ÇATALHÖYÜK 1999 ARCHIVE REPORT

Archaeobotany and Related Plant Studies

Arkeobotani Ve İlişkili Bitki Çalışmaları

Eleni Asouti (EA), Aylan Erkal (AE), Andy Fairbairn (AF), Christine Hastorf (CH), Amanda Kennedy (AK), Julie Near (JN), Arlene Miller Rosen (AMR)

Abstract

The earlier deposits in the South Area produced a rich assemblage of charred and mineralised remains. In them we discovered flax, stabling deposits with evidence of fodder and lime burning deposits with evidence for the use of dung as fuel.

In the BACH area three floors excavated here this season revealed evidence of crop processing in the form of chaff and cereal remains, feasting and post-consumption discard, a child burial with a cache of hackberries inside (possibly an offering) and a myriad of fire pits for cooking food, smoking and cleansing, with wood and chaff used as fuel.

Charred and mineralised plants were recovered in the KOPAL area. They included the same taxa as those found on-site, though the assemblage here was smaller and certainly not as well preserved. One of the more unique finds was a cache of mineralised awns providing our first piece of good evidence for Neolithic cereal processing off-site. A find of molluscs and algae remains suggested that there was also a wet area here or nearby in the Neolithic.

Charcoal analysis identified twenty four tree species and phytolith analysis produced evidence for domestic activities within buildings and identified different types of plants used for baskets and matting which suggest that particular types of grass or sedge were used for particular functions. Experimental work on the use of bulrush was undertaken.

Özeti

Güney alanının erken tabakalarını zengin kömürleşmiş ve mineralize olmuş kalıntı vermiştir. Bunların içinde keten, saman kalıntılarının gösterdiği üzere ahır alanları ve tezek kalıntılarının bulunmasıyla da yanık tabakalar keşfedilmiştir.

BACH alanında bu sezon kazılan üç tabanın verdiği sonuçlara göre: tahıl ve tahıl kabuklarının gösterdigi kadarıyla bu alanda ürün işlenmiş ve tüketimi yapılmıştır. Ayrıca çitlenbikle dolu bir çocuk mezarı ile çok sayıda yemek pişirmek için kullanılmış çukurlar yakacak için kullanılmış tahıl kabukları ve ağaç kalıntıları ile birlikte bulunmuştur.

Kömürleşmiş ve mineralize olmuş bitkiler KOPAL alanında da ele geçmiştir. Burdaki buluntular yerleşim alanıyla aynılık göstermesine rağmen daha az miktarda ve aynı derecede korunmuş değildir. Buradaki en önemli buluntulardan biride işlenmiş olarak bulunmuş tahıldır. Bu ilk defa olarak bize Neolitik dönemde yerleşim dışındada tahıl işlemenin yapıldığını göstermektedir.

Kömürleşmiş bitkilerin analizleri bize yirmidört farklı ağaç türünün burda mevcut olduğunu ve fitolit analizleri ev içindeki aktiviteleri kanıtlamış ve sepet yapımında kullanılan üç değişik ot ve saz çeşidini ortaya çıkarmıştır.

1. Introduction (AF, CH, AK, JN)

The main focus of research at Çatalhöyük in 1999 was the six month excavation of the South Area. Excavation of several buildings and deposits at the base of the mound provided an excellent sample series to investigate temporal and spatial patterns of plant remain preservation and plant use in this part of the site. Of particular interest was determination of the groundwater-level in the mound and investigating the effect on archaeological preservation of recent hydrological changes in the region. There was the possibility of organic preservation in the lower levels of the site and in the KOPAL trench as a result of anaerobic conditions caused by waterlogging. In the case of plants this meant the possible preservation of a complete range of plant remains, including the fragile elements, such as leaves, that are unlikely otherwise to survive.

The BACH team, now in its third year of excavation, continued excavation of building 3 and auxiliary rooms, excavating through the floor levels in building 3. The archaeobotany team followed the same collection, processing and analysis procedures as in the other excavation areas, with blanket soil collections from every unit excavated. 1999 saw an increase in the number of samples processed because the project excavated over a longer period than in previous seasons.

The KOPAL project opened a large excavation trench to the north of the east mound in order to expand on the information gathered from smaller off-site exposures in previous years. The archaeobotany team sampled for plant remains in this area in order to establish the types of botanical preservation that exist here as well as to address questions of off-site plant related activities and environmental conditions during the occupation of the mound.

The 1999 field archaeobotany team consisted of Meltem Ağcabay, Eleni Asouti, Hatice Bilgiç, Aylan Erkal, Andy Fairbairn, Christine Hastorf, Amanda Kennedy, Katy Killacky, Harpreet Malhi, Arlene Miller Rosen, Slobodan Mitróvic, Julie Near and Michele Wollestonecroft. Contributions to the heavy residue processing were also made by Banu Aydınoğlugil, Serdar Cengiz, Makbule Ekici, Aslı Kutsal, Elif Kavas and Perihan Kösem.

The main field tasks for the archaeobotany team included providing sampling advice and field team support; flotation of soil samples; heavy residue sorting; field analysis of botanical remains and direct feedback of the results to the rest of the Çatalhöyük team. These tasks were shared throughout the team. Christine Hastorf was the team co-ordinator and was in attendance during the main South/BACH/KOPAL season, taking responsibility for the BACH area with Julie Near, who also took main responsibility for KOPAL and provided cover for the South Area. Andy Fairbairn and Amanda Kennedy were in attendance for the six-month season and were responsible for the South Area archaeobotany and general management of field processing. Meltem Ağcabay, Katy Killacky, Harpreet Malhi and Slobodan Mitrovic supervised flotation and heavy residue sorting, Slobodan taking the main responsibility for heavy residue processing during most of August and September.

Other team members’ contributions were made during visits of between 3 days and six weeks and most volunteered to help with the core field tasks as well as completing specific research projects. Arlene Miller-Rosen (phytoliths) contributed to the botanical effort, with an extended visit during August. Eleni Asouti (charcoal analysis) visited between mid-July and August. Aylan Erkal and Michelle Wollestonecroft completed ethnobotanical work, including extended observations and experiments on the use of Scirpus maritimus in July and August. Wendy Matthews (micromorphology), whose work contributed to many archaeological interpretations, had considerable input to the archaeobotanical research through her microscopic analysis of site deposits, including their botanical components. Wendy Matthew’s work is included in a separate section of the archive report.

As with previous years, workers from the local settlements of Küçükköy and Çumra were important team members, working on the flotation tanks and in the residue processing area throughout the season. They included: Riza Büyüktemiz, Mevlüt Sivaz, Saliha Sivaz, Hassan Tokyağsun, Hatice Yaslı, Hülia Yaslı, Hüsein Yaslı and Gazi Yaslı as well as Havize Sarikaya, Fatima Yaslı, Rabiye Yaslı, Suna Yaslı, Fatimana Yaslı, Saliya Yaslı, Saliya Sivas and Mustafa Tokyağsun.

2. Plant macro-remains (AF, CH, AK, JN)

2.1 Methods

Although most of the methods employed for recovery of plant remains were the same as in previous seasons (see archaeobotany archive reports 1996, 1997, 1998), some changes were introduced during the 1999 season in an ongoing effort to improve the system. Additional procedures were also employed to assess the potential for waterlogged preservation in the lower deposits of the South and KOPAL areas.

2.1.1 Sampling

Sampling for flotation of charred plant remains continued as in previous seasons; a flotation sample of approximately 30 litres was collected from each unit where possible. Additional samples from units, also of 30 litres, were often required after consultation between excavators and specialists. Samples from floors were collected on a grid system and some spatial control of samples was also achieved in middens and dumps. A total number of 1604 flotation samples, totalling 26,188 litres of soil, were collected from 1042 units and processed during the 1999 season. Most of these had plant remains recovered from both the light and heavy fractions.

In addition, a parallel series of 15 litre samples were collected from units in which waterlogged preservation was possible. These 15 litre botanical ('BOT') samples were collected from the South Area (22 samples) and KOPAL trench (15 samples) and processed by wet-sieving. BOT samples were sealed in plastic bags and transported to the processing area in flour sacks, the former preventing dessication and the latter preventing damage to the delicate plastic bags. In addition to the South Area and KOPAL trench samples, several samples were also provided for analysis from a KOPAL core collected from between the west and east mounds.

2.1.2 Recovery

Charred remains were recovered using Ankara and SMAP type machines as described in previous archive reports. Samples from potentially waterlogged units were wet-sieved by hand. A frame with a 4mm mesh bottom was set up over the largest settling tank of the SMAP flotation machines. It was designed to hold firmly three medium sized 500 µm mesh brass test sieves. A subsample of 1 litre was taken from each 15 litre 'BOT' sample and wet sieved by hand using a hose with a spraygun attachment fed by the flotation machine pump. Samples were gently disaggregated and the sediment matrix washed through the sieve. The mesh size was small enough to collect most potential waterlogged plant remains, but was large enough to let the matrix pass through.

Material collected in the sieve was then temporarily stored in water in resealable plastic bags. Subsequent processing and storage depended on the outcome of microscopic examination. If waterlogged organic remains were present, ethanol (IMS) was added to these samples (up to 5%) and the bags were stored in sealed plastic containers until further analysis could be undertaken. The remaining 14 litres of the sample was then sent for wet sieving and further sampling was determined by consultation with the dig team. In samples containing no waterlogged organics, the remaining 14 litres of the sample were dry sieved for artefacts following the standard site procedure.

