Lecture                                                                                                                   

Geoffrey D. SUMMERS & Douglas BAIRD
Department of Political science and Public Administration, Middle East Technical University, Ankara, TR. / University of Liverpool, UK.
summers@metu.edu.tr / D.Baird@liverpool.ac.uk

CONCERNING THE IDENTIFICATION, LOCATION AND DISTRIBUTION OF THE NEOLITHIC AND CHALCOLITHIC SETTLEMENTS IN CENTRAL ANATOLIA

by Geoffrey D. Summers                                                                                            

Introduction

It might perhaps be fitting to start out by examining what should be understood by the term ‘Central Anatolia’. (The term Central Anatolia is here used in the way well defined by Ian Todd (1980:13–25)) The upland central plateau is physically defined, to a very considerable extent, by geographical features: mountain ranges, the Taurus Mountains and the Pontic Mountains, to north and south, and by the Anti-Taurus Mountains and the upper reaches of the Euphrates River to the east. The western limits might be taken as defined by the Lake District and, further north, by the western edge of the so-called Phrygian Highlands. The eastern part of this region was, more or less, the area of Persian Kapakta which became, of course, Roman Cappadocia (hence the use of the term Cappadocia to denote merely the region of spectacular volcanic landscape in the Kayseri–Nigde–Nevsehir triangle is somewhat misleading to students of later periods). (The shifting boundaries of Roman Cappadocia and the creation of Galatia is complex, for good summaries see Mitchell 1993, Sevin 1998 and Strobel and Gerber 2001:215–234. For the more restricted modern usage of Cappadocia see Sözen [ed.] 1998.) The western part of the region, west of the Kizilirmak, was central and eastern Phrygia.

This large region is characterised by systems of drainage that feed internal lake basins, of which the largest are the Konya Plain and Tuz Gölü. Only two major rivers, the Kizilirmak and the Sakarya, rise in and flow through this region, both eventually debouching into the Black Sea. Large volcanic mountains, Erciyes Dag and Hasan Dag, and many smaller volcanoes, bring stark beauty and great variety to the landscape. Central Anatolia, then, appears to be a place of great diversity, full of ecological niches and micro-climates. Set this diversity against maps pertaining to geomorphological development during the course of the Holocene and the complexity becomes overwhelmingly bewildering. We cannot, perhaps, see the wood for the trees, so I will engage in a little deforestation.

Assume for a moment that the entire Central Anatolian Plateau is an elevated, flat, level plain and then consult the rainfall temperature maps, and peruse the statistics for maximum and minimum temperatures, the number of days with frost, the amounts of sunshine and so forth. (This evidence is often summarised, e.g. by Todd 1980, and the Türkiye Yeni Atlasi.) These statistics are, of course, derived from measurements taken at the places where people live today, not on mountaintops. What is striking in these maps and statistics is the very opposite of diversity, the general, overall similarity. The parameters are more or less the same over this entire region, from north to south and from east to west. I can see no good reason to think that there was significantly more variation over this whole region of Central Anatolia at any time in the middle to late Holocene, that is, over the last 10–12,000 years. I am not suggesting here that the climate did not change, but rather, that such climatic changes as have taken place during this time period would have been more or less uniform over this region that lies between the southern flanks of the Pontic Mountains and the northern edge of the Taurus range. I am not a climatologist, but in so far as I am able to understand the maps and such specialist papers as I have read, I perceive a situation in which the climate, or the weather, of the Central Anatolian Plateau is dominated by the Mediterranean system, barely impinged upon by the Black Sea system. (A useful overview of recent studies can be found in Roberts and Wright 1993.) It might thus be argued that, whatever the effect of any dramatic rise in the level of the Black Sea (Ballard et al. 2001), and regardless of the actual date of any such sea level change, whether in the Black Sea or in the Mediterranean, climatic conditions across all of the Central Plateau would have been more or less uniform. Whether or not there might have been more significant effects on particular microclimates I am not qualified to judge. It seems doubtful to me, however, that any microclimatic changes which might possibly have occurred would have made any significant impact on wider pattern of settlement in the Neolithic period.

The distribution of currently known, permanent Neolithic settlements over the Central Anatolian Plateau does not, however, reflect any such climatic uniformity. (This conclusion has not been substantially altered by the results of many years of survey by Sachihiro Omura and his team, although some Neolithic material, obsidian and pottery, was reportedly found to the west and to the north of Tuz Gölü (Omura 2000:46). The known settlements are to be found in a variety of locations: the marshy margins of lakes (Çatalhöyük, Pinarbasi – Karaman), in the valleys of small, swift-flowing rivers (Asikli Höyük) or by copious springs on high exposed plateau (Pinarbasi – Bor and, more sheltered, Kösk Höyük). Each particular location can, doubtless, be seen to have offered its own individual attractions. If correctly identified, these attractions may offer some rational explanation that might, in turn, provide partial understanding of some of the factors that lay behind both the initial choice and lengthy success of individual settlements. (Explanations are not, however, easy or clear-cut. Few scholars would now see command of obsidian trade as providing an explanation for the location and persistence of Asikli Höyük.) One of many other ways of looking at the evidence to hand is to look at the blanks on the settlement map, and to look at the blanks bearing in mind that very large areas of the Central Plateau have, to some degree or other, been surveyed. It is striking that there are no known sites north and east of the Kizilirmak, including the Seyfe Gölü Basin, no sites in the northern and western parts of the huge Tuz Gölü basin (The only site within the entire basin recognised by Todd (1980 passim) was Sapmaz Köy. Omura, cited in n. 4, has apparently found some traces of Neolithic obsidian and pottery.) and no sites in the Kizilirmak valley. (No archaeological survey of the central portion of the Kizilirmak Valley made before the construction of the Hirfanli Dam has been published.) Any meaningful explanation of the distribution of known Neolithic and, for that matter, Chalcolithic, sites on the Central Anatolian Plateau has to take account of the negative evidence. (These difficulties are not, of course, new. Ian Todd struggled with the same problems in the 1960s, as have many people at this gathering.) There are two possibilities. One, that there was no settlement north of the line from Konya to Sivas. Two, that the early settlements have not (yet) been found. In the remainder of this paper I will rehearse some of the possible reasons for the failure to recognise or locate Neolithic and early Chalcolithic settlements in the northern half of Central Anatolia. My principle aim is to stimulate discussion, rather than to come to any particular conclusion or set of conclusions. It would perhaps be useful to list the possibilities before discussing each in turn. If there is any rational basis for the order of the list, it is to migrate from natural circumstances, such as geomorphology, to archaeological shortcomings (such as the failure to correctly identify pottery sherds).

1. The geomorphological burial and erosion of sites.

2. The failure to identify prime locations as a result of marked differences in hydrology or forest cover.

3. The existence of settlements that did not lead to the formation of (conspicuous) mounded sites.

4. The possibility that early settlement sites buried below later occupation.

5. The focus of surveys in lake and river basins.

6. An inability to recognise early material even when it is present.

7. An obsession with obsidian.

1. Geomorphology

Although the impact of geomorphological change has long been recognised by certain scholars, such as Hans Henning von der Osten who conducted the pioneering and in many ways exemplary excavations at Alisar Höyük, it is only quite recently that some archaeologists have become aware of the scale of the changes. The lowest excavated cultural levels at Alisar Höyük were found to be some 11m lower than the modern surface of the valley floor (Von der Osten 1937 frontispiece, Pl. X and pp. 28–30).

