The Bronze Age Animal Economy from Akrotiri: A Preliminary Analysis

Studies of animal bones from archaeological sites have, until recently, concentrated upon the history of the present domestic animal community that is first encountered as a recurrent entity in the Neolithic. The aim of studying animal bones from later prehistory is to monitor the subsequent subsistence behaviour of early agricultural societies and in particular to observe the effect of social processes and developments such as urbanisation, central control and early state organisation on the food producing base. The exploitation of animals for their products of meat, milk, wool, hides, traction; manure, protection and as hunting or herding aids represent a series of goals, the attainment of which is achieved by the implementation of management strategies whose primary aim is to secure a return with the minimum expenditure of effort. These goals are realised in two main ways; by arranging the combinations of species and by managing individual species through their age and sex structures.

The Archaeozoologist is equipped with a variety of models and techniques designed to elucidate the animal husbandry patterns represented by his data. Such interpretative models have been proposed by among others Payne (1973) for domestic sheep and goats, Wright and Miller (1976) for wild sheep and Higham (1967) with reference to cattle. This paper will outline the basic structure of the animal community from the settlement of Akrotiri and indicate the particular opportunities that are offered by this site for the better understanding of animal bone data and Aegean bronze age subsistence economies.

It is important however to stress that the animal bone data must in the final analysis be fully integrated into the wider archaeological picture of which it is a part, so that a full appreciation of the constraints that operated upon, and the possibilities that were open to, prehistoric husbandry practices can be achieved.

The fragmentation of archaeological data into narrow specialities each with their own independent explanatory framework is an unwelcome trend and the aim of the Theran Congress in promoting intra-and inter-disciplinary discussion of common problems is to be applauded.

The small non-random sample of animal bones that was studied can be used to establish the following aspects of the prehistoric animal community at Akrotiri;

1. The properties of the sample-biases of recovery and representation.
2. Butchery and disposal patterns.
3. Intra-specific age structures.
4. Inter-specific relationships.

1.     SAMPLE PROPERTIES.

The sample is dominated by the 2845 specimens that could be identified to the four main species, sheep, goat, pig and cow (TABLE I). This represents 40 % of the total amount of bone fragments which numbered 7155 studied specimens.

The weight of the total sample was 32.225 kilos. of which identifiable specimens accounted for 62 % thus indicating a relationship between the size of a fragment and its potential for identification to the species level. The non-identifiable fragments were organised into three groups allocated on the basis of size into small, medium and large size animal fragments. As can be seen in Tables (II, III and IV) the structure of the sample for identifiable and non-identifiable specimens is very similar whether calculated by weight or number. In a site such as Akrotiri where the faunal list is dominated by four species with three discrete size ranges such a comparison is possible and helpful in ascertaining sample bias due to recovery factors, selective destruction and hence under-representation in the faunal list. The conclusion here is that recovery, even though sieving was not carried out, does not appear to have biased against the smaller size classes as might have been expected, thus permitting some confidence to be placed in the later conclusions of the analysis.

Table I: Identified species from Akrotiri

Table II: Relative proportionsof the Primary Economic species

TABLE III: Relative proportions of the Non-identifiable specimens

Table IV: Average weight (grammes) of individual fragments

2.      BUTCHERY AND DISPOSAL PATTERNS.

There was very little evidence for butchery with only 5 % of ovicaprine, 4% of pig and 1.2 % of cattle bones bearing traces of cut marks. These were mostly found in sheep and goat on the forelimb - humerus, radius and ulna (Table V). While it might be argued that such evidence points to the special treatment of this part of the skeleton the differential representation of the skeletal elements is more economically explained as the result of differential preservation linked to the age, fusion and hence durability of bone as has been detailed by Brain (1976). The rarity of butchery marks points to a non-intensive attitude to processing the animal carcasses which possibly indicates that meat was an occasional resource butchered by the household rather than an economic staple that was centrally processed by a specialist.

The two red deer antler specimens both bear traces of working. The small beam fragment is sawn and the large antler fragment has knife marks on the attached skull together with thin flakes shaved off the beam. This evidence of antler working, which is also known from the bronze age site of Phylakopi on Melos is interesting as there are no antler artifacts yet known from either site.

Evidence for chewing and gnawing by dogs is also very rare on the Akrotiri material, which accurately reflects the low numbers of this species in the site.

The disposal of the bones is one aspect which deserves further treatment.

The recent work of Meadow has shown that the role of context has to be taken into account when dealing with differential representation of skeletal elements (1975). Akrotiri with its clearly defined contexts of streets, houses, floors and spaces together with the associations of artifacts that permit functional reconstruction will provide an ideal testing ground for such taphonomic questions and it is proposed to follow up this line of enquiry.

