Version classiqueVersion mobile

On salt, copper and gold

Catherine Marro
Thomas Stöllner

Early mining and metallurgy within their broader economic context

The Kura-Araxes economy: mobile pastoralism or sedentary agriculture?

Perspectives from Sos Höyük

Catherine Longford et Antonio Sagona


L’économie du complexe culturel kuro-araxe a souvent été décrite comme fondée sur un pastoralisme transhumant. Le pastoralisme nomade était présenté comme un mécanisme expliquant l’expansion kuro-araxe au Proche-Orient. Pourtant, l’importance de la mobilité pastorale dans l’économie kuro-araxe a souvent été remise en question, à la fois par des analyses archéozoologiques et par une nouvelle évaluation de la transportabilité de leur culture matérielle. Cet article présente tout d’abord différents types de stratégies pastorales et agricoles, avec des exemples ethnographiques ; puis il propose un cadre d’analyse permettant de distinguer pastoralisme nomade et agriculture mixte sédentaire. Ces critères sont ensuite appliqués pour identifier, grâce à des données archéobotaniques, l’économie de subsistance de Sos Höyük, site kuro-araxe du nord-est de la Turquie. À Sos Höyük, les cultures semblent avoir été exploitées localement ; quant à la saisonnalité, tant la flore sauvage que la fabrication de combustible à base de bouse suggèrent que le site était occupé toute l’année. Sos Höyük aurait donc été un village sédentaire mixte, agro-pastoral. L’identification de Sos Höyük comme site agricole permanent a des implications sur notre compréhension de l’économie kuro-araxe et de la diffusion de cette culture.

Texte intégral

  • 1 Sagona has since revised his opinion and regards the Kura-Araxes as settled agro-pastoralists (Sag (...)

1The Kura-Araxes culture has been characterised as a cultural complex founded on, or at least containing a significant element of, mobile pastoralism. This interpretation was derived from the seemingly transient nature of their architecture and the portability of their material culture (Burney, Lang 1971; Cribb 1991, pp. 220‑223; Frangipane, Palumbi 2007; Frangipane 2014; Işikli 2012; Kushnareva 1997; Palumbi 2003, 2010; Rothman 2003, 2005; Sagona 1984, 1993; Shimelmitz 2003; Yakar 2006).1 The wattle and daub construction of certain Kura-Araxes house types was thought by some to indicate temporary occupation, while the standard internal arrangement of features and the plan of Kura-Araxes houses have drawn parallels to the internal domain of nomadic tents (Cribb 1991, p. 222; Palumbi 2012; Sagona 1993). Similarly, the portable andirons, a distinctive and ubiquitous artefact type, and ceramics with large lug handles used for easy transportation were also associated with a mobile lifestyle (Kushnareva 1997; Shimelmitz 2003; Smogorzewska 2004). Transhumance within the Kura-Araxes culture has been used to explain the culture’s mobility, as well as the wide distribution of the Kura-Araxes “cultural package” across the Near East. In particular, both Rothman’s and Palumbi’s models for the expansion of the Kura-Araxes horizon stress the importance of transhumant pastoralism. They view it as fundamental to the Kura-Araxes economy and the key factor behind population movements, which saw Kura-Araxes groups and their ideas expand rapidly into new regions (Palumbi 2009, 2010, 2012; Rothman 2003). How the Kura-Araxes economy is interpreted therefore has important ramifications for understanding both the spread of the Kura-Araxes culture across the diaspora and the social and economic structure of their communities.

2Despite this emphasis on mobility, a re-examination of the apparent portability of Kura-Araxes material culture has led some to question the identification of the Kura-Araxes as mobile pastoralists (Kohl 2007, 2009; Paz 2009; Summers 2014; Wilkinson 2014). For this group Kura-Araxes communities practiced a low risk economy and are best described as settled agro-pastoralists (Kohl 2007; Sagona 2011, 2014; Smith 2005; Summers 2004, 2014).

3The connection between the Kura-Araxes and pastoralism is not the only link that has been questioned in terms of mobility. Across the Near East, the antiquity of transhumant pastoralism in the Near East, often used as an ethnographic parallel for Kura-Araxes mobility, has been queried with some suggesting it is a product of Medieval migrations (Potts 2014; Thevenin 2011). Until relatively recently, discussions of the Kura-Araxes economy have all relied on indirect evidence of transhumance and rarely incorporated archaeozoological or archaeobotanical evidence from Kura-Araxes sites. In recent years, this lack of environmental studies has been offset, and new bioarchaeological evidence is informing economic interpretations (Badalyan et al. 2008; Hovsepyan 2010, 2011, 2015; Howell-Meurs 2001a, 2001b; Kakhiani et al. 2013; Longford, Drinnan, Sagona 2009; Messager et al. 2015; Monahan 2007; Piro 2009; Piro, Crabtree 2017; Ristvet, Bakhchaliyev, Ashurov 2011).

4This paper will explore the nature of the Kura-Araxes economy and attempt to determine whether it was based on mobile pastoralism or settled mixed farming. Firstly, different types of pastoral and agricultural strategies known from ethnographic studies will be discussed and parameters developed for defining mobile pastoralism and sedentary agriculture. These criteria will then be used to interpret the archaeobotanical assemblage from Sos Höyük, a Kura-Araxes site in north-eastern Anatolia.

Defining principles: mobility, pastoralism, transhumance and agriculture

5In order to investigate whether the Kura-Araxes complex was based on mobile pastoralism or sedentary farming, it is first important to provide a framework to understand the concepts of pastoralism, agriculture, mobility, and transhumance. Inherent in these concepts are the notions of community mobility versus herd mobility, which are often conflated but in reality quite distinct. Pastoralism is a specialised form of subsistence where management of domesticated animals forms the primary, although rarely sole, basis of the economic and cultural lives of the people that exploit them (Arnold, Greenfield 2004; Cribb 1991). It is important however, not to confuse indications of animal herding as evidence for a mobile lifestyle as ruminant exploitation is also an integral part of sedentary mixed farming (Halstead 1996). Agriculture is the cultivation of crops, which necessitates a high level of residential stability to maintain fields and often occurs in tandem with household herding of animals. Animal herding and crop cultivation are often integrated in a mutually beneficial manner to enhance productivity both as part of the same farming community, or through interaction between mobile pastoral and agricultural groups. These interactions can involve cultivators using their own animals, or those of pastoralists, to plough fields, and letting the beasts graze the crop stubble after harvest which, at the same time, manures the farmers’ fields. In agricultural communities, animals can be fattened with grain as a way of storing surplus for future consumption either in the form of milk or meat (Halstead 1989, 1996). Competing demands for access to land and resources can, however, lead to tensions between pastoral and farming communities that need to be managed.

6The varied forms of pastoralism can be defined in many different ways depending on the parameters applied (Arnold, Greenfield 2004; Makarewicz 2013). Khazanov (1994, pp. 19‑25) defines different levels of pastoral mobility according to the relative importance of pastoral and agricultural subsistence in the economic system, whereas Cribb (1991, pp. 15‑20) uses the level of mobility within the society to categorise different types of transhumance. In both models, these two distinguishing concepts, mobile versus settled and pastoralism versus agriculture, are intricately linked; the stronger the degree of pastoralism the greater the degree of mobility, and conversely, the greater the reliance on crop agriculture the more sedentary the population becomes. While within this spectrum there are many varied and nuanced forms of economic subsistence, some coexisting on a local and regional level, three broad categories can be identified to describe the type of herding and cultivating strategies used: nomadic pastoralism, transhumant pastoralism, and settled mixed farming. These forms of economic subsistence can be applied at different scales in society – household, village, or regional level – and groups or individuals may vary in their mode of subsistence becoming more or less settled as environmental or political pressures significantly alter the availability of resources and access to land (Cribb 1991; Honeychurch, Makarewicz 2016; Salzman 2002; Wendrich, Barnard 2008).