2.1.3 Heavy residue sorting and density calculations

The previous year's methods were altered slightly, after lengthy consultation with the relevant specialists, to improve the system. These changes are recorded in detail in the diary entries of the 26th and 31st of May. Briefly they include the introduction of a riffle box for sub-sampling, changes in the type and quantity of bone recovered and the disposal of fully-sorted residue to reduce storage requirements.

2.1.4 Sorting and field analysis of flots

Preliminary analysis in the field consisted of a mixture of estimating plant remain abundance, preliminary identification and, for priority samples, sorting, weighing and counting of plant remains in the larger sample fractions (2mm and 4mm) using a low powered dissecting microscope. Plant remains from both the light fraction (‘flot’) and heavy fraction (‘heavy residue’) were recorded in this way. Charcoal was picked from the samples at this stage for analysis by Eleni Asouti (see below). The procedure (‘field sort’)was similar to that used in previous seasons, but was also slightly altered to suit the aims of the 1999 six-month season. The major changes included the addition of a ranking system to enable rapid estimation of the abundance and diversity of suites of botanical remains without the need for time-consuming counts and sorts. The procedure formalised established scanning procedures, allowed us to assess sample contents more rapidly, and was useful for evaluating the composition of the smallest flot fractions and rich heavy residues that could not realistically be fully sorted.

Identification and quantification was limited in the field by the facilities and time available. Only the larger sample fractions were sorted and quantified using rough plant remain groups (e.g. cereal grain). Identifications were, where possible, noted on record sheets with reference to standard identification manuals. A visit to the British Institute of Archaeology at Ankara (BIAA) in September provided an opportunity for secure identification of a range of specimens using the botanical seed reference collection in the Institute’s research laboratory. Table 5 provides the latest species list for the site using field and lab data from all areas sampled during the main excavation seasons.

The 'flotlog' (the flotation log used to record all processed samples) was maintained on an Excel spreadsheet stored on the site computer network. Basic sample information was enhanced with the addition of unit and matrix descriptions, building, space and feature numbers (not immediately available due to a backlog of unit data input). It enabled the archaeobotanical team to keep a record of priority units and helped to ensure that examples of all unit interpretative categories were analysed in the field. The flotlog was also used by other specialists and proved to be a useful management tool.

2.1.5 Field assessment of 'BOT' samples for waterlogged remains

After processing, the 'BOT' samples were scanned using a low-powered dissecting microscope to assess the preservation of waterlogged and mineralised remains in the deposits. Samples were first poured into a 500 µm mesh sieve and distributed across the mesh surface. After the sample had drained, the receiver (base) was attached and the sample was then scanned. The assessment records were made on pro-forma sheets (Figure 30) using the relative abundance (ranking) scale also used for the charred remains (recorded in the quantity column). After scanning, the samples were returned to the storage bag and stored in water with ethanol.

2.1.6 Priority tours and liaison with the field team

Priority tours by the specialists continued through the season every second excavation day. The tours provided an opportunity for the specialists to keep up-to-date with developments on-site, consult with the field team about specific sampling problems, give feedback to the field team about preliminary analytical results and to decide processing and analytical priorities. At least one archaeobotanist was present on each tour. The results and notes on discussions during each 'priority tour' were recorded in a communal notebook - the ‘priority book’ - which was taken by the botanical team's attendant on each trip. As well as the formal priority tours, liaison between the archaeobotany team members, excavation and specialist teams continued informally on a daily basis and contributed to generally good relations between all concerned.

2.2 Area reports (AF, CH, AK, JN)

2.2.1 South Area (AF, AK)

Sample suite

A total of 1270 samples from 923 units in the south area were floated during the 1999 season. Samples were collected from the occupation and abandonment phases of buildings in phases VIII, IX and X (note that phasing is at this stage preliminary and may change). Samples from phase VIII included 195 from building 6 (consisting of spaces 163 and 173), much of which had been excavated in the 1960s. Buildings 17 and 2 were excavated in phase IX. A total of 333 samples were collected from building 17 lying under building 6 and 75 samples were collected from building 2, completing the sample set started in earlier seasons. Buildings 18 and 23 were sampled from Phase X, 92 and 170 samples collected respectively. Two spaces outside of buildings were sampled in the south area. A total of 25 samples were collected from space 115, filled with midden deposits, possibly from phases VII and/or VIII and associated with building 2. In space 181 (the ‘deep sounding’) 232 samples were collected from levels X, XI, XII and pre-level XII. Samples were also taken from other partially excavated and truncated buildings, but as these were not analysed in the field they will not be discussed in this report.

From the sample set, 47 were assessed to phase I level and 185 were partially sorted (‘field sorted’) to phase II level (see the archaeobotany archive report 1998 for further details of the analysis phases). Therefore, approximately 18% of the total number of samples underwent some kind of analysis and discussion in the field. These samples were chosen during consultation with the team on priority tours. Units were also chosen retrospectively if, for example, some interpretative categories/unit types (e.g. mud-bricks and mortar) were overlooked by the selection process. This resulted in the phase II analysis of a good sub-sample of units representing all the interpretative categories assigned to units during the season. It is this sub-sample that this archive report was based on. All of the processed plant remain samples were exported from Turkey for phase III analysis (full identification and quantification) to be carried out in 2000-2001.

Finds

The majority of the plant remains recovered by flotation were preserved by charring. Limited field identification confirmed that in the lower part of the mound a rich variety of plants were exploited and enjoyed by the people of Çatalhöyük (see Table 5). They include both domestic and wild plants, concurring with work from earlier seasons at the site (archaeobotany archive reports 1996, 1997, 1998; Helbaek 1964) and findings from earlier and contemporary sites in Anatolia (Helbaek 1970, van Zeist & de Roller 1995, 1994, Rosenburg et al. 1995). As with earlier investigations, the domestic assemblages were dominated by the wheats, especially emmer (Triticum dicoccum), with smaller quantities of naked barley (Hordeum vulgare var nudum) also present. Wheat glumes were present throughout the deposits and grains less so. Domestic pulses were common and included vetches (Vicia sp.), lentils (Lens sp.) and pea (Pisum sp.) with occasional chickpeas (Cicer sp.).

One of the more interesting finds this season was that of flax/linseed (Linum sp.) in a single sample from the lower levels of space 181. This taxon had not previously been recovered at Çatalhöyük and we have yet to confirm whether it was wild or domestic. If the latter its cultivation could have provided the inhabitants with both linseed oil and flax fibres, although the restricted distribution of the finds suggests that its use may have been for only a limited period. Flax was cited as a possible source for the fibres in charred textiles recovered during the 1960’s (Burnham 1965), although the use of flax had previously been discounted (Helbaek 1963), partly on the basis of a lack of flax seeds at the site. This interpretation has now to be reviewed in the light of these finds.

Among the wild plants, the fruitstones of hackberry (Celtis sp.), shells of wild almond (Amygdalus sp.) and seeds and tubers of sea club-rush (Scirpus maritimus) were most ubiquitous. They were accompanied by the seeds of many wild plants (Table 5) including several potential wild resources, such as another new discovery in 1999, fig (Ficus carica). Several seeds were discovered in an oven rakeout in space 173 (building 6) and charcoal has also been found (see Eleni Asouti’s archive report). This plant does not grow in the Konya Basin today and indicates trade, or long-distance transport of plant resources to the site as do several other more common finds (see below). A single seed of sumach (Rhus coriaria) was also found in space 181, the first example on the site, the seed finds being paralleled by a find of its wood (see Eleni Asouti’s archive report). It is uncertain where the plant came from and it may be another import possibly used for tanning or cooking, as it has been used since at least classical times. Several whole Pistacia fruitstones were taken for comparison with the BIAA seed reference collection and were identified as Pistacia atlantica/terebinthus, being most similar in size to P. terebinthus. Acorn cup and cotyledon fragments were also identified sparsely, but regularly, in the samples.

Weed seed floras in association with cereal chaff were sparse, although a wider seed flora was found in samples from space 181 where dung may have been used as fuel. Many grasses (Poaceae) and small seeded legumes (Fabaceae) were present as were chenopods (Chenopodiaceae), docks and knotweeds (Polygonaceae) and several other taxa. Among the grasses were seeds of a rye species. Several wetland plants, including those from deep water and wetland margins, were also identified for the first time to accompany the ubiquitous seeds of sea club-rush. These included alder (Alnus sp.), small reed (Phalaris sp.), pondweed (Potamogeton sp.), Aeluropus sp. and water dropwort (Oenenthe aquitlalis/fluviatalis), all present in small numbers.

Other plant remains from the site included numerous silicifed plant structures (silica skeletons), especially those of the grasses (field identifications by Arlene Miller Rosen). They were found in discrete patches in dumps and burials as well as preserving the structure of baskets. Silicified cereal awns were also occasionally found in the samples (see section 2.3 below). Seeds preserved by mineral replacement were also present in several samples from space 181, mainly the ‘stabling’ units (e.g. unit 4822). The tough, mineralised fruitstones of hackberry and seeds of Boraginaceae, especially Lithospermum sp., were ubiquitous on site, preferentially preserved because of the toughness of their seed coats. The results of the search for waterlogged organic remains are discussed in section 2.3.

