Writing Wednesday: A Kiln at Polis

As readers of this blog know, I’ve been using my winter research leave to work on a paper writing up an article that documents the pottery productive installation at Polis (ancient Arsinoe). It’s been a nice project after the toil of finishing a book. 

So far, I’ve shared a section on a lamp deposit and a levigation pool. This is a very-drafty section of this same article that describes the kiln and offers some very preliminary summative remarks. 

The Kiln

In 1990, excavations through the material that filled the levigation pool revealed the upper levels of the beehive kiln. These rubbly fill levels continue to appear to date to the Late Roman period with Cypriot Red Slip sherds (T06.1990.Level 47: CRSK3 [L47P1B40], CRS Body Sherds [L47P1B37]),Late Roman cook pots ([L47P1B13 and B18]) and what what appear to be lamp wasters (L47P1B19 and B20) reflecting how disturbed this entire area was in antiquity.

In 1991, excavations of the kiln began in earnest when the removal of the south wall exposed more of the upper levels of the kiln. The superimposition of the south wall and the levigation pool over the kiln allow us to date the kiln itself to no later than the 2nd century AD as the wall and the levigation pool must be no earlier than this date. As the kiln and levigation pool run atop the kiln, it is obvious that the kiln must have been out of use by the time of their construction. The base of the kiln is a meter below the the lowest level of the levigation pool.

The truncated shape of the kiln this indicates that the builders of the levigation pool removed the highest courses of the kiln to create a flat space of the pool and the south wall, whether contemporary with the pool or not, cut across the top of the leveled kiln. It seems probable that kiln builders cut the apsidal shaped kiln into the natural slope of the ground on the east side the ravine to that the basilica construction fill took pains to level. The preserved courses of the kiln were constructed of rough field stones and opened to the west. The presence of mud brick fragments in the kiln suggest that this material was used to either line the kiln’s stone walls or for its upper courses. This is consistent with the proposed construction methods of the kilns at Zygy-Petrera and Dhiorios (cf. Manning et al. 2000; Catling 1972, 29). The lowest levels of excavations in the kiln did not discern a distinction between the firing chamber and the lower combustion chamber. It may be that the excavators did not reach the lowest levels of the kiln, that the kiln was cleaned out after it fell out of use, or, as at Dhiorios, the fuel was burned inside the kiln amid the pots (Catling 1972, 31). Because the excavations occurred at the very end of the 1991 season, they were hastily recorded. The absence of significant quantities of kiln debris within the kiln suggest that the kiln was cleaned out after it went out of use.

The excavators excavated the interior of the kiln in a series of arbitrary levels. What is interesting is that most of the material in the upper levels of the kiln dates to the Hellenistic period. The lowest levels of the kiln, however, include a small number of Roman period sherds including fragments of Eastern Sigillata A and B and Cypriot Sigillata. The absence of any significant traces of Roman period material in the upper levels of kiln and the location of the kiln on the eastern side of north-south drainage suggests that kiln was filled deliberately. If the filling of the kiln was part of constructing a terrace or level area upon which to build the levigation pool, this would account for the reversed stratigraphy as the upper level of the kiln is filled with material cut from deeper below the slope of the ravines surface. The process of creating terraced surfaces along the slope of the ravine during the Roman period appears to have occurred on the western side of the drainage as well. Whatever accounts from this inverted stratigraphy, the Roman material deep in the kiln provides a terminus post quem for the abandonment of this structure. This date is largely consistent with the date of the fills associated with both South Wall phases and the material behind the tiles wall of the levigating pond.

Conclusion and Comparanda

The small number of excavated Roman pottery production sites on Cyprus make it challenging to identify clear comparanda both for the kiln and its associated features. The kiln appears to have had a beehive or conical shape with a diameter of around 1 m at the lowest excavated courses. This makes this kiln a good bit smaller than best preserved ancient kilns at Dhiorios, but approximately the same diameter and shape as the kiln documented eroding from the scarp at Zygi-Petrini in the Maroni valley on the south coast of the island. In other words, the size of the kiln is appropriate for ancient pottery manufacturing on Cyprus, but perhaps at a small scale or designed to fire table wares or lamps.

The proximity of the kiln to the later levigation pool and treading basin appear to be a common assemblage associated with kilns in the Levant. Sites such as Legio X Fretensis Kilnworks at Binyanei Ha’uma (Jerusalem), and Horbat ‘Uza, and Tel Yavne preserve some combination of kilns, pools, and treading surfaces used to process raw clay. While the levigation pools very in size and depth they are generally stone lined and sealed with water proof cement. The size of the pool at Polis, of course, remains indeterminant because its northern side was lost. That said, the preserve section of the pool finds parallels with the pools at the ceramic production site in the Levant. It suggests that this is not the site of small scale production but part of a larger complex that the limits of the excavation failed to reveal. The combination of these pools in close proximity to kilns, installation for wheels, and other features associated with ceramic production indicate a kind of integrated production site where clay is processed, refined, and fired into objects and vessels. The presence of clay deposits both in the proximity to Polis and in the broader region as well as a market for ceramic vessels make the area to the north of the city a suitable location for production at the scale assumed by this levigation pool.

The kiln is not contemporary with the levigation pool or other production activities in the area. The use of an area for ceramic manufacturing across multiple phases is consistent with the remains at Dhiorios where earlier kilns were buried beneath a so-called “Potter’s House.” Catling argues that the Potter’s House must not date much earlier than the 6th century and effectively assigns the early kilns, house, and later 7th-8th century kilns to a two century span of time. The relationship between the kiln and later levigation pool at Polis is likely chronologically closer than the phases at Dhiorios but nevertheless reflects the tendency for continuity of use in ceramic production areas. This likely has to do with the proximity of resources — clay and also fuel for kilns, the presence of a slope to support kiln walls, and location far enough from settlement to avoid subjecting residents to the smoke and commotion of the production area, but close enough to be convenient for the distribution of good and access to labor. Indeed, as the final section of this article will show, it appears that most Roman period activity in the area of E.F2 involved production of some description.