7Although trying to define categories of pastoral and agricultural subsistence may create artificial divisions in economic behaviour, it does highlight the complexity of subsistence and the importance of clearly identifying what is meant when these labels are applied to archaeological sites or populations. In the case of the Kura-Araxes, it is also important to consider what proportion of the community is thought to have been involved in these activities. Kura-Araxes groups have been variously described as transhumant pastoralists, mobile pastoralists, semi-nomadic agro-pastoralists, settled mixed farmers, and as alternating between a transhumant and settled lifestyle. In his survey of archaeobotanical data from Kura-Araxes sites in the southern Caucasus, Hovsepyan (2015) concluded that the Kura-Araxes economy had a “strong emphasis on mobile or semi-mobile pastoral production”, despite finding that the Kura-Araxes economy was focused on the cultivation of cereals. The implications of these labels in regards to Kura-Araxes lifeways, the scale of movement involved or economic structures are rarely explained in the literature. Many of these categorisations have been applied based on the artefactual assemblage of a site in the absence of archaeobotanical or archaeozoological evidence. Indeed Rothman and Kozbe recognised, in 1997, that only with the systematic recovery of plant and animal remains would Kura-Araxes subsistence economies be understood (Rothman, Kozbe 1997). It is only since the early 2000s that routine analysis of plant remains and animal bones has become common at Kura-Araxes sites. The environmental data that plants and bones provide, such as the presence and varieties of crops, evidence for agricultural production, herd composition and the seasonality of site occupation and animal slaughter, form the basis of any interpretation of ancient subsistence economies.

8The next section will define and describe the three broad categories of pastoral and agricultural subsistence using ethnographic and archaeological examples. These examples simply outline the possible economic strategies Kura-Araxes communities may have utilised. While it cannot be assumed that Kura-Araxes communities or other ancient populations would exhibit the same behavioural and economic activities as modern pastoralists and farmers, these ethnographic examples at the very least provide a basis for interpreting and framing the archaeological record.

Models of economic subsistence

Nomadic pastoralism

9Nomadic pastoralism is a specialised form of mobile animal herding where the whole household moves with the stock over an extended area (Arnold, Greenfield 2006; Cribb 1991; Khazanov 1994; Makarewicz 2013). These groups have a high degree of mobility with flexible routes of migration and shifting temporary camps between pasture areas. Pastoral nomadic societies are also characterised by an absence of agricultural production (Khazanov 1994, p. 19), though agricultural resources are often gained through interaction with settled communities (Cribb 1991, p. 16). Indeed no nomadic pastoral group is able to be completely disconnected either directly or indirectly from agricultural products (Goldschmidt 1979). The nomadic Sarakatsani, sheepherders of Greece, for instance, would sell milk, lambs and wool products to buy flour which provided the main calorific component of their diet (Halstead 1996). Alternatively, East African cattle herders like the Maasai, Turkana and Samburu traditionally gain almost all their calorific intake through the milk, blood and meat of their animals and supplement their diets in times of drought with wild plants they collect or grains from agricultural groups (Grillo 2014; Little 1989). Nomadic pastoral groups move through arid and semi-arid steppe lands across the globe which are often regarded as too marginal for agricultural cultivation but can provide plentiful seasonal grassland pasture for herders (Makarewicz 2013; Spooner 1973).

10Ethnoarchaeological studies of the material remains of pastoral nomadic camps indicate that few traces other than stone foundations of pens and animal dung might be preserved after herders moved on, with most of their possessions being perishable or extensively recycled (Biagetti 2015; Cribb 1991). Pastoral nomad sites, by their nature ephemeral, have proven very difficult to identify in the archaeological record (Honeychurch, Makarewicz 2016; Wilkinson 2003). Their location has sometimes been tentatively inferred through the presence of circular anomalies in vegetation caused by variation in soil chemistry (Franchetti 2008). In southern Arabia, potential camps of nomadic pastoralists have been identified at Wadi Sana (5200-4200 cal. BC) where multiple ash and alluvium have laminated hearths in a dated sequence that attest to repeated reuse and abandonment of the site (McCorriston et al. 2012). The presence of burnt domesticated cattle bones associated with these hearths indicates cattle were butchered and feasted upon during these intermittent visits, although the seasonality of slaughter could not be discerned owing to the maturity of the cattle. No domesticated plant remains were found at Wadi Sana; indeed this site predates the introduction of crops into southern Arabia (McCorriston 2013). The timing of water availability at the site and the alluvium laminations suggest seasonal winter occupation. Evidence of alternating periods of occupation and abandonment, the presence of cattle bones and the lack of crop remains support the identification of Wadi Sana as a pastoral nomadic site. For a Kura-Araxes site to be considered a pastoral nomadic camp there would need to be similar evidence of intermittent temporary occupation and a reliance on domesticated animal resources with no archaeobotanical evidence for agricultural production at the site.

Transhumant pastoralism

11Transhumant pastoralism, often referred to as semi-nomadic pastoralism, involves the movement of herds between fixed locations that can be altitudinally separate and/or seasonally complementary pasture zones (Arnold, Greenfield 2004; Chang 1993; Geddes 1983; Halstead 1987). Vertical transhumance often occurs in mountainous regions; herders optimise access to grazing and water resources by migrating between high altitudes for the summer months and lower altitudes over winter. In continental mountain regions of the Near East and Europe, high summer temperatures and the associated lack of adequate water and pasture in marginal semi-arid lowlands encourage the movement of herds to higher altitudes which have cooler summer temperatures and more plentiful grazing lands. Over winter, however, the freezing temperatures and extensive snowfall in the high mountains (800 m altitude and above) require herds to be foddered, or moved to lower altitudes to survive (Chang 1993). In more temperate low and mid-altitude pastures the climatic variation is not as severe and year-round grazing is possible with horizontal transhumance sometimes occurring between different vegetational zones. The distances travelled by transhumant groups can vary greatly from short seasonal movements (ca 20 km) between highland and lowland pastures to migrations of many months over distances in excess of 500 km (Ocak 2016). Pastoralists with herds of sheep and goats, such as the Ozghurs who migrate from the Mediterranean to Central Anatolia, can travel at a rate of 8‑10 km each day on foot through the mountains (Ocak 2016). Large-scale transhumant pastoralists tend to focus on one species (sheep or goat or horse) to maximise productivity and ease competing grazing pressures (Halstead 1996). Ethnographic studies have recorded many different types of transhumant pastoralism. It can involve the whole community moving with their animals between winter and summer pastures (Ansari-Renani et al. 2013; Barth 1961; Thevenin 2011), or partial group transhumance where some members of the community take the herds to summer grazing lands while the rest of the group remain in the lowland winter settlement growing crops and fodder (Cribb 1991, p. 25; Tapper 1977).

12Similarly, the level of agricultural activity varies between transhumant pastoral groups. The Raen cashmere goat herders of Iran, for instance, do not cultivate crops despite their herds gaining 15% of their graze from crop stubble, a diet that is supplemented with barley grain over winter (Ansari-Renani et al. 2013). This is facilitated by negotiating access to fields in autumn with farmers who want animal manure as fertilizer, and buying barley when needed (Ansari-Renani et al. 2013; Ansari-Renani 2016). Other transhumant groups practice some form of crop cultivation which can involve minimal or considerable investment of time and labour. In Central Asia some transhumant herders were historically recorded as practising low input cultivation, sowing millet and barley in river valleys or in lowland fields and leaving them unattended over summer until the crop was ready to be harvested (Spengler, Chang, Tourtellotte 2013; Spengler, Franchetti, Domani 2014). Similarly, Mongolian transhumant pastoralists planted wheat, barley or rye near rivers at their wintering camps before they left for their summer pastures and returned to harvest their crops in the autumn and then stored their grain in shallow pits (Di Cosmo 1994). Ethnographic accounts record a similar pattern for the Qashqaaii and Bamadi transhumant tribes of Southwest Iran, who sowed crops in December at their lowland winter camps and harvested the wheat and barley in April before they left for their summer pastures in the mountains (Alizadeh 2008). If their winter crops were late for harvest, the Qashqaaii and Bamadi would hire local farmers to undertake the harvest, or leave some members of their tribe behind temporarily to harvest and bury the grain in pits for their winter return (Alizadeh 2008). Alternatively, the Basseri in Iran would plant wheat as they arrived at their highland summer camps and harvest it before they departed, or paid local, settled, highland villagers to plant the crop in the spring before they arrived so they could harvested it during the summer (Barth 1961, p. 9). Other transhumant pastoralist groups, like the Pashtun, who began to settle in the 1880s, are able to practice more intensive agriculture by splitting their households and sending a third of each community into the distant mountains with the herds over summer while the majority remain in the lowland winter settlement cultivating cereals and fodder during the summer (Tapper 1977).

