Contents

 

THE SITE

The Setting

The site is situated on a granitic mountain (alt. c. l,400m.) at the northwestern edge of the Cappadocian plain (Fig. 1). It dominates the east-west trunk road that today links Iran with Europe and is close to one of the natural routes linking the Black Sea with the Mediterranean; it thus lies on a natural cross roads close to the center of modern Turkey. The city wall was skillfully set out around the rim of the eroded granitic dome in such a way as to take optimal advantage of the terrain (Front Cover). The dimensions of the city are approximately 2.5 by 1.5 km. and the length of the city wall is some 7 km., making it the largest known pre-Hellenistic site in Anatolia so far discovered. Most remarkably, the whole of the interior of the city would seem to have been utilized for buildings of various sorts with only the steepest slopes apparently avoided. This intensity of building is the more surprising since it is clear that the life of the city was very short, perhaps less than a generation, and that large parts, including the very massive defensive system, were never completed. The sheer size and urban complexity, however, is not necessarily indicative of a huge, permanent, urban population, and the understanding of these problems remain central goals to the project. A further area of interest, and one that cannot be divorced from the urban structure and the internal functioning of the city, is the economic base that supported, or was intended to support, the city population; and in order to address these problems intense study of the immediate environs and consideration of the wider historical and geographic setting is necessary.

The Historical Background

The date of the site is clearly somewhere in the late pre-Hellenistic Iron Age, the period that goes under the obscure archaeological name of Alişar V (a term derived from the fifth level up from the bottom of the near-by site of Alişar Hoyuk that was excavated by E.F. Schmidt and 11.11. von der Osten in the 1920’s and 30’s). It is now clear, both on the basis of the pottery that was recovered by Schmidt from a number of the small test trenches that he dug within the site in 1928 (Schmidt 1929) and for wider reasons of historical geography and architectural history, that the site in fact pre-dates the western extension of the Persian or Achaemenid Empire, ultimately as far as the Aegean, by Cyrus the Great in or a few years after 547 B.C. Further, the ceramics are not of the (earlier) Alişar lV period which is presumably to be dated to the eighth century B.C. By a process of elimination, therefore, the city on the Kerkenes Dağ has to have been constructed and deserted in the seventh or sixth centuries B.C. In view of the magnitude and strength of the site it might be considered legitimate to assume that somewhere in the scant records of the ancients some reference to the city might be found. If this line of reasoning is accepted and followed, there is only one candidate, the city of Pteria mentioned by the “Father of History”, the Greek historian Herodotus.

The background is fascinating. After the destruction of the Assyrian Empire by the Babylonians and Medes, acting in uneasy alliance, a territorial division of interests was agreed upon. The Babylonians got Mesopotamia and the Medes got the area to the north, including of course Anatolia. The nature of the early Median state, indeed the very existence of a Median “Empire” has been vigorously questioned (e.g. Sancisi-Weerdenburg 1988; Brown 1990) but there is written testimony (indisputably correct, whatever the problems that lie concealed in the scanty details and the reconciliation of western and eastern sources) that the Medes fought a war for five (or six) years with the Lydians and that this war, or perhaps more properly a series of annual campaigns, took place in Central Anatolia. The war came to an end on the afternoon of May 28, 585 B.C., a date known to both ancient Greeks and modern astronomers since it was an eclipse of the sun that frightened both sides to the extent that peace was declared and a peace treaty between the two protagonists brokered by the King of Cilicia and (or on behalf of) the King of Babylon (see Huxley 1965 for evidence that there may have been subsequent hostilities). The treaty fixed the ancient Halys river (mod. Kizilirmak) as the border between the warring parties and the treaty was sealed by the marriage of the Median King Astyages to the Lydian princess Aryenes, daughter of Alyattes and sister of Croesus. There is thus a high probability that the city on the Kerkenes Dağ was constructed shortly after the conclusion of this peace treaty as the western Royal City of the newly expanded Median Empire. Indeed, it is not impossible to imagine that the palace was designed for the newly wed royal couple. The references to Pteria in the histories of Herodotus (1.76) describe later events. Astyages is overthrown in a palace revolt at the heart of the Iranian Empire and Cyrus the Great becomes the first of the Achaememd kings. Croesus, by now King of Lydia and gaining fame as the richest man in the world, sought to flex his imperial muscles on the grounds, doubtless convenient, that Cyrus had been responsible for the murder of his brother-in-law Astyages. Envoys were sent to the most reliable oracles of the time and, in due course, the answer came back: if Croesus were to cross the Halys river an empire would be destroyed. It was, naturally, just what Croesus wanted to hear and after the crops had been harvested he led his army northeastwards, crossed the river and sacked Pteria. According to Herodotus, Pteria lay on a line south of Sinop and was the strongest place in that part of Cappadocia, a geographical description that fits exceptionally well with the remains on Kerkenes Dağ. Croesus, Herodotus goes on to relate, enslaved the population of Pteria and chased away the inhabitants of the surrounding villages. Cyrus the Great rose to the threat and marched rapidly from the east. The two empires fought an inconclusive battle on the plain outside Pteria as a result of which Croesus withdrew to his capital at Sardis to await additional forces from his international allies before, as he naturally thought, the renewal of hostilities in the spring. Cyrus, however, was a man of action and rather than return home or winter on the Anatolian Plateau he led his forces speedily against Sardis with the inevitable result. The traditional date is 547 B.C., although there is some evidence that it may have been a few years later (Beaulieu 1989: 80-82; Burstein 1984; Cargill 1977). But whatever the exact year, the historical picture fits the observed evidence at Kerkenes so closely that it falls little short of proof the geographical location, the strength of the fortifications, the Imperial nature of the newly founded city, perhaps (a subject of ongoing study) eastern influences in the domestic architecture and in the plans of the temple and palace, the short period of occupation followed by total abandonment and the date of the pottery are consistent with the identification. Moreover, there is no alternative historical context that we know of in which the site might fall (although other suggestions have been posited from time to time).

