ÇATALHÖYÜK 1995 ARCHIVE REPORT

Animal Bone Report

Louise Martin

In the 1995 field season, zooarchaeological work concentrated in four areas:

In addition, a database specifically tailored for the Çatalhöyük faunal remains was designed and programmed by Nerissa Russell and Tim Ritchie respectively. This forms part of the Çatalhöyük integrated database, and will be used for recording all faunal material from 1996 onwards.

1995 excavation areas

The two main areas excavated in the 1995 season were building I on the north mound (hereafter referred to as building 1), and an area on the main mound, which was bounded on all sides by separate mud-brick walls (of different houses) at ground level (hereafter termed MELL). Parts of the latter have tentatively been interpreted as an external area or courtyard.

Faunal analysis aims at comparing all aspects of the animal bones from these two areas, to see how results may inform on, or be evaluated in the light of, their eventual interpretation. All material studied so far comes from 'fill' deposits, but these could potentially have very varying derivations, for example they could represent either intentional filling, structural collapse, waste dumping, post-occupational erosional fill, or temporary later occupation. Within MELL area, for example, it has been suggested that deposits are of two main types: building fill (units 1022, 1023, 1026) and midden (units 1028, 1033, 1034, 1037). These divisions have been used in the following analysis.

In order to explore patterning both between the two areas and within MELL area (comparing the 'building fill' and 'midden' deposits), data has been collected on the following:

It should be stressed the study of the animal bone from these two excavation areas is incomplete: analysis stopped at an arbitrary cut-off point, and therefore any patterns must remain tentative.

Density of bone

A comparison between building I on the north mound and the MELL area shows that the former produced a total of 547 animal bone fragments, while the latter yielded 2941 (table 1). When the volumes of deposit excavated are compared, the density of animal bone in building I is (preliminarily) 0.6 fragments/litre of soil; the 'fill' deposits in MELL area produced 1.9 fragments/litre, while the 'midden' has an average of 2.9 fragments/litre.

Clearly, building I has the lowest density of bone; the 'midden' has the highest (with great variation between individual units), and the 'fill' in MELL area falls between.

Bone fragmentation

A size count of all the animal bone material showed that in building 1, most of the fragments were less than 5cms in length, and hardly any fragments over 5cms were found (although some large bones were clearly visible but had not been lifted by the time this study ceased).

MELL area also had a lot of bone in the less than 5 cm category, but in addition the 'midden' deposits contained more fragments in the 5-10 cm class. It is interesting that these results in some way reflect the relative proportions of medium-sized mammals and large sized mammals present among the identifiables (see Relative proportions of taxa below). The 'midden' area actually has more cattle and equids identified than either of the other areas (see figure 3).

A full analysis of the fragment types is pending, but two further coarse methods of assessing bone fragmentation are used here. Firstly, the average weight per fragment within an assemblage has been calculated, and secondly, the percentage of diagnostic fragments (assuming that as fragmentation increases, bones become less identifiable).

In building 1 the average weight per bone fragment is 1.5 g, which is very similar to that of the 'fill' deposits in the MELL area, where the figure is 1.4 g. Bones from the 'midden' deposits in the MELL area, however, have a much heavier average weight of 4.1 g, again supporting the picture of more bones of large-sized animals here.

The percentage of identifiable bone clearly depends on what is classed as 'identifiable'. In this initial sort of the material, only certain bones/zones were counted as diagnostic, but since this was consistent across areas, a comparison of identifiable bone may be informative, Building I produced only 4% identifiables from the total bone recovered, compared to the MELL area, which has 9% in the 'fill' deposits and I 0% in the 'midden'. All these figures are relatively low, but the variations are interesting in suggesting any combination of faunal compositional, processing or taphonomic differences between areas and deposits.

Burnt bone

A much higher percentage of burnt bone was recorded from building 1 (10%) than from MELL area, where both the 'fill' and 'midden' deposits have only 1% of bone exhibiting burning.

Surface condition of bone

The observation was consistently made that bones from building 1 had surfaces that were matt and dull, and frequently described as 'dusty' despite having been washed. This subject clearly requires experimental work in order to explain these taphonomic patterns, but it is likely that such surface conditions relate either to the way the bone has been heat-treated (cooked, or burnt in disposal), or to specific weathering conditions. Some bone from building 1I is also described as containing very eroded and battered bone, although fragment edges always showed sharp and spikey breakage (as opposed to smooth/rounded edges which are characteristic of attrition and movement).

