Çatalhöyük 1995 Archive Report

ÇATALHÖYÜK 1995 ARCHIVE REPORT

Excavations at Pinarbasi, 1995 season

Trevor Watkins

The second season of excavations at Pinarbasi took place in September 1995. This season the excavation was concentrated on the Late Neolithic occupation in one of the, rock-shelters (Area B, where soundings were begun last year). The excavations were funded by the British Academy, the, British Institute of Archaeology at Ankara, the Society of Antiquaries of London and various funds at the University of Edinburgh. Our collaborator on the excavation (carried out jointly by the University of Edinburgh and the Karaman Museum) was Cengiz Topal, Assistant Director of the Karaman Museum. We are most grateful to him for all his work and indispensable assistance, and to the Directorate General in Ankara for facilitating the fieldwork. Because of the great pressure on living and working space at Çumra, the Edinburgh team was accommodated in Karaman, and had working space provided by the Karaman Museum.

The cluster of sites at Pinarbasi was noted in 1993 in the first season of survey work in the region around Çatalhöyük in the Konya plain. As well as a series of small-scale occupations ranging from the Early Bronze Age to the Byzantine periods, there is evidence of substantial early prehistoric occupation in several rock-shelters and on an open village site below the cliffs and at the very edge of a small, spring-fed lake. The purpose of investigating the sites is to recover evidence of human occupation and use of the Konya plain earlier than and contemporary with the settlement of Çatalhöyük during the period when hunter-gather groups in some parts of western Asia first became sedentary and then turned to cultivation and finally to herding.

The purpose of the second season of excavation was to continue and extend the soundings in one of the rock-shelters, begun last year. It was clear from last season's sounding that early prehistoric occupation of the rock-shelter described as Area B was well-preserved, and potentially deeply stratified. An isolated hearth at the back of the shelter has given a radiocarbon date in the late fourth millennium BCE (uncalibrated), but the last significant occupation of the shelter has now been radiocarbon dated to the middle to late sixth millennium BCE (uncalibrated). that is, the late Neolithic towards the beginning of the Chalcolithic periods.

Last season's sounding showed us part of a substantial, stone-built structure, dug into already existing archaeological deposits. It was carbon from the fill of this structure that yielded the late Neolithic dates. The objective of this season was to see more of the stone structure, reach its floor, to excavate more material from the, archaeological levels into which the structure had been dug, and if possible sample the material below the floor of the stone structure. Because we needed to pass all the material excavated through either wet-sieving and flotation or dry-sieving, it was decided that we should postpone further work on the open village site (Area A), and concentrate more effort on the processing of materials produced from a single excavation area.

A new, improved design of flotation and wet-sieving machine was built in Karaman to a design that is now also in use at Tell Brak in Syria, Kilise Tepe and Çatalhöyük in Turkey, Similar sampling strategies are now in use with identical machines giving harmonised recovery strategies across a broad range of sites, covering periods from the Neolithic, through the Chalcolithic to the Bronze Age and beyond. The new machine seems to be more efficient and productive than its predecessor. In the event, the small excavation team processed more than four metric tonnes of archaeological deposit, of which more than one tonne was passed through the wet-sieving and flotation process. All the resultant heavy residues from the wet-sieving were separated into four fractions (>10mm, >5 mm, > 3 mm and > 1 mm) and sorted according to a regime tested in the first season. The >10 mm and >5 nun residues were completely sorted; 20% of the >3 mm heavy residue was sorted; the >1mm heavy residue was weighed and abandoned. The decision to limit excavation to Area B proved to be correct, as the team was just able to process that volume of excavated material through the complicated and labour-intensive wet and dry-sieving and sorting processes.

Last year's 4 x 2 metre trench (Trench 1) was re-opened, and a second trench (Trench 2) of the same size was opened beside it, giving a nominal area of 4 x 4 metres. A narrow baulk was retained between the two trenches to preserve the main east-west section across the excavation area. So much new material was obtained from this comparatively limited area, and there were such restrictions on the time available in general, that no attempt was made to re-open excavations on Area A, the open village settlement beside the lake below the cliffs.

