Preliminary Report 1996

TEST EXCAVATIONS

Some of the test trenches by Erich Schmidt in 1928 were cleaned out. Trenches along the western side of the city and which appeared to be free of later levels were selected (numbers 4, 5, 6, 8, 10 and 11). By beginning in this way it was possible to obtain an understanding of the nature of the archaeological and geomorphological deposits with the minimum of disturbance. It also enabled us to familiarise the workmen with the basic requirements of excavation.

We then excavated a number of new test trenches (TT15-19 following on from Schmidt’s numbering) which were each carefully sited to answer particular problems and to test hypotheses. The approach was to maximise the amount of information recovered with as little disturbance to the site as possible. Trenches were excavated in several parts of the site, all but one where geomagnetic maps had raised questions.

At the end of the excavations new courses of stone were added to the wall tops with broken glass between the ancient and modem work, strips of plastic were laid along the bases and up the sides of the trenches and the earth was put back (Fig. 7). In this way, and following the precedent set at Bogazkale, the lines of the buildings are made visible while the original structures and the sides of the trenches are preserved.

1. Geophysical
Test excavations showed that the geomagnetic maps were highly accurate. The results have greatly helped in the interpretation of existing maps, and in the formulation of research design for further seasons of geophysical survey. Large area geomagnetic survey at Kerkenes will provide an exceptionally clear and almost complete city plan.

Detailed comparison of geomagnetic maps and excavated structures is continuing. It has become clear that the geomagnetic maps contain more information than we have heretofore been able to extract from them and that much is to be gained by careful comparison of geomagnetic maps and balloon photographs with features visible on the ground. Exactly what the geomagnetic maps show is now better understood, allowing for greater confidence in their interpretation; but the reasons for the strength and weakness of particular signals, the relationship between signal strength and burning, and for a constant north-south discrepancy between the position of buildings on the maps and their actual position on the ground are all matters for future research. Another area for experiment is geophysical determination of differing surface material, especially identification of stone paving and burnt clay floors.

2. Ground Penetrating Radar
GPR (Fig. 6) was tested against the excavated profiles while the trenches were still open. Despite great effort and considerable ingenuity the results were disappointing. Theoretically GPR has great potential at Kerkenes and, conversely, the site ought to aid the development of GPR interpretation. The reasons for the poor results are not at all clear.

3. Destruction
The intensity and uniformity of the burning confirms that the city was put to the torch (e.g. the burnt floors in Fig. 3).

4. Dendrochronology
The catastrophic fire that destroyed the city was of such intensity that beams and timbers used in the construction of buildings had burnt completely away. Some pieces of charcoal have been recovered but it is doubtful that they contain sufficient annual growth rings. The potential for confirming the date of Kerkenes through dendrochronology remains extremely high and the increased understanding of the geophysical maps obtained in 1996 will make it easier to pinpoint potential locations for test trenches where less intense burning might yield charred beams.

5. Micromorphology
Samples for micromorphological study were taken by Wendy Matthews. Slides will be made in Cambridge and these will be studied over the winter. Phosphate analysis will be done at Swansea. The immediate aim is to see whether the archaeological deposits, soils and geology at Kerkenes are such that this relatively new technique will be of help in answering questions relating to the sorts of activities that went on in particular types of buildings and open areas. The results will enable us to evaluate the potential of the techniques for future research at Kerkenes. In addition, it is expected that they will answer specific questions about some of the structures revealed in 1996. This particular aspect of the project fits into Dr Matthews’ wider programme of research in the Near East.

6. Architecture
A long 2m wide trench was put through a large complex at the north-west side of the site (Fig. 1 Area B, IT 15, Figs 3, 4, 5 and 6) which did, as anticipated on the basis of the geophysical map, contain a columned hall (Fig. 3 between the two scales).

Two-roomed structures, prominent on the geomagnetic maps (Fig. 8) in several areas of the site, are not “megarons”; rather, they comprise a smaller roofed unit, with wide central doorways, leading to a walled but unroofed unit (Fig. 1 Area D, TT16 & TT18; Figs 9, 10 and 11).

The large complex to the north of the “Cappadocia Gate” (Fig. 1, Area A, TT 19) has the characteristic plan and paving associated with animals (Kroll, 1992) and were perhaps the imperial stables.

In general, many structures had timber frames filled with mud-brick on top of stone footings, although some walls appear to have been built of stone to roof height. Internal surfaces were laid clay floors, many external surfaces were stone paved. There seems to have been a development in building methods during the short life of the city. The architectural forms are strongly reminiscent of an Eastern tradition. The construction techniques, devoid of embellishment, also echo pre-Achaemenid Iran and show no discernible influence, e.g. drafted masonry, from Lydia or Ionia (see the comments by Stronach, 1978, 10-11).

7. Objects
A number of objects were found during the excavation. Most important is an exquisitely carved ivory plaque, perhaps an inlay from a couch (Figs 12, 13, 14 and 15). The design echoes East Greek art and the piece may have been made in a Lydian workshop. The total length is preserved and holes for attachment survive at the right side. There is an animal frieze with traces of applied gold leaf on the horns, a meander pattern along the base and an applied line of bead and reel comprising amber beads and gold covered ivory reels along the top. The luminosity of the beads was increased by small reflecting plates of silver, or perhaps tin, set behind them. At the left end is a deer facing right whose body was once adorned with enamel or paste set into double drill holes. Facing left is a procession of four domestic animals, a billy-goat followed by a she goat, a ewe and a ram respectively. The leading billy has its head lowered in front of the opposing deer. The sheep have carefully depicted fleeces.

Other objects of note include a composite object comprising a (?boars) tusk, preserved wood, copper (?alloy) sheet and pins (perhaps a cheek piece for a horse, Fig. 16), an ivory “handle” (Fig. 17), a large quantity of very well preserved iron (e.g. Fig. 18), a large piece of lead (probably an ingot) and a plain gold “strap end”.

The pottery, although not prolific, mirrors that found by Schmidt in 1928. There are a few sherds with coarse white slip panels and polychrome paint, some small fine black polished bowls with flat or disc bases and a very few incised or impressed grey ware sherds. The great majority of pottery is plain, table ware being predominantly red, kitchen ware grey. There is a small repertoire of forms. The complete and closely dated ceramic corpus is distinctive. It is not yet possible to ascertain changes in the ceramics during the short life of the city. The only recognisable import is a thumb-nail sized rim sherd from what appears to be a Lyconian cup.