Mesolithic 1 Economy and Society

Chapter 2 (Pages 44-54)

The great environmental diversity of the Levant meant that a wide range of fauna, flora and raw materials could be exploited by man but these resources would have varied considerably from zone to zone. This diversity of resources is reflected in the different artifact inventories from supposedly approximately contemporary deposits on Mesolithic 1 sites. The artifacts reflect above all the activities practised by the inhabitants and it is reasonable to expect that these activities would have varied according to the resources available although other factors such as human preference would also have played a part.

The animal bones from the largest and most important category of direct evidence for the economy of Mesolithic 1 and the varied collectlons from individual sites give some idea of the wide range of resources that was used. The catchment areas of the sites would have contained a variety of species in different proportions and the environmental factor can be detected in the excavated faunal remains.

Preliminary information from Ein Gev I indicates that gazelle comprised 43% of the bones of the main food animals, deer (fallow, red and roe together) 36% and ovicaprines 15.5%. There were also some cattle bones (4.5%) and a little pig (1%) [See Page 120ff in *1.5 and Page 372 in *1 Below] while a number of other species, including birds, were represented in small quantities. Many of these species were also found in Ein Gev II and III.

The fauna from the Kebaran layers at Nahal Oren was quite similar and the species were represented in much the same order but here the preponderance of gazelle (77.4%) was even more striking. Fallow deer were second in importance (15.2%) but the other species were much less significant: cattle 3.3%, pig 2.6%, red and roe deer 1.1% and goat hardly at all (0.1%) (See Table 4 in *2 Below).

At Kebara itself gazelle dominated the fauna in layer C (See Figure 3 in *3 Below) while fallow deer were much less important than in the Aurignacian. Other species were killed only in very small quantities. The pattern was similar at Hayonim even though the faunal sample was small. Gazelle was again the most numerous species (See 129ff in Page *4 Below) with fallow deer second. Cattle, red deer and caprines were present in small quantities as well as numerous other species of rodents and carnivores. Hares were very common and it appears that terrestrial molluscs were also eaten.

When the fauna from these sitess is compared it is clear that gazelle were eaten in much greater numbers in Palestine than any other species. Deer, especially fallow deer, were second in importance and hares seem to have been a regular supplement. A wide range of other species was hunted or collected for food but these were of minor importance in the total diet. The consumption of gazelle rapidly increased in proportion to that of fallow deer after the Aurignacian (See Figure 9 in *2 Below). This trend continued throughout Mesolithic 1 as the evidence from Nahal Oren makes clear; fallow deer declined from about 30 or 40% in layer IX early in the Kebaran to 14.9% in layer VII at the end while gazelle increased reaching 82.6% in layer VI transitional between the Kebaran and Natufian (See Table 3 ibid). Such a pattern of exploitation depended partly on environmental factors. The forest in central and northern Palestine thinned out in response to more arid conditions during the last cool phase of the Pleistocene. This probably led to a natural decline in the deer population and a rise in that of gazelle. As gazelle became more plentiful so man hunted more of them rather than other species which were much less numerous. Legge has suggested that the people of Mesolithic 1 may have practised gazelle husbandry (See Page 123 in *5 and Page 91 in *2 Below). Certainly it appears that this species was pursued more intensively and that care was taken to kill a high proportion of immature animals but this may have amounted to no more than selective hunting. It is probable however that a close dependence on gazelle was established at this time and that this relationship was intensified later.

This concentration on gazelle was not a universal pattern of exploitation in Mesolithic 1. At Ksar Akil for example the percentage of fallow deer killed was very high (53%), a higher proportion even than in the Aurignacian. Goat was second in importance (30%) and roe deer third (15%). Some cattle were taken (1%) but even fewer gazelle than in the preceding stage (1%). The emphasis on fallow deer throughout the Ksar Akil sequence is as striking as that of gazelle on sites in northern Palestine after the Aurignacian. It must equally represent selective intensive hunting of the species near the site that was most numerous and easiest to kill.

Goats (the bezoar Capra hircus aegagrus and beden Capra ibex nubiana) were the main animals hunted at Wadi Madamagh (See Page 67 in *6 Below), their bones accounting for 82.7% of the animal bones at the site although other food animals such as aurochs (Bos primigenius), gazelle, pig (Sus scrofa), an equid and hare were also taken. Again one species was much preferred to all the others and it was probably the one that was especially numerous around the site. Selective hunting was practised on most sites in Mesolithic 1, the species chosen being those that were readily available in the vicinity and which provided ample meat and other products.