Methods of survey

The 1994 season was largely devoted to finishing all essential balloon photography. As before, grid crosses and control points were marked on the ground with white lime and plotted with a total station. In some areas data was also collected for the production of contour plans and digital terrain models. Levent Topaktaş and his team in METU have the results in LandCAD and AutoCAD and are combining them with plans digitized from photographs over the winter. The overall grid was subdivided into blocks each defined by a coordinate (e.g. A.2). Each block is then enlarged and the positions of the photographs and the digitized plans produced from them are plotted. Figures 5 and 6 are working examples and some corresponding photographs are presented in Figures 3, 4 and 5.

The Defensive System

The defensive system is constructed from uncut granitic bedrock. The nature of the stone is such that it often gives the superficial impression of having been shaped because the stone has a tendency to cleave along flat planes. The wall itself is some 4.5 to 5m. wide at the top. Built against the outer face of the wall, at places where the lie of the ground made additional protection desirable, have been added rectangular towers, buttresses and, in at least two places, larger and more amorphous additions. The whole has then been encased in a stone glacis that slopes at an angle of around 60° and runs up so that it is flush with the top of the wall (not to the base). Despite the great width of the wall, the number and size of the towers, the skill of the military architect and the labor involved in construction, the outer sloping face does not provide much of a deterrent to a determined aggressor since many stretches are little more than two meters in height. it is therefore thought that the original scheme envisaged a huge mud-brick wall (as is usual at Near Eastern sites), but that only the stone base or footing was ever built.

The original city gates can be seen and seven have so far been identified with certainty; perhaps the original total since the wall can be clearly seen for most of its length, although later breeches (Figs 3, 4, 5, 6, 7) may obscure small gates or openings. Each gate was individually planned so that it provided the strongest possible defense in its particular situation (in contrast, e.g. to Hittite gates which seem always to follow a standard plan). Special and complex provisions against attack were made where the stream flows through the west wall: perhaps the Persian entrance of Babylon by means of the river outlet was an old Iranian trick?

It is specially noteworthy that there is no inner system of defenses, nothing to separate off a royal quarter, no citadel or acropolis; and that there was never any intention of constructing internal barriers is clear from the layout of the city as a whole (see below). It is thus evident that no strife was expected within the city itself, a conclusion that has important implications relating to the loyalties, civil and political cohesion and ethnicity of the inhabitants. This apparent unity is reinforced by the striking conceptual conformity of the enclosures and buildings within the city (described below).

No outworks or extramural defenses have been recognized, perhaps not surprisingly given the unfinished state of the city wall. It is quite possible, however, that watchtowers lie obscured beneath later remains on surrounding hilltops.

The Urban Space

The use of space and the structure of the urban fabric is becoming clearer as work progresses, both in the field and with the photographs taken from the blimp in 1993 and 1994. This photographic coverage is now complete and comprises vertical or near vertical shots from a range of altitudes between 60 and l,000m., both black and white and colour and all with considerable overlaps and numerous ground control points. In 1994 a new innovation for the project was oblique photography from the blimp; this was done by adjusting the camera sling so that the camera itself was at an angle of 45° (Fig. 10).