Trench 2

The large pit dug illicitly before September 1993 in the back of the rock-shelter was mostly within the area of the new trench. The pit was emptied and the sides cleaned; it was found to be about 1 metre deep, and dug entirely through stratified archaeological deposits. The rock-face exposed at the back of the rock-shelter was seen to be blackened by smoke, and the slope on the surface of the rock was still dipping backwards. These observations suggest that there is still some considerable depth of deposit below the bottom of the pit. As in last year's trench, where there were two shallow, stone-lined pits, so this year in the new trench, a third, similar pit was found close to the surface. Like the others, it seems to have been used as a hearth, with large, blackened stones in the bottom, covered by a layer of fine ash. Between the looters' pit and the south side of the trench a series of strata could be observed, including at least two deposits of a light grey, fine material but the area available for excavation was too limited to be useful in seeing what these deposits were and how they functioned.

Immediately below the surface of the ground, in the NW quarter of the new trench, were found the tops of several large slabs of rock set on edge. This setting of large rocks proved to have considerable depth. Within the setting there was a deep layer of black ash, above which there were many large stones deposited in more ash- However, it is difficult to see this structure as some sort of deep hearth, since the large slabs that formed its sides were conspicuously unmarked by fire, smoke or heat. However, if the hypothesis that the structure was some form of fire-place is dismissed, it is even more difficult to think what purpose it may have served. Until the surrounding deposits have been excavated down to the level of the base of the slabs, and the whole structure can be fully investigated and dismantled, further speculation is fruitless. Elsewhere in the new trench the disarticulated bones of the skeleton of a small child were found, indicating the former presence of a burial that had been disturbed in antiquity.

Because of the complications of the stratigraphy and the degree of disturbance in part of the trench, it was not possible to penetrate the stratigraphy as deeply as we had wished in this area. An alignment of large stones was visible at the bottom of the excavated area in part of the trench at exactly the point that we would expect to see the continuation of the wall of large stone structure in last year's trench. Further exploration of this stone alignment must wait until next year, when the later strata have been dug and removed and the large structure of slabs has been dismantled.

Trench 1

One of the differences between the deposits dug last year was that chipped stone (obsidian and flint) was more common in Area A, the eighth millennium BCE open village site, than in the rock-shelter, Area B. Large whole earth samples, especially from the deposits around the large stone structure, were therefore passed through the flotation and wet-sieving process in order to recover the maximum amount of chipped stone material from the rock-shelter site.

Last season had explored the latest phases of occupation of the rock-shelter (radiocarbon dated to the middle to late 6th millennium BCE), and had located an earlier, large stone structure, This curvilinear stone structure represents the revetment wall around a very large depression cut into the earlier deposits. The structure was filled in deliberately with lenses of almost clean soil and stone interdigitated with thin lenses consisting almost entirely of charcoal and animal bone.

Excavation around the outside of the large stone structure revealed a succession of thin strata that all contained considerable amounts of carbonised plant remains (mostly wood charcoal) and animal bones, and are putatively of earlier date. Amounts of chipped stone varied from one layer to the next, but a larger sample of chipped stone has certainly resulted, Preliminary analysis of the chipped stone has resulted in tables of statistics that await detailed analysis against the stratigraphic evidence, but first impressions are that these strata represent rubbish dispos4 and the chipped stone may well represent an earlier phase in the occupation of the site, Further radiocarbon dates will be obtained before next season, which, together with the careful analysis of the records of the assemblage, may show the degree of chronological difference between this set of strata and the large stone structure that was inserted into these levels.

The deep deposits of material thrown into the interior of the large stone structure were investigated to their base. The deposit consists of deep strata of brown soil and thin but dense lenses of charcoal and animal bone. It appears that the filling of the structure was a rapid process, during which the deposits did not have time to weather or erode. The whole extent of the wall of the structure within Trench 2 was revealed. It consists of three or four courses of very large boulders piled one above another. Its purpose was not as a free-standing wall, but as a revetment against the soft deposits of rubbish into which the structure had been cut.

