Tepe Düzen: July 30-August 3, 2006

This week test soundings were initiated at Tepe Düzen under the supervision and in collaboration with the Museum of Burdur (director Haci Ali Ekinci), represented by Mustafa Erol Erbay. The K.U.Leuven team, supervised by Hannelore Vanhaverbeke, consisted of archaeologists Kim Vyncke, Sarah Linten, and Jonas Danckers. The main aim of these test soundings was to gain insight into the material culture (architecture, pottery, small finds, faunal and floral remains, etc.) of the people living at Tepe Düzen during the EIA (ca. 800-400 B.C.). Sabri Aydal delineated a 20-by-40 m test area that included part of the ground examined by Branko Music and his geophysical survey team. The area, according to the geophysics, seemingly spans two buildings intersected by a street. This area was subdivided into 5-m squares, of which the southeastern one was excavated first, but was soon extended 5 meters north, and east over a length of 10 m, thus encompassing 4 sectors of 5 by 5 m. Quickly, it became clear that the surface remains, which Sabri Bey continues to map for us, only represent the last phase of a complex succession of occupation layers. Based on the position of the different wall sections, and their stratigraphic context, the following 4 phases could be distinguished in the excavated trench:

Phase 1 is constituted by the actual surface remains embedded in the topsoil. Belonging to this layer are quite carefully erected walls; large limestone rubble stones are set on two sides (no mortar was used), forming a double face with in an infill of smaller stones. These walls rest on a second layer, into which they were partially dug, which at its surface is quite compact. The current surface must have been a former walking or floor level corresponding to this phase. The walls in Phase 1 are oriented SW-NE. Within the area of 20 by 20 m (composed by 4 sectors of 5 by 5 m), we could trace the external walls of a large apsidal building. The south wall runs over a length of at least 12 m, with an average wall thickness of ca 0.80 to 1 m. The north wall is much shorter. In between the two (orange arrows on the illustrations) a wall perpendicular toward the two outer ones divided the building in two. Its width corresponded with 7 m. This kind of apsidal (or elsewhere even oval structures) houses ideal for thatched roofs were especially quite common during the Geometric Period (e.g. in the oldest Greek colonies at Old-Smyrna, Miletos, Ephesos; in many sanctuaries or houses on the Greek mainland, such as at Eretria, Perachora, etc). They were very popular during the 8th century B.C., before the introduction of the fired terra-cotta tile ca. 700 B.C., gradually replaced them by orthogonal structures better fitting the roof shapes of the latter. But they may have continued into the Early Archaic period, certainly in indigenous non-Greek communities, like Tepe Düzen, where the majority of the houses already are orthogonal as well.

Phase 2 lies immediately under Phase 1 and is often difficult to distinguish from it. We can often identify Phase 2 walls by their different orientation, running W-E, and in some places they run under the walls of Phase 1. In general, the wall construction seems to be less careful (average width is 0.65 to 0.70 m). This Phase also had an apsidal building, the apse of which runs just east of the apse of the Phase 1 building, but is oriented differently. The width of the building seems to be ca. 5 m, the greatest preserved long wall has a length of 9 m. Related to this phase, another layer (Layer 2), very compact at its surface and ca. 0.10 m thick was identified both inside and outside the building. The compactness identified it as a second walking level/floor, the walking level of Phase 2. Finds in this layer were scarce. Apparently old floors were cleaned before laying out a new floor.

Phase 3 rests on top of Layer 3 (average thickness: ca. 10 cm), which is yet another compact layer, distinguished by its red color and forming the floor of phase 3. Walls related to this building phase seem to be carefully constructed as well, being composed of medium-sized limestone rubble, forming a wall without external facing and with an average width of 0.75 m. Finds from this layer were slightly more numerous, and contained a number of rims, which will allow us to construct a relative typology of the pottery found at Düzen. The walls of this phase are again oriented SW-NE and seem represent the inner area of an orthogonal building with at least 2 rooms (together at least 8 m by 10 m). A narrow alley (ca. 0.75 m wide) separated this building from one to the north, of which two small rooms (2.40 by min. 1.80 m; 3.20 by min. 2.70 m were exposed.

Very few wall segments can as yet be attributed to Phase 4 (mainly because we only reached this phase in one sector). This phase seems to be built directly on top of the virgin soil or making use of natural limestone outcrops, but it contained a floor of beaten earth, in some places 0.40 m thick covering the irregular original surface. The only wall segment belonging to this phase runs SW-NE. Finds associated with this building phase are again slightly more numerous than in Phase 1 and 2, but still not frequent. The structure itself is mostly destroyed by those of later phases, but seems to have had walls of ca. 0.70 m wide forming at least one corner of a NW-SW oriented building of min. 2.30 by 1.50 m.

On Thursday, the last day of our first excavation week at Düzen, the team identified a dump, outside of the apsidal buildings of Phase 1 and 2 (it is as yet unclear whether this pit related to Phase 3 or 4), which contained quite some bone fragments (sheep, goat, and cattle according to our archaeozoologist Bea De Cupere) and lots of pottery, including several rims, handle fragments and bases. Some of the sherds contained within this pit contain traces of paint (dark red lines) that will become clearer only after the sherds have been cleaned by Emine Kocak's conservation team. The painting and pottery are so brittle, however, that it took more time to clean a single one of these sherds than to manual clean a coin on either side. The pit has a depth of 0.60 m and is 3.90 by 2.50 m large. It is dug into Layer 4, but material from Layer 3 seems to be mixed with it. If that is the case, it probably was dug at the time of Phase 3, as otherwise no mixing would have occurred, unless the people from Phase 3 had been looking through it. A burnt area found outside the apsidal wall of the building of Phase 2, but set within the area contained by an older wall (Phase 3), may be a hearth, but no structural elements (such as stones encircling the fire place) remain.

Our first week at Düzen thus revealed a rather complex archaeological reality, of a succession of building phases, with associated floor levels. This not only clearly illustrates the settlement's longevity, but also its very dense occupation, since four different building phases could be identified in an area of less than 100m. In the coming week, we will further excavate the area in order to reach virgin soil in all sectors and clarify the relation of all wall segments encountered. The excavation area will also be extended farther south, in order to incorporate the complete garbage pit and to trace the SW external walls of the two apsidal buildings.