Marine shells have been found at a number of inland Mesolithic 1 sites such as D5 near Ein Avdat (See Page 20 in *7 Below), Wadi Madamagh (See Page 56 in *8 Below) and Ein Gev I and II (See Pages 121 and 125 in *1.5 Below) and also nearer the sea at Hayonim (See Page 137 in *4 Below). Many of these were decorative objects brought up frcm the Mediterranean but some were edible species. It seems likely that fish and molluscs would have been eaten, if only in small quantities, where they were available. One instance of this is Ein Gev where fish bones were found at En Gev IV only, a site transitional between Mesolithic 1 and 2, but not at En Gev I, II or III (See Page 109 in *1.5 Below). The latter sites were all in use when the Pleistocene Lisan Lake still existed. It had much of the character of the Dead Sea today and fish could not have lived in it. At the close of the Pleistocene the Lisan Lake shrank and eventually the Sea of Galilee was formed from it as a freshwater lake. Fish gradually established themselves and by the time Ein Gev IV was founded on its shore man was able to catch them for food. It is possible that some Mesolithic communities depended to a much greater extent upon marine resources but their sites will have disappeared as the sea level rose at the end of the Pleistocene. During Mesolithic 1 the sea lay about 15 kilometres west of the present coast of Falestine.

There is some archaeological evidence that plants were eaten in Mesolithic 1. 64 seeds were recovered from the Kebaran levels at Nahal Oren which included fig and grape pips as well as vetch and grass seeds (See Table 6 in *2 Below). There is also rather enigmatic evidence that cereals were collected as one barley and three emmer (Triticum dicoccum) grains were found in the deposits. One cannot unreservedly accept the presence of cultivated emmer grains in such an early deposit despite the apparently unequivocal archaeological evidence that they were in situ (See Page 93 ibid) until supporting evidence is forthcoming from other sites. Not too much should be made either of such a small sample but it would not be surprising if cereals were being collected at this early date as part of the vegetable diet. Further, it is not impossible that emmer, which is present in some early Neolithic contexts, was being collected in the wild in such a way that a domesticated form had already developed.

We do not know what proportion of the diet of the people of Mesolithic 1 consisted of vegetable foods but the probability is that they were at least as important as the meat obtained from hunting. Recent studies have shown that among most contemporary hunter-gatherers vegetable foods form a major part of the diet. Australian aborigines select food from almost all edible plant and animal species in their habitats but they eat more vegetable foods than anything else. Woodburn has calculated that 80% of the Hadza diet by weight is composed of vegetable foods and 20% of meat and honey although the calorific value of the plant foods is proportionately rather less. The !Kung bushmen also eat similar quantities of vegetable foods. It is only in certain extreme environments where vegetable foods are rare such as the north-west coast of North America or the Arctic that man relies on hunting and fishing for subsistence. The wide range of plant and animal foods available to man in the Levant is typical of many other regions in lower latitudes where hunter-gatherers now and in the recent past have subsisted on a largely vegetable diet though eating a great variety of other foods from time to time. It seems highly probable that this was also the case in Mesolithic 1; vegetable foods were almost certainly collected and eaten in quantity and may have formed the basis of the diet.

The artifacts and technology of Mesolithic 1 communities differed significantly from those of the Aurignacian. The most obvious development was in the chipped stone industries although the innovation of microlithic tools remains difficult to interpret in the Levant where there is no accompanying evidence of a major change in the basis of subsistence. One of the new tools was the sickle blade (See Page 156 in *1.5 Below), that is a blade with gloss on the cutting edge. This was quite a rare tool on Mesolithic 1 sites, as were blades and blade tools in general. The gloss an the edge indicates that the tool was used for cutting plants with siliceous stalks, reeds, cereals or the like. This innovation signifies that specialised stone tools were being developed to process plants though there is no indication yet that anything resembling conscious agricultural techniques was being practised. One other new group of tools was stone pestles, mortars and grinders. These were found first at Ein Gev then recently at a number of other Mesolithic 1 sites (See Page 368 in *1 Below). Stone grinders have been found at the Aurignacian site of Ein Aqev in the Negev (See Page 4 in *9 Below) but it appears that these tools were first made in quantity during Mesolithic 1. We do not know if they were used in some new industrial process or for crushing plant foods but the development of such tools is significant not only as a technological innovation but because they were the first bulky heavy items made by man in the Levant. They would have been an impediment to mobile groups and as such were harbingers of a later more sedentary existence.

