eitun and the transition to agriculture in Central Asia

J eitun and the transition to agriculture in Central Asia David R. Harris Most research on the origins and early spread of agriculture in the Old World has focused on Southwest Asia and Europe. Since 1 989, Institute and other British archaeologists have been working east of the Caspian Sea in southern Turkmenistan, and it is now possible to draw conclusions about how agriculture and village life began in Central Asia.

T he questions of why and how, some 10,000 years ago, hunter gatherers living in Southwest Asia began to cultivate crops and raise domestic herd animals has fascinated and puzzled archaeologists since V. Cordon Childe first proposed his con cept of the Neolithic Revolution.1 Even before the technique of radiocarbon dating had been invented, he had argued, in a series of widely read books, that the tran sition from hunting and gathering to set tled village life based on agriculture took place in the Near East soon after the end of the last glacial period. He envisaged that fr om there the new way oflife spread west wards to Greece and the Balkans, from where "Danubian" farmers colonized cen tral and eventually northwest Europe. Although this diffusionist model for the spread of farming during the neolithic period has been modified by later archae ological research, it has been broadly con firmed by many radiocarbon dates and it continues to fit the available evidence bet ter than alternative models of independent agricultural origins in Europe.2 Childe's main concern was with the Near East as "the oriental prelude to European prehis tory", 3   The finds of pottery, bone and stone tools (particularly many sickle blades), together with impressions of barley and wheat grains in the mudbricks, indicated that the inhabitants cultivated cereals and herded sheep and goats. The presence of the bones of a variety of wild animals showed that the people also depended to some extent on hunting. No radiocarbon dates were obtained for Jeitun, but at two other sites on the piedmont -Togolok and Chag ylly ( Fig. 1) -levels judged on pottery styles to be somewhat later than Jeitun were dated, respectively, to 5370±100 be and 5050±110 bc.6 By the late 1960s, the existence in the mid-sixth millennium be of a series ofneo lithic agropastoral settlements on the pied mont had been convincingly demonstrated ( Fig. 1). They came to be referred to collec-tively as the Jeitun Culture 7 and their exist ence raised challenging questions about how the beginnings of agriculture and settled life in Central Asia related to the equivalent processes in Southwest Asia. The cardinal question was: did agriculture originate inde pendently in Central Asia or was the tran sition there the result of the spread of crops, domestic animals, agropastoral techniques and people fr om Southwest Asia?
Masson and his colleagues drew atten tion to similarities in certain types of arte fact, such as clay figurines (Fig. 6), pottery and small stone axes, between the Jeitun Culture sites and finds at such Southwest Asian neolithic sites as Jarmo, Tepe Guran and Tepe Sarab in the Zagros mountain region, and they also raised the question of whether the origins of the Jeitun Culture should be sought in the southern Caspian area, where several mesolithic cave sites had been discovered (Fig. 1). But the ques tion of how agriculture originated in the piedmont zone remained unresolved.

Recent investigations at Jeitun
No further investigations took place at Jeitun until the late 1980s, when Professor Mass on invited the author and his archaeo botanical colleague at the Institute of Ar chaeology, Cordon Hillman, to undertake palaeoenvironmental research at Jeitun in conjunction with renewed excavation of part of the site. We gladly accepted this in vitation and first visited the site in April 1989 for a short season of small-scale ex cavation (Fig. 5) and off-site ecological sur vey. Our main aim that year was to see if we could confirm Masson's conclusion that cereals had been cultivated at neolithic Jeitun. We hoped to do so by recovering charred remains of the plants (grains and chaff) by means of flotation -the technique whereby lighter materials, principally charred seeds, other plant fragments and small bones, are separated in water from the heavier finds and sedimentary matrix of the excavated deposits. If we were suc cessful, we hoped also to have some ofthe grains directly dated by the new accelera tor mass spectrometric radiocarbon (AMS) method, which allows samples as small as a single seed to be dated. During this first season it became apparent that organic re mains were well preserved in the Jeitun de posits, and then and in subsequent seasons many samples of charred cereal grains and chaff were recovered, 11 of which have now been directly dated. 8 Our investigations, fr om 1989 to our final season at Jeitun in 1994, have shown conclusively that domestic (6-row) barley and domestic (einkorn) wheat were culti vated, and domestic goats and sheep raised, at Jeitun when the site was first occupied at c. 5000 be (= c. 6000 cal BC) . The radio carbon dates suggest that the settlement was occupied for perhaps no more than 500-600 years, not necessarily continu ously; but its size, layout and architectural features, which broadly resemble those of early agricultural neolithic sites in South west Asia, indicate that, in the later phase   (Fig. 1). The results of these new investigations are not yet avail able, but it is possible meanwhile to offer a speculative answer to the question of whether agropastoralism developed inde pendently in western Central Asia or as a result of diffusion from Southwest Asia.
How did agriculture originate in western Central Asia?
We can first approach this question by con sidering the likelihood of any of the do mesticated cereals and herd animals found at )eitun -barley, einkorn wheat, goat and sheep -having been domesticated locally. Their wild ancestors (progenitors) are known, and it can be inferred from their present, and probable former, distributions that neither einkorn wheat nor sheep would have been domesticated locally. Barley, too, is very unlikely to have been domesticated in Central Asia, whereas local domestica tion of the goat cannot be excluded because its wild progenitor, the bezoar (Capra aegagrus}, has a range that extends from Southwest into Central Asia.9 These bio geographical considerations suggest that the agropastoral economy is much more likely to have been introduced from the west than to have developed autonomously in Cen tral Asia, as does the evidence that, from the beginning of neolithic settlement on the piedmont at c. 5000 be, the founder crops and domestic animals of neolithic South west Asian agriculture are found together at J eitun. If we add to this inference the ar chitectural and artefactual similarities be tween the Jeitun Culture sites and neolithic sites in Southwest Asia already referred to, that conclusion is further strengthened. This line of reasoning has led us to search the published literature for evidence ofneo lithic settlement in northeastern Iran, where, it transpires, there are several such sites, e.g. Yarim, Tureng and Sang-i Chakmak (Fig. 1), whose architecture, pottery and other artefacts resemble to varying degrees those of the Jeitun Culture sites.1° Farther west across the northern Iranian plateau there is a dearth of excavated neolithic set tlements for over 400km, until the site of Sialk is reached, the lowest levels of which are said culturally to resemble the Jeitun CultureY The overall conclusion to be drawn from our investigations is therefore that the tran sition to agriculture in Central Asia was the result of the spread of already domesti cated crops and animals from the Fertile Crescent. Whether this process was brought about wholly or mainly by colonizing neolithic farmers (primary diffusion), or whether it was the result more of hunter gatherers adopting the crops, animals and techniques of agropastoralism (secondary diffusion), cannot at present be resolved. It is likely that both processes operated, but, given the inherent tendency of agrarian populations to increase more rapidly than hunter-gatherer groups, and the relatively sudden appearance of the Jeitun Culture exhibiting many of the features of the neolithic economy and settlement pattern found earlier in the Fertile Crescent, the probability is that agriculture began in Cen tral Asia primarily as a result of coloni zation. Although Gordon Childe did not concern himself with the aftermath of the Neolithic Revolution east of the Fertile Crescent, this conclusion mirrors, in a re markably symmetrical way, his model for the westward spread of agriculture into Europe.