Nevertheless, the observations at Alisar were only seen in relation to local circumstances, with the result that the scale of the changes and, therefore, their full implications for understanding archaeology were not, and very often are still not, recognised. We can to some extent replicate the evidence from Alisar at the site of Kusakli Höyük (Yozgat) which lies within the same drainage system. Even on those occasions where there is some recognition, geomorphology tends to be a cushion to fall back on when all other attempts at explanation have drawn blank. Perhaps I am being over harsh given that the geomorphology of lake beds has long been studied, and I suppose that most scholars realised that most of the mound of Hacilar was below the modern lake bed. But what filled the lakes’ basins came from the erosion of higher ground. It is, for instance, only in the last few years that the Gordion expedition has come to understand that the Sakarya River, which now flows at the foot of the City Mound on the western side, followed a course to the east of the mound in the Iron Age. Phrygian remains have been reported under 12.00m of silt and one recent notice mentioned locating Chalcolithic sites in the vicinity despite the scale and extent of geomorphological change (Marsh 1999; Greaves and Helwing 2001). Asikli Höyük, as was only recognised from an aerial photograph taken from the Cappadocia Hot Air Balloon team, is not just cut by the river but more than half has been washed away by a shifting bed that once ran around the mound and which now runs through it (Esin 1998). Further, it is reported that the lowest levels of Asikli lie below the present level of the streambed. One more example will suffice to make the point. The Iron Age city on the Kerkenes Dag was burnt and destroyed c. 547 BC. Yet within two-and-a-half millennia much of the city has been reduced to stone foundations and bare rock. Sites three times as old in similar upland situations stand little chance of survival. In recent Japanese studies (Naruse et al. 1997; Ishimaru and Kashima 2000) of the Tuz Gölü basin 14C dates may suggest that at least 8m of alluvial fan have accumulated in the last 5,000 years, or since the beginning of the Early Bronze Age. One possibility, then, is that sites remain undetected beneath alluvium in lakebeds and broad valleys. Another possibility is that sites may have been completely eroded away.

2. Rivers, lakes and forests

Changes in drainage system since the Neolithic have been well studied in the Konya Plain. Similarly comprehensive and accessible studies have yet to be produced for the Tuz Gölü basin (but see Naruse et al. 1997), or for the smaller basin of Seyfe Gölü. When Ian Todd made his survey of Central Anatolia in the 1960s he was, of course, aware of changes. But Ian Todd was looking at evidence from all periods and, in addition, he was aiming to identify regions or areas for further, much more intensive, study. In the end, however, other circumstances prevailed with the unfortunate result that to this day the gaps very largely remain unfilled. Perhaps it might be, then, that intensive and carefully targeted survey in areas that could perhaps have been favoured locations in the early Holocene, but which now appear barren, would possibly reveal traces of early, i.e. Neolithic, settlement. The identification of such locations would entail the investment in considerable effort at reconstructing ancient landscapes through intensive geomorphological studies. (A beginning has been made from Kaman (Ishimaru and Kashima 2000; Kashima 2000; Naruse et al. 1997, and also Sayhan 2000) but all the results are still preliminary and require full integration with the archaeological data.)

A related consideration is the extent of dense forest in the Neolithic period, and the consequence of dense tree cover for permanent settlement. Whether or not forestation played a significant role, it is not credible that forest of sufficient density as to exclude Neolithic settlement extended up to the right bank of the Kizilirmak.

Perhaps the most obvious example of the failure to find sites, for whatever reason, is in Southeastern Anatolia. Nevali Çori was not located by traditional survey. Now that we know what to look for things are different, as, for example, the recognition of a carved pillar from the slopes of Nemrut Dag (Hauptmann 2000).

3. Non-mounded settlements

It is difficult not to be transfixed by the concept of mounds, aware though we are that settlement mounds (höyükler) might be largely buried or hugely eroded. Perhaps this conception that settlements formed mounds is correct because early settlement in the Balkans also led to the formation of mounds, e.g. Karanovo. It remains possible, however, that building largely in timber on wooded slopes was of considerable importance, yet is not recognised in our surveys. Any such failure to locate non-mounded sites might perhaps be explained by burial, erosion, or inappropriate research design. (It could be suggested that permanent settlement in the Neolithic period on the Anatolian Plateau would have needed to be sufficiently proportioned to avoid protection from bears and wolves, particularly once domestic animals began to be kept in or near to settlements.) Chipped stone might provide the only obvious clue.

4. Burial below later settlement

An old idea is that the early settlements are not visible because they are buried beneath later occupation that, very naturally, grew largest at the prime locations. Perhaps. Maybe there is an unrecognised Neolithic settlement under the core of the City Mound at Gordion, but I rather doubt it. This particular explanation may have greater validity in areas such as the Konya Plain or North Syria, areas where later settlements grew to huge proportions, but neither Asikli nor Çatalhöyük became Bronze Age cities. If burial of sites is one explanation, in the northern half of the Central Anatolian Plateau it would surely have been burial through geomorphological change, rather than by later settlement, in all but the smallest number of cases.

5. The foci of earlier surveys

The discovery of Çatalhöyük completely overthrew the idea that there was no Neolithic settlement on the Anatolian plateau because it was too cold. The new wisdom focused on the attraction of lakes. This was very convenient because mounds could be seen protruding above the dry lakebeds and could, for the most part, be easily reached even if the drive was circuitous. Survey in upland areas is slower, more strenuous, generally less rewarding (for a variety of reasons) and, therefore, less attractive – particularly to scholars whose training and experience is in the ‘Near East’. As far as the British are concerned, the one man who did look in the hills was Charles Burney – and he rediscovered Urartu.

There is, of course, nothing new here. Todd, in his search for his own ‘Çatal’ found Tepecik-Çiftlik, and was told about Asikli; proof, should it have been needed, that the search for mounds need not be restricted to lakebeds and river valleys. In the Yozgat region we have very small Chalcolithic and EBA mounds in high and exposed locations and a similar pattern pertains in the Bogazköy region. Some of these sites are by northern Anatolian standards quite substantial, like the site of Taslik Höyük which has been sliced to the base in the course of village road improvement. (For Taslik Höyük see Summers et al. 1995:59–61 and Pl. V.)

If there are early sites up in what are today barren hills, and they have not washed away, how are they to be found?