13As seen by this survey, when cereal agriculture is practiced by vertical transhumant groups, the winter lowland site is usually used for cultivating crops (Alizadeh 2008; Di Cosmo 1994; Spengler, Franchetti, Domani 2014; Tapper 1977), therefore the products and by-products of crop processing should be present at the winter village, whereas only agricultural products, cleaned grain, or processed foods should be found at the summer occupation site. Alternatively, in rare cases where the summer pasture camp is also the location of cereal cultivation, the Basseri for example (Barth 1961), crop products and by-products should be recovered from the summer camp, and only grain and processed foods should be present at the winter camp. Irrespective of whether crop cultivation is a low or high input activity, the by-products of cereal processing (chaff and weed seeds) and the tools used for ploughing, harvest, and threshing are only likely to be found at the site of agricultural production.

14Some modern transhumant pastoral summer camps appear similar to nomad camps in that they contain few if any permanent features except occasional low stone wall foundations for tents and animal corrals (Cribb 1991; Thevenin 2011). Alternatively, Thevenin (2011) describes two Kurdish transhumant tribes who used permanent culverts to pool or to channel water from springs to campsites and pits that were used to store food at locations visited by the same group each summer. In the Zagros mountains a different pattern is seen, namely the reuse of permanent dwellings by transhumant herders for their winter and summer quarters (Alizadeh 2008). Based on these ethnographic examples, it would be expected that shifting transhumant summer camps would leave few archaeological traces, whereas campsites that were repeatedly reused each summer would be better preserved with yearly accumulations of midden material and the gradual development of more substantial landscape modifications, including buildings, over years of frequent site use. All this reminds us of one important point – that the apparent temporary or permanent nature of architecture by itself therefore is not a reliable indicator of mobility or sedentism. Rather it is the seasonality of site occupation, with distinct periods of repeated use and total or partial abandonment, which enables the identification of transhumant pastoralist sites.

15The presence of animals at a site can be determined by identifying the age range of youngest and the most accurately aged deaths (Halstead 1996; Jones, Sadler 2012; Legge, Williams, Williams 1991; Payne 1973). Thus summer camps would only have evidence of activities during the summer, including herd animals killed over the summer months. Lowland winter camps, often villages with permanent buildings (Cribb 1991, p. 108), could have similar architecture to the settlements of sedentary mixed farmers. Based on the season of occupation however, unlike mixed farming villages, it would be expected that the winter villages would only have evidence for the presence of herd animals over winter; that is no herd animals with an age at death between the spring departure and autumn return. Moreover, unless a portion of the community remains at the winter village throughout the year, as in the case of the Pashtun (Tapper 1977), there should be no evidence for occupation of the winter site over the summer migration period. The Kura-Araxes have often been described as transhumant pastoralists (Burney, Lang 1971; Frangipane, Palumbi 2007; Palumbi 2010; Rothman 2003), but the type of transhumance (vertical or horizontal), distance of migration, and scale of community involvement, is rarely defined. As we have seen, the identification of transhumant pastoralists, would require evidence of distinct seasonal pattern – occupation in either the summer or the winter or, in the case of partial group transhumance, the presence of herd animals only during the winter months.

Settled mixed farming

16Settled mixed farming is the most common form of agro-pastoral production practiced in the Near East and Caucasus today. A reliance on crop agriculture promotes a high level of sedentism within a community. Even within communities practicing shifting cultivation, the village location remains constant while the swidden fields are moved around the landscape over a multiple year cycle (Dutrieux et al. 2016; Höhn, Neumann 2012; Mertz et al. 2013). Agricultural production can be intensive or extensive and continue throughout the year with multiple crops grown depending on the seasonality of each crop species and the local climatic conditions (Bogaard 2004a; Jones 2005).

17High levels of agricultural production does not inhibit herding of livestock, instead crop cultivation and animal pastoralism can be integrated at a household or group level within sedentary mixed farming communities. Households in mixed farming villages maintain a variety of animals for meat, milk and wool production, ploughing, transportation, threshing and milling grain rather than specialise in one species, like transhumant pastoralists, for the production of one valuable exchange commodity (milk, wool, meat or transport) (Halstead 1996). The amount of herd animals kept per household in agricultural communities is less than for transhumant pastoral households, with transhumant and nomadic flocks on average being five times larger than village herds in the same region (Cribb 1991, p. 35). Herds can be managed by individual households or communally with designated shepherds (Halstead 1996). Village flocks are moved around the landscape between pasture zones, their distance from the settlement dictated by the availability of pasture and the distribution of cultivated fields in a form of village tethered pastoralism. Animals can either be returned to the village each day to be milked and stabled, or kept overnight in stone corrals at a short distance from the settlement where villagers travel to milk them each day (Chang 1993; Hopkins 2003; Thevenin 2011). Those with large flocks need to increase the pasture zone of their animals, often requiring further movement away from the cultivated fields particularly in the period of crop growth prior to harvest (Cribb 1991). Over winter, in areas of prolonged snow cover, village herds are stabled indoors and provided with fodder and grain supplements which have been cultivated and/or collected over the summer. Sedentary mixed farming settlements are ubiquitous across the Near East and can be identified by evidence of year-round occupation, crop cultivation and the presence of herd animal throughout all seasons. Yet only a small number of Kura-Araxes sites have been interpreted as settled agro-pastoral villages based on the seasonality indicators of archaeozoological remains (Crabtree 2011; Howell-Meurs 2001b; Piro 2009).

18This brief summary using ethnographic examples, while not exhaustive, highlights the key features of nomadic pastoralism, transhumant pastoralism and settled mixed farming and may provide a basis for distinguishing different economic strategies employed by archaeological cultures. These traits will be employed to investigate the subsistence economy of Sos Höyük using archeobotanical evidence.

Archaeobotanical case study: Sos Höyük

Archaeology of Sos Höyük

19Sos Höyük is located in the village of Yiğittaşi on the southern bank of the Çoğender stream in the Pasinler Valley (fig. 1). The site is at an altitude of 1,771 meters above sea level and is 24 km east of the town of Erzurum in eastern Turkey. The Pasinler Valley is a broad plain flanked by the Karapazarɪ Mountains to the north, the Palandöken Mountains in the south and the Deve Boyun ridge in the west. The mound of Sos Höyük rises 12 m above the plain and is 1.2 hectares in size but extends further beneath the village of Yiğittaşi to the west, south and east (Sagona, Sagona, Özkorucuklu 1995). Excavations at the site were conducted in 1987 by Atatürk University and from 1994‑2000 by the University of Melbourne (Sagona, Sagona, Özkorucuklu 1995; Sagona et al. 1996; Sagona et al. 1997; Sagona et al. 1998; Sagona, Sagona 2000, 2004).

Fig. 1 – The Pasinler Valley, Erzurum Province, Turkey, indicating the location of Sos Höyük and other places mentioned in the text; elevation intervals are every 100 m (C. Longford).

Fig. 1 – The Pasinler Valley, Erzurum Province, Turkey, indicating the location of Sos Höyük and other places mentioned in the text; elevation intervals are every 100 m (C. Longford).