We know the area of some Mesolithic 1 sites and from this it is possible to estimate approximately the number of people which inhabited them. When this information is combined with the data we have for their economy it enables us to reconstruct in outline the social structure of these human groups. Some sites were very small indeed; Nahal Lavan VI was 15 square metres and Nahal Lavan II 40 square metres (See Page 477 in *10 Below) while Hofith-68 (75 square metres) and Kefar Darom 28 (70 square metres) were only a little bigger (See Pages 63 and 92 in *1.5 Below). Most of the other sites for which we have information were between 120 and 300 square metres in area. Wadi Madamagh was originally about 120 square metres (See Page 55ff in *8 Below) and the Mesolithic 1 occupation at Jiita II about 130 square metres [NOTE 1]. The maximum extent of site D5 at Ein Avdat in the Negev was also 130 square metres (See Page 20 in *7 Below). Moshabi XIX was only a little larger (160 square metres) [See Page 483 in *11 Below] but Moshabi I was almost twice as big (300 square metres). The estimated area of Kebara itself was the same as Moshabi I (See Page 156ff and Plate XXIII in *12 Below). The total area of occupation at Kefar Darom 8 was 500 square metres (See Page 102 in *1.5 Below) and is apparently greater but as it is thought that this site was used over a long period the total area occupied at any time was probably less and so within the range of the other sites. One site (Ksar Akil) may have been bigger than the others I have mentioned since the total area of the shelter and terrace in front was well in excess of 500 square metres [NOTE 2]. The area occupied in Mesolithic 1 may have been 500 square metres or more but we do not have enough information about the excavations at Ksar Akil to be sure.

While the size of camps used by hunter-gatherers today varies widely they frequently come within the same range as Mesolithic 1 sites. It is known that camps of Australian aborigines are often from 200 to 300 square metres in area while Bushmen camps may be a little larger, ranging from 230 to 790 square metres. These are open sites in which groups of the same size might be expected to occupy a slightly larger area than in a cave. These groups or bands have 20 to 30 members and groups of this size are commonly found among hunter-gatherers today. Using data from the Birhor, Kung Bushmen and some Australian aborigines Birdsell (See Page 235 in *15 Below) has suggested that the average band has 25 members. Steward has found (See Page 331 in *14 Below) that this is also true for the Athapaskans, Hadza and Western Shoshoni. Given that Mesolithic 1 sites are approximately the same size as those used by groups of hunter-gatherers today it seems likely that the basic unit of social organization, a band of about 25 people, was also about the same.

It is possible to suggest in a little more detail how these bands may have been organized. Some years ago in a discussion of social organization among primitive peoples Service stated that most hunter-gatherers today were grouped in patrilocal bands. He also believed that this type of social structure was characteristic of prehistoric hunter-gatherers. Recent research has demonstrated that this particular band structure is much less common than was previously supposed (See Page 7ff in *13 Below) and indeed Service has considerably modified his earlier views. It is now clear that Primary (See Page 331 in *14 Below) or Composite (See Page 8 in *13 Below) bands are more typical of hunter-gatherer societies today. Bands of this type may fluctuate considerably in size but usually consist of about 25 individuals. The members will be related to each other but there is great variety in the kinship structure.

This form of social organization has considerable adaptive advantages. Composite bands are small so they do not rapidly exhaust the food supply in the vicinity of their camps. They are also extremely flexible: they can move easily from one camp to another in order to exploit fresh resources and can also split up or amalgamate to adapt to seasonal variations in food supply. This type of adaptation is well-suited to a region such as the Levant with very varied geographical conditions and seasonal differences of climate, vegetation and fauna. Such a model of social organization fits the archaeological evidence for Mesolithic 1 quite well. While it is no doubt over-simplified and will need modification when more evidence is available, I believe it indicates in outline how society was organized in Mesolithic 1.

Although most Mesolithic I sites fall within the same size range as camps of composite bands of hunter-gatherers today a few, as we have seen, were smaller than the average, that is less than 100 square metres in area. Such sites were too small to have accommodated a complete Composite band. They may have been the camps of single families or small hunting parties which had separated from a band. Some sites from 100 to 200 square metres were intermediate in size between these transitory stations and the larger camps; they may have been used by two or three families or groups smaller in size than the full composite bands. All Mesolithic 1 sites and thus the groups which inhabited them were generally quite small particularly when compared with some Mesolithic 2 camps and earlier Neolithic settlements.