6. The inability to recognise early material

What, however, if the sites are not only there, but were known all along? Simply, we failed to recognise their true age. In 1963 James Mellaart published the second part of a paper on the prehistory of the Konya Plain in which a great deal of material was ascribed to EB II (and EB II was everything before the Indo-European storm, but that is a different story). It is becoming clear that much of the material that mellaart, and those who followed him thought, was diagnostic of EB II is in fact somewhat earlier. The problems include: 1. how much of the material is earlier and, 2. how much earlier is it? For example, some of the sherds collected by Ian Todd from mounds on the edge of the Seyfe Gölü basin are definitely earlier than EB II. With them there is a very little obsidian. This material can be replicated for a number of sites in the Todd survey. Certainly there is much Late Chalcolithic (Alisar fruit stands), and some Middle Chalcolithic. Ronald Gorny's excavations at Çadir Höyük have yielded a 4m sequence of Chalcolithic occupation with a range of 14C dates that stretch back in to the fifth millennium (pers. comm.). A longer stratified sequence is desperately needed and, until such is at hand it appears to me that further discussion will, of necessity, be over-speculative.

7. Obsidian

Dark, sharp and ugly, yet instantly recognisable – the hallmark of the Neolithic in Anatolia. Does the distribution of obsidian really mark routes southwards from the plateau? Why did it not also travel northwards? Are we so infatuated with obsidian that we are not recognising Neolithic settlement in which obsidian is a rarity? Surely we would have recognised other types of chipped stone, even though flint and chert are less iridescent than obsidian, and Neolithic settlement without chipped stone industries are implausible.

Conclusions

I suspect that we will eventually recognise Neolithic settlement in the northern half of Central Anatolia, and I also expect that we will slowly reveal complete Chalcolithic sequences. How much further to the north and north-west they will be found I would not like to guess. A perhaps more pertinent question is the extent to which new discoveries will add to or alter the existing picture and, thus, modify the various theories that are being developed in attempts to offer explanations. It is perhaps improbable, although not impossible, that there are sites of similar proportions to Asikli and Çatal which have not yet been noticed. Large flat sites, on the other hand, remain a possibility – as do sites which have been heavily eroded or buried. Small settlements there must surely be, sparse, elusive, and not perhaps very exciting. There must also be, although even harder to find, sites of a more temporary nature.

References

Ballard, R., F. Hiebert, D. Coleman, C. Ward, J. Smith, K. Willis, B. Foley, K. Croff, C. Major and F. Torre, 2001. Deepwater archaeology of the Black Sea: the 2000 season at Sinop, Turkey. American Journal of Archaeology 105, 607–623
Esin, U., 1998. Paleolitik’ten Ilk Tunç Çagi’nin sonuna: tarihöncesi çaglarin Kapadokyasi. In M. Sözen [ed.]. Kapadokya. Istanbul: Ayhan Sahenk Vakfi, 62–123
Greaves, A. and B. Helwing, B., 2001. Archaeology in Turkey: the Bronze and Iron Ages, 1997–1999. Journal of American Archaeology 105, 463–511
Hauptmann, H., 2000. Ein frühneolithisches Kultbild aus Kommagene. In J. Wagner [ed.]. Gottkönige am Euphrat: neue Ausgrabungen und Forschungen in Kommagene. Mainz: Phillipp von Zabern, 4–9
Ishimaru, K. and K. Kashima, 2000. Geomorphological environment and the distribution of archaeological sites in the region of Kaman-Kalehöyük, Central Turkey. Anatolian Archaeological Studies IX. Kaman-Kalehöyük 9, 167–175
Kashima, K., 2000. The geo-archaeological program for palaeoenvironmental reconstruction in Central Anatolia, 1995–1999. Anatolian Archaeological Studies IX. Kaman-Kalehöyük 9, 177–192
Marsh, B., 1999. Alluvial burial of Gordion, an Iron Age city in Anatolia. Journal of Field Archaeology 26, 163–175
Mellaart, J., 1963. Early cultures of the South Anatolian Plateau, II. Anatolian Studies 13, 199–236
Mitchell, S.,1993. Anatolia, land, men and gods in Asia Minor. Oxford: Oxford University Press
Naruse, T., H. Kitagawa and H. Matusbara, 1997. Lake level changes and development of alluvial fans in the Lake Tuz and the Konya Basin during the last 24,000 years on the Anatolian Plateau, Turkey. In Y. Yasuda [ed.]. Changes in the environment and civilisations in Turkey and Syria (= Japan Review 8), 65–84
Omura, S., 2000. Preliminary report of the general survey in Central Anatolia (1999). Anatolian Archaeological Studies IX. Kaman-Kalehöyük 9, 37–96
Osten, H.-H. von der, 1937. The Alishar Hüyük seasons of 1930–32: part I. Chicago (OIP XXVII)
Sayhan, S., 2000. Geologic and geomorphological characteristics of Kaman and its surroundings. Anatolian Archaeological Studies IX. Kaman-Kalehöyük 9, 193–200
Sevin, V., 1998. Tarihse cografya. In M. Sözen [ed.]. Kapadokya. Istanbul: Ayhan Sahenk Vakfi, 44–61
Sözen, M. [ed.]. 1998. Kapadokya. Istanbul, Ayhan Sahenk Vakfi
Strobel, K. and C. Gerber, 2000. TAVIUM (Büyüknefes, Provinz Yozgat) – ein regionales Zentrum Anatoliens. Bericht über den Stand der Forschungen nach den ersten drei Kampagnen (1997–1999). Istanbuler Mitteilungen 50, 215–265
Summers, G., M. Summers and K. Ahmet, 1995. The regional survey at Kerkenes Dag: an interim report on the seasons of 1993 and 1994. Anatolian Studies 45, 43–68
Todd, I., 1980. The prehistory of Central Anatolia I: the Neolithic period. Göteborg: Paul Åströms Förlag
Türkiye Yeni Atlasi, 1987. Ankara: General Directorate of Mapping
Wright Jr., H. and N. Roberts, 1993.Vegetational, lake-level, and climatic history of the Near East and Southeast Asia. In H. Wright Jr., J. Kutzbach, T. Web III, W. Ruddiman, F. Street-Perrott and P. Bartlein [eds.]. Global climates since the Last Glacial Maximum. Minneapolis: University of Minnesota Press, 194–220

_________________________________

EARLY HOLOCENE SETTLEMENT IN CENTRAL ANATOLIA, PROBLEMS AND PROSPECTS AS SEEN FROM THE KONYA PLAIN

by Douglas Baird                                                                                                        

Introduction

The nature of the appearance of sedentary communities and the adoption of agriculture in Central Anatolia still remains in many ways an obscure topic. The role of the extensive plains, typified by the Konya basin, rather than hilly areas and their associated valleys, like Cappadocia, in these processes is even less clear. The nature of the aggregation and dispersal of population in the early Holocene has also tended to be viewed from the perspective of one or two larger sites like Asikli and Çatalhöyük. To be fully understood such larger aggregations need to be put in a broader historical and contemporary context in terms of both environment and settlement. In this paper I will consider the problems facing attempts to map and interpret the distribution of Neolithic and Early/Middle Chalcolithic sites with specific reference to our experience on the Konya Plain Survey. I will then, in the same vein, consider the evidence for factors underlying observed settlement distributions and locations, population aggregation and dispersion, and changes in those distributions from the early Holocene to Early/Middle Chalcolithic (Periods ECA I–V in the terms of the CANeW Table Ronde) in the southwest Konya basin.