Different areas of the city can now be discerned and clues as to function are becoming apparent. It is also clear from simple observation that certain complexes were located in particular places for very specific reasons or, in other words, there was an overall city plan and the planner(s) had good reasons for the placing of any particular complex. It is not perhaps overstating the case to suggest that what we are able to see is a colonial, Imperial foundation, built as a visible symbol of the new Imperial power, as a sign not only of wealth and strength but also of domination. Thus, in a sense, the site is an “ideal city”. The concepts that lay behind the planning are thus of the greatest interest, both in themselves and for any influence that there might (or might not) have been on later Achaemenid and Hellenistic ideas on cities. The problem of where the influences might have come from, fascinating and intriguing as it is, must await renewed research to the east of Turkey since it is already obvious that the city on Kerkenes Dağ owes nothing to Hittite or to Neo-Hittite tradition, nor to any civilization that was native to the Anatolian Plateau.

Since the difficult task of producing a plan of the entire city is not going to be achieved for at least another twelve months from the time of writing, it is perhaps premature to try and describe the whole city. What follows, therefore, are a selection of rather disjointed observations that are deliberately intended to display, on the one hand, some of what we already know, and on the other, to emphasize the potential that the project still has to offer.

To begin, as it were, from the outside, the position and the functions of the city gates are a logical starting point. Not only (as above) is each gate designed to maximize the defensive potential of its own particular position, but each gate leads to and from specific places and the places from which they provide entry to the city is reflected by the type and function of the urban areas and individual building in proximity to each gate. Perhaps the most important of the gates is the southern one. The plan of the gateway itself reflects the importance, having an internal passage and chamber with towers at the innermost limit. The position affords a magnificent view southwards over the northern part of Cappadocia and snow capped peak of Erciyes Dağ (Fig. 1) can be seen rising above the haze on a clear morning. A broad and gentle road winds gently up from the plain and passes through the gate to the most important area of the city. Immediately inside the cities most important streets cross. One major main branch of the road makes its way downwards along the steep side of the “kale” and joins the northeast gate, another important street runs from the east gate towards the great stone facade of the palace. Opposite the south gate is an area of large terraces and long narrow structures the function of which is currently enigmatic, it seems likely that they were only built to foundation level (unless they were robbed in Byzantine times) and it is possible that this was a complex of storage magazines situated, as might be expected, just inside the “Cappadocian Gate” through which the agricultural products of the plain and caravans coming from the Mediterranean would have passed. Opposite the gate, just a little to the left, is the At Göl or Sülük Göl, the latter name being derived from its population of medicinal leeches which have given it some local notoriety and which attract the sick from considerable distances. Today a frog filled pond, it was an artificial, stone lined reservoir or settling pool at the head of an elaborate and centrally organized system of water control and distribution.

To the left (west) gate is a large enclosed space which extends from the east-west street to the city wall and stretches from the gate to the corner of the palace. The western limit of this enclosure, dictated by a sharp drop in the level of the bedrock, comprises a long narrow (c. 3 m.), corridor like building. The function is unknown but stables, storage or barracks are possibilities and we expect the geophysical survey might provide an answer. On a raised outcrop of rock that physically pervades the enclosed area is a major building and between it and the city wall a rectangular reservoir. One other rectangular building is clearly visible but otherwise the area was empty, an observation confirmed by geophysical survey in 1993. The whole of this enclosed complex, immediately inside the gate and flanking the approach to the palace had some very special and important function that has yet to be determined: it is possible that it was the military headquarters, perhaps with barracks and/or stables for the Median cavahy, exercise yards and parade grounds, but this intriguing possibility has yet to demonstrated.

Turning left inside the south gate, joining the street from the east gate and passing between the enclosure and the Sülük Göl just described, would take the visitor (ancient and modern) to the imposing monumental entrance to the Palace (Front Cover). The situation is ideal because the roof would have provided a view over much of the city and there was sufficient flat terrain for construction on a scale commensurate with the importance of the building. Few alternatives were available to the planners, it would have been possible to build it further to the west but this area was prescribed for different purposes and, in any case, would have neither afforded the view over the city nor, of greater importance, have provided a setting which maximized the imposing grandeur that was the symbolic function of the palace. Today the most imposing area of the city is that Kale (castle), known as Keykavus Kale after a Seljuk sultan of that name, which dominates the mountain from the modem trunk road to the north and which overshadows many areas of the city (Front Cover, Fig. 1). From Hellenistic times (if not Persian) this small rocky outcrop was fortified and occupied. There is no evidence of the role it played within the Iron Age city but the general layout of the city tends the skirt the rock outcrop on all sides and there is no hint that later Kale was of any significance in earlier times. Indeed it is clear that the secular and religious monuments of the Iron Age city lay elsewhere and the bare granite fingers had not yet been adapted for occupation. In 1993 it was ascertained the palace was destroyed by fire.

