ÇATALHÖYÜK 1993 ARCHIVE REPORT

Analysis of field section from 1960's excavations

Wendy Matthews

Introduction

In the first season of renewed investigations at Çatalhöyük a total length of ca. 51m of field sections, 175m2 in area, were cleaned, photographed, drawn at 1:20 and described, in Area A and the northern half of Area E of the 1960s' excavations. Selected areas of these sections were in addition drawn at 1:5, described in further detail, and sampled for micromorphological and pollen analysis.

The aim is to learn and record as much as possible from the surviving sections before they erode any further. The cleaned sections were correlated with plans of Levels VII-III published by Mellaart in his preliminary reports and general book (Mellaart 1962, 1963, 1964, 1966, and 1967). The features present in these sections enabled study of the architecture, occupation sequences, and site formation processes. The experience gained from these analyses will be of value in future excavations.

Section conditions

The sections in Area A and the northern half of Area E are better preserved than those in the southern half of Area E and were selected as the first target for section cleaning. The sections in Area A and the northern half of Area E were 3-4 metres high. The sections in the southern half Area E are higher, standing at ca. 8 metres, and were more dangerous with large segments of wall about to fall away from the section face. During the almost thirty years since Mellaart's excavations the upper areas of the sections have been eroded and the base of the sections buried behind steeply sloping banks of eroded materials which stand up to 2 metres high. In many instances in both Areas A and E Mellaart excavated up to the edge of building units and left the walls standing along section faces. In the uppermost levels, the walls of buildings have often eroded and fallen away exposing first the walls and then occasionally the internal room fills of unexcavated buildings behind. At the base of sections, building walls are better preserved, protected behind the banks of eroded materials. White wall plaster still adhered to the face of walls at the base of the sections. Room fills and ashy deposits in middens were more susceptible to erosion than mud brick walls.

Section cleaning

Each section was photographed to record the condition of the section face prior to cleaning. The thin skim of silty wash which adhered in cracked patches to the uppermost area of the sections was then removed by trowel, artist's palette knife and plasterer's leaf blade to expose a freshly cut section face. The steep sloping banks of eroded material which had accumulated against the bases of the sections were carefully removed with a shovel by the foreman of the workmen Ismail Yasli who was exceptionally dextrous and sensitive to the texture, structure and consistency of deposits, and any buried features. He stopped removing earth at the interface of the eroded deposits and the extant section face, leaving grains of eroded material clinging to the section face which were then cleaned by trowel, palette knife and leaf blade. The banks of eroded material were thus removed in a series of platforms to the very uppermost fragments of surviving wall plaster inside rooms, at which point no further earth was removed and the whole of the section was recorded at this level. No areas of wall plaster were uncovered. The plaster remained protected in its current buried environment. At the end of this season all of the earth was banked back up against the base of the sections, often to higher levels than before, to minimise further erosion.

Section recording

All of the cleaned sections were photographed, drawn at 1:20. Depositional features or stratigraphic units were described recording: unit number, deposit type, moisture, Munsell Colour Reference, texture, structure, and inclusions.

Floor sequences, fire installations and fixtures within these sections were photographed, drawn at 1:5, and described in greater detail in order to record the microstratigraphy of these activity areas.

Sampling

Micromorphology

Nineteen samples were collected for analysis in large thin sections, 13.5 x 6.5 cm. Each sample comprises a solid block of sediment ca. 13.5 x 6.5 x 8cm which was cut out of the face of a section using a Swiss Army knife and saw. Each block was supported by a tightly fitting plastic box which was inserted before the block was detached from the section face. Once detached, each block was wrapped tightly in strong laboratory tissue and parcel tape, and. labelled. The samples were then air dried on an open shelf in the expedition workroom, before packing for export.

In the Department of Archaeology, University of Cambridge, the samples have been impregnated with an unsaturated crystic polyester resin in styrene monomer, which takes six weeks to harden. The thin sections will be ready for analysis in January 1994.

In large thin sections it is possible to study the type, form and contextual relationship of all constituent mineral, organic and artefactual components in floors and occupation deposits, and to identify agents and contexts of deposition. In this first season the aim has been to sample a representative selection of the range of deposits and sequences present in the cleaned sections, namely:

Pollen

A small series of pollen samples were taken for pilot study in order to ascertain whether or not pollen was present in occupation deposits both within rooms and in open areas within the settlement. Samples were also taken from mud brick and floor plaster in order to examine whether there was any pollen in architectural source materials. All of these samples were taken adjacent to thin section samples. It is hoped the latter may provide some indication of the depositional context and history of any pollen grains present. The pollen analyses will be undertaken by Dr Neil Roberts and his team.

Correlation of sections with plans from 1960's excavations

It has been possible to correlate the sections closely with the plans of the 1960's excavations by analysis of building levels, and room size, layout and associated features. Plans from both the preliminary reports and the general book were reproduced at uniform scales and were overlain to reconstruct the series of building levels excavated by Mellaart. The correlations refer to section numbers and wall letters assigned during the 1993 season, and level and building numbers assigned by Mellaart.