20The Kura-Araxes sequence at the site, based on a many radiocarbon dates from secure contexts, spans from the mid 4th to the early centuries of 2nd millennium BC beginning with Sos VA at the beginning of the Early Bronze Age and concluding in Sos VIB at the end of the Middle Bronze Age (tab. 1) (Sagona 2000, 2004, 2014). That the Kura-Araxes continued in the 2nd millennium should come as no surprise if we do not adhere to the implicit evolutionary assumption, common in Kura-Araxian studies, that cultural developments occurred in a linear fashion more or less simultaneously over its vast territory. In reality, adjustments in complex social systems do not occur at the same rate everywhere (Van der Leeuw, McGlade 1997), and hence the Pasinler Plain does not mirror the southern Caucasus in this instance. Throughout this lengthy period, the material culture changed, and among the most notable in the late phase is the appearance of multi-roomed structures and red-and-black ceramics with distinctive white grit inclusions (late gritty ware). Although extensive horizontal exposure was not possible at Sos Höyük owing to the encroachment of the modern village, it is clear that during the Kura-Araxes occupation building styles alternated between rectilinear, circular or sub-rectangular houses with stone, mudbrick or wattle and daub walls of varying combinations in shape and construction. These buildings frequently contained either fixed or portable central circular hearths. A substantial curved stone wall was built early on around the centre of the mound in Sos VA and appeared to demarcate a discrete zone within the settlement. What it lacks in complete house plans, Sos Höyük makes up in a rich assemblage of artefacts that is consistent with the Kura-Araxes cultural package (Kiguradze, Sagona 2003; Sagona, Zirmansky 2009; Sagona 2014).

Tab. 1 – Kura-Araxes chronology of Sos Höyük adapted from Sagona 2000 and Sagona 2014.

Middle Bronze Age SOS IVB
2000-1500 BC
2200-2000 BC
Early Bronze Age SOS VD
2500-2200 BC
2800-2500 BC
3000-2800 BC
3500/3300‑3000 BC

Environment and modern economy in the Pasinler Valley

21Eastern Anatolia is characterised by an extremely harsh continental climate with long cold winters and dry hot summers. In the Erzurum region, including the Pasinler Valley, frosts begin in October and heavy snow covers the ground from November to March or April (Yakar 2000). The average temperature from December to March ranges from –1 to –16oC and may drop as low as –37oC (Bulut et al. 2010). In summer the average temperature ranges from 15oC in June to 19oC in August but can reach as high as 34oC on hot days, although sudden changes can drop the temperature to less than 5oC and snow often remains on the high peaks surrounding the Pasinler Valley throughout the year (Bulut et al. 2010; Hopkins 2003). Annually, the Pasinler Valley receives approximately 430 mm of rain which mostly falls towards the end of autumn and at the beginning of spring with the dry season lasting from late June to September (Hakgören 1972). Snow melt supplements the spring rains. Today, the Pasinler Valley, which lies at the border of the Euro-Siberian and Irano-Turanian floristic regions (Davis, Cullen, Coode 1965‑1988; Zohary 1973), is a broad flat treeless plain with rocky hillocks rising above the cultivated fields, dry stream beds and occasional marshy ground.

22The village of Yiğittaşi, modern Sos Höyük, is inhabited by about 300 people from 40 different households (Hopkins 2003). The villagers practice settled mixed farming. Wheat, barley, potatoes, lentils, sunflowers, sugar beet, capsicums and garden vegetables are grown by the villagers (Hopkins 2003). A local landrace of bread wheat (Triticum aestivum), known as Kirik, is cultivated in Yiğittaşi. Although Kirik can be sown in either spring or autumn, it produces its highest yield when sown in early autumn, and has a high cold tolerance (Bardsley 2001; Öztürk, Çağlar, Bulut 2006). Cereal crops sown in spring need to be planted near to the village and require irrigation since their main growth occurs after the spring rains and snow melt (Çağlar et al. 2011; Hopkins 2003). The fields are left fallow when possible to restore moisture and nutrients to the soil, but increasing agricultural and commercial demands often mean that fallowing cannot be practiced and crops are instead rotated to preserve soil productivity (Hopkins 2003).

23Animal husbandry is the main economic concern of the village. Sheep, goats and cattle are the dominant herd animals in Yiğittaşi (Hopkins 2003). Some animals are sold for meat and the sheep are shorn for their wool, but the animals are primarily kept for their milk production, which is either sold as milk or turned into cheese as a year-round protein source (Hopkins 2004). During the summer, the animals are taken out to pasture each morning and returned after a day’s grazing to be milked and stabled in the households’ barns. Some sheep and goats are kept overnight during summer in a circular stone sheepfold one kilometre from the village. Over winter, animals are kept in each households’ barns and fed fodder which was harvested over the summer. Two forage crops are utilised in Yiğittaşi. Wild grasses from fallow fields and marginal land are harvested for fodder and the legumes Trifolium and Medicago are specifically grown as fodder crops but need to be fully ripened and dried before being fed to animals (Hopkins 2003). Livestock dung is collected, dried and used as fuel for heating and cooking in the village (Hopkins 2004). The villagers of Yiğittaşi do not practise transhumance and are able to graze their livestock locally within the Pasinler Valley and on the slopes of the Karapazarɪ Mountains. This summary of agricultural practices in Yiğittaşi is not to suggest that the Kura-Araxes inhabitants of Sos Höyük behaved identically to the modern villagers of Yiğittaşi but to give an insight into how a community could operate in the environment and climate of the Pasinler Valley.

Sos Höyük compared with the nomadic pastoralist, transhumant pastoralist and settled mixed farming models

24When Sos Höyük was first excavated it was interpreted as a site intermittently occupied by Kura-Araxes transhumant pastoralists (Sagona et al. 1996). The intermittent use of wattle and daub architecture and the concentration of pits were seen as signs of temporary occupation of the site. Using the ethnographic criteria outlined above for recognising different types of pastoral and agricultural systems we can define parameters for locating Sos Höyük within a mobile/sedentary economic framework. If Sos Höyük was a temporary camp of nomadic pastoralists, based on the ethnographic examples, there would be few architectural traces and no evidence of crop cultivation at the site. Yet the architectural remains present throughout phases VA to IVB appear inconsistent with the traces left by the campsites of nomadic pastoralists (Biagetti 2015; Cribb 1991) and this can be further tested by assessing the evidence for crop cultivation.

25Owing to its high altitude location at 1,771 m asl, it is mostly likely that, if Sos Höyük was part of a vertical transhumant seasonal cycle, it would have been a highland summer camp (Cribb 1991; Piro 2009; Yakar 2006). Modern transhumant pastoralists migrate from as far south as Şanlɪurfa, Elazɪğ and Bitlis through the Palandöken Mountains to graze their livestock on the foothills of the Karapazarɪ Mountains near Yiğittaşɪ over the summer (Hopkins 2003; Longford, Sagona, pers. obs., 2002; Thevenin 2011). Conversely, Sos Höyük could have been a lowland winter village for transhumant pastoralists who migrated during the summer to the high alpine steppe on the Karapazarɪ Mountains flanking the Pasinler Valley. This possibility seems less probable, as a separate summer camp for grazing livestock on Karapazarɪ Mountains appears unnecessary since the heights of the Karapazarɪ Mountains are within five kilometres of the site and can be reached within half a day’s walk (Longford, Drinnan, Sagona 2009; Newton 2004). If Sos Höyük was a seasonal transhumant camp, the evidence for occupation and any indicators of crop cultivation, particularly the timings of crop sowing and harvest, should correspond with the period of site use which, in this case, was most probably over summer.

26Archaeozoological analyses, carried out by Howell-Meurs (2001b, 2001a) and Piro (2009), of sheep/goat mortality profiles have instead indicated that Sos Höyük was probably permanently inhabited throughout the year and that the economy was most likely based on settled mixed farming rather than specialised transhumant pastoralism. Based on the sheep/goat evidence, if lambing happened in the spring then the death of juveniles at 5‑6 and 9‑12 months shows that animals where kept at the site in autumn, late winter and early spring; foetal and neonatal bones indicate pregnant ewes and newborn lambs over the spring and summer (Piro 2009; Piro, Crabtree 2017). If Sos Höyük was a settled mixed farming village, archaeobotanical data should provide evidence for year-round occupation and crop cultivation at the site. Analysis has recently been completed on the archaeobotanical assemblage of Sos Höyük (Longford 2015) and the results of this study will be utilised to apply an archaeobotanical approach to investigate Kura-Araxes subsistence.

