Kissonerga-Mylouthkia: the Edinburgh excavation of the Aceramic Neolithic wells

Two shafts cut into the rock at Kissonerga-Mylouthkia, found when a quarry edge was cleaned in 1994, proved to be wells. AMS dates on short-lived carbonised seeds place well 116 in the mid-later 9th millennium cal BC and well 133 about a millennium later¹. Edgar Peltenburg's team argue that this Period 1 occupation represents a newly recognised Cypro-PPNB period of Cypriot prehistory, preceding the Khirokitian and possibly giving it the long-sought ancestry¹.

Three seasons in plot 58/496

The work was the third phase of the investigations at Kissonerga-Mylouthkia, after earlier phases of investigation, and was run by the Lemba Archaeological Project of the University of Edinburgh^1,2. Edgar J. Peltenburg directed. Gordon Thomas supervised excavation in plot 58/496 until he returned to Edinburgh, and Paul Croft excavated the wells and continued work in and around Building 200¹. The final volume counts three short formal seasons with small groups of students in 1994–6; 1996 was also a study season for processing and conserving the finds^1,3.

The aim was to test whether the original Chalcolithic ground surface survived in plot 58/496, and to find Early Chalcolithic structures contemporary with the recorded pits, so as to learn the plan of the settlement's habitations¹. The five-week 1994 season worked away from the eroded coastal slopes, where Thomas hoped for intact deposits. Random trenches showed an undulating prehistoric surface with structures in hollows. They located a 6 m circular building, B 200, beside the earlier pit Feature 5, and a second structure in a 5 m hollow to the north^1,4. The work was funded by, among others, the British School of Archaeology in Jerusalem, the British Academy and the National Museums of Scotland^1,4.

Building 200

B 200 is a circular stone-founded building 6.0 m in diameter, with a mud-bonded rubble wall about 0.60 m thick, preserved to 0.57 m. It has four phases of construction, occupation and site formation. A hearth, a quern, socketed-stone settings, sunken vessels and pits lay inside, and the building ended in catastrophic destruction¹. On the same spot lay the earlier timber structure B 290, 3.60 x 2.60 m in an oval scoop, and the stone-built B 330, whose southern half was destroyed when B 200 was built¹. Herscher's survey reports small ceramic vessels, stone and bone tools, lids, antler beads and figurines on the floor, and juvenile human bones crushed by collapsing roof debris⁵.

Thomas and Herscher first reported B 200 as Early Chalcolithic. The final volume places it in the early Middle Chalcolithic on its pottery^1,4,5. AMS dates on it include OxA-7463 (4,710 +/- 50 BP) and OxA-7462 (4,650 +/- 50 BP)¹. The second structure to the north had postholes for roof support, a low earth and clay bench, a pot support and a clay basin; Thomas suggested a food-processing installation^4,5.

The wells

Well 116 is 8.50 m deep, with a 0.9 m cylindrical shaft belling to 2.25 m, cut into havara to intercept an underground stream, and has more than two dozen hand- and foot-holds. It was excavated as a matter of urgency¹. Well 133 lies about 11 m to the south, with hand- and footholds and pick-marks in the niche roofs that are interpreted as made by fallow deer antler picks¹. Croft describes his reading of the stream and channel history as tentative, since the observations were made by torchlight at the bottom of a very deep, dark hole¹.

The upper fill of well 133 held a human mandible and Skull 1 (KMyl 1181), then articulated caprine bone through some 4.25 m of fill, interpreted as many unbutchered whole sheep and goat carcasses, with scattered disarticulated human remains, some lightly burnt, in the fill below¹. The counts of carcasses differ: Peltenburg and colleagues report 22 originally whole skeletons, Croft 2002 reports 23 complete sheep and goats^2,6. Well 133 also held 449 ground-stone artefacts, 63% of them stone vessel fragments and 25.4% hammerstones; well 116 gave abundant obsidian^7,8. Mollusc shells from the wells gave results far too early and were left out of the chronology¹. Herscher's survey notice said the wells might be aceramic Neolithic, and Peltenburg writes that the AMS dates dispelled Herscher's reservations about the date^1,5. Steel summarises the wells as requiring a reassessment of the earliest Neolithic occupation, pushing it back to a period contemporary with domestication on the mainland⁸.

Finds, archive and publication

Material from the Lemba Archaeological Project's investigations at the site is mainly in the Paphos Museum, with a few display objects in the Cyprus Museum, Nicosia, some sherds in the Ashmolean Museum, Oxford, and other material in the Curium Museum at Episkopi¹. The site records are held by the School of Arts, Culture and the Environment at the University of Edinburgh, with duplicate plans, sections and unit sheets at the Lemba Archaeological Research Centre in western Cyprus¹. The final report is Peltenburg's edited volume, The Colonisation and Settlement of Cyprus: Investigations at Kissonerga-Mylouthkia 1976–1996 (Lemba Archaeological Project III.1, SIMA 70:4, 2003), in which Croft wrote the chapter on the wells¹. In 2002 the southern part of the site lay within the Paphian Sun Holiday Village and a central portion had been destroyed by the Queen's Bay Hotel¹.