The setting

It is useful to outline a number of the distinctive features of the component of Central Anatolia in which the survey is located and of the survey area specifically. The Konya basin is a high altitude, relatively level, inland drainage basin. It is relatively arid in terms of moisture introduced to the basin by precipitation, with less than 250mm towards the centre of the basin and c. 300mm around its edges (Driessen and De Meester 1969:7–8). Several larger and smaller rivers and also sub-surface sources introduce water to the basin from surrounding hills. These rivers lay down alluvial fans, two of which, the May and Çarsamba fall within the survey area. The relatively level character of the basin might indicate a degree of uniformity but this is deceptive. A volcanic massif with surrounding sedimentary rocks, Karadag, characterises the centre of the basin and dune systems and gravel ridges, shorelines of Pleistocene lakes introduce elements of relief. The flanks of the plain are characterised by rising terrain. Soil types vary significantly as well, with lake marls and soft lime soils forming areas where plant cultivation would be less productive, bajada and hillside soils with significant potential, to alluvial soils with high potential for cultivation, laid down in fans by the larger rivers (Driessen and De Meester 1969). Of course these geomorphological units point to a varied environmental history of which more shortly.

The geomorphological evidence from the plain has implications for key past environment reconstructions, those relevant here concern late Pleistocene and early Holocene environments. It also has implications for a chronology of the depositional record and its role in site exposure, burial and destruction. The main work in the Konya Plain has been carried out by Kuzucuoglu (Kuzucuoglu et al. 1998) and Roberts’ KOPAL project (Roberts et al. 1996). The KOPAL project and Konya Plain Survey have been working closely together to reconstruct the geomorphological history of the Çarsamba fan. The distinct phases of deposition have been dated by relating them to archaeological occupations by coring through site and alluvial sediment sequences and in geoarchaeological trenches in relation to Çatalhöyük. It appears that on the Çarsamba alluvial fan, alluvial deposition commenced very early in the Holocene sometime before Çatalhöyük was occupied and continued at least during the early occupation of the Neolithic site at Çatalhöyük East (Roberts et al. 1996:39; Boyer et al., forthcoming). Perhaps in later phases of that occupation or in the Early Chalcolithic the area around Çatalhöyük saw less flooding, indicated by a buried soil horizon (Boyer et al., forthcoming). This period may relate to the occupation of Early Chalcolithic Çatalhöyük West as two 14C dates from the buried soil horizon suggest. The lower alluvial unit associated with these periods apparently continued to be deposited, further down stream on the edges of the fan, until the earlier part of the Early Bronze Age at least. What is currently unclear in this reconstruction is whether a down stream shift in deposition of alluvial sediment meant the fan grew in area during the Neolithic and later or whether it reached its maximum extent during the Neolithic. A later alluvial unit was deposited under slightly different conditions and ultimately less extensively in the southern area of the fan, at least from the Iron Age onwards (Roberts et al. 1996:39; Boyer et al., forthcoming). Perhaps up to 3m of alluvium masks some of the Pleistocene marl in the area immediately around Çatalhöyük. However, given clear irregularities in the Pleistocene marl deposits and variations in the extent of alluvial sediment distribution, very significant areas of the Çarsamba fan have much lower depths of alluvium, particularly the peripheries. It is unclear how similar the May alluvial sequence is.

Palaeoenvironmental work in the survey area suggests the following; earlier work by Roberts suggests the large Konya palaeolake receded after the height of the last glaciation (Roberts et al. 1996:20). The lateglacial period may have witnessed the presence of a number of smaller water bodies in the peripheries of the basin (Roberts et al. 1996:20). In the early Holocene before the onset of true alluviation the area was probably characterised by a number of standing water bodies in limited depressions in the area of the later alluvial fan (Boyer et al., forthcoming). This landscape of marsh and standing water could potentially be contemporary with the occupations in the ‘Microlithic’ phase outlined below, since 14C AMS dates on the organic clay formed in these widely spread depressions indicates they existed c 79-7200 BC calibrated fan (Boyer et al., forthcoming). Probably from some time before 8200 BP uncal, in the range 7500–7000 cal BC, an environment in which there were relatively regular extensive seasonal floods characterised the area of the Çarsamba alluvial fan (Roberts et al. 1996:39). One of the problems for us is in translating this geomorphological scale of events to a more human scale. Thus the periodicity of regular extensive flooding is a key question, once in five years would have a different significance than almost every year. A number of convergent lines of evidence points to standing water bodies and marshy areas close to Çatalhöyük in the Neolithic (Asouti et al. 1999). Clearly these reconstructions have significant if as yet unresolved implications for Neolithic agricultural practices. The soils may be of high potential, but extensive spring floods and high water tables might have made agriculture problematic in some of the area of the developing fan. However, we should not imagine, particularly in an early phase of alluvial deposition, that the whole area of the Çarsamba fan offered the same prospects and problems and we need also to conceive the potential role of land management strategies like drainage.

Survey strategy and methods

Given the potential for alluvial burial in some areas, survey strategies that include the aim of locating a representative sample of early sites must sample a range of geomorphological units. They must include an understanding of local geomorphological histories to take masking effects of alluvial burial into account in interpretation, and they must have strategies for revealing early occupations buried by sediment and later occupational material. Perforce, this must include an element of intensive survey of both landscape and sites themselves. These factors influenced our survey methodology.

The survey area includes c. 1000km2 of the area around Çumra and south of Konya. This area was designed to encompass a range of soil and terrain types and precipitation belts, so alluvial fan, marl/soft lime, hillside bajada and colluvial soils, and gravel ridge soils (which run through the marl and alluvial fan areas) were included (Driessen and De Meester 1969). Two alluvial fans were incorporated to see if there were different settlement histories on different alluvial fans.

The survey methodology had intensive as well as extensive aspects to gain coverage of ground and simultaneously address problems of identifying small, early and buried sites (Baird 1996). There were two components to the survey strategy, the locating of sites and the investigating of the histories of individual sites. Six methods were employed for locating sites:

1) remote sensing using satellite imagery.

2) Canal walking yielded sites.

3) Field walking recovered sites and artefact scatters, the latter probably relating to ancient agriculture.

4) Visual inspection of the topography to identify raised features or soil marks revealed new sites.

5) Visiting distinctive topographic features marked on maps.

6) Asking local people about the occurrence of höyüks and artefacts in their fields. These methods have added significantly to the site record detailed by previous archaeologists. Site comprehension involved contour survey and intensive collection from a grid of squares evenly distributed across sites (Baird 1996). This produced vital and detailed information about site histories on the many multi-period sites. We have had particular success in isolating site components of particular periods or types and documenting the complex and early histories of particular sites.

The paper will demonstrate the vital role of such intensive survey methods and such associated geomorphological work in locating a representative sample of late Pleistocene to early Holocene sites, partly through comparisons with earlier survey in the area. In addition, the preliminary indications of the nature of settlement distribution and changes in such distribution in the early Holocene will be considered.