Assemblages from inside buildings

Units sampled from spaces within buildings included fire installations/ovens (including structural elements and demolition debris), associated rakeouts or spreads, floors, bins and bin fills, pits (including post retrieval and clay ball pits), obsidian clusters, walls (including mudbricks, mortar and plaster), burials and building infills. Structural features, such as ovens, bins and walls, contained a low density of charred remains (see Figure 31), much lower than corresponding demolition fill layers. This is unsurprising and indicates the use of prepared and selected material for the construction of house features and not simply raw dumped material from earlier buildings.

Burials contained no specific plant offerings in the form of discrete deposits of ash or charcoal, although one concentration of hackberry stones was recorded in association with a burial beneath building 6 (and see BACH archaeobotany archive report 2.2.3.). It remains to be seen whether this concentration is distinctive from the surrounding room fill that did contain moderate concentrations of plant remains. The remains of a partially burnt wooden vessel or plank made from a trunk of hackberry wood was found lying over a burial beneath building 6 in association with a bundle of reeds (F492). Silica skeletons of plants were found in several other burials. This data suggests that plant products had some importance in the burial sphere and concurs with earlier evidence for deliberate inclusion of plants in graves (Mellaart, 1967: 208).

Burial fills, building demolition fills, pit fills, bin fills and mudbricks usually contained a range of plant remains seen throughout the site (see Table 5) such as wood charcoal fragments, cereal grain fragments, glume wheat chaff, various wild seeds (particularly sea club-rush nutlets) and nut or fruit shell fragments (particularly hackberry and almond stones). The plant remain signatures from these units varied (see below) but in most cases the plant remains from these units were probably accumulated as the result of secondary or tertiary deposition and so have little association with activities in the location of deposition. For example, the seed and fruitstone assemblages from building fills usually consisted of homogenous collections of taxa ubiquitous to the site rather than consisting of discrete assemblages derived from specific activities. They also, like several oven base and floor assemblages, contained far fewer non-wood remains than rakeouts.

Floor samples characteristically contained a very low density of plant remains and the assemblages were more characteristic of ‘background noise’ rather than the remains of specific activities. Indeed, intensive gridding of floors in buildings 6 and 17 revealed only slight variation in the patterning of plant remains. It is likely that the only useful information from the gridded areas will be the distribution of total charred plant matter as all of it is secondary. Although the archaeology does suggest that specialisation of buildings did occur the plant remains to date have not been sensitive to this (e.g. building 2 largely consisted of ovens as opposed to building 17, which contained both oven and bin areas).

Unlike previous seasons at Çatalhöyük, we found no concentrations of charred plant remains in primary storage contexts within buildings (i.e. in storage bins or bowls). The richer assemblages, and those most likely to be preserved in-situ, came from secondary contexts that were more likely to contain the results of numerous burning episodes, namely oven/fire installation bases and associated ash rakeouts. There was considerable variation in the density of macrofossil assemblages preserved in oven bases and there was no apparent characteristic botanical signature. Unsurprisingly, wood was often a major component in these contexts and the seed assemblages often revealed a variety of taxa. In some oven bases the proportion of food seeds was very large in the assemblages and it was common for the assemblages mainly to contain complete, large cereal grains and pulses. Plant remains in oven bases were no doubt food spillages that were preserved by charring and slipped down in the large cracks seen in the oven structure, sealed as the oven was remodelled. The presence of whole grains in these units may support the suggestion that the inhabitants had a non-ground diet (Çatalhöyük Newsletter, 1998).

For reasons of space and time we have summarised a large data set and focused on the general botanical signatures of unit interpretative categories excavated within the buildings. Some of these units do have characteristic botanical signatures, although the majority show large variability (see Figure 31 and Figure 32). The patterns noted in 1999 also correspond with patterns noted in earlier seasons in other excavation areas. Evidence for differences in plant use within and between buildings was slim, because most charred material was swept up and thrown away and, therefore, does not survive in-situ. There may have been some slight differences between the oven base samples from buildings 6 and 17 and those in building 2 (space 116) relating perhaps to function and the use of rooms, but that remains to be investigated. Overall, the macro-botanical investigation of the use of space in these buildings was difficult, but there is at least some potential for interpretation of individual spaces and comparison of patterns between spaces via analysis of the rakeout and oven sequences.

Assemblages from spaces 115 and 181

Spaces 115 and 181 contained units quite distinct from those inside the buildings, providing us with clues as to life and events both inside and outside domestic structures. Most units in these spaces contained large concentrations of charred remains and had formed as a result of dumping activity. The plant remain assemblage from the midden (4121) that occupied space 115 consisted largely of wood charcoal and tubers, with the occasional hackberry stone. The tubers were of sea club-rush and the flotation sample was supplemented by a large sample of remains from dry-sieving.

Deposition in space 181 was more varied and included middening, possible stabling layers, in situ burning events, episodes of lime burning, post foundations, burials, pits, gullies and scoops, clay bands and basal alluvial deposits (see Figure 33 and Figure 34). The plant remain assemblages in some of these units were quite distinctive. There were also noticable concentrations of cereal grains and hackberries (e.g. 5306) and in others mineralised seeds and the remains of dung. Interestingly, there was an increase in the total quantity and proportion of assemblages composed of fruitstones and nutshells accompanied by a noticeable increase in size of pulses and cereal grains lower in the sequence.

As with space 115 the midden/dump layers had a high density of charred remains, usually because of the high wood charcoal component. Tubers were less evident in space 181 and the dump/midden layers were more variable. The seed assemblages were also much more diverse. There was no unique signature for the middens and they appear to have been formed as a result of the dumping of remains from various activities such as food preparation and fuel burning. In space 115 unit 4121 contained large quantities of sea club-ruch tubers, recovered by both flotation and dry-sieving. While possiblly used for food, these tubers may also have been the burnt by-products of craft activities or burnt demolition debris from roofs (ethnobotanical work showed that the plant is still commonly used for roofing).

Midden 4121 from space 115 had a greater density of remains than those in space 181. This may have been due to the fact that the 115 midden was formed by the deposition of charcoal rich rubbish dumps immediately on top of older rubbish dumps whereas the midden layers in space 181 were formed by a mixture of charcoal dumps with material from other sources. Nevertheless, the middens in both spaces appear to have been household tips containing in single units the conflated traces of multiple dumping episodes. Variation in type and rate of dumping also may explain the diverse condition of the remains, some being poorly preserved and others well preserved suggesting faster burial.

Although much of the sequence in space 181 was composed of dump/midden deposits other activities also left a botanical trace. Two discrete groups of possible stable deposits were found in level XI and XII. These samples contained little charcoal and abundant seed assemblages of small-seeded herbs, including Scirpus maritimus, many grasses and small-seeded legumes, mixed with cereal chaff. In 4822, these were accompanied by a mineralised seed assemblage containing a similar range of species. The seed remains were a probable fodder assemblage and suggest the deposition of burnt dung. The mineralised flora and the chemical environment suggested by the mineralisation provide support for the presence of uncharred seeds in large numbers in the sediments. This data with faunal and especially micromorphological data contributed to the identification of the layer, and several others as stabling.

From level XI to the base of the mound, large charred assemblages of small seeds and chaff were common. Small particles of charred dung were recorded in several units. Some of the most interesting deposits were the lime-rich layers in pre-level XII. The plant assemblages consisted again of mainly small seeds and chaff and wood charcoal was sparse. The charred material was accompanied by mineralised dung pellets and it was concluded that the deposits represented lime-burning using dung as fuel. Assemblages dominated by large numbers of small seeds continued deep into the space 181 sequence and may indicate the use of dung as fuel throughout the period of deposition, although wood was sometimes abundant deeper in the sequence.

Pits, in-situ burning events and the fills of scoops from space 181 also had varying botanical signatures. The discrete burnt horizons from in-situ burning events often contained large cereal grain assemblages, of greater abundance than in any other unit excavated in 1999. Burials and post foundations contained few remains which were ubiquitous in nature and similar to the assemblages found in floors, bins, pits, clusters and the walls of buildings. The clay bands and the alluvium deposits right at the base of the deep sounding contained a greater density of charred remains that the midden/dump layers immediately above them. We are yet to tackle the taphonomy of this situation. It should also be noted that plant silica skeletons were also present throughout space 181, often in extensive continuous layers.

Discussion

The 1999 South Area archaeobotany project fulfilled its objectives by providing a large, processed archive of samples from the excavated sequence and evaluating the remains preserved within. On-site analysis demonstrated that abundant, well preserved plant remain assemblages, including charred, mineralised and silicified plant remains, were preserved throughout the buildings and from the earliest excavated phase of site formation. As with earlier seasons, units from the external spaces contained a much higher density of plant remains than those in the buildings (except for rakeouts and the occasional oven base or fill). Buildings were generally cleaned of charcoal which was dumped with much of the other household rubbish in the external spaces. The people of Çatalhöyük were very clean, at times almost obsessive about how they left buildings, leaving only very small assemblages of plant remains. Those remains were, perhaps, the larger, tougher and more ubiquitous taxa incorporated by chance after avoiding the broom. This highlights the value of any deposits from the final phases of building occupation that may have been left in-situ or sealed within phases of house re-modelling.

