologically disturbed area of large rocks. No in situ finds were made, but the prevalence of restorable ceramics in some areas suggests that we were close to original areas of occupation, convincingly proving habitation before Dhaskalio. Initial fabric analysis of the pottery suggested most of it was imported, a surprising and significant result indicating connectivity as a factor before the establishment of the sanctuary. Second millennium structures were excavated in larger Trenches 17, 23 and 8. Trench 17 proved to contain a major portion of a relatively large and well-constructed building of the second millennium BCE, specifically of the Middle Cycladic period (ceramically MM III) and perhaps ‘Minoan’ in nature, in the sense that the nearby and contemporary settlement of Akrotiri on Thira is ‘Minoan’ (Fig. 3). This substantial building suggests a small settlement was probably located in this area, and further excavation in this area is probably necessary to understand the nature of any such settlement. Trench 23 contained an interesting rough stone structure set on top of sloping bedrock incised with petroglyphs. These, four in number, may well date to the end of the fourth millennium BCE, based on comparisons with Vathy on Astypalaia and Strofilas on Andros; although petroglyphs have been discovered elsewhere on Dhaskalio and Keros, these are the first to be found on fixed
bedrock. In Trench 8 a Late Cycladic grave (perhaps around 1400 BCE) proved to contain a single crouched male burial in a grave previously used for up to four other funerals, as shown by the collected bones placed in a niche at the side (Fig. 4). The grave contained three Mycenaean pots and bronze tweezers. This grave interestingly showed local as well as ‘Mycenaean’ features, the latter shown by the items buried with the last burial, but the local features including the mode of
burial (crouched rather than extended as would be normal with Mycenaean burials) and the architecture of the tomb, consisting of a roughly delimited cist with a stone platform to one side.