Section numbers 1 and 2

Sections 1 and 2 are at right angles to one another in the south of Area A. Section 1 faces west, Section 2 faces north. The following correlations have been made:

It should be noted that when excavated, Level VIA-VIB Shrine 61 had a bench with bulls horns set along its length (Mellaart 1963, 50-52, Plate VI), and Level III Shrine 1 had elaborate wall paintings which depicted scenes of men dressed in leopard skin loin cloths hunting a large bull and deer (Mellaart 1962, 62-65, Plates XIV-XVIII).

Section numbers 3,4 and 5

Sections 3, 4 and 5 lie in the northern half of Area E of the 1960's excavations. Section 3 joins the western edge of Section 2 at right angles, facing west, and is 21 metres long, running up to the edge of deeper excavations in the southern half of Area E. Section 5, facing south, includes a short stretch of wall running east-west which connects Section 3 to Section 4 which also faces west. The following correlations have been made:

Level VII Shrine 44 when excavated was decorated with leopards in relief which were plastered and painted (Mellaart 1966, 176-178, Plates XXXV-XL).

Observations

A few brief observations on the information gained from analysis of these sections in the field will be made here as a preliminary report. More detailed observations are pending analysis of the thin section samples from January 1994.

Architecture

The walls at Çatalhöyük are constructed from large mud bricks which are generally 76-100cm in length, 29-30cm wide, and 8-10cm high, and are laid in walls a single brick thick. The bricks would have required considerable care in handling, and perhaps some form of leather carrier to transport them from the place of manufacture to the site of the wall. It has been suggested that the bricks may have been moulded in-situ on the walls, but this requires experimental testing. The bricks range in colour and texture from:

They were on the whole made from natural sediments exploited from the surrounding region, and stabilised with up to 10-20% vegetal remains represented by linear and curvilinear voids in the mud brick matrix.

The mortar is often dark- brown in colour, 10YR 5/2 greyish brown - 10YR 3/2 very dark greyish brown, and has many anthropogenic inclusions and charred fragments suggesting it was simply excavated from surrounding occupation debris, probably from open areas. This practice continued for as long and as far afield as the third millennium BC in Tell Abu Salabikh in central southern Iraq.

Study of the wall plaster really lies beyond this particular research project and requires analysis by conservators. A few pieces of fallen wall plaster were studied under a hand lens at Çatalhöyük, and described. A thin layer of slightly creamy silty clay plaster, 10YR 712 light grey - 10YR 8/2 white, with vegetal stabilisers is often laid as a foundation to a finishing fine skim of whiter plaster, 10YR 8/11 white, without stabilisers. The surface of this upper coat appears to be smoke darkened in the samples examined. Up to 25 layers of plaster were counted in a series 8mm thick. The total thickness of these plaster fragments was 2.9cm.

A preliminary reading of literature on the environment surrounding Çatalhöyük, suggests source materials for the wall plasters may be found today within 5km of the site, and perhaps even closer ca 6,000 BC when backswamp and fan deposits were perhaps not as extensive. Dutch soil surveys carried out in the Konya Basin and Çrumra area record chalk soils of "almost pure soft calcium and magnesium carbonate" which have formed on soft Neogene Limestone 50-100cm below the surface, and are classified as the Kasinhani soil series west of Çatalhöyük, and the Ürünlü series to the south-cast (de Meester 1971, 57; Driessen and de Meester 1969,42-46). The soft chalk ranges in colour from 10YR 8/2 white -5Y 8/1 white. The surface soils overlying this soft chalk include a clay-loam 10YR 7/2 light grey in colour, which is comparable to the thin layers of foundation plaster observed in the samples examined under the hand lens. The soils have a clayey texture and are very sticky and plastic. The soft chalk and chalk soils are likely to have mixed readily with water to form a pliant plaster, without the use of combustion, for example, as in the preparation of quicklime plasters. The water table is less than 160cm below the surface and in the area of the Kasinhani series sometimes occurs as surface water or pools (Driessen and de Meester 1969,42). The chalk soils are mainly used for cattle grazing today. It would be exciting next year to examine a profile of soft chalk soils, to test their properties, and to investigate by augering whether they occur closer to the site buried by more recent fan deposits.

Sample 6 was taken at the juncture of floor plasters with the wall plaster in the building at the southern end of Section 3. The walls were plastered much more frequently than the floors.

Double wall construction. The walls as already stated were constructed from single widths of mud bricks up to one metre in length. The load bearing outer walls of buildings often abut walls of adjacent buildings, separated by a thin ca. 2-8cm layer of vertical packing which includes lumps of assorted hard-core. Dr. J. Coulton suggested that the characteristic abutting or double walls at Çatalhöyük and earlier sites such as Asikli Höyük may have been critical design elements in sharing or supporting the weight of the roofs , on the basis of ethnographic parallels.