In contrast, surfaces of bone from the MELL area were smooth, more shiny, and were generally described as being in good condition. It is notable that bone from here showed a great variety of surface colours, which again relates to the specific histories of the material. Modem experimental work, and internal comparisons of deposits and bone are needed to understand these variations.

Relative proportions of taxa

A very small sample of identifiable bone derives from building 1 (n=24). The most commonly represented animals, however, are sheep and goat (figure 1). No cattle are yet recorded (although this is certainly a factor of the arbitrary cut-off point of this study since a number of cattle bones were visible in the trench). Other taxa which are represented in low frequencies are equids, boar/pigs and small canids; a single bird bone was also found.

The 'building fill' deposits in the MELL area have a similar pattern of faunal representation, with a larger sample size (n=116). Again, sheep and goat are dominant (figure 2), and equids and boar/pig are represented, but some further taxa are present in small numbers: cattle, deer (fallow?) and hare.

The 'midden' deposits in MELL area have a different character, as mentioned in the discussion of fragmentation above. Here, cattle and equid remains are almost as commonly found as those of sheep/goats. The boar/pig counts are also higher than in the other areas. The contrast between the two sets of deposits in the MELL area - where one is dominated by medium-sized herbivores (e.g. sheep/goat), and the other having a much higher proportion of large herbivores - is striking, and requires further investigation.

Comment on the actual species represented, and the domestic/wild status of some of the animals awaits larger samples.

Sub-surface scrape areas

The findings of the three 10x10 m2 areas scraped this season add to those from the 30 already scraped in 1993 and 1994, and results will be discussed as a whole in the forthcoming volume on the surface material from Çatalhöyük.

The animal bone patterns appear similar to those from the scraped areas immediately to the north (e.g. units 708, 709, 710 in square 1030/1170), with similar counts and densities.

Bone from 1960s deep-sounding backfill

Over 5000 fragments of bone were collected from backfill deposits from the deep sounding (which included both human and animal remains). Since this is unstratified and mixed material, no form of analysis was undertaken, except to sort through the bone for possible reference material. Two specimens are worthy of comment. Firstly, part of a fused distal humerus of Bos sp. (cattle) was found (from unit 1005) and is in very good condition. Measurements (following von den Driesch 1976) are as follows:

Bd: 113.00mm     BT: 110.6mm     HTC: 47.0mm

The interesting point about the large size of this humerus is that it exceeds the size range given by Perkins (1969) for any cattle humeri previously found at Çatalhöyük. (the range for the Bd measurements from Layer VI is 63.0-105.0mm - interpreted as 'domestic'; that for Layers X-XII is 85.3-108.0mm - argued to be also possibly domestic, although the size range overlaps with wild cattle (see Perkins 1969:178)). This cattle hurnerus, assuming that it does indeed derive from the tell, would therefore extend the size range of the cattle known from the site.

The second specimen worthy of note is a very large deer (Cervus elaphus) antler. The antler was shed, and the brow tine (closest to skull) had been removed by a series of even v-shaped cuts circumscribing the base of the tine. The antler is obviously undated and could presumably derive from any period represented at the site, but the large size makes it interesting for comparative purposes. The measurements are as follows:

Max. width of pedicle: 82.4mm

Max width beam above brow tine: 80.0mm

Animal bone reference collection

The partial skeletons of 10- 12 animals were collected from modern dumping grounds in the region, in order to start a comparative collection for work at Çatalhöyük. These skeletons were of horse, cow, sheep, goat, dog and cat. They were cleaned firstly by soaking in a solution of biological (enzyme) washing powder and water; any remaining sinews and cartilage were removed with knives; and specimens were finally degreased in a dilution of bleach.

These specimens represent the beginning of a collection to be housed in the Çatalhöyük laboratories which will be expanded in future seasons. Such material is very important for identification purposes in zooarchaeology.

References

Perkins, D. (1969). 'Fauna of Çatal Hüyük: Evidence for Early Cattle Domestication in Anatolia' in Science 164, 177-179.