Specialist activity areas were identified outside the buildings with the external spaces, especially space 181, providing some of the most interesting assemblages, including stabling, burning episodes, dumps and lime-burning deposits. The burning deposits often contained high concentrations of mainly cereal grains with charcoal and may indicate cereal processing or preparation. Evidence for the burning of dung is strong in the lower deposits and wood charcoal was less abundant than in the upper middens. This suggests either a change in fuel use over time, or that the units excavated in space 181 contain the residues of suites of activities carried out at the site fringe and previously unsampled. Initial work also suggests that other changes are seen over time, including an increase in cereal grain size, increase in abundance of fruitstones and the presence of a different range of plant species in the lower site deposits. These differences may be due to different collection and subsistence strategies linked to changes in the flora, evolution of economies and/or development of exchange networks. Although no further examples of possible botanical feasting signatures were recorded during the 1999 South Area project, it confirmed the presence of plant use within spaces with elaborate symbolic art (e.g. building 6 - the former shrine 10) and suggests some significance of plant products in the burial context.

The analysis completed on the South Area during 1999 supports other work showing that while the people of Çatalhöyük extensively used crops, the gathering of wild plants was still widely practised and played a major role in the life of the site. The plant species came from a wide range of environments. Not only did the inhabitants make use of the marshy lands in the immediate vicinity of the site, but also imported fruits including acorns, pistachios, almonds, hackberries and figs from nearby steppe and more distant habitats. Evidence from other botanical studies (see other contributions to this report) also support this picture, comparable to other archaeobotanical studies in the region. The inhabitants of the site obviously had a keen understanding of the products of their lands and the lands of others’ surrounding the basin, using this knowledge when building their homes, fuelling their ovens, feeding their people and crafting their baskets, boxes and bowls. The 1999 season has demonstrated that this knowledge existed from the earliest phases of occupation excavated so far and has provided an archive that, with samples from other areas, will allow detailed investigation of the details of this knowledge throughout the period of site occupation.

Other activities and some comments on the archaeobotanical process at Çatalhöyük

Attendance at Çatalhöyük for six months allowed us to engage in several side-projects and activities previously impossible during the limited time-scale of the normal two month season. A display for visitors to the laboratory was set up to convey the importance of plant remains for interpreting life at Çatalhöyük in the Neolithic and to highlight some of the ongoing work. This display provoked questions and interpretations from visitors and improved the visibility of a subject that usually only has a visible field presence at the flotation tank. The display highlighted interpretative aspects of the archaeobotany team’s work and was improved by input from several team members during the season.

Close collaboration with the site illustrator, John Swogger, also prompted the drafting of a set of reconstruction drawings depicting possible past exploited vegetation and the uses of plants, with the emphasis on showing plants actuallly being used and gathered in specific habitats. An outline set of explanatory panels was also drafted and it is hoped that these will be completed and made available for public display in later years. It is also hoped that these will prompt wider public discussion about the site’s environment and the role of plants in Neolithic lives. On several occasions the drafts were shown to visitors and team members from Küçükköy, provoking questions and comment. Discussions with John Swogger about the content of drawings also prompted detailed questioning of the level of detail achievable through archaeobotanical research and helped to highlight gaps in interpretations.

Three botanical fieldtrips (two with John Swogger, Julie Near and Eleni Asouti) greatly informed our understanding of modern environments by providing information on plant forms and associations, and also providing specimens for the lab display of plants found on the site. The visits were to Karadağ, Pınarbaşı and Hotamis gölü and the Taurus mountains between Side and Seydişehir. A visit to the plant reference collection at the BIAA also helped to refine some of our species identifications and their associated environmental niches.

Overall the system of priority tours and constant site feedback was laborious but very rewarding. Relationships developed between the excavators and specialists in the South Area that, in our opinion, provided for much of the time the most efficient feedback and level of information attainable during an ongoing excavation. This was provided at some cost, as the equipment and labour required to provide a two-day turn-around period from sample to data is considerable. The process did, however, greatly improve our understanding of the site as specialists, who otherwise would have been isolated at the dighouse involved in the ever present backlog of work. The chance to work with specialists from associated fields (palaeoenvironment, soils, fauna, etc.) also informed our interpretations and, considering the limited level of analysis possible in the field, provided a considerable quantity of information. The experience convinced us that fuller integration of archaeobotanical specialists with the excavation team and other specialists provides not only a better sample suite, but also a more well informed analysis.

Sampling was very large-scale, a point that has provoked private criticism from some workers on other sites. Apart from accumulating an impressive archive of bone, shell, plant and small artefact data, this scale of sampling was required if the reflexive and flexible field analysis process was to work. It allowed the widest possible range of archaeological and archaeobotanical questions to be addressed at any time. Questions arose before, during and after samples were collected. In many cases deciding to sample only a limited range of contexts in advance would have precluded research into those questions. The results of this sampling effort will be evaluated in due course and while sometimes results were disappointing, the whole site analysis process often provided new angles and unexpected pieces of information that would not otherwise have been unattainable.

Acknowledgements

We’d like to thank Ian Hodder and Christine Hastorf for allowing us the rewarding opportunity of being part of the six-month team at Çatalhöyük; to Julie Near for her help and support and the chance to work in a team, rather than as another lone floater; to all the students and villagers for their help and enthusiasm and most especially to the other members of the six-month team for their professionalism and friendship.

Cited texts

Burnham, H. B. 1965. Çatal Hüyük - the textiles and twine fabrics. Anatolian Studies 15: 169-174.

Hastorf, C. & Near, J. 1998. Çatal News: The Newsletter of the Çatalhöyük Research Project 5: 17.

Helbaek, H. 1963. Textiles from Çatal Hüyük. Archaeology Spring 1963: 39-46.

Helbaek, H. 1964. First impressions of the Çatal Hüyük plant husbandry. Anatolian Studies 14: 121-3.

Helbaek, H. 1970. The plant husbandry of Hacilar: A study of cultivation and domestication. In: J. Mellaart, Excavations at Hacilar 1: 189-244.

Mellaart, J. 1967. Çatal Hüyük: A Neolithic Town in Anatolia. Thames and Hudson: London.

Rosenburg, M., Nesbitt, R. M., Redding, R. W. & Strasser, T. F., 1995. Hallan Çemi Tepesi: Some preliminary observations concerning Early Neolithic subsistence behaviours in Eastern Anatolia. Anatolica XXI: 1-11.

van Zeist, W. & de Roller, G. J., 1995. Plant remains from Aşıklı Höyük, a Pre-Pottery Neolithic site in Central Anatolia. Vegetation, History and Archaeobotany 4: 179-185.

van Zeist, W. & de Roller, G. J., 1994. The plant husbandry of Aceramic Çayönü, South East Turkey. Paleohistoria 33/34: 65-96.

van Zeist, W. & Bakker-Heere, J. 1982. Archaeobotanical studies in the Levant: 1. Neolithic sites in the Damascus Basin: Aswad, Ghoraifé, Ramad. In: Palaeohistoria: Acta et Communicationes Instituti Bio-Archaeologici Universitatis Groninganae, 24 Rotterdam/Boston: A. A. Balkema, 165-256.

Zohary, D. & Hopf, M., 1994. Domestication of plants in the Old World: The origin and spread of cultivated plants in Western Asia, Europe and the Nile Valley. 2nd edn. Oxford: Clarendon Press.

Figures

Figure 30: Pro-forma for assessment of waterlogged and mineralised remains from waterlogged ‘BOT’ samples

Figure 31: Histograms showing plant remain density for oven bases, rakeouts, fills and floors (numbers denote [sample size/unit/building]; * denotes calculation made without heavy residue data)

Figure 32: CH99 plant remain density field data: Pie charts comparing the composition of samples from oven bases, rakeouts, fills and the floors of living spaces (The heading of each chart shows [sample size in litres/unit number]; Densities of plant remains of less than 1% are not shown)

Figure 33: Histograms of plant remain densities from space 181 showing examples of some important unit interpretative categories (numbers denote [sample size/unit/building]; * denotes calculation made without heavy residue data)

Figure 34: Pie charts of plant remain densities from space 181 showing examples of some important unit interpretative categories (The heading of each chart shows [sample size in litres/unit number]; Densities of plant remains of less than 1% are not shown)

2.2.2 KOPAL Area Botanical Summary (JN)

Within the KOPAL excavation area, supervised by Neil Roberts and field directed by Pete Boyer with National Geographic sponsorship, soil samples were taken systematically from each context as in all other areas of the site. Since the excavation was centred in a region considered to be "off site" most units were either arbitrary divisions within layers of sediment or fill of pits, rather than sediments associated with buildings and structural features. For the botanical team this was significant in that a majority of the samples were close to the desired 30 liter size, and samples were easy to compare since almost all were from similar contexts. A total of 49 samples were floated from KOPAL, of which 6 were prioritized and examined in the field.

The botanical team aimed to inform field investigations regarding several specific topics. Firstly, we hoped to determine the type of preservation for botanical remains from these soil samples. Secondly, we wished to identify the range of taxa represented in samples from KOPAL in order to help reconstruct the environment and possible plant activities occurring here. Thirdly, we looked at the density of botanical remains in attempts to further the interpretations of contexts.

In regards to the first topic, Andy Fairbairn assessed specific samples from KOPAL for waterlogging as is discussed elsewhere in this report. The most common form of preservation for botanical remains was charring, but mineralized materials were recovered from some units. This pattern is similar to that seen elsewhere at Çatalhöyük, although the observation of large quantities of mineralized cereal awns from one sample was unique to the KOPAL area. This may reflect different post depositional conditions even though the general pattern (predominantly charred preservation) is the same across the site.