Chronological framework

It is obviously essential to have a chronological framework in which to study settlement distribution and changes in distribution. In the Konya Plain Survey area we can currently recognise five distinct phases of settlement defined by peculiar artefact assemblages for the period with which we are concerned in this symposium.

1. ‘Microlithic’ phase, 17,000–8000 cal BC. The earliest assemblages recognisable in the survey area are characterised by microliths but may cover a number of periods. During this phase chipped stone assemblages had a significant component of microlithic tools, including a high proportion of obsidian microliths. Such assemblages have been best defined at Pinarbasi where ‘scalene’ bladelets dominate and small single and multiple platform cores indicate the nature of the reduction strategy. This microlithic phase at Pinarbasi is dated in the early Holocene to c. 8500–8000 cal BC (Watkins 1996:52). Microliths obviously also typify Epi-Palaeolithic assemblages in the late Pleistocene. We can therefore state that survey sites with microliths could date anywhere from Epi-Palaeolithic to early Holocene. Microliths also occur at Asikli and Çatalhöyük apparently in more classic Aceramic Neolithic contexts. The microliths at Asikli are larger than those from Pinarbasi and produced from opposed platform cores (Balkan-Atli 1994, Fig. 2), in contrast to those from Pinarbasi. At Çatalhöyük the microliths are not distinct from those at Pinarbasi but occur alongside more classic late Aceramic Neolithic material (Carter 2000). Given that at Çatalhöyük they were recovered from a small sounding into the earliest levels (Pre-Level XII) it is unclear whether they are part of a later Aceramic Neolithic assemblage or residual material from earlier occupation on the site. There is therefore the possibility that some later Aceramic Neolithic assemblages may have a microlithic component, although microliths are not recorded at Canhasan III (Ataman 1988) or Musular (Özbasaran 1999:152). Survey assemblages with microliths on the Konya Plain may therefore span the period c. 17,000 to 8000 cal BC and possibly some of the Aceramic Neolithic (the last approximately 8000–7000 cal BC to judge by the Asikli, Canhasan III and Suberde dates).

Fig. 3: Chipped stone from Konya Plain Survey sites

2. Late Aceramic Neolithic, 7500–7000 cal BC. Assemblages akin to those at Canhasan III (Ataman 1988) and Musular (Özbasaran 1999) – late Aceramic Neolithic assemblages without microliths, but with distinctive point types and pièces ésquillées can be distinguished e.g. Sancak. These must date 7500–7000 cal BC, within which period the Aceramic Neolithic occupation in Pre -Level XII at Çatalhöyük falls.

3. Ceramic Neolithic, c. 7000–6200 cal BC. Ceramic Neolithic assemblages are well defined by the distinctive range of ceramics (Last 1996) and lithics (Connolly 1999) reported from Çatalhöyük.

4. Early Chalcolithic, c. 6200–5500 cal BC. Early Chalcolithic assemblages were defined by their distinctive painted pottery, particularly typified by red painted linear geometric motifs (Last 1996:152; Mellaart 1965), called by Mellaart ‘Çatal West ware’ (Mellaart 1975:120).

5. Middle Chalcolithic c. 5500–4500? cal BC. Middle Chalcolithic is used as defined at Canhasan I Layer 2A by French (cf. Mellaart 1975:120–122) in terms of its distinctive painted pottery. This phase has also been called Early Chalcolithic II by Mellaart (1965), followed by Last (1996:152) in relation to similar pottery revealed at Çatalhöyük West (Mellaart 1975:120–122). Lattices and more tightly meshed and finer lined designs are more common with different paint colours also evident. We have no indication of the duration of this phase on the Konya Plain, nor any indication as to whether it is synchronous with ‘Middle Chalcolithic’ period designations used elsewhere, such as in the Cappadocia area.

Effectiveness of approach

The significance of the employment of intensive survey techniques to both detect and investigate early Holocene settlement can be gauged by comparing the figures for settlement detected by Mellaart and French’s work in the SW of the Konya basin (French 1970). In the Konya Plain Survey area French’s 1970 publication indicated no sites earlier than the Ceramic Neolithic sites e.g. falling within phases 1 and 2 above. The Konya Plain survey has found seven.

It has been suggested that there are two Ceramic Neolithic sites in addition to Çatalhöyük. On one, a large site, Kerhane, Mellaart (pers. comm.) found an obsidian Neolithic type point. The Konya Plain Survey found another such point at Kerhane. Given the limited evidence and the absence of Neolithic ceramics it would be premature to assign this site to the Ceramic Neolithic or indeed to the Neolithic at all with any degree of confidence. A second site which Mellaart suggested was Neolithic, Reis Tümsegi, has been investigated by the Konya Plain Survey and large ceramic assemblages indicate a Late Chalcolithic occupation, some of whose pottery may have been mistaken for Ceramic Neolithic material. Both sites are in the CANeW sites database in Appendix 2 as Ceramic Neolithic, based on the TAY inventory, but given the state of the evidence cannot be assigned to the Ceramic Neolithic period.

French (1970, Fig. 7) indicates five Early Chalcolithic sites, over one of which there is some question. We have discovered 15 (Figs. 6 and 7). One Middle Chalcolithic (Çatalhöyük West) site was recognised from Mellaart’s excavations (Mellaart 1965) and we have recognised five.

Perhaps most telling is the importance of intensive methods in revealing the presence of the earliest settlement phases of which we were unaware without such methods. However, it is also clear we would not have a representative sample of Chalcolithic sites without such methods either.

Environmental context of early Holocene settlement

We can start to understand something of the ecological and landscape context of this early settlement by combining the site distribution evidence with the reconstructions of the environment and landscape offered by KOPAL and the Çatalhöyük project. Sites with a microlithic component in their assemblages show a wide range of putative site settings. These range from plain edge settings, at ecotone between plain, marsh and upland (Pinarbasi actually just outside the survey area) to three sites in the area of the early Holocene alluvial fan. Given the potential dates of the assemblages with microliths these occupations are likely to pre-date the formation of alluvial fan north of the sand ridge which runs across the Çarsamba alluvial fan, and were likely to be located in areas of marsh and open water bodies. One is located on the sand ridge with such putative marshy and pond areas to the north and riverine and seasonally flooded alluvial settings to the south. The current palaeoenvironmental reconstructions suggest that whatever the predominance of alluvial fan settings for later settlement, such environments were not significant in the earliest phases of settlement detected by the survey. Thus the predominant environmental context for settlement later in the Holocene was not relevant for these communities. Given that we may be considering a lengthy period of settlement and the potential for these sites to span Epi-Palaeolithic to 8000 cal BC or even later, a range of subsistence strategies may well have been practised. Hunting and gathering is a likely strategy for much of this period. However, even if cultivation was introduced during this period to the Konya Plain the site settings, that is limited development of alluvial soils and marshy settings, may suggest hunting and gathering probably retained very considerable importance, as it may well have done at Aceramic and Ceramic Neolithic Çatalhöyük (Asouti et al. 1999).