Investigating the subsistence economy at Sos Höyük through archaeobotanical evidence

27During excavations at Sos Höyük soil samples for archaeobotanical analysis were collected from recognised floor levels rich in organic material, with hearths being preferentially sampled. The soil was floated using a modified Ankara flotation machine (French 1971; Nesbitt 1995), the floated material was collected in cloth chiffons and air dried in cloth bags. Plant remains were studied at the University of Sheffield using a Leica MZ6 stereomicroscope with a magnification of up to x63. Identifications were made with reference to seed atlases and illustrative archaeobotanical texts (Anderberg 1994; Bejerinik 1947; Berggren 1969, 1981; Cappers, Beeker, Jans 2006; Hillman 2001; Jacomet 2006; Neef, Cappers, Bekker 2012; Van Zeist, Bakker-Heeres 1985, 1986, 1988) and comparison with comparative modern material housed in the Department of Archaeology at the University of Sheffield. Cereal nomenclature follows the traditional binomial system for Triticum and Hordeum taxa as outlined in Zohary, Hopf, Weiss (2012, p. 29, 57) and wild taxa nomenclature follows the Flora of Turkey by Davis, Cullen, Coode (1965‑1988). Samples with at least 100 or more cereal items or those from contexts of potential archaeobotanical interest, such as hearths, were selected for analysis. In total, 70 samples from the Kura-Araxes period at Sos Höyük were designated for full analysis on the basis of sample richness and context. Of these 70 samples, 42 were from the Early Bronze Age and 28 were from the Middle Bronze Age. A more detailed methodology, sample by sample assemblage data and discussion of results are available in Longford (2015).

28Throughout the sequence at Sos Höyük the most common cereal crops were bread wheat (Triticum aestivum L.), and two-row barley (Hordeum distichum L.). These two crops are free threshing cereals and appear in approximately equal proportions in all phases apart from Sos VA where bread wheat is dominant and Sos VB when two-row barley is the more common (fig. 2). Although no food storage contexts were found, bread wheat and two-row barley commonly occur throughout the Kura-Araxes period in considerable quantities which suggests they were the primary arable products used at Sos Höyük. Emmer wheat (T. dicoccum Schübl) is present in trace quantities in the Early Bronze Age and increases in proportion later in the Middle Bronze Age although it remains a small component of the crop assemblage. Considerable amounts of wheat and barley were recovered; a total of 4,375 cereal grains and 6,626 items of cereal chaff were identified in the Kura-Araxes samples. Very few cultivated pulses, mainly lentil (Lens culinaris Medlik.) and pea (Pisum sativum L.), were found in the assemblage and, due to their rarity, it is uncertain whether they were deliberately cultivated at Sos Höyük. The scarcity of pulses at Sos Höyük is consistent with Hovsepyan’s (2014, 2015) observations that Kura-Araxes agriculture is focused primarily on cereals and lacks pulse crops. The only clear non-cereal crop species at Sos Höyük is the oil plant, Gold of Pleasure (Camelina sativa [L.] Crantz.), which was found in high concentration in a rather pure Middle Bronze Age sample. All archaeobotanical samples contained wild plant seeds, straw culm nodes, wood charcoal and many also contained fragments of amorphous animal dung. In total, 18,403 wild items were recorded for the Sos Höyük assemblage, representing seeds from 26 different plant families. Taxa present in more than 10% of samples are listed in tab. 2, this also indicates in which phase each taxon was present together with ecological information relating to each taxon.

Fig. 2 – Graph of crop item percentages for each period at Sos Höyük.

Fig. 2 – Graph of crop item percentages for each period at Sos Höyük.

Tab. 2 – Most common seeds of wild/weedy taxa in the Sos Höyük assemblage indicating presence by phase (ecological information from Davis, Cullen, Coode 1965‑1988).

  Va Vb Vb/Vc Vc Vd IVa IVb      
Number of samples 17 8 6 3 8 8 20      
Taxa               Flowering Germination Life habit
Polycnemum arvense x x x     x x June-July Spring Annual
Atriplex cf. lasiantha x x x   x x x May-July Spring Annual
Chenopodium cf. album x x x x x x x May-August Spring Annual
Kolchia prostata / scoparia           x x June-August Annual/ Perennial
Buglossoides arvense x x x x x x x Feb.-June Autumn/ Spring Annual
Lithospermum officinale   x   x x x x May-July Perennial
Cardaria / Lepidium type x x x x x x x April-May Autumn/ Spring Annual/ Perennial
Thlaspi arvense x       x x x Feb.-May Autumn Annual
Euclidium syriacum x x   x x x x April-June Spring Annual
Silene spp.   x   x x x x May-August Autumn/ Spring Perennial
Vaccaria pyramidata x   x x x x x April-July Spring Annual
Convolvulvus sp. x     x x x x April-Sept. Perennial
Bolboschenus maritimus x x x x x x x May-Sept. Perennial
Carex spp. x   x   x   x   Perennial
Eleocharis sp. x x       x x April-Sept. Perennial
Cyperaceae x x x x x x x    
Trifolium / Melilotus / Medicago x x x x x x x April-August Annual/ Perennial
Trigonella sp. x x x     x x April-July Annual
Vicia / Lathyrus sp. x x x x x x x    
Lallemantia iberica / canescens / peltata x x x x x x x June-August Perennial
Teucrium sp. x x     x x x May-Sept. Perennial
Malva sp. x x x   x x x May-October Annual/ Perennial/ Biennial
Bromus cf. japonicus x x x x x x x May-July Spring Annual
Lolium cf. perenne / multiflorum x x x x x x x April-Sept. Annual/ Perennial/ Biennial
Fallopia convolvulvus x x x   x   x May-Sept. Perennial
Persicaria sp. x x   x x x x June-August Spring Annual
Polygonum arenastrum / bellardii x x x x x x x May-Nov. Spring Annual
Rumex sp. x x   x x x x June-August Perennial
Adonis sp. x x x   x x x March-June Annual/ Perennial
Rosa spp. x   x x x   x June-July Perennial
Asperula cf. involucrata x x x   x x x May-Sept. Perennial
cf. Crucianella sp. x x     x   x May-August Spring Annual
Galium spp. x x x x x x x May-August Annual/ Perennial
Hyoscyamus niger   x     x x x April-August Spring Annual/ Biennial
Thymelaea sp. x x x   x x   April-August Annual

29Archaeobotanical data can be used to investigate the type of economic strategy at a site, in this case whether Sos Höyük was a nomadic pastoral, transhumant pastoral or settled mixed farming site. There are three key avenues of investigation of the archaeobotanical assemblage that are useful for assessing the mobile versus settled nature of the site: the evidence for cultivation, crop seasonality and dung use and content. Firstly, the presence or absence of crop processing products and by-products can demonstrate whether a site was a producer or consumer of agricultural products, that is if crop cultivation occurred locally or if grains were imported (Hillman 1984; Jones 1984, 1987, 1990; Van der Veen, Jones 2006). Seasonality of site use can be determined from the period of wild seed dispersal and, if crop cultivation occurred locally, the timing of crop sowing and harvest as shown by modern local crop cycles and the germination and seed shed of wild taxa in the assemblage (Behre, Jacomet 1991; Bogaard 2004b; Bogaard et al. 1999, 2001; Bogaard, Jones, Charles 2005; Charles, Jones, Hodgson 1997; Jones 1992; Jones 1981; Wasylikowa 1981). Finally, animal dung can be used to show livestock diet, seasonality of livestock presence, and type of fuels used at a site (Akeret, Jacomet 1997; Charles 1998; Charles, Bogaard 2005; Miller 1984; Miller, Smart 1984; Riehl 2006; Shahack-Gross 2011; Valamoti 2007).