Information relating to the second and third topics was generated through analysis of samples chosen as priorities from this area. Although few priority tours occurred in the KOPAL area, the discussions between specialists and excavators yielded interesting information. On the first priority tour 2 samples were examined to get a general picture of the types and quantities of remains recoverable from the samples. Units 6011 and 6013 were both identified as pit fills. Although the contexts were the same, the samples appeared different. 6011, a fill above 6013, had a moderately high density of botanical remains relative to samples from across the site. It was filled primarily with wheat (Triticum sp.) grain fragments and chaff. Both free threshing (T. aestivum) and hulled wheat (T. dicoccum/T. monococcum) species were noted Also seen in this sample were remains of barley grains (Hordeum sp.), pulse remains (Pisum sp. and Vicia sp.), sea club-rush (Scirpus sp.), a fragment of hackberry (Celtis sp.) ,wood charcoal, and a bit of unidentifiable fruit. Of interest was the abundance of mineralized cereal awns in this sample. As noted above, this was a unique find and perhaps represents our first good evidence for cereal processing at Çatalhöyük. The form of preservation makes it difficult to compare the sample to others since preservation by mineralization is restricted to a few taxa on site. Therefore it is impossible to say that this is the only processing areas so far uncovered, but we can hypothesize that our lack of on site processing evidence is the result of the fact that processing was taking place off site, in areas surrounding the mound like KOPAL .

In contrast to unit 6011, 6013 which was also a pit fill, was less dense. While some of the same taxa were recovered (cereal grains and awns, some mineralized, pulses, wood, and Scirpus sp.), the quantities were lower and additional material, namely parenchyma tissue and mineralized tubes from carophytes (green algae), were also seen. If we interpret the unit 6011 as crop processing residue, it may be possible that this unit too is the result of processing activities though further away from the center of the action.

The second priority tour to the KOPAL area aimed to compare 4 samples from an arbitrary layer of sediments in order to help understand what may have been going on off site while these sediments accumulated. The first unit (6025) was from an upper spit and the remaining 3 (6029, 6030, and 6031) were smaller units all from the underlying lower spit. The results from the botanical analysis showed that all of these samples were significantly less dense than the pit fills previously discussed. Ranging from .001 to .006 grams of charred botanical material per litre (compared to, for instance, the pit fill 6013 with .017 grams per litre) these samples yielded very little information regarding plant related activities. The remains found in this sample group included cereal fragments of both chaff and grain, Scirpus sp. nutlets, a small seeded legume, a Bromus type grass seed, a few fragments of parenchyma, and small wood pieces. The remains were averagely to poorly preserved, and were broken up by mechanical damage either pre or post deposition. More variable between the upper and lower spits were the quantities of ostracods (small bi-valve mollucs), amphibian bones, and charophytes. The unit from the upper spit contained higher quantities of all of these remains, while the lower ones had them, but in much smaller amounts. Though it is difficult to accurately interpret these finds without specialist analysis, the presence of both ostrocods and charophytes does indicate that the area was at least semi permanently inundated during this time period. The difference in quantities between the spits might indicate a difference in water level, duration of inundation, or conditions of the water present. Alternatively, we might hypothesize that while 6025 was clearly a wet area, the very small quantities of ostrocods and charophytes in the lower spit could have been carried into the sediments from elsewhere, and therefore not be significant indicators of a wet environment at this period.

Generally, the KOPAL botanical materials can be described as similar to those found on site in terms of the taxa present (see taxa list). Many fewer taxa were identified here, but this is at least in part due to the number of samples examined. It is quite likely that many of the species found on site would eventually turn up in the KOPAL area if enough samples were analyzed. For now though we can only conclusively say that none of the off site taxa are unique to this area. Densities are lower overall compared to on site samples. None of the samples comes close to reaching the densities seen in midden or dumps on site (middens are often between .500 g/l and 2.000 g/l), but middens are generally more dense than pit fills so this contrast may be explained by context related variation. Densities are variable within KOPAL units which points to spatial segregation of activities, and/or changes in what sorts of plant activities occurred here at different times. Taphonomic variations might also explain these differences, so further research will be aimed at accounting for as many taphonomic variables as possible between samples. Most of the remains from the KOPAL trench were in fairly poor condition Almost no materials were recovered that were larger than 4 mm indicating that fragmentation here is high. This might be interpreted as showing that these were not charred remains from primary or even secondary deposition events where often larger remains are still recoverable, particularly wood. If they were, the people here were burning without wood as fuel, or they were dumping different types of remains in different places. Finally, crop processing residues, up until this point undiscovered at Çatalhöyük, may now be in evidence off the mound itself. More research on non priority samples will help to clarify our understanding of the plant remains from the KOPAL area as well as the relationship between on and off site contexts.

2.2.3 Berkeley Archaeology at Çatalhöyük (BACH) Archaeobotanical Report for 1999 (CH)

This is the third field season that this northern zone of the site has been under investigation by the UC Berkeley team directed by Drs. Ruth Tringham and Mira Stevanovic and funded by the National Science Foundation. This year they have been continuing their focus on Building 3, an elaborate and large building with a series of platforms, burials as well as what is called a screen that was built during the occupation along the western wall, to hold back a collapsing wall. This 1999 field season was a slightly longer excavation season spanning about six weeks with 259 soil flotation samples collected and processed. Of those, 81 were identified as priority units with high interpretive value. These proveniences are primarily the floors of the building, debris on the floors and fill above the floors, especially including the platform floors, as well as a series of pits, fire installations, in situ burnings, niches, and one child burial. To date, 34 samples have been studied in the field, using our field sorting procedure. All members of the archaeobotanical field team helped with the processing, but the bulk of the work was completed by Katy Kilacky, Harpreet Malhi, Julie Near and Christine Hastorf.  The excavation covered three superimposed floor surfaces within Building 3 (see the BACH section for plans). At this point in the analysis, we can only report on several specific areas within these quite detailed floor and plaster excavations.

The first concerted effort concerned what was called the uppermost floor of the 1999 field season excavations. This group of flotation samples includes central floor samples with surrounding floor samples, and some platform surface samples. The central floor unit, 3573 was the most dense botanically, with a high diversity of taxa, much wood, as well as the range of plant taxa found at the site, including pulses, parenchyma, cereals, chaff, but also nut and hackberry. The ashy layer sitting on top of this surface, unit 3582, contained mainly wood, with sparse pulse and nutshell remains, but some burnt microfaunal bone.  The samples surrounding this locale are homogeneous, having lower seed densities, with more of a wood dominance. Analyzed floor units have plant densities and contents that suggest crop processing, with lots of cereal grains, wild seeds and some chaff. There is a niche off to the west of this floor, along the screen area in space 158. Unit 3533 was extremely dense and diverse, with a high count of burnt, edible taxa. It was considered a good candidate for feasting due to this diversity, including nutshell and parenchyma (most likely edible tuber) evidence. I think however that it is more likely to include rubbish fallen in from another area, hence not specifically a location to hold burnt remains of a feast, but rather an out of the way spot to harbour such material from sweeping or trampling. A more likely feasting or special offering deposit unit from this level is 3545, in space 89 associated. This unit is nearby and associated with a bucranium fragment. Despite its low diversity, unit 3545 contains many cereal grains but low counts of seeds, as well as it being burned in situ.. These analyzed surface samples can be grouped into two categories, the northern, clean floor group (units 3579, 3561, and 3570) and the southern, dirty floor area (units 3567, 3568, and 3577). Although the excavators felt they could clearly see differences in these surfaces, light, whitish plaster in the north and darker, more organically stained in the south, the flotation samples from both groups showed no differences however. This suggests that material used across the surface was scattered across the whole floor, which had been prepared and perhaps used differently, but which received the same blanket of secondary or tertiary refuse.

The next lower floor also had analyzed samples from the central floor and surrounding platforms of space 86. The platform samples overall had a low density of plant matter, as the data suggested that they were quite trampled. They contained moderate densities, mostly wood, lots of cereal chaff and grains, some parenchyma and some pulse fragments. Intriguingly, they contained a high phytolith content, and no dung, suggesting they were composed of post consumption discard. These floor compositions are likely to be made up of food preparation and consumption remains ground in and kicked around. Included amongst this surface are two caches of higher density plants. One, unit 6191, is near the child burial, 6211, which has a plethora of hackberry remains. The other, unit 6187, is a little pit, perhaps even a fire installation , i.e. in situ burning locale, with denser, edible material. It is across this floor that three fire installations began to be evident. These features each had distinct constructions and contents. Most intriguing of these is the central feature, Feature 606. This "brazier", seems to have been a small oven, made of a series of bent reeds with a likely mud covering. This construction is suggested because the firing area is surrounded by a circle of small holes, the size of reeds. Within is much ash and wood charcoal. The other diagnostic pattern to the north is a series of small burning pits, one being stone lined. These have low densities overall, but contain a high density of Scirpus tubers, some whole cereal grains, some chaff and few wild grasses. The phytoliths from this pit displays dense cereals, chaff, whole tubers, and sedges (unit 6191). This unit received a detailed reporting on wood identification by Eleni Asouti as well. It included oak, Pistacia, Phragmites, Salix, and Chenopodium twigs. With no evidence for dung, it is suggested that only wood and chaff was the fuel for this food preparation. These combined results suggest that at least two events are recorded in this fire pit, a low cereal firing and a tuber roasting, perhaps using chaff for fuel.