It is notable that three, later Aceramic Neolithic sites probably occur in areas in which alluvial deposition was on-going (including Çatalhöyük) and spring flooding was significant. Two may well have been situated in more generally marshy settings beyond the areas of alluvial deposition. An aceramic occupation that may also represent an Aceramic Neolithic site was situated on the sand ridge, probably overlooking marshy areas, but was certainly a significant distance from alluvial settings as it is today. Site locations thus remained diverse and marshy settings remained important. It seems likely the communities in this area in this period practised cultivation and probably herding (as at Canhasan III) (French et al. 1972; Mellaart 1975:97–8; Payne 1972) as well as hunting and gathering, successfully adjusted multi-component subsistence strategies to the opportunities and difficulties of the alluvial regime of the Çarsamba at this time.

Ceramic Neolithic period Çatalhöyük was located within an area of active alluviation (Roberts et al. 1996:39). At Pinarbasi the Area B occupation represents a rather different, small scale and temporary rock shelter occupation (Watkins 1996) at the plain/upland/marsh ecotone. Substantial sedentary population seems highly concentrated in the alluvial zone in distinct contrasts with earlier periods.

Early Chalcolithic sites are thus found much more dispersed than in the Ceramic Neolithic period, but almost all still within the alluvial zone of the Çarsamba, both on sand ridges and on the alluvium. However, with a putative down stream shift in alluvial deposition in this period the fringes of the zone might have seen more active flooding than the area around Çatalhöyük. More variety is, therefore, likely within environments across the fan. Nevertheless, such changes are unlikely to be sufficient to explain distributions as sites are located on fan periphery, and possibly just beyond, as well as fan centre. Thus sites were certainly located in those areas where alluvial deposition and seasonal flooding were most likely in the Early Chalcolithic. Middle Chalcolithic sites are found with lower frequency in the same zones.

Factors underlying population distributions and fluctuations

We can also consider evidence of variations in site frequencies and sizes as well as settings from period to period to appreciate something of changes in factors influencing settlement and population distributions through time.

Aggregate site area rather than site frequency is probably the better indication of relative overall population levels. It is unlikely that all sites or the complete area of each site will have been occupied simultaneously through each period, given that these periods occupy minimum of several centuries. We can take into account some of these complicating factors and whilst changes in aggregate site area cannot be used as a direct measure of the degree of demographic flux it is probably a useful guide of relative changes, e.g. significant increase or decrease in population. The trend indicated by the graph of aggregate site area in the different periods would suggest that overall population increases from early Holocene through Aceramic and Ceramic Neolithic into Early Chalcolithic after which there appears a significant drop in the Middle Chalcolithic. Both the site frequency and aggregate area graphs probably overstate the scale of increase from Ceramic Neolithic to Early Chalcolithic. Whilst the one major Ceramic Neolithic period site was occupied throughout that period it is unlikely that all the small Early Chalcolithic sites were occupied synchronously throughout the Early Chalcolithic period although it seems likely Çatalhöyük West was.

It is difficult to gauge the size of many of the sites with microlithic phases, masked as they are by later sites. At sites where there are some indications they are unlikely to have exceeded a hectare. Some of the later Aceramic Neolithic sites are certainly 2 hectares or so in area and (Aceramic Neolithic Çatalhöyük may be more) although others are smaller, settlement remained of a modest size and dispersed.

The situation of the Ceramic Neolithic is thus in marked contrast to earlier and later phases, with extreme concentration of population at one large site. The only other site apparently belonging to this period is Pinarbasi where the rock shelter occupation clearly represents a temporary camp, probably of people engaged in herding and hunting amongst other activities, perhaps from a sedentary settlement like Çatalhöyük or part of a mobile community. We cannot rule out the existence of permanent settlement bases within the survey area contemporary with Çatalhöyük, but they are likely to be both very small and low density. Even if they exist the contrasts with Çatalhöyük would have been extremely marked.

This concentration of population cannot be explained by developments in the environmental setting, i.e the development of seasonally flooded alluvial areas. This is because it is likely that a number of later 8th millennium cal BC settlements, that is of the later Aceramic Neolithic settlements, including one at Çatalhöyük itself, are founded in areas of active alluviation before the growth of Çatalhöyük and apparent concentration of population there.

Our ability to detect small sites earlier than the Ceramic Neolithic makes the apparent relative isolation of Çatalhöyük more plausible. Equally it requires to be understood in the light of those previous patterns of settlement. It suggests it had little to do with adaptation to life on a regularly and extensively flooded alluvial fan and more to do with developments within and between a network of pre-existing communities. Relatively large sites, of the size of Çatalhöyük, are often understood as central sites within the context of relationships that extend beyond individual settlements. Such centrality might be defined in economic, social or political terms, and in terms of asymmetries in consumption or production. It seems unlikely that Çatalhöyük can be understood in terms of such centrality in the absence of a network of sedentary communities within close proximity with which it might develop such relations. It is unlikely that relations with more distant communities, perhaps on the Karaman fan or in the lake area, respectively 70 and 80kms away, involved such intimate connections that we can see Çatalhöyük playing a central role for them. It remains possible of course to see Çatalhöyük as a central site for mobile communities in the general area. However, if we cannot understand Çatalhöyük as an economic, ritual, or administrative centre, then perhaps its large size becomes even more intriguing.

Given densities of Neolithic structures at Çatalhöyük in both Mellaart’s excavations (Todd 1976) and Hodder’s scraped areas (Matthews 1996) and evidence from Mellaart and Hodder’s excavations for the nature of Neolithic households on the site, relatively dense occupation occurred on the mound. It may be the case that at a number of periods, perhaps at the end of the Neolithic occupation and potentially before Level VI, not all areas of the mound were occupied simultaneously. Nevertheless, given the evidence for occupation around the time of Level VI on both northern crests and southern contemporaneously it seems likely that there were periods when much of the mound was occupied simultaneously. Even allowing for putative open areas and the fact not all structures may have been occupied simultaneously, minimum populations are likely to number in the low thousands with up to 6000 quite possible (Matthews 1996:86; Todd 1976:122–3).

Wobst (1974, 1976) has pointed out that mating networks require populations in excess of 500 people to avoid the ill effects of inbreeding. Communities below this threshold would, thus, usually be exogamous. It is clear that Çatalhöyük with population levels as proposed above had the potential to become an endogamous community. The endogamy of such a large community may have discouraged the establishment of other communities in the vicinity since this situation would have restricted the ‘marriage’ options of exogamous communities in the neighbourhood. A community’s control of proximate territory and resources is typically enhanced by endogamy (Bintliff 1999:532–3) as are its options for managing those resources corporately. Such factors may have helped maintain the large size of a community at Çatalhöyük and discouraged emigration. That Çatalhöyük had a subsistence and other resource base involving an extensive geographical area might be inferred from several lines of evidence. These include firstly the large size of the community (Todd 1976:124–5), especially if the extensive seasonal flooding made some proximate areas unsuitable for cultivation in some years at least. Secondly there is the evidence of materials, edible and other, from plain edge, steppe, and hill areas, particularly tree related products (Asouti and Hather 2001) and fruit and nuts (Asouti et al. 1999), but also minerals and stones, e.g. volcanic ground stone materials (Baysal 1998). Thirdly some of the wall paintings indicate hunting deer (and possibly other animals) in what were presumably wooded hilly areas. Fourthly, animal management strategies for sheep and goat herds are likely to have required their movement from the immediate vicinity of the site at some seasons, particularly during extensive spring floods. Such a far-reaching exploitation of the fan and adjacent areas may also have discouraged the growth of proximate communities. The large potential warrior force that the community could call on might also have discouraged near neighbours if the inhabitants of Çatalhöyük had been so minded. We might see in this context why the combination of large site and the absence of contemporary smaller communities might not be unsurprising. The intriguing question then becomes why Çatalhöyük became so large in the first place and why this situation changed into the Early Chalcolithic (see below).