The wall was found to be standing on a deposit of light brown soil and gravel, mixed with frost-fractured limestone that originated from the roof of the rock-shelter. The lower part of the depression that formed the structure had been dug into this deposit, which is obviously quite deep. This underlying stratum of light brown and stony material was also found at a higher level outside the stone structure, Here, to the south-west of the stone structure and below the stratified deposits of ash and bone described above, the light brown, gravelly material appeared as a ridge, which had been cut away to form hollows to the south-east and the west. These hollows await further investigation next year, when the trench can be extended, and when the wall of the stone structure are dismantled.

It is clear that this light coloured, gravelly deposit, full of frost-fractured stone, is mostly of natural origin. At present we do not know how thick it is. A small test was dug into this material at the base of the large stone structure. This test showed that this material contains some fragmentary animal bone. The first hypothesis is that it represents a natural accumulation of material during an extended period with no human occupation. There are enough indications among the chipped stone materials recovered both last season and this year that there is an earlier occupation very close by, from which some characteristically earlier pieces were accidentally finding their way into the late Neolithic deposits,

Cultural assemblage

Given the radiocarbon dates (in the sixth millennium BCE), and the clear resemblance between the chipped stone tools in the latest levels at Pinarbasi and those from Çatalhöyük, the cultural assemblage from Pinarbasi is distinctly eccentric. In particular it is deficient in both pottery and heavy ground stone tools. The only pottery recovered this season, as last season, came from the latest deposits immediately below the surface in Trench 2. There was very little, and only some of it may be Neolithic in date. The late Neolithic chipped stone assemblage is very heavily dominated by obsidian, and flint is a considerable rarity. It is striking that, although cores of obsidian are very rare, there are plenty of preparation flakes. This suggests that the occupants of the site were at least resident long enough to need cores in order to make tools. The absence of other tools than those of chipped stone, especially food preparation equipment such as mortars, pestles and querns, suggests that they either did not use cereals and pulses or that they brought them to the site already prepared. There is a distinct shortage of bone tools, also. So far, only one fragment of spatula and one bone point of a needle have been recovered. Two very small disc-shaped beads complete the very limited repertoire of the site in the late Neolithic period.

The central question relating to the later Neolithic occupation concerns the, nature of the occupation and the relationship of the occupants to the communities that lived on höyüks on the plain. Were the people who occupied the rock-shelter in the sixth millennium BCE using it seasonally, for example, as a hunting station for killing and butchering wild equids? Or were they a small permanent community at the margins of the society that occupied the plain, earning their living by supplying, for example, meat of wild animals? Detailed study of the cultural assemblage, the animal bones and the botanical evidence will be needed to help us to address this question. The second question, which is unlikely to be answered conclusively simply from what has been excavated to date, is whether the stratigraphically earlier deposits dug in Trench 1 this year represent earlier occupations of significantly earlier date. Radiocarbon dates from the strata earlier than those dated last year must be awaited.

Environmental and economic data

The botanical material from last season is still being examined by Mark Nesbitt at the Institute of Archaeology in London. The great bulk of the botanical material seems to consist of wood charcoal, often in microscopic pieces, and there are very few seeds. This season, flotation produced just over 10 kg of carbonised plant materials. Plans are now in hand to begin the study of the wood charcoals.

The zoological material from last season is also still under study by Dr Nicola Murray in Edinburgh University. It is very well preserved, and is proving to be rich in potential information. This season, a further 34 kg of excellently preserved animal bone was recovered. Denise Carruthers, a postgraduate student in the Department of Archaeology at Edinburgh University, who has now completed a year of training in zoo-archaeology, will undertake the detailed study of the animal bones under the supervision of Dr Murray. The material from the, late Neolithic deposits in the rock-shelter are particularly prolific and well-preserved. What is immediately apparent is the very large quantities of bones of an extinct equid species. Other species represented include very large wild cattle, sheep and goat, fox, hare, many birds and some reptiles, The study of the late Neolithic material will be of particular interest in conjunction with the nearly contemporary material now being excavated at Çatalhöyük. It is already apparent that there is a considerable degree of difference in the bone assemblages from the late Neolithic in the rock-shelter and the early Neolithic in Site A, the open village settlement beside the lake.