The third floor sampled in 1999 contained a series of fire installations as well as floor samples, continuations of some of the above firing areas. Most notable in these small pits is the low density of material overall, which are highlighted by wood with some cereals. The features 604, 605, 624, and 621 have low densities of seeds, and are dominated by wood. However, the feature 619 pit has more tubers, nutshell, wheat husks, and ceramics, as well as wood. It should be noted that is it often the case that in locations of direct fire that most plant remains have turned to ash and thus the macrobotanical evidence is not as prevalent as is the microscopic signatures. The impression these contemporaneous fire installations give is that they were each used for something different, probably not all for food preparation, some perhaps for smoking, or cleansing.

These firing feature compositions in Building 3 suggest that dung was not used as fuel, but wood and perhaps cereal chaff was the major cooking and roasting fuel source. Also it does seem to have been the case that material generated from activities completed on the floors were swept into the fires periodically, since all of these installations contained bits of obsidian, and many small burnt bone fragments. All samples analyzed, with the exception of the burial, are most likely tertiary deposits. Fortunately, excavations in this building should continue in the next summer field season. This will help us further place these floors and fire installations into a better context.

2.3. Waterlogged remains and the affects of de-watering (AF)

A major aim of the 1999 excavation season was the establishment of the current groundwater level and determination of the presence and extent of organic preservation in and around the east mound. Plant remains are a major indicator of organic preservation and a programme of monitoring and sampling was implemented to elucidate the extent of plant remain preservation and help fulfil these aims.

The presence of waterlogged organic remains was monitored by the site excavation team, who were familiar with the signs of organic preservation, and by specialist liaison during priority tours. Additional samples (BOT samples) were collected from Space 181 of the South Area and from the KOPAL trench, and were subject to scanning (see methods section). A total of 40 samples were scanned, including 22 from Space 181, 15 from the KOPAL trench and 3 from a KOPAL core collected from between the east and west mounds. These represented a representative sub-set of the units excavated in both areas and included units from a range of absolute heights.

Waterlogged plant macro-remains were not found in any of the samples from the South Area or KOPAL trench. Samples from the KOPAL core contained waterlogged rootlets that are most likely to be later contaminants than remains that were contemporary with the period of deposition.

Several reasons for the lack of organic preservation low in the mound can be suggested, including recent local dewatering. This can be discounted as sediments in the lower levels and base of the mound were wet during the 1999 excavation at the height of summer, suggesting that recent hydrological changes have yet to affect the base of the mound. A more plausible explanation is that, at least in the excavated area, there was not a permanent high water-table during the initial stages of höyük formation. Such a scenario would suggest that there never was any preservation of waterlogged organics over much of the tell.

It should be noted that this interpretation DOES NOT preclude the preservation of waterlogged plant remains in other low-areas of the mound. It is also possible that any features cut deeply into the sub-höyük soils would act as a preservational basin for organic remains if the feature lay below the Neolithic water-levels. There is also some evidence for localised anaerobic conditions in the höyük. Below building 23 a large, partially mineralised wooden post was found identified as oak (Quercus sp.) by Eleni Asouti. This, with the finds of mineralised seeds in stabling layers of Space 181, suggests that at some point in the past decay was sufficiently delayed for replacement of at least some plant structures by mineral salts. It is possible that this preservation was caused by localised high ('perched') water levels. It is possible that such perched water tables still remain in currently wet areas of the höyük and would be damaged by dewatering.

Although waterlogged plant macro-remains were absent in the excavated deposits, rich assemblages of charred, mineralised and silicified plant remains were present. In the lower water-affected layers of the höyük, preservation of these remains was excellent, especially the charred remains. Preservation of charred remains in the KOPAL core were also found to be good. The good level of preservation may indicate the maintenance of stable soil water conditions since deposition as charred remains are delicate and prone to physical damage caused by repeated wetting and drying and its side effects, including the shrinkage and expansion of soils and growth of salt crystals in voids (the site soils contain a considerable salt content). It is uncertain how mineralised and silicified remains would react to such pressures.

The plant remain record at Çatalhöyük is a valuable one. Considering the potential damage to the charred and other plant remains, the possibility of localised waterlogging elsewhere and the value of the site as a whole, it is recommended that stabilisation and maintenance of the current water levels are necessary if the plant remain record at the site is not to deteriorate.

3. Charcoal analysis report (EA)

This report summarises the results of the 1999 field season with respect to wood charcoal analysis and the investigation of present-day vegetation in the region. Fieldwork was undertaken between 15^th of July and 1^st of September and was supported by a travel grant from the Institute of Archaeology and the Graduate School of University College London.

3.1. Brief summary of the archaeobotanical results

In total, 30 flotation samples deriving from the deep sounding area were examined for charcoal analysis. Table 6 summarises the taxonomic information obtained.

Contexts sampled included rubbish and stabling deposits, plus what the excavators have identified as the remains of distinct burning episodes. Overall, more diverse appear to be the burning deposits, especially in what concerns small-sized taxa (e.g., woody chenopods and legumes, wormwoods and mints). One possible explanation for this may be that these contexts represent the remains of localised fires, lit for specific purposes, including lime production. The available evidence suggests that, for lime burning in particular, firewood was most probably used indiscriminately, without consideration of the burning qualities of each particular taxon. Further analysis will either refute or substantiate these preliminary observations.

After this season, it is possible to give a comprehensive list of the taxa found so far in the Çatalhöyük charcoal assemblages (Table 7).

A re-examination of the fragments previously identified as fir (Abies) (see Charcoal analysis report 1998) with the aid of scanning electron microscope showed that they should be recorded as juniper. The two taxa are very similar in what concerns their anatomical characteristics.

At present, research has focused on comparing this season’s results with the charcoal data made available from previous seasons in order to reconstruct, at this stage, the potential woodland types and tree habitats during the Neolithic. To this end, ecological knowledge of present-day vegetation in this area and comparable environments in other regions will be used, together with palaeoenvironmental evidence (i.e., pollen and geomorphological data). Preliminary results are forthcoming.

3.2. Vegetation survey

Investigation of present-day vegetation focused on two distinct areas: the north facing slopes of the Taurus range and various localities in Karada— . Survey of Taurus included selected localities covering a transect from the hillslopes on the margins of the Konya plain up to the Central Taurus massif. The following woodland types were described: Open oak coppice woodland, woodland pasture, riverine forest and marshlands, oak-pine woodland, coniferous-oak montane forest, fir-juniper-cedar montane forest, Mediterranean montane forest. Ecological descriptions of each locality were made and samples of wood were collected as reference material.

In the Karadag area, I visited the village of Madensehir, the area of Melendic Dagi and the villages of Vassitepe and Ucküyü. During this trip, observations were made involving the architecture of the houses and the use of construction timber, the location of the fields and their relationship to woodlands and the ecology of woodlands. The latter included recording of evidence for past woodland management both directly (i.e., form of trees and tree stands as a result of coppicing/pollarding) and indirectly (e.g., use of trees as fences, field boundaries, structure and floristic composition of woodlands used as pastures, effects of modern conservation on former pastures, abandoned common lands, etc.). Details of these observations were kept according to location and type of observation and a photographic record accompanied them.

4. Phytolith Archive Report for 1999 Season at Çatalhöyük(AMR)

Phytoliths are microscopic silt-sized particles usually composed of opaline silica. They form in the cells of plants, primarily grasses and other monocotyledons and are often distinctive of plant family, genus and more rarely species. Phytoliths occur in most types of woody trees and herbaceous dicotyledons as well, yet in these types of plants they are more difficult to identify. The marshy environment around Çatalhöyük in the Neolithic period was ideal for the accelerated formation of phytoliths in the plants which were exploited by the inhabitants of the town. We therefore find remains of phytoliths in all of the archaeological sediment contexts at the site, as well as thick white silica skeletons of former in situ baskets and matting which were buried in grave pits, middens and living surfaces.

Given this good phytolith preservation and abundance of remains, the phytolith analysis at Çatalhöyük has taken two main directions. One is the standard analysis of individual silica bodies from almost all archaeological sediment contexts, and the other is the identification of plant genera used in the basketry and matting found in variable contexts. The goals of phytolith analyses from sediment units at Çatalhöyük include an understanding of plant usage, activity areas, and building functions from different living contexts. These include room floors, platform bases, bin contents, pit fills, hearths, middens, and possible animal pens. Animal pens also furnish valuable information on the foddering or grazing activities of herd animals from the analyses of phytoliths contained in decomposed dung. This points to additional information on land use in the site vicinity. The study of the extensive remains of basketry and matting materials at Çatalhöyük provides a unique opportunity to understand both the technology and cultural context of this important, yet fragile aspect of the Neolithic toolkit.

4.1. Phytoliths from Sediment Contexts

Sediment samples for phytolith analysis are collected by both the phytolith analyst and excavators at the time of excavation. Due to time constraints only minimal processing is conducted on site. The field processing gives a preliminary overview of the phytolith content of any unit which allows comparison with on-site analyses by other specialists and provides immediate feedback and first-hand information to excavators at the time of unit excavation. Additional sediment samples are saved for more thorough phytolith processing in a conventional phytolith laboratory after the end of the excavation season.