The evidence above allows two alternative processes in the development of the relatively large community at Çatalhöyük. Firstly, the growth of one late Aceramic Neolithic community on the fan at Çatalhöyük as others disappeared and thus potentially at the expense of other communities. Secondly, we might envisage the amalgamation of a number of pre-existing late Aceramic Neolithic communities at Çatalhöyük. This perhaps led to precocious growth and explains the enduring concentration of population at Çatalhöyük along with the possible factors enumerated above. Some have argued that the size of Çatalhöyük is unlikely to be accounted for by the demographic growth rates of 0.25–0.35 % typically encountered in historical times within traditional Central Anatolian communities (Yakar 1998:668–9). But it is far from clear such growth rates should be taken as a guide, much greater growth rates have been sustained historically in pre-industrial communities in some parts of the world, up to 1 % (Livi-Bacci 1997:22). In addition, Angel (1971:82) in his analysis of the Çatalhöyük human remains from Mellaart’s excavations postulated growth rates of 0.8 %. Todd (1976:125) thus calculates that even with modest founder populations Çatalhöyük could easily have reached the estimated maximum population figures within the span of occupation of the site. However, the argument for an amalgamation of later Aceramic Neolithic communities at least has the advantage of obviating such doubts about growth rates. Such a scenario might even help explain the demography of the community, such an amalgamation resulting in changing marriage practices that could have boosted population increase. Given the ability of late Aceramic Neolithic communities to exploit the fan on a dispersed basis, ecological conditions are unlikely to explain the concentration of population at Çatalhöyük. We must rather see it as a peculiar function of distinctive social and political alliances within late 8th millennium cal BC communities. These clearly persisted for a lengthy period given the several hundred year duration of Çatalhöyük as a particularly large site in a landscape where nearby sedentary settlement sites did not exist or were small, low density, and relatively transient. As indicated above we might understand this persistence in terms of the nature of a continuing corporate management of the community’s extensively distributed resources.

Early Chalcolithic settlement is represented by a large settlement of c. 8 hectares at Çatalhöyük West alongside numerous (14) dispersed smaller sites. Clearly this represents a further radical change in settlement and social relations on the plain. Boyer and Roberts (Boyer et al., forthcoming) have indicated that the extensive flooding characteristic of the central area of the Çarsamba fan is less evident at the transition to the Early Chalcolithic, at least in the geoarchaeological trench next to Çatalhöyük. This may correlate with other evidence for down stream shift in the depositional centres of the fan suggesting that the fringes of the fan saw more extensive flooding at this period. It is not impossible that this encouraged the dispersal of settlement across the fan area that we see in the Early Chalcolithic. But we should not see environmental conditions as a dominant feature of distribution in that Early Chalcolithic. Early Chalcolithic sites are found toward fan peripheries and perhaps just beyond them as well as in the central area, so again seasonally flooded areas would have been an important zone within the subsistence practices of the Early Chalcolithic occupants.

It is tempting to correlate this Early Chalcolithic spread of small settlements with the decline in size of the area occupied in the Çatalhöyük setting. This correlation would imply a colonisation of the surrounding territory by groups from Çatalhöyük. There are interesting implications deriving from, and questions about, such a conclusion. Presumably this would reflect change in the social relations at the larger community of Çatalhöyük itself, but also possibly the carrying over of some of the relationships found within one settlement to integrate the more dispersed settlements. Such substantially larger sites such as Çatalhöyük West are often conceived as central sites. The classic economic model would see such central sites supplying goods and services to surrounding communities, and representing a concentration of administrative or economic functions. We may be able to examine some of the survey evidence with these questions in mind, although evidence is unlikely to be conclusive in the absence of excavation. But whether or not we can document such centrality we should consider the possibility that Çatalhöyük West is central in other ways. This might be as the large parent community where certain social or religious institutions existed not found at other sites and which might occasion disproportionate visits to such a community. None of the other 14 Early Chalcolithic sites exceeds 4 hectares in area and most are significantly smaller, mean size of these Early Chalcolithic sites is 1.6 hectares. Very few are likely to have been inhabited by more than 500 people and any that were would only just have exceeded such a total. Whether or not Çatalhöyük West was the parent of the initial group of such smaller communities, given what we have said earlier about the size of mating networks communities of this size would have needed to exchange marriage partners. It seems likely that Çatalhöyük West was a much larger community with a bigger pool of partners and thus may have had a central role to play in the exchange of reproductive partners across the Çarsamba fan. Broad similarities in the decorated ceramics between Çatalhöyük West and the other sites may indicate those regular personnel flows especially if the movement of marriage partners involved individuals who were potters. Alternatively it may be that the community at Çatalhöyük West was endogamous, it may have been large enough. In such case we might expect contrasts between the material culture at Çatalhöyük West and the shared material culture of other Early Chalcolithic sites.

The Middle Chalcolithic sees a decline in both the number of sites and aggregate site area. This is because it would appear that Middle Chalcolithic pottery occurs not just on fewer sites but also is more restricted in its distribution at a number of sites at which it has been identified. At Çatalhöyük West Middle Chalcolithic (Early Chalcolithic II) pottery is not found as extensively as the classic Early Chalcolithic pottery (Last 1996:152) suggesting reduction in the area of the settlement. At Musluk, one of the Konya Plain survey sites, only a few Middle Chalcolithic sherds are documented compared to many of the Early Chalcolithic style. It seems likely that some settlements contracted, as well as there being a lower number of sites of this period. It seems possible that the Middle Chalcolithic may be a shorter period on the Konya Plain than the Early Chalcolithic. However, there is no direct evidence for that and indeed unless the Late Chalcolithic in the Konya plain was exceptionally long the Middle Chalcolithic period is likely to be at least as long as the Early Chalcolithic. Whilst some disparity in settlement numbers may be related to differing lengths of the two periods, the scale of decrease in numbers of settlements and reduction in site areas, seems to argue for settlement retraction and population decrease. Within the geoarchaeological record documented by the KOPAL project there are no coincident events that might suggest significant changes in environmental conditions, and thus whilst we cannot rule this out it seems any such putative events were minor. A further reason for discounting a significant impact of widespread environmental change on Middle Chalcolithic settlement in the SW Konya plain is the thriving of Middle Chalcolithic Canhasan I on a similar alluvial fan only 70kms to the southeast. It may well be that there were significant shifts of population between such alluvial fans and between plain and hills in the early Holocene as well as those documented above on one part of the plain itself.

