Fig 14	Fig 15	Fig 16	Fig 17
Fig 18	Fig 19	Fig 20	Fig 21
Fig 22	Fig 23	Fig 24	Fig 25

POST-EXCAVATION STUDY

Simulation of the 'Cappadocia Gate'
New architectural reconstructions and graphic simulations of the 'Cappadocia Gate' (Fig. 14) provide realistic visualisations of the original appearance of this impressive monument. These visualisations also highlight a number of outstanding architectural problems that will perhaps be resolved through the complete clearance of the gateway passage and the adjacent internal chamber over the next three seasons. Outstanding issues include the nature of the original road surface, which was presumably paved with stone, and the way in which the gate passage was drained. It may be presumed that the outer section of the gate passage was roofed in such a way as to provide access between the flanking towers, particularly if the passage was narrower than shown in Figure 14. Such an arrangement would have greatly enhanced the defensive properties of the gate. It is not yet clear, however, whether such a walkway over the passage would have been vaulted or carried on long, horizontal beams. It seems possible that the sandstone battlements along the front of the towers might also have been carried across the passage. Depictions of Iron Age city gates from Assyria and Urartu suggest that the passage battlements would have been at the same height as the city wall rather than at the greater elevation of the towers.

A detailed proposal for conservation and limited reconstruction, which addresses the twin concerns of preservation and the safety of visitors, has been drawn up in advance of plans for further clearance and excavation in and near the gate.

Understanding the 'Palace Complex'
Last year's excavation at the eastern end of what has very tentatively been called the 'Palace Complex' included the recovery of a group of complete pottery vessels and lids from the floor of the smaller of the two rooms in Structure C. This year Noël Siver was able to reconstruct these unusual vessels (Fig. 15). They comprise a small two-handled pithos, a large conical bowl which turned out to have two bands of thick white paint on the outside and further paint on the base and handles, and two large flat lids with single handles (Figs 16 and 17).

These coarse vessels were hand-made, all but the pithos being fired at a fairly low temperature. The large bowl was smoothed to a burnish on the inside, the pithos and lids on the outside. The same room contained one fine juglet and fragments of several others, the remainder of which may perhaps await excavation in the other half of the room. The purpose of these large vessels and lids, and therefore the function of the room in which they were found, is enigmatic. Samples have been taken from this pottery in the hope of finding traces of organic residue that might provide further clues.

The same room also yielded a small number of Cornelian cherry stones. Of an altogether different order are the small bone inlays which were perhaps set into small wooden containers (Fig. 20). Both plain petal-shaped and square inlays bearing delicate incised decoration, all discoloured by fire, were found. Whatever precise interpretation might be placed on of the contents of the room, a domestic context would appear to be most unlikely.

Charred timbers found last year in Structure D turned out to be Oriental Beech, which is not suitable for dendrochonological dating.

The narrow space between Structure D and Structure A was crammed full of broken pottery that, in addition to the funnel and tripod bowl illustrated last year, included several hand-made, red ware jugs with cut-away beak spouts. In 2001 Noël Siver was able to restore one complete example (Fig. 24) and to reconstruct partially several others. Although these jugs are all of the same general shape they are not of a standard size and exhibit some variation in the precise form of the spout and handle.

Special Sherds and Particular Objects
In addition to the surprising range of hand-made pottery (Figs 16,17 and 18), portions of fine-ware vessels that have good parallels at other sites on the Central Anatolian Plateau, such as Boğazköy and Gordion, have been recognised (although none were recovered in situ, nor are any examples complete). Collaboration with scholars at Boğazköy revealed that the range of excavated pottery from levels that appear to be close in time to the occupation of Kerkenes contain wheel-made fine-ware vessels that are extremely similar (e.g. black-polished relief ware, Fig. 19), but there does not seem to be a significant hand-made component to the Boğazköy assemblage. The extent to which these observed differences in the ceramic repertoire of these two sites, only 50 km apart as the crow flies, represent slight chronological variation, regional traditions, cultural preferences, and trade remain highly fruitful topics for further research and continued collaboration.

Other diagnostic objects include an incomplete two-pronged bronze pin with a double looped head and an incomplete bronze fibula (Fig. 21), both of well known types and a polished and perforated stone harness trapping (Fig. 22)

Evidence for Writing
The foundation and administration of this great city, as well as that of the territory that fell under its rule, would have required written communication and record keeping. The first evidence for written language at Kerkenes has been found in the form of 10 marks, most or all of which appear to be letters, incised into the surface of pottery vessels recovered from the 'Palace Complex' (Figs 23, 24 and 25). The marks are usually single, although one appears to comprise two signs, and it can be seen that they were scratched in after the pots had been fired. Several have been found high on the shoulders of jugs, where they were incised behind or slightly to the right of the handle. This careful placement argues against idle doodling and suggests that they may be representative of some kind of deliberate record keeping. The presence of a mark on the funnel and another on a base shows that not all, if any, of these marks were used to indicate capacities or contents. Whether they represent potters marks, owner's marks or some form of administrative notation connected with the use of the vessels is uncertain.

The marks, whatever their function, demonstrate that there was some level of record keeping, whether or not it was truly literate. Further, it might well be expected that the language of this particular system was a local (Anatolian) one. If this can be substantiated, it raises the possibility that these few signs represent the first ever evidence that the Cappadocian language may perhaps have been written in an alphabetic script with close affinities to Phrygian.