In the laboratory the sediment is processed by a series of cleaning procedures designed to remove pedogenic carbonates, very coarse sand and clay particles, extraneous organic matter, quartz and heavy minerals. The remaining fraction primarily consists of plant silica bodies. These are mounted and the forms are counted at 400 x magnification. The process is designed to acquire an absolute count of phytoliths per gram sediment rather than relative percentages. In this way the results are standardized and samples can be compared across a variety of archaeological contexts. Separate counts are performed for Single-celled phytoliths consisting of solitary silicified cells versus the Multi-celled phytoliths also known as silica skeletons. The latter type allows identification down to the genus due to the distinct characteristics of the suite of adjacent silicified cells (Rosen 1992). Also, high counts of large multi-celled phytoliths from cereals and other grasses are one indicator of a farming environments in moist alluvial soils. The size and quantity of multi-celled forms can indicate if fields were located in the marshy areas around the site, or if farming was conducted on higher ground in better-drained soils (Rosen and Weiner 1994). Many multi-celled forms also come from the sedges and reed grasses which grow naturally in moist marshy conditions.

Full analyses have been completed on 19 samples collected in the 1998 Season (see Table 8). Some general results of the analyses of the 1998 samples are as follows. Four different archaeological contexts are represented by these samples. These include the midden deposits in the South Area, Space 115, the living floor surfaces of North Area Building 5, Spaces 156 and 154, the floor of the "bin room", Space 157, and the base of Bin 1 (F.232, Unit 3802). Predictably, the midden deposits were much richer in phytoliths than those of room floors, and far more diverse in composition. Figure 35 illustrates the weight percentages per gram of sediment for each of the samples analyzed. Not shown here are the samples from the Bach Area roof context (Space 86, Unit 2271, S.5 and S6 ) which exceed even the midden deposits in weight percent (up to 15% phytoliths), and are dominated by phytoliths from the common reed, Phragmites sp. These reeds are presumably the remains of fuel from the oven rake-out on this roof described by Matthews in the 1998 Archive report.

In addition to the large quantities of phytoliths, these midden samples were rich in multi-celled forms (with the exception of CH-98-25) (Figure 36) suggesting the exploitation of marsh resources and wet-land cultivation. There were also large quantities of phytoliths from the seed husks and floral parts of grasses (Figure 37). Husks from wild grasses (also present in these midden deposits in large quantities) are one possible indicator of seasonality. If these wild grass husks came into the midden as ash from dung fuels then they are indicative of a spring accumulation in this midden deposit. This, however, is not the case for these deposits since the wild grass husks are positively correlated with both wheat and barley husks (Corr. Coefficient = 0.93 for wheat/ wild grass and 0.97 for barley/ wild grass). It is most likely, then that the wild grasses were field weeds which were harvested with the cereals and ended up in the midden as waste products from food processing. This could have occurred in any season since cereals are stored year round.

Cereals were far more numerous in these midden samples than in most other locations analyzed (Figure 38). This was especially true of Sample 98-28, a gray ashy lens within the midden. These results emphasize the fact that the house floors at Çatalhöyük were for the most part very clean with respect to the remains of food processing residues. These clean house floors might be a result of the fact that room surfaces were resurfaced and white-washed many times throughout the life of the structure, thus precluding the accumulation of plant residues on the living floors. Cereal straw was also abundant in one of the samples of the midden (98-26). This is a surprisingly rare element in the samples from other localities at Çatalhöyük. It suggests that cereal grains arrived to the village in a relatively clean state, and the chaff was generally not used for domestic purposes.

Another distinctive characteristic about these samples was the presence of vast numbers of "polyhedral" phytoliths from the leaves of an as yet unknown tree or woody herbaceous plant. Such plants generally produce a very low yield of phytoliths so their presence here in large numbers suggests massive quantities of this particular plant. These plants are rare or nonexistent in other contexts at the site. More basic research is needed, however, before we can identify the plant source of these "polyhedral" phytolith forms.

Finally, although sedges (Cyperaceae) are abundant in other contexts at the site, they only occur in low numbers in the midden samples. By contrast, there are large numbers of phytoliths from the common reed Phragmites. These are sometimes represented by the leaves of the plant and sometimes by the stems. At Çatalhöyük sedges were used for basketry and matting as well as for their edible tubers. Reeds were used as roofing material, sometimes as fuel and for making objects of cane.

The phytolith samples from Building 5 in the North Area were not as rich as those from the South Area midden of Space 115, yet they still displayed significant differences which provide us with preliminary clues to the use of space in the Neolithic houses. The room represented by Space 154 had varying distributions of phytoliths of wheat and barley across the surface with highest concentrations in the location of Sample 98-48 and 98-50. In Space 157, the bin room, there were varying concentrations of cereal phytoliths on the floor next to specific bins. None of these concentrations were overwhelmingly high again emphasizing the cleanliness of the floors at the time of abandonment, but their relative differences may be significant. There was a peak of barley husk phytoliths in Sample 98-59 (number phytoliths per gram sediment = 604), but low numbers of wheat near the lip of Bin 4, suggesting that this bin contained a relatively pure quantity of barley grains. Other samples such as 98-53 contained a mixture of cereal husk phytoliths from both wheat (n/gm=1098) and barley (n/gm=1244) suggesting mixed or alternate storage of both cereals.

Probably the most significant results from these preliminary Building 5 phytolith samples are those from Sample 98-35 (Space 156). This sample contained the highest peak of wheat husk phytoliths from any non-midden samples examined to date (n/gm=5736). Although there were no barley husk phytoliths, there were a significant number of weed-grass forms. The sample also contained large numbers of "bilobe" phytolith shapes. Bilobes are relatively rare in sediment samples at the site, but are commonly found in the silica impressions of certain types of coil baskets at Çatalhöyük . These results suggest that this sample represents the presence of a former basket filled with wheat grains. The sample occurred immediately under the fallen wall (F.227). This wall fell onto the floor in one piece and seems to have buried a basket full of grain. This implies that the room had not been abandoned before the wall fell and was still in use at the time of the structural collapse.

4.2. Phytoliths from Silica Impressions of Baskets and Matting

One of the unique aspects of Çatalhöyük is the good preservation of basket and matting impressions in a number of different archaeological contexts. This provides a unique opportunity to understand the use of specific plants for distinct purposes. During both the 1998 and 1999 seasons the remains of baskets and matting were subsampled and analyzed for their phytolith content. The samples collected in the 1999 season are listed in Table 9.

Many of the wild plants used for baskets and matting can not yet be identified to genus because of the lack of reference material for this region. However, they can be identified at least to plant family or sub-family which in itself has provided us with useful insights. Even at this preliminary stage there are a few trends in plant use which suggest cultural and environmental factors in the selection of plant shoots for basketry, matting, and floor construction. Matting seems to be constructed almost exclusively of sedges derived locally from the surrounding marshes. These have spongy leaves and stems and would have provided a comfortable floor covering or bedding material. All of the adult burial mats examined from the 1998 season were composed of sedges. The sedges were abundant, and an easily obtained source of plant fiber. They were used as temper in floor packing, and may also have been used as animal fodder.

Baskets appear to be constructed of several different plant types. Occasionally sedges were used for baskets, but for the most part basketry materials seem to have been more carefully selected for specific purposes. One of the materials is an as yet unidentified dry-land grass in the Panicoid sub-family. These grasses generally grow in dry and warm environments, and were probably collected away from the marshy areas in better drained soils. The baskets constructed from this material were presumably used for storage of food and other items. There was only one basket (or coiled mat) made of cereal straw, a material which is surprisingly rare in the phytolith assemblages at Çatalhöyük.

Finally, there was one unique type of grass phytolith which may have only been used in the production of neonatal burial baskets. This type of multi-celled phytolith is similar in appearance to the wild grass Agropyron sp. Although not all neonate baskets are made of this grass, when it does occur it is only in these infant burial baskets. Another significant aspect of these phytoliths is the fact that they are mixed with phytoliths from the floral portions of the plant. Other baskets are composed of the stem and leaf portions. It is possible that the admixture of grass inflorescences was intentional and related to the ideology surrounding the burial of infants, although more baskets will need to be analyzed in order to establish this as a true pattern. However, it is possible to say that due to the presence of the floral components these baskets were most likely made in the spring, which might be indicative of the season of death for these infants.

4.3. Field Examination of Phytoliths from Archaeological Sediments

Preliminary field analyses were done on samples from priority units in the field. Most of these were examined jointly with Dr. Wendy Matthews who identified the spherulites mentioned below and contributed to the preliminary interpretation of these deposits. Some of the results are the following:

4.3.1. South Area

"Lime burning deposits," Units 4880, 5274, 5276 are characterized by phytoliths from wild grass stems and husks (possible Bromus-type and Hordeum sp.), with moderate numbers of multi-celled phytoliths from unidentified grass stems; The samples include some phytoliths from the leaves of woody plants; Many phytoliths display pitting from alkaline chemical dissolution. The significant numbers of spherulites in these units suggest that the phytoliths were probably derived from dung fuels. The phytoliths indicates that the dung was collected from animals which were grazing off-site on wild grasses and woody herbaceous plants.