Crop cultivation

30A scarcity of cultivated plant remains at a site has often been used as a key indicator for identifying nomadic pastoral sites (Franchetti 2012; Motuzaite-Matuzeviciute, Telizhenko, Jones 2012; Rühl, Herbig, Stobbe 2015; Shishlina, Gak, Borisov 2008; Stevens, Fuller 2012). Even so, a lack of crop remains at a site is not necessarily evidence for an absence of agricultural activity in a society (Jones, Rowley-Conwy 2007; Bishop 2015). Issues of taphonomy, changes in crop-processing strategies, in waste disposal and in settlement structure could lead to differential preservation of crop remains at a site. Nomadic pastoral sites, if crop remains are present, should only contain cleaned and processed grains that have been imported since pastoral nomadism does not involve the cultivation of crops (Cribb 1991; Franchetti et al. 2010; Khazanov 1994; Spengler, Franchetti, Domani 2014). Crop processing, separating the grain from the chaff and weeds, has been studied ethnographically by Hillman (1984) and G. Jones (1984, 1987, 1990) to identify the products and by-products of each crop-processing stage (fig. 3). The main crops at Sos Höyük, bread wheat and two-row barley, are free threshing cereals, that is when these crops are threshed after harvest their grains are easily removed from the chaff. Winnowing separates the grain from the lighter chaff and straw fragments and sieving with coarse and fine meshes in the final stages separates the weeds from the cleaned grain. Glume wheats (Emmer, Einkorn and Spelt) require an extra stage of pounding to release the grain from the spikelets. Sites of agricultural production are characterised by a high proportion of early stage crop-processing by-products (Hillman 1981). This is especially true of free threshing cereals because they are usually processed to remove chaff before exchange or transportation to reduce the bulk weight of grain (Jones 1981).

Fig. 3 – The products and by-products of crop processing for glume wheats and free threshing cereal (free threshing wheat and barley); note that hulled barley is a free threshing cereal and is processed the same way as bread/durum wheat (for barley the “hulls” are the palea and lemna rather than the glumes) (based on Charles 1984; Van der Veen, Jones 2006).

Fig. 3 – The products and by-products of crop processing for glume wheats and free threshing cereal (free threshing wheat and barley); note that hulled barley is a free threshing cereal and is processed the same way as bread/durum wheat (for barley the “hulls” are the palea and lemna rather than the glumes) (based on Charles 1984; Van der Veen, Jones 2006).

31Crop-processing stages can be determined by comparing the physical characteristics of weed seeds found in ethnographic samples of crop-processing products and by-products with weed seeds in archaeological samples (Jones 1984, 1987). The weed characteristics used are seed size (relevant to fine sieving), tendency to remain in heads or spikes, etc. despite threshing (relevant to coarse sieving) and aerodynamic properties including density and shape (relevant to winnowing) (Jones 1984, 1987). This approach uses discriminant analysis based on a combination of these weed seed characteristics to discriminate between the four known crop-processing groups from Amorgos, Greece. With this analysis (fig. 4), all the Sos Höyük samples were classified as fine sieving by-products. Another method to assess crop-processing stages is to compare archaeobotanical samples with the known proportions of cereal grain, chaff and wild seeds in ethnographic samples of crop-processing products and by-products. Sos Höyük archaeobotanical samples are compared to ethnographic samples of free threshing cereal processing from Amorgos (Jones 1990), in fig. 5. Samples with cleaned grain, the final product of crop processing, would plot at the top of the triangle. The Sos Höyük samples instead have a mix of cereal chaff and wild seeds showing a mixing of winnowing and sieving by-products. The abundance of crop-processing by-products, particularly bread wheat and two-row barley rachis internodes (fig. 2), in all phases indicates local cultivation of cereals throughout the Kura-Araxes period at Sos Höyük. Sos Höyük was therefore not a nomadic pastoral camp.

Fig. 4 – Discriminant analysis plot comparing the Sos Höyük samples to ethnographic crop-processing groups from Amorgos, Greece, using physical characteristics of weeds seeds (Jones 1984).

Fig. 4 – Discriminant analysis plot comparing the Sos Höyük samples to ethnographic crop-processing groups from Amorgos, Greece, using physical characteristics of weeds seeds (Jones 1984).

Fig. 5a – Sample composition of ethnographic crop processing samples from Amorgos, Greece (Jones 1990) for comparison with fig. 5b.

Fig. 5a – Sample composition of ethnographic crop processing samples from Amorgos, Greece (Jones 1990) for comparison with fig. 5b.

Fig. 5b – Sample composition of Sos Höyük based on the proportions of grain, rachis internodes and weed seeds.

Fig. 5b – Sample composition of Sos Höyük based on the proportions of grain, rachis internodes and weed seeds.

Crop seasonality

32Evidence of crop cultivation at Sos Höyük indicates that the site could have been either a transhumant pastoral camp or a settled mixed farming village throughout the Kura-Araxes period. Vertical transhumant pastoralism has a distinct seasonal pattern where camps are either occupied in the summer or winter, and were Sos Höyük a transhumant pastoral camp it would most probably have been a summer grazing camp (Cribb 1991; Piro 2009; Yakar 2006). Crop seasonality can help determine the period of site use. Harvesting time can potentially be identified from the flowering and fruiting times of the weed species accompanying crops (Wasylikowa 1981; Jones 1981; Behre, Jacomet 1991; Jones 1992; Charles, Jones, Hodgson 1997; Bogaard et al. 1999, 2001; Bogaard, Jones, Charles 2005). In Yiğittaşi, irrigated fields close to the village are used for the cultivation of both cereals and cash crops, and Kirik wheat is generally sown in the spring (late April/early May) and harvested in the autumn (October) (Hopkins 2003). Fields further from the village are sown with Kirik in the autumn (October/November) and harvested in the summer (July/August) (Bardsley 2001). So, if seeds from wild taxa that fruit during the summer harvest are missing from the assemblage, it would indicate that crops were not harvested in the summer and, similarly, if no wild taxa that fruit in the autumn are present, autumn harvest would be unlikely. All of the wild taxa in the Sos Höyük samples can flower and fruit early (between April and June) and there are no taxa that flower only in the late summer to autumn (between August and September) (tab. 2), most samples contain taxa with long flowering periods that set seed from summer to autumn. It is therefore not possible to exclude either of the crop cultivation cycles on the basis of the flowering and fruiting times of wild taxa: seeds from some of the wild taxa could have been collected with an autumn harvest of crops sown in spring, and all could have been collected with summer harvested crops sown in the previous autumn. Crop-sowing time can also be inferred from the germination period of weed species (Charles, Jones, Hodgson 1997; Bogaard et al. 2001; Bogaard 2004b). But based on the taxa for which germination time is available (tab. 2), both autumn and spring germinating taxa are present amongst the Sos Höyük taxa, so neither autumn-sown nor spring-sown crop cultivation cycles can be excluded.

33Nevertheless, some idea of seasonality can be gained from the modern crop cycles. While it cannot be assumed that the cereal taxa grown at Sos Höyük from the Late Chalcolithic to Middle Bronze Age would have had exactly the same growth patterns as the modern Kirik wheat, these crop cycles provide a guide for the possible sowing and harvesting times in the Pasinler Valley, constrained as they are by the long heavy snow season. Both cycles necessitate occupation of Sos Höyük during autumn, since autumn would have been the season for sowing or harvesting of crops, and either summer harvesting or spring sowing of crops depending on the cycle. From November to March, heavy snow fall in the Erzurum highlands, including the Pasinler region, makes the mountain passes impenetrable and some villages inaccessible (Yakar 2000, pp. 392‑398). If Sos Höyük were a summer camp, pastoralists would have left the Pasinler Valley before the late autumn snow falls blocked the high mountain passes to the south or north, which conflicts with both of the currently practiced cultivation cycles, that is with both autumn sowing (and summer harvest) or autumn harvest (with spring sowing). Their return would have not been possible until after the snow melt in early April, which would have delayed spring sowing until after the spring rains and required irrigation of the crops. In cold years snow can remain on the high mountain passes for many months; Pollington (1840) recorded snow still covering the high roads to the south of Pasinler as late as June. Today, pastoralists bringing their flocks to the Pasinler Valley from south of the Palandöken Mountains can arrive as late as June for summer grazing (Longford, Sagona, pers. obs., 2002). The scheduling of activities related to the cultivation of crops conflicts with the available window for transhumance and argues strongly against Sos Höyük being a summer grazing camp. Similarly, the extended flowering and fruiting time of wild taxa found throughout the archaeobotanical sequence means that Sos Höyük must have been occupied from spring until autumn and therefore could not have been a winter transhumant camp. The seasonality of crop cultivation and wild seed shed, then, suggests that site was continuously occupied throughout the year and therefore that Sos Höyük was a settled mixed farming village.