Conclusions

The Konya Plain Survey evidence indicates the importance of including intensive survey combined with geomorphological and geoarchaeological approaches in survey design and interpretation, particularly in relation to the detection and recording of a representative sample of early Holocene settlement evidence. Without such intensive survey the picture of early Holocene settlement will be substantially deficient. The palaeoenvironmental evidence suggests environments in the plain changed significantly over the earlier part of the Holocene. Early Holocene sites were thus located in settings different from those found on the plain today. Their distributions must be examined and understood in relation to those past environments. In relation to 9th–8th millennium cal BC settlement it is notable that a series of marshy settings, and from the later Aceramic Neolithic onwards seasonally flooded alluvial landscapes as well, formed the context from which the first sedentary and agricultural communities exploited the plain.

The evidence revealed by such survey can play an important part in understanding changes in the social landscape of settlement in the early Holocene and in the alluvial fan contexts of the Konya basin. It suggests that considerable fluctuations in population aggregation and dispersal were common and quite independent of specific environmental developments. In addition, it is possible to start to examine the role of the larger aggregations in these settlement distributions. At least in the case of Ceramic Neolithic Çatalhöyük we should not look for explanations framed in terms of an understanding of such sites as economic, administrative or religious centres for networks of proximate communities. Rather we might consider them aggregations bound together by social and possibly political practices that made them distinct from many Neolithic communities.

References

Angel, J., 1971. Early Neolithic skeletons from Çatal Hüyük: demography and pathology. Anatolian Studies 21, 77–98
Asouti, E., A. Erkal, A. Fairburn, C. Hastorf, A. Kennedy, J. Near and A. Rosen, 1999. Archaeobotany and related plant studies. Online: Çatalhöyük 1999 archive report http://catal.arch.cam.ac.uk/catal/Archive_rep99/fairbairnkennedy99.html
Asouti, E. and J. Hather, 2001. Charcoal analysis and the reconstruction of ancient woodland vegetation in the Konya Basin, south-central Anatolia, Turkey: results from the Neolithic site of Çatalhöyük East. Vegetation History and Archaeobotany 10, 23–32
Ataman, K., 1988. The chipped stone assemblage form Can Hasan III: a study in typology, technology, and function. Unpublished PhD, Institute of Archaeology, London
Baird, D., 1996. The Konya Plain Survey: aims and methods. In I. Hodder [ed.]. On the surface: Çatalhöyük 1993–1995. Cambridge and London: McDonald Institute for Archaeological Research and British Institute of Archaeology at Ankara, 41–46
Balkan-Atli, N., 1994. The typological characteristics of Asikli Höyük chipped stone industry. In H. Gebel and S. Kozlowski [eds.]. Neolithic chipped stone industries of the Fertile Crescent. Berlin: ex oriente, 209-222
Baysal, A., 1998. Provisional report on geological surveys in relation to groundstone study, Çatalhöyük 1998. Online: Çatalhöyük 1998 archive report http://catal.arch.cam.ac.uk/catal/Archive_rep98/baysalstone98.html
Bintliff, J., 1999. Settlement and territory. In G. Barker [ed.]. Companion Encyclopedia of Archaeology Vol 1, 505–545
Boyer, P., N. Roberts and D. Baird, forthcoming. Holocene environment and settlement in the Konya Plain, Turkey: integrating geoarchaeology and field survey
Carter, T., 2000. Chipped stone. Online: Çatalhöyük 2000 archive report http://catal.arch.cam.ac.uk/catal/Archive_rep00/carter00.html
Conolly, J., 1999. The Çatalhöyük flint and obsidian industry. Oxford: BAR International Series 787
Driessen, P. and T. de Meester, 1969. Soils of the Çumra area, Turkey. Wageningen: Centre for Agricultural Publishing and Documentation
French, D., 1970. Notes on site distribution in the Çumra area. Anatolian Studies 20, 139–148
French, D., G. Hillman, S. Payne and R. Payne, 1972. Excavations at Can Hasan III, 1969–70. In E. Higgs [ed.]. Papers in economic prehistory. Cambridge: Cambridge University Press, 180–90
Kuzucuoglu, C., M. Fontugne, M. Karabiyikoglu and C. Hatté, 1998. Evolution de l’environnement dans la plaine de Konya (Turquie) pendant l’Holocène. In M. Otte [ed.]. Anatolian prehistory: at the crossroads of two worlds. Liege: Liège University, 605–624
Last, J., 1996. Surface pottery at Çatalhöyük. In I. Hodder [ed.]. On the surface: Çatalhöyük 1993–1995. Cambridge and London: McDonald Institute for Archaeological Research and British Institute of Archaeology at Ankara, 115–172
Livi-Bacci, M., 1997. A concise history of world population. Oxford: Blackwell
Matthews, R., 1996. Surface scraping and planning. In I. Hodder [ed.]. On the surface: Çatalhöyük 1993–1995. Cambridge and London: McDonald Institute for Archaeological Research and British Institute of Archaeology at Ankara, 79–99
Mellaart, J., 1965. Çatal HüyükWest. Anatolian Studies 15, 135–156
Mellaart, J., 1975. The Neolithic of the Near East. London: Thames and Hudson
Özbasaran, M., 1999. Musular: a general assessment on a new Neolithic site in Central Anatolia. In M. Özdogan and N. Basgelen [eds.]. Neolithic in Turkey, the cradle of civilization. New discoveries. Istanbul: Arkeoloji ve Sanat Yayinlari, 147–155
Payne, S., 1972. Can Hasan III, the Anatolian Aceramic and Greek Neolithic. In E. Higgs [ed.]. Papers in economic prehistory. Cambridge: Cambridge University Press, 191–4
Roberts, N., P. Boyer and R. Parish, 1996. Preliminary results of geoarchaeological investigations at Çatalhöyük. In I. Hodder [ed.]. On the surface: Çatalhöyük 1993–1995. Cambridge and London: McDonald Institute for Archaeological Research and British Institute of Archaeology at Ankara, 19–40
Todd, I. 1976. Çatal Hüyük in perspective. Menlo Park: Cummings Publishing Company
Watkins, T., 1996. Excavations at Pinarbasi: the early stages. In I. Hodder [ed.]. On the surface: Çatalhöyük 1993–1995. Cambridge and London: McDonald Institute for Archaeological Research and British Institute of Archaeology at Ankara, 47–57
Wobst, H., 1974. Boundary conditions for Paleolithic social systems. American Antiquity 39, 147–78
Wobst, H., 1976. Locational relationships in Paleolithic society. Journal of Human Evolution 5, 49–58
Yakar, J., 1998. The question of social structure in Neolithic Anatolian communities. In M. Otte [ed.]. Anatolian prehistory: at the crossroads of two worlds. Liège: Liège University, 665–671

__________________________

Go to the related debate (Panel : Sevil Gülçur, Catherine Kuzucuoğlu)