The "gully fill" (Unit 4884) and "scorched layers" (5277, 4881) under the lime units were typified by abundant spherulites also indicating dung deposits. The phytoliths here were characterized by phytoliths from wild grass husks and stems, again indicating the type of plants consumed by the animals which produced the dung fuels. These phytoliths were better preserved than those in the limey deposits.

In Building 17, the floor make-up, Unit 5020, consisted of thin sets of gray lenses alternating with thin white plaster lenses. The white calcareous material was poor in phytoliths, but the gray sub-units contained many disarticulated long cells, and thin rods typically found in marsh sedges (Cyperaceae). It is possible that the ash from burned sedges was added to the floor make-up as both temper and a calcareous bonding agent, with a finish of a white plaster wash. Sedge phytoliths also dominated the oven rake-out from Unit 5038, Building 17. Some common reed stem phytoliths were also present in rake-out sediments from Unit 5021.

Sedges were abundant in the marshes around Çatalhöyük and provided an easily acquired source of plant fiber. These plants seem to have been used in place of cereal straw for building material temper, tinder and/or fuel. Cereal straw is the more common plant fiber used for construction fiber in other prehistoric and historic period agricultural villages. The cereal straw also does not turn up in the dung fuels at Çatalhöyük suggesting that it was not stored and fed as fodder to animals. The only places where cereal straw phytoliths have occurred in significant numbers from the samples examined to date, are in the construction material of a coiled mat or basket (South Area, Building 18, Unit 4686.X1), and in the contents of an off-site pit containing what appeared to be the sweepings from a threshing floor (KOPAL Area, Unit 6011). It is possible that there were ideological reasons for not using cereal straw for mundane purposes, or perhaps there were some as yet undetermined functional reasons why other plant materials were used for these purposes.

4.3.2. BACH Area

A number of samples were examined in the field from the BACH Area. These include small pits within surfaces from Building 3, such as Unit 6191, which contained a number of phytoliths from the husks of wheat and stems of sedges. The sedge stems may have been used for lining the base of the pit. Since macrobotanical remains from this pit were dominated by tubers and not cereals, the cereal phytoliths found here might indicate a previous use of the pit, rather than the last use. Other pit fills examined include Unit 6161 (F. 624) containing phytoliths from unidentified grass stems, plant trichomes, and silica skeletons similar to those from Juncus sp., perhaps remains of the pit lining. Feature 621 Unit 6214 contained a wheat husk and single-celled phytoliths from grass husks and stems. Unit 3591 (F. 604) was dominated by wild grass husks including many wild C₃ grasses from cooler upland areas, as well as phytoliths from dicotyledon (woody plant) leaves, and Cyperaceae sedge stems. Finally, the F. 619 pit fill (Unit 6212) contained numerous phytoliths from sedge stems and wheat husk phytoliths.

Several samples came from hearths in this building including Units 6202, S.8 (upper level of central hearth), 6202, S.9 (lower level of central hearth), 6208 (small hearth), and 6187 (small stone-lined hearth). For the most part the hearth samples were dominated by stems and leaf phytoliths from the common reed (Phragmites sp.). Since reed stems are not a common animal fodder, and there were only minimal amounts of spherulites, it appears as if these reeds along with some sedges were the primary fuel for these household fire installations. Phragmites also dominated the samples taken from small post holes circumscribing fire installation F.613 (Unit 6160, S.11 and S14). This suggests that a cane frame surrounded this hearth.

Figures

Figure 35: Weight percent of phytoliths per gram sediment from Çatalhöyük 1998 Samples.

Figure 36: Ratio of Multi-celled to Single-celled phytolith forms.

Figure 37: Numbers of grass husk phytoliths per gram sediment from selected samples.

Figure 38: Number of wheat and barley phytoliths per gram sediment for selected samples.

4.4 References Cited

Rosen, A. M. (1992). Preliminary identification of silica skeletons from Near Eastern archaeological sites: An anatomical approach. Phytolith Systematics: Emerging Issues. G. J. Rapp and S. C. Mulholland. New York, Plenum: 129-147.

Rosen, A. M. and S. Weiner (1994). "Identifying ancient irrigation: A new method using opaline phytoliths from emmer wheat." Journal of Archaeological Science 21: 132-135.

5. Experimental Work at Çatalhöyük 1999 (AE)

The overall purpose of plant processing experiments at Çatalhöyük is to better understand the uses of a wild plant species, Scirpus maritimus (bulrush) in the diets of pre-agrarian and early agrarian peoples of Anatolia. The experiments involved first, the collecting and processing of the plant species to produce modern samples for nutritional analyses. This part has been important to produce comparative materials that can be used as in the identification of archaeological specimens. These comparative materials include charred specimens and SEM recording (by Michele Wollstonecroft) of unprocessed, processed charred and uncharred material.

Scirpus maritimus’ recovery from numerous sites throughout the Near East, such as Epipalaeolithic sites of Hallan Çemi, The Neolithic and Chalcolithic contexts of Çatalhöyük, Turkey; Abu Hureyra, Syria, and Wadi Kubayani, Egypt has shown its important role in daily life during prehistoric times.

During the 1998-99 field season, I carried out my MSc. thesis research that was based on ‘The Ethnobotany of Wild Plant Foods In The Konya Basin.’ In this research I found that Scirpus maritimus, which is one species included in my taxa list, was not reported to be eaten in the region today. Also, due to heavy drying of the wetland areas in the Konya Basin in the last ten years, Scirpus sp. as well as many other plant and animal species in the region has decreased in number dramatically. This research also showed that not only do people continue to use Scirpus sp. for weaving mats and other items, but additionally those living in wetland areas around Hotamış Lake (approximately 20km. to Çatalhöyük) were not able to recognize Scirpus tubers as coming from this plant. This unfamiliarity with the many parts and uses of Scirpus sp. is partly due to drying. However, Scirpus tubers are known and recognized among people living in the non-wetland (steppe) village of Küçükköy, the closest village to Çatalhöyük.

In the Hotamış Lake region of the Konya Basin, local people still harvest Phragmities and Scirpus to use them as mainly building materials. During harvesting, they cut the plant leaving the roots on the ground. On the other hand, in Küçükköy, seasonally Scirpus is harvested especially for its tubers, because the tubers are used in the mudbricks on the garden walls, as it is a good binder.

Local people harvest the plant for its fresh tubers because it is easy to pull out, and it is also a good binder in mudbricks. However, it only has these properties while fresh. In fact, during 1999 experimental work, it was found that the plant was very difficult to harvest when not in season.

Apparently, the villagers who are not familiar with these tubers do not use them as building materials except for its stalks. It can be observed that houses in this region are made from stone and wood foundations with Phragmities sp., Juncus sp. and Typha sp. the roof material are used. Local people explain the use of these plants as building materials, since straw was difficult to find in old days. Today, because of drying lands, straw has gradually taken their place in building materials. In these wetland villages, Scirpus is used only on the floors, and in the foundations.

On the other hand, people in steppe villages, such as in Küçükköy, who are familiar with Scirpus tubers, never use its stalk as building material. Instead, they fill in their cushions with the stalks. During my research, I found that Scirpus has been one of the most important exchange materials until recently. People in wetland areas used to exchange wood for bulrush with people in forest areas.

This may explain why it is also possible to see Scirpus used as building materials or mats in forest areas where the species does not grow. There is no evidence yet that Scirpus is used as fuel in contemporary settlements in the Konya Basin. However, in some steppe villages, such as Taşağıl, which were previously wetlands, Juncus, a plant similar to Scirpus, is used as tinder, since the dry plants cover a large area in steppe today. Local people in the areas where Scirpus grow explain why Scirpus cannot be used as fuel as it burns quickly and does not give a long lasting heat energy, this is why it is used only as a tinder.

In villages like Küçükköy where Scirpus is used seasonally, dry tubers inside of mudbricks fall out of the walls due to deterioration. People then burn them along with the fallen soil to get rid of the dirt created on their garden floors. This, however, not be considered of use of the plant for fuel, but is a simply a form of cleaning or waste disposal.

In the Neolithic context of Çatalhöyük, Scirpus sp. might have been used in the human diet as well as for weaving baskets, mats and as building materials. Even today, especially along the irrigation ditches near Çatalhöyük Scirpus sp. grows as ocassionally dense groups (some 10 m. in length) between Phragmities groups.

In the 1999 Field Season, I was joined by Michele Wollstonecroft who conducted in depth research on Scirpus maritimus. Our joint experiments focused on harvesting and processing Scirpus sp. in a variety of ways.

The future aim of my research is to continue to record knowledge and information gathered from local villagers and to use this to better understand the archaeological record at Çatalhöyük and other excavations. For example, in the preliminary discussions during the 1999 Çatalhöyük Excavation Season, the idea of Scirpus’s being used as fuel or food in the Neolithic, since Scirpus tubers are found in several ovens, and also it may have been used in mudbrick construction were suggested. From ethnobotanical information which here gathered up to this point, these archaeological findings may be explained as tubers used in mudbrick and from similar to what is seen in Küçükköy today. One possible explanation of these charred tubers is their use in human diet (as roasting tubers in ovens) and other one might be to get rid of waste disposal of the construction material or of some food which is not edible (overroasted, rotten, etc.)

Also, it is needed to know the distribution on the species level in the cultural contexts, to better understand the use of plants at Çatalhöyük excavation site.

© Çatalhöyük Research Project and individual authors, 1999