34Unlike the modern Basseri in Iran who sometimes employ local non-Basseri farmers to sow crops in the spring prior to their arrival at their summer camps (Barth 1961), there is no indication for any contemporary non-Kura-Araxes settlements in the Pasinler Valley (Sagona 1999); therefore any tending of the Sos Höyük crops would have required the assistance of neighbouring settled Kura-Araxes communities. This would have necessitated the Kura-Araxes having distinct pastoral transhumant and settled agricultural communities as separate elements of the Kura-Araxes oikumenè (Rothman 2005). Based on ethnographic work with many transhumant groups in the Caucasus and Central Asia, Khazanov (2009, p. 125) noted that the material culture of seasonal shepherds while they were away with their flocks was very different to the artefacts and architecture found in their permanent home settlements. Furthermore, Khazanov felt that the material culture of seasonal camps and permanent settlements would be so different that it would be difficult to know that they were two elements of the same culture. If there was a Kura-Araxes mobile/settled dichotomy then, according to Khazanov’s ethnographic work, there should be a marked heterogeneity between Kura-Araxes groups whose lifestyles diverged between transhumant pastoralism and settled agriculture. The Kura-Araxes horizon across the Near East is, however, distinguished by its shared cultural traditions which accommodate variations derived from regional styles and household level production, but do not exhibit large differences, which could be the product of a mobile and sedentary economic duality.

Animal dung

35Recognisable animal dung, either fragmented or in pellets, is found in the majority of Sos Höyük samples throughout the Kura-Araxes period. Amorphous dung fragments are found in 59% of samples throughout the assemblage (Longford 2015). Some of the amorphous dung fragments from Sos Höyük contain charred seeds of wild taxa and dung fragments were also observed adhering to wild seeds and chaff, particularly two-row barley rachis internodes, throughout the assemblage. These amorphous dung pieces may represent the interiors of sheep/goat faecal pellets, fragments of burnt dung fuel cakes or burnt stable waste. Animal dung can be used to study the diet and grazing patterns of livestock which may give insight into landscape use and transhumant cycles (Akeret, Jacomet 1997; Charles, Bogaard 2005; Riehl 2006; Miller 1984; Valamoti 2007). Unfortunately no seeds were visible in the sheep/goat faecal pellets from Sos Höyük so direct evidence for livestock diet is unavailable. Plant remains embedded in amorphous dung fragments or with dung adhering to their surface may indicate the mixing of dung and crop-processing by-products and so cannot be used to infer animal diet. Based on the mixed composition of the Sos Höyük samples with their combination of dung fragments, mixed crop-processing by-products, free threshing cereal rachis internodes, culm nodes and bases, and grain, the dung in samples is likely to represent the remnants of burnt dung fuel cakes (Anderson, Ertug-Yaras 1998).

36The use of animal dung as fuel has implications for determining seasonality of site occupation. For animal dung to be an abundant and viable fuel resource at Sos Höyük, livestock must have been kept either on site or in close proximity to the site for at least part of each year. The seasonality of this occupation may be inferred from the scheduling of dung cake making activities. Manufacture of dung cake fuel in central and eastern Anatolia commonly occurs over spring and summer, when winter stable waste is processed and dried over a three to four month period before being cut into cakes and stored (Hopkins 2003; Anderson, Ertug-Yaras 1998). Today in the village of Yiğittaşi, animal dung is accumulated primarily during winter stabling of animals, the floors of barns being designed to collect animal droppings by channelling them into a hollow in the floor for later retrieval (Hopkins 2003). From the seasonal cycle of dung fuel collection and manufacture, the use of dung fuel may indicate occupation of Sos Höyük from at least November, when the snow begins and livestock are stabled indoors to produce an accumulation of dung, through to August for the drying and forming of dung cakes. This suggests that the site was utilised during winter, spring and summer which is inconsistent with the seasonal cycle of transhumant pastoralists, supporting again the view that throughout the Kura-Araxes period Sos Höyük was a settled mixed farming village.

Kura-Araxes: mobile pastoralists or settled agriculturalists?

37Archaeobotanical analysis of the Sos Höyük assemblage indicates that the Kura-Araxes inhabitants of the site were probably settled agriculturalists who practiced mixed farming with localised household herding of animals. The seasonality of crop cultivation and dung fuel production were found to be incompatible with the migration patterns of transhumant pastoralists indicating the occupation of Sos Höyük in all seasons. These interpretations are consistent with the results of archaeozoological analyses of animal bones from the site (Howell-Meurs 2001a, 2001b; Piro 2009; Piro, Crabtree 2017) which also determined that the site was settled year round and that the Kura-Araxes economy was based on sedentary mixed farming.

38Furthermore, there is archaeobotanical evidence from other Kura-Araxes sites in eastern Anatolia and the southern Caucasus that the Kura-Araxes economy has a strong agricultural component. Similar to Sos Höyük, crop remains including free threshing wheat and/or barley rachis internodes have been recovered from several Kura-Araxes sites, including Chobareti, Dilkaya, Tsaghkasar, Aşvan Kale, Taşkun Mevkii and Tepecik which suggest that there was local cultivation and processing of grain at these sites (Hovsepyan 2011; Messager et al. 2015; Nesbitt et al. 2017; Nesbitt 1991; Van Zeist, Bakker-Heeres 1975). Further analysis of the weed flora associated with these sites will help to identify the seasonality of site occupation. Moreover, most Kura-Araxes sites are located in agriculturally productive land, often fertile intermontane valleys that are highly suited to cereal cultivation (Batiuk 2005). At Kura-Araxes sites in the Caucasus, artefacts provide extensive evidence of agricultural activities: clay models of plough boards, sickle blades made from obsidian and flint and quernstones are found at many Kura-Araxes sites throughout the Caucasus (Kushnareva 1997, p. 183‑186). Rather than being a highly mobile pastoral society, the Kura-Araxes appear to have been a community of sedentary agro-pastoralists cultivating crops and managing household herds. At the very least the Sos Höyük evidence suggests that we need to approach models of Kura-Araxes expansion that rely solely on mobile pastoralism as a mechanism to explain the spread of the Kura-Araxes horizon with a degree of caution.

39Archaeobotanical and archaeozoological studies however, cannot reveal the relative importance of agricultural to pastoral resources within the economy or human diet (Halstead 1996). Plant and animal remains at archaeological sites are discarded remnants of economic production subject to a myriad of pre- and post-depositional processes and so cannot be used to measure the absolute quantities of different foodstuffs consumed or produced by a population. Instead a multidisciplinary approach for investigating ancient subsistence at Kura-Araxes sites is needed combining archaeobotanical, archaeozoological, and isotope analyses. Isotope analysis of human skeletal remains can reveal dietary patterns and give an indication, through carbon and nitrogen levels, of the relative proportions of animal and plant products in the diet (Messager et al. 2015; Froehle, Kellner, Schoeninger 2012; Muldner 2013; Pearson et al. 2013; Styring et al. 2014). Similarly dietary isotope studies of animal bones can help determine the different vegetation zones of livestock grazing in the landscape (Makarewicz 2014; Makarewicz, Sealy 2015). Animal and human migration patterns, including vertical transhumance and movement across different geological zones, can be identified through oxygen and strontium isotopes analyses (Knipper et al. 2008; Viner et al. 2010). This type of integration will enable a holistic understanding of Kura-Araxes economy, society and mobility.