Some information about the internal arrangements of Mesolithic 1 sites has been recovered in excavations at Ein Gev which enables us to deduce in a little more detail the organization of Mesolithic 1 communities. Excavations here revealed four sites in close proximity, three of them Kebaran (Ein Gev I, II, III) and one transitional between the Kebaran and Natufian (Ein Gev IV). Ein Gev I was 150 square metres in area but the density of stone tools found suggests that only 50 square metres was intensively occupied (See Page 182 in *1.5 Below). Ein Gev II was very small but may have been inhabited about the same time as Ein Gev I ibid, so these two sites came within the size range of the intermediate sites discussed above. Ein Gev I consisted almost entirely of a large round hut 5 to 7 metres in diameter floored with pebbles (See Figures 89 and 90 ibid) with associated hearths and occupation debris. Ein Gev III probably represented a similar hut (See Figure 102 ibid) while En Gev IV was also apparently a habitation structure (See Page 126 ibid). These large single structures suggest that each site may have been occupied by an extended family rather than a band composed of several families. The total number of people in the group was probably less than the average of 25 or so suggested for composite bands; perhaps no more than 10 or 15.

A little more evidence for a similar type of small camp has now been found at Moshabi XIV in the Jebel Meghara (See Page 483 in *11 Below). A pit dwelling has been discovered here associated with post-holes and hearths. This camp would appear to resemble those at Ein Gev and may have been inhabited by a group of similar size.

Most Mesolithic 1 sites have very thin occupation deposits indicating that they were occupied briefly perhaps on a seasonal basis. The presence of amphibian species but not reptiles in the fauna from Hayonim may be taken as evidence that the cave was too wet in the winter for human occupation (See Page 138 in *4 Below); it was probably only inhabited during the summer. The absence of certain rodents closely associated with man is a further indication that human occupation was intermittent during Mesolithic 1. The hut at Ein Gev I had six floor levels separated by sterile sand (See Page 110 in *1.5 Below), clear evidence that the structure was repeatedly occupied for short periods of time. Ein Gev III seems to have had a similar series of floor levels (See Page 124 and Figure 2 ibid) emphasising that this was a regular practice over several years.

Composite bands or smaller groups may have lived on each site for a short time varying between a few days and several weeks until the easily available food in the neighbourhood began to diminish. They would then have moved on to a new site. On modern analogy their movements would probably have taken place within a defined area or territory and may have been restricted by the presence of other groups who were exploiting neighbouring territories. This pattern of movement would have been related to the marked seasonal variation in resources typical of the region. The length of time spent at each site would have varied during the year depending on the resources available in the vicinity. A band might have returned to certain favoured sites regularly for a few years but then have moved away into a new territory.

Although Mesolithic 1 sites were usually insubstantial, clusters of them have often been found together, as around Nahal Poleg (See 10ff in *16 Below), Kefar Darom, Ein Gev itself and in Wadi Malih. These areas were sufficiently attractive for the people of Mesolithic 1 to visit them frequently. The same was probably also true of some of the shelter sites with more substantial occupation deposits such as Ksar Akil and Jiita II, themselves close together, and Yabrud III. Higgs has suggested that transhumance may have been practised between Nahal Oren and Rakafet (See Page 95ff in *2). This would have been a specialised form of seasonal movement between a lowland and an upland site quite near each other which offered complementary resources. Such a pattern of movement may have been typical of some other areas of the Levant where contemporary sites have been found only a short distance apart in different environmental zones. This practice may have begun much earlier during the Palaeolithic as has been postulated for Epirus in Greece (See Page 18ff in *17 Below).

The constant movement that was one of the dominant features of this way of life would have imposed a severe constraint on the rate of population growth. The mothers would be unable to look after more than one or at the most two infants at a time and so could only support children born at least three years apart. Without some form of birth control children would often be born more frequently than this so some other limitation would be necessary. The solution adopted among many hunter-gatherers today is infanticide (See Page 236ff in *15 Below) while among communities in extreme environments even more severe demographic controls are exercised. Where infanticide is not practised a high infant mortality rate severely restricts population growth. By means of these and other constraints the Mesolithic 1 population would have remained stable or grown only very slowly indeed. On modern analogy it would have remained for much of the time well below the maximum carrying capacity of the region ...