The east gate leads up from the village of Şahmuratlı, where the expedition has its base, by most gentle route. Modern tracks carved out with bulldozers have obliterated parts of the original route and changed the aspect of the eastern approach. Travelers from the east would have approached this way and have followed the street past the south gate to the palatial area described above or have turned northwards at the crossroads inside the south gate and made their way down town. In moments of high fancy this gate is nicknamed the “Hamadan” or “Ectabana” Gate.

The gate on the northeastern side would have received traffic from the north and west which doubtless came via the large mound at Kuşaklı (described later in this report). It also gave access to and from the extramural temple at Karabaş (also described in a separate section below). This gate gave direct access to the main residential area of the city and the main thoroughfare ran, as described above, to the southern gate.

A small gate in the western wall led to pasturage and to the groups of artificial reservoirs in the valleys to the west that were discovered in 1994 and are described below. Further to the south is a heavily fortified gateway with complex outworks, much altered by later tumuli and shepherds shelters (Fig. 13), through which the road to the prominent “tumulus” at Gözbaba, the highest part of the Kerkenes Dağ, clearly led. This gate, in spite of the strength of the fortifications, does not obviously connect with major routes from further afield, courtyards or open areas and small pools to collect rain water. Some of these structures appear not to have risen above foundation level, an observation that geophysical survey should be able to confirm. These large compounds, it is surmised, were or were intended to be aristocratic residences. To the south a grid like system of streets gives access to a second row of similar enclosures and structure before the ground falls steeply away to the south. The street to the south of the palace has a slight kink, associated with a pool for the collection of water.

Turning again to the major areas of the city, the palace is flanked by narrow streets with large enclosures on either side (Fig. 12). Each enclosure contains a major structure or structures with then broadens out to a width of some 18m. At this point it is aligned east-west and at the western end opens into a square or plaza with a large rectangular building on the western side built just inside the line of the city wall. Contrary to expectations, the gate is not at the end of this street but a little to the north. It is, in fact, the gate that leads to Gözbaba. It is tempting to interpret this wide street as having had some ceremonial function because of its obvious alignment with the rising and setting sun, and with the palace and the “tumulus” at Gözbaba. In 1993 it was established that there was extensive and intense burning associated with structures at the end of the wide street. Later tumuli, constructed on the ruins of the Iron Age buildings, have superficially changed the appearance of the complex relationship between the street, gate, buildings and enclosures, but the blimp photographs (Figs 12 and 13) taken in 1994 clearly show the unity and the overall layout of these features. Further geophysical prospection in this area of the city is a priority for 1995.

The task of planning and understanding the rest of the city plan is in hand and description here would be premature. One preliminary observation that can be made relates to architecture: there is no known parallel for any of the architectural features observed, not the city walls, the palace, the “ceremonial” street or the enclosures with their spacious buildings, and there is nothing that appears to belong to an Anatolian tradition (although it has to be admitted that we know almost nothing of the Anatolian tradition in the late seventh and sixth centuries B.C. on the Plateau). Many of the enclosures and buildings can be easily seen on the photographs and surface observation has shown that many buildings, perhaps all of those that were finished and roofed, were destroyed by fire and that there was considerable use of mud-brick for internal (but not external ?) walls. There is, therefore, great potential for high resolution magnetic survey of these structures which will greatly aid interpretation.

There are a number of outstanding problems which future study will address: the nature and function of the different type of enclosures and buildings; the extent to which open areas within the enclosures were used, if at all, for flocks and herds and what other activities did the populace engage in (industrial, agricultural, domestic etc.). It will be possible, when the city plan is complete, to estimate the area of roofed residential structures and to use this estimate to calculate the likely population of the city. ft is already apparent, however, that space was plentiful and that although there are no large areas that were not utilized the density of population may have been far less than in other ancient cities of similar size. Understanding of the urban framework is dependent on a number of difficult questions of a general nature. It is imperative to discover whether or not the city was fully inhabited all the year round or whether it was in some sense a summer city, perhaps the base for annual military campaigns, the collection of tribute or tax, control of seasonal international caravan trade and so forth. An Iranian tradition of annual migration and seasonal campaign could be postulated and it should be possible to prove or disprove the hypothesis by determining the function of the various areas of the city through a combination of geophysical prospection, coring and, perhaps in 1996, a very limited number of carefully placed test trenches. The function of the city crucial to an understanding, and the impression gained so far is that of an imperial foundation that was intended as a strategic military and administrative base rather than as a self-sustaining city economically dependent on the productivity of the surrounding countryside. This attractive suggestion also needs to be clearly demonstrated and a program of research is being initiated to examine the agricultural potential of the surrounding area and to attempt to reconstruct, as far as is possible, the ancient landscape (forest cover, pasture, arable land and gardens). Modem mechanized farming has lead to the tilling of much marginal land so that superficial impressions of agricultural production must be discounted.