3.      INTRA-SPECIFIC AGE STRUCTURE

This provides a coarse sorting of the data into relative age groupings and is only used to indicate the general age structure. The histograms show three contrasted management patterns with few pigs surviving past 3 - 3½ years whereas over 50 % of the cattle specimens reach bone fusion maturity. For the pig this suggests that it is being kept for meat and slaughtered at c. 2 years, after the major period of growth has ended. The articulated skeleton in Δ 9.1. at Akrotiri has a mandible with the M3 almost fully erupted. In the forelimb the late fusing bones - proximal humerus and ulna, distal radius, ulna and the metapodials - were all unfused indicating an animal killed sometime in its third year. It therefore serves as a more complete example of the management strategy adopted toward this species that is also substantiated by the bulk of the fragmentary data. It has also been suggested (Doumas pers. comm.) that the pig in  Δ 9.1. was hanging from the roof thus making it an intentionally stored item rather than the victim of earthquake or pumice.

The cattle data is remarkable for the rarity of juvenile specimens amongst the bones and this is also so for the teeth. The small numbers of this species indicate its use for draught purposes where milk would have been a useful bonus.

The small numbers of surplus animals together with the older beasts that could no longer be used for traction would have been used for meat.

The various estimates for the sheep: goat ratio are listed in Table VI below.

The goat horn cores all come from large adults and some carry traces of cutting around their bases, indicating selection for some special purpose. The other elements all point to the dominance of sheep over goat and the age data will be taken to refer primarily to this species. The histogram in Fig. 1. shows the steady exploitation of the ovicaprid population at all three fusion stages with 40 % of the specimens surviving beyond the late fusion stage. Tooth wear analysis on the lower M1 shows that 46 % of all specimens come from worn or heavily worn specimens, 27 % are in full wear and 27 % are in the process of reaching the full wear stage. This is consistent with the structure suggested by the bone fusion data. The dominance of sheep in the fauna together with this age structure for its exploitation suggests that it was subject to a mixed management strategy with the accent being on dairy products, together with the removal of surplus lambs for meat and the retention of small wether flocks for wool.

4.      INTER-SPECIFIC RELATIONSHIP 

The dominance of ovicaprids is a common expectation for island sites in the Aegean Neolithic and Bronze age. The available data is shown in Fig. 2. from which it is apparent that with the exception of Keos (Coy 1974) pig maintains a marked consistency in representation that rarely exceeds 20% either by weight or by frequency. This suggests a standard pig keeping strategy where probably each household had its own pig which it kept in or near the settlement, fed on a low quality diet and slaughtered between 2 to 3 years old. Fig. 2. also poinrs to a similarity between Akrotiri and the final Late Bronze II phase at Phylakopi, with respect to the relative proportions of the main domestic species. The comparability of the two islands in size and their importance in Cretan and Mainland trading networks, together with the fact that both sites are small towns, will make them useful units of comparison in later analyses of island subsistence economies.

Three other aspects require brief comment. The three red deer (Cervus elaphus) specimens attest the presence of this species on the island. One specimen is a massacred antler with a portion of skull attached. It comes from a young animal, has both brow and bay tines, and is cut above the tray tine. A first phalange provides a stronger argument for the existence of this species on Santorini rather than antler being just a traded resource. In the absence of palaeontological specimens to indicate an indigenous cervid population we must assume thar the animal was brought to the island by the inhabitants of Akrotiri. Jarman considers that the Minoans imported both the fallow deer and badger to Crete and he suggests that the former was kept as park deer (Jarman mss.).

The single tibia assigned to the genus Equus is a small, immature specimen that on size inspection is equivalent to donkey rather than to mule or horse. A small equid is also known from Late Bronze II levels at Phylakopi, and in both cases their role in the economy would have been for draught purposes.

The supplementary resources are poorly represented which is probably due to the lack of small scale recovery; a situation that future excavation will remedy. Fish is very rare and when encountered is only from small species. There have as yet been no indications of a substantial resource such as tunny and the rarity of fish hooks and fishing equipment (Doumas pers. comm.) supports this view. 

SUMMARY

A consistent pattern is beginning to emerge of island animal economies in the prehistoric Aegean. As might be expected the dominant element is provided by the ovicaprids whose main use is to be seen in terms of milk and wool. Pig provides, as it does today in the islands, an easily husbanded meat source, whereas cattle which are not suited to the small island environments of the Cyclades would have been important for traction, although under certain conditions their milk potential might have been more fully realised. It will be argued elsewhere that the shift from a dispersed to a nucleated settlement pattern in Melos resulted in a change in animal management, primarily as a result of increasing cost factors in exploitation (Gamble, in preparation). By analogy with the Melian data  Akrotiri can be grouped with those animal economies associated with a nucleated settlement pattern. Obviously such a hypothesis needs testing by further excavation to determine elements of nucleation within the settlement itself. For the archaeozoologist such future work would provide the opportunity to observe variation in the bone material and in particular to examine the effect of context on the composition of samples. In this way the site of Akrotiri provides an excellent opportunity to investigate the structure and nature of one category of achaeological material while permitting its integration with other classes of data and this providing information on a critical period of Aegean social development.

---------------------------------------

---------------------------------------