Modifications. One quite widely spread practice was the emplacement of a second mud brick wall in front of the plastered surface of an earlier wall, which was also observed during the surface clearance operations. The new interior wall was then itself replastered on many occasions. This for example took place in Level VIB Shrine 61 Section 2. In this instance it covered a burnt wall in the western half of the building. Such practices obviously extend the use-life of any building, and argue for considerable longevity in the history of individual buildings. In other instances the plaster on the original wall was very thick and irregular, and may have no longer have been suitable for plastering or decoration, and was therefore blocked in behind the new wall which could be replastered smoothly.

Rebuilding

At least four methods of rebuilding were practised at Çatalhöyük:

Fixtures

Few fixtures or plastered features were exposed during the section cleaning operation.

Two fire-installations from Section 2, were probably plastered hearths or ovens with some collapsed superstructure. These fire-installations were constructed with a mud brick base and sides, and lined with successive applications of plaster 1-3cm thick. At least three layers of plaster were applied in the eastern fire-installation, twelve in the western fire-installation. The fire bowls were cleaned out before the application of each layer of plaster. No fuel residues were visible between each layer of plaster in the field. Each layer of plaster was rubified by the heat at the base of the fire.

The bench in Section 3 was constructed from a variety of mud brick sizes, and was covered in layers of plaster to a thickness of 1.3cm.

A number of moulded features such as horn fittings were visible in the sections, particularly in Section 3, associated with a Bos jaw and 3 horns. These were manufactured from a distinctive reddish sediment which was heavily stabilised with added vegetal remains represented by voids in the sedimentary matrix.

Occupation sequences

All of the occupation sequences within buildings exposed by section cleaning comprised sequences of thin floors with little overlying occupation debris which had accumulated to depths of less than 15cm. Analysis of the sequence of floors at the southern end of section 3 suggests there was a change in the use of the building. The first plaster floors are 1-2.5 cm thick and are made from a sandy clay loam 10 YR 5/4 yellowish brown. Each floor is overlain by trampled occupation debris with burnt aggregates, in layers 2-3cm thick. This sequence abuts a plastered scoop which may have been a placement for a pot or wooden vessel, measuring 27cm in internal diameter, and was probably associated with domestic activities. The plastered scoop and these coarser floors are sealed and succeeded by a very different sequence floors and occupation deposits. These plaster floors comprise fine orange sandy clay, 7.5YR 6/6 reddish yellow, 0.5-0.6cm thick, or a thin skim of white plaster 2.5Y 8/2 white, <0.l cm thick. Intervening accumulated deposits are loam-ashy loam, 10 YR 6/2 light brownish grey- 10 YR 4/1 dark grey in colour and less than l cm thick. The high quality of these later floors and the scrupulous standards of maintenance, in association with a Bos jaw and three horn cores and moulded plaster features in overlying collapse, suggests the last phase of the use of this building was probably related to more formal and perhaps ritual activities.

The floor sequences in the adjacent room to the north in the same building, suggest a similar change from domestic to more formal uses of space. The earliest floors are thick and overlain by coarse burnt trample. These are succeeded by the construction of a bench which is then plastered with successive layers of thin white plaster which also cover the edge of the floor.

Open areas

Much of the western half of Section 2, and areas of Section 3 comprise layers of ash rich in charred remains up to 3 cm in diameter, and bone. These layers variously cut or abut buildings, indicating that they accumulated both in large open pits, and in open areas or courtyards. The sediment input in the open layers of ash varies in colour and sediment concentrations according to the presence and nature of the surrounding mud brick.. The layers of ash in Section 2 therefore include reddish brown sediments from the cut stubs of Wall N, brown sediments from the face of Wall Z, and orange sediments from the face of Wall X.

Some of the ash layers in the northern half of Section 3 accumulated from in-situ burning in shallow scoops, 60-180cm in diameter. The remaining ash layers presumably accumulated from hearth and oven rake-out and from layers of burnt rubbish swept or thrown out of the houses. The clean sequences of floors and occupation deposits within buildings must be studied in the light of the large quantities of refuse dumped outside of them.

Site formation processes

The entire sequence of floors and occupation deposits in any building level yet studied is less than 15 cm thick, and comprises <10% of the build up of deposits associated with an entire building phase. Most building walls stand to an extant height of 1.25-2.15m, and are usually infilled and levelled with collapsed building debris, which constitutes >90% of the build up of deposits from an entire building phase.

There has been considerable post-depositional disturbance by root action, animal burrows and insect activity, particularly in the upper 50cm of the sections. There are also concentrations of calcareous nodules which require further study, in thin section.

The largest disturbances occur in Section 3 in the form of intrusive pits of Hellenistic or later date. A large shaft in particular occurs more than 2 metres below the excavated surface of Level III, and is 7m in length. Todd suggested these large shafts may have been dug in search of the sandy materials employed in the construction of the Neolithic mud brick walls.

Future work

The aim in the future season will be to clean the remaining sections from the 1960's excavations, where safe to do so, in order to increase the sample of building types and occupation sequences.

List of Figures

© Çatalhöyük Research Project and individual authors, 1993