ÇATALHÖYÜK 2003 ARCHIVE REPORT

MACRO BOTANICAL REMAINS

Meltem Agcabay, Amy Bogaard, Mike Charles, Glynis Jones & Nicola Stone

Abstract

This was the first year under the direction of the new team leaders, Mike Charles and Glynis Jones (Sheffield University) and Amy Bogaard (University of Nottingham). The team leaders present were Amy Bogaard and Mike Charles. The flotation officers were Meltem Agcabay and Nicola Stone, supported by Riza Buyuktemiz and Mevlut Sivas.

With thousands of archaeobotanical samples processed since 1995, the inherited system is tried and tested and is impressive in its scale and resources (expert local workforce, lab space etc.). Taking into account the recommendations of the previous archaeobotanical team and other team leaders, certain aspects of the system were modified, especially the laboratory analysis phase for priority and non-priority samples (see below).

In addition to these changes, it was decided that the flotation team would no longer supervise heavy residue processing. Again, this decision was informed by the views of the previous archaeobotanical team as well as other team leaders. In 2003 Meltem Agcabay supervised heavy residue sorting but in future years a separate team will organise this process.

Since excavation this season consisted mostly of finishing trenches (BACH) or beginning new ones (4040), the number of priority samples was much lower than in previous years. This report is mostly concerned, therefore, with clarifying methodological changes to be followed through over the current cycle of excavation.

Özet

Bu sezon, çalismalarin yeni ekip baskanlari Mike Charles ve Glynis Jones (Sheffield Üniversitesi) ve Amy Bogaard (Nottingham Üniversitesi) tarafindan yönetildigi ilk sezon oldu. Sahada hazir bulunan ekip baskanlari Amy Bogaard ve Mike Charles olup, yüzdürme sorumlulari Meltem Agcabay ve Nicola Stone, yardimcilari Riza Büyüktemiz ve Mevlut Sivas idi.

1995 yilindan beri binlerce arkeobotanik örnegin islemden geçirildigi sistem, zamanin testinden geçmis, gerek boyutu gerek de yerel uzmanlari, laboratuvar alani gibi kaynaklariyla son derece etkileyici bir sistemdir. Önceki arkeobotanik ekibinin ve diger ekip baskanlarinin önerileri göz önünde bulundurularak, sistemin çesitli yönlerinde, özellikle de öncelikli ve öncelikli olmayan örneklerin laboratuvar analizi asamasinda degisiklikler yapilmistir.

Bu degisikliklere ek olarak, yüzdürme ekibi bundan böyle agir çökeltilerin islemden geçirilmesini denetlemeyecektir. Bu degisiklik de yine arkeobotani ekibinin ve diger ekip baskanlarinin önerileri dogrultusunda yapilmistir. 2003 yilinda Meltem Agacabay denetlenen agir çökelti islemleri, önümüzdeki yillarda ayri bir ekip tarafindan organize edilecektir.

Bu yil genel olarak, süregelen kazilarin tamamlanmasi (BACH) ya da yenilerinin baslamasi (4040) ile geçtigi için, öncelikli örneklemelerin sayisinda önceki yillara kiyasla büyük bir düsüs olmustur. Bu sebeple, bu rapor daha çok varolan kazilar üzerinden sürdürülecek olan metodolojik degisikliklerin açiklanmasina egilmektedir.

Methodological changes

The major change introduced was the application of a rapid assessment procedure to all samples (priority or not) in order to estimate their botanical composition and richness. This rapid assessment method will form the basis on which we decide which samples merit further analysis. In order to make use of the archive of samples that results from large-scale, systematic sampling and flotation, an efficient way is needed of identifying samples rich enough to be statistically representative of different events and deposit types. It may often be the case that priority samples, chosen in collaboration with other teams, are poor in botanical remains and so do not provide an adequate basis for assessing variation in the deposition of botanical remains and the activities that produce them.

Table 6 sets out the differences between the new rapid and priority assessment methods and the phase 1 and 2 procedures applied (to priority samples) in previous years; elements that have remained the same are not shown in the table. The rapid assessment method represents a compromise between time expenditure and desired accuracy. Initial scanning of flots proved unreliable, particularly for small items such as chaff and wild/weed seeds. Scanning, therefore, was replaced by sorting a subsample, but extraction of a small, random subsample was considered too time consuming. Instead, the method adopted was to take a non-random subsample (about a teaspoon, 5 ml) of the coarse (> 1mm) flot and to sort this rapidly under the microscope. The count for each category was then multiplied up based on the total flot volume and each category was scored on an abundance scale. The sorted material was then returned to the flot bag. Full sorting of some samples showed that the abundance estimates from rapid assessment were reasonably accurate.

Table 6: differences between the new rapid and priority assessment methods and the phase 1 and 2 procedures applied (to priority samples) in previous years

The key elements of the new rapid assessment are:
-it is applied to all samples in the field
-identification is to crop type and plant part
-it enables an evaluation of sample richness in identifiable plant material

On this basis it is possible to make an informed decision about the suitability of samples for further analysis and the sort of deposition they represent (e.g. single or mixed crops, single or mixed crop processing stages, predominantly non-crop material etc.).

The key elements of the new priority assessment are:
-as for the rapid assessment method, it produces detailed information on sample composition in terms of crop type, plant part etc.
-it provides more accurate estimates of quantities and density than the rapid assessment method

For priority samples we can quickly evaluate the overall status of the deposit – its density in charred plant remains and the extent to which these remains appear to represent a recognisable activity or event.

Archaeobotanical results for 2003

The team processed 481 samples, which break down by area as shown in Table 7. The target of c. 30 litres of soil from each context, where possible, was requested; the average sample size was 22 litres. Only c. 13 contexts were prioritised during site tours and these were either very poor in botanical remains or contained a mixture of crops and processing stages (with varying amounts of parenchyma, wood and dung).

Table 7: Processed samples by area

Table 8: Summary of the richness of the 481 samples, as estimated by the rapid assessment,

The majority of the samples are moderate in density (1-30 items per litre); even the density of the richest samples amounts to one or two teaspoon’s worth of charred seeds/chaff per litre of soil processed. Nevertheless, processing of relatively large quantities of soil has generated over one hundred samples (i.e. 117 containing a minimum of 100 items) that are rich enough to be considered representative of the deposits from which they derive, and so warrant full detailed analysis. This reasonably large assemblage can potentially provide the basis on which to investigate variation in the deposition of botanical remains. The rich samples tend to contain hundreds of glume wheat glume bases but relatively little barley or free threshing wheat material; some of them are also rich in wild plant seeds, especially sedges (Cyperaceae). The abundance of glume bases is consistent with frequent dehusking of stored glume wheat spikelets (grains still enclosed by glumes). The origin of the sedge seeds (including sea club-rush, Scirpus maritima) is a matter of some debate but may reflect the contribution of animal dung burned as fuel (see reports by the previous team).

We hope that statistical analysis of a large number of rich samples will eventually help us to tease apart the different sources of archaeobotanical material at the site. It is clear that variation between samples and contexts can be rather subtle. Multivariate statistical approaches to a large dataset have the potential to identify underlying trends in composition through time/space. It should be noted that many of the samples processed this season will be studied as part of ongoing work on the BACH and TP assemblages.