Conclusions

This season's excavation has produced an excellent representation of the material from the late Neolithic, and possibly earlier, occupations in the rock-shelter, Area B. We now have a much better picture of the large stone structure, its rapid re-filling with occupation debris, and its stratigraphic relationship to the last stages of occupation above it, although we still have to extend the excavation in order to recover the full plan of the structure, and find if it was part of a larger-scale occupation. We hope that we have the first representative samples from the earlier strata into which the large stone structure was dug, whose age will be determined shortly by AMS dating. The sampling and recovery strategy has produced excellent samples of carbonised plant remains, and plans for extensive research on the mammal bones and the wood charcoals are now in hand. Examination of this year's archaeobotanical remains will tell us if we have seed remains as well as the ubiquitous wood charcoal. We intend to have first results from the carbonised plant remains and the animal bones, together with the second group of radiocarbon dates, ready for reporting at the 1996 Symposium. The study of the wood charcoals should be taking place in the summer of 1996, by which time we expect to have heard the results of the analysis of the lake-bed core from nearby Süleymanhaci Gölü from Dr Neil Roberts' team.

Publication

The report on the first season of survey is already published in the Annual Symposium series. A report on the first season of excavations was given at the 1995 Symposium, and the publication copy left with the organisers. A preliminary paper is planned as part of Ian Hodder's first volume of Çatalhöyük Research Project papers. This will cover the first two seasons of excavation, is due to be completed before Christmas, and to be published early in 1996. After the 1996 season, an interim report will be written in the autumn and offered to Anatolian Studies., Specialist papers can be expected to begin to appear in journals in 1997 on the cultural assemblages, and the zooarchaeological and archaeobotanical materials. Since the work is part of the Çatalhöyük Research Project, it is anticipated that most of the definitive publication of the excavations will appear in that publication series. As the lead researcher in this project, I would expect the results of the research to feed the series of my publications in recent years on the nature of the processes involved in the shift from mobility to sedentism, and from hunting and gathering to intensification, storage, cultivation and herding. These articles are likely to appear in regional journals such as Palaeorient and prehistoric journals such as Proceedings of the Prehistoric Society.

Next season

For next season, we need to operate on a more substantial scale in order to be move forward the research at a reasonable speed. Now that the excavation conditions are known, and we have recovery techniques and sampling strategies tested and in place, we may begin to use a number of archaeology students from Edinburgh, Liverpool and Istanbul as contributors to the field research team. In order to continue the work on the rock-shelter we need to extend the area under excavation at the same time as we dig deeper into the deposits. In particular we should plan to excavate through the stratum of nearly sterile, light coloured material into which the large stone structure has been cut in order to reach any underlying and earlier strata. The field research has divided into three stages, each absorbing about one third of the team's efforts: excavation; processing of the deposits by wet-sieving, flotation and dry sieving; and sorting of the resulting heavy residues. We need to add a fourth stage to the process, allowing us to do more effective preliminary analysis of the materials recovered while we are in Turkey. It is unreasonable to expect Turkish authorities to allow us either to export all the materials for study in Britain, or to delay study until some future 'study season'.

Useful contacts were made with the international team based at Kara'in and working in that cave and at the site of Okuzini (led by Prof. Yalcinkaya of Ankara University and Prof. Michel Otte of Liege). It is hoped that next season we shall be able to exchange research visits with this team and work towards ensuring direct comparability of the results from the two teams. In particular, the Kara'in team is interested in learning more of the recovery techniques (flotation and wet-sieving) and sampling strategies being employed. The opportunity to co-ordinate work on the important stratigraphic sequences near Antalya and those on the southern plateau is something worth working for.

Dr Trevor Watkins,
University of Edinburgh,
Department of Archaeology