Akeret, Jacomet 1997: Ö. Akeret, S. Jacomet, “Analysis of plant macrofossils in goat/sheep faeces from the Neolithic lake shore settlement of Horgen Scheller: an indication of prehistoric transhumance?”, Vegetation History and Archaeobotany 6, 1997, pp. 235‑239.

Alizadeh 2008: A. Alizadeh, “Archaelogy and the question of mobile pastoralism in late prehistory”, in W. Wendrich, H. Barnard (dir.), The archaeology of mobility. Old world and new world nomadism, Los Angeles, Cotsen Institute of Archaeology, 2008, pp. 78‑114.

Anderberg 1994: A. Anderberg, Atlas of seeds and small fruits of Northwest-European plant species with morphological descriptions, Part 4, Stockholm, Swedish Museum of Natural History, 1994.

Anderson, Ertug-Yaras 1998: S. Anderson, F. Ertug-Yaras, “Fuel fodder and faeces: an ethnographic and botanical study of dung fuel use in central Anatolia”, Environmental Archaeology 1, 1998, pp. 99‑109.

Ansari-Renani 2016: H. Ansari-Renani, “An investigation of organic sheep and goat production by nomad pastoralists in southern Iran”, Pastoralism: Research, Policy and Practice 6/8, 2016, pp. 1‑9.

Ansari-Renani et al. 2013: H. Ansari-Renani, B. Rischkowsky, J. Mueller, S. Seyed Momen, S. Moradi, “Nomadic pastoralism in southern Iran”, Pastoralism: Research, Policy and Practice 3/11, 2013, pp. 1‑25.

Arnold, Greenfield 2004: E. Arnold, H. Greenfield, “A zooarchaeological perspective on the origins of vertical transhumant pastoralism and the colonisation of marginal habitats in temperate southeastern Europe”, in M. Mondini, S. Munoz, S. Wickler (dir.), Colonisation, migration and marginal areas, Oxford, Oxbow Books, 2004, pp. 96‑117.

Arnold, Greenfield 2006: E. Arnold, H. Greenfield, The origins of transhumant pastoralism in temperate south eastern Europe. A zooarchaeological perspective from the Central Balkans, BAR International Series 1538, Oxford, Archaeopress, 2006.

Badalyan et al. 2008: R.S. Badalyan, A.T. Smith, I. Lindsay, L. Khatchadourian, P. Avetisyan, B.H. Monahan, R. Hovsepyan, “Village, fortress, and town in Bronze and Iron Age Southern Caucasia: a preliminary report on the 2003‑2006 investigations of Project ArAGATS on the Tsaghkahovit Plain, Republic of Armenia”, Archäologische Mitteilungen aus Iran und Turan 40, 2008, pp. 97‑101.

Bardsley 2001: D. Bardsley, A comparative study of in situ agrobiodiversity conservation in Switzerland, Turkey and Nepal, PhD thesis, The University of Melbourne, 2001 (unpublished).

Barth 1961: F. Barth, Nomads of South Persia. The Basseri tribe of Khamseh Confederacy, Boston, Little Brown and Company, 1961.

Batiuk 2005: S. Batiuk, Migration theory and the distribution of the Early Transcaucasian culture, PhD dissertation, University of Toronto, 2005 (unpublished).

Behre, Jacomet 1991: K.E. Behre, S. Jacomet, “The ecological interpretation of archaeobotanical data”, in W. Van Zeist, K. Wasylikowa, K.E. Behre (dir.), Progress in old world palaeoethnobotany, Rotterdam, Balkema, 1991, pp. 81‑108.

Bejerinik 1947: W. Bejerinik, Zadenatlas der Nederlandsche Flora, Wageningen, H. Veenman and Zanen, 1947.

Berggren 1969: G. Berggren, Atlas of seeds and small fruits of Northwest-European plant species with morphological descriptions, Part 2, Stockholm, Swedish Museum of Natural History, 1969.

Berggren 1981: G. Berggren, Atlas of seeds and small fruits of Northwest-European plant species with morphological descriptions, Part 3, Stockholm, Swedish Museum of Natural History, 1981.

Biagetti 2015: S. Biagetti, “Desert pastoralists: the Kel Tadrart Tuareg from South-West Libya”, Pastoralism: Research, Policy and Practice 5/6, 2015, pp. 1‑4.

Bishop 2015: R. Bishop, “Did Late Neolithic farming fail or flourish? A Scottish perspective on the evidence for Late Neolithic arable cultivation in the British Isles”, World Archaeology 47, pp. 834‑855.

Bogaard 2004a: A. Bogaard, “The nature of early farming in Central and South-East Europe”, Documenta Praehistorica 31, 2004, pp. 49‑58.

Bogaard 2004b: A. Bogaard, Neolithic farming in Central Europe: an archaeobotanical study of crop husbandry practices, London, Routledge, 2004.

Bogaard et al. 1999: A. Bogaard, C. Palmer, G. Jones, M. Charles, “A FIBS approach to the use of weed ecology for the archaeobotanical recognition of crop rotation regimes”, Journal of Archaeological Science 26, 1999, pp. 1211‑1224.

Bogaard et al. 2001: A. Bogaard, G. Jones, M. Charles, J. Hodgson, “On the archaeobotanical inference of crop sowing time using the FIBS method”, Journal of Archaeological Science 28, 2001, pp. 1170‑1183.

Bogaard, Jones, Charles 2005: A. Bogaard, G. Jones, M. Charles, “The impact of crop processing on the reconstruction of crop sowing time and cultivation intensity from archaeobotanical weed evidence”, Vegetation History and Archaeobotany 14, 2005, pp. 505‑509.

Bulut et al. 2010: S. Bulut, I. Coruh, H. Zengin, A. Ozturk, “The effects of different sowing time and seeding rates on weeds in wheat”, Erciyes Üniversitesi Fen Bilimleri Enstitüsü Dergisi 26/4, 2010, pp. 362‑368.

Burney, Lang 1971: C. Burney, D.M. Lang, The peoples of the hills. Ancient Ararat and Caucasus, London, Weidenfeld and Nicolson, 1971.

Çağlar et al. 2011: Ö. Çağlar, M. Karaoglu, S. Bulut, H. Kotancilar, A. Öztürk, “Determination of some quality characteristics in winter and facultive bread wheat (Triticum aestivum L.) varieties”, Journal of Animal and Veterinary Advances 10 (supplement), 2011, pp. 3356‑3362.

Cappers, Beeker, Jans 2006: R. Cappers, R. Beeker, J. Jans, Digital seed atlas of the Netherlands, Eelde, Barkhuis, 2006.

Chang 1993: C. Chang, “Pastoral transhumance in the southern Balkans as a social ideology: ethnoarchaeological research in northern Greece”, American Anthropologist 95/3, 1993, pp. 687‑703.

Charles 1984: M. Charles, “Introductory remarks on the cereals”, Bulletin of Sumerian Agriculture 1, 1984, pp. 17‑31.

Charles 1998: M. Charles, “Fodder from dung: the recognition and interpretation of dung-derived plant material from archaeological sites”, Environmental Archaeology 1/1, 1998, pp. 111‑122.

Charles, Bogaard 2005: M. Charles, A. Bogaard, “Identifying livestock diet from charred plant remains: a Neolithic case study from southern Turkmenistan”, in J. Davies, M. Fabis, I. Mainland, M. Richards, R. Thomas (dir.), Diet and health in past animal populations, current research and future directions, Oxford, Oxbow Books, 2005, pp. 93‑103.

Charles, Jones, Hodgson 1997: M. Charles, G. Jones, J. Hodgson, “FIBS in archaeobotany: functional interpretation of weed floras in relation to husbandry practices”, Journal of Archaeological Science 24, 1997, pp. 1151‑1161.

Crabtree 2011: P. Crabtree, “Animal remains at Godin period IV”, in M. Rothman, H. Gopnik (dir.), On the high road. The history of Godin Tepe Iran, Costa Mesa, Mazda Publishers, 2011, p. 178.

Cribb 1991: R. Cribb, Nomads in archaeology, New York, Cambridge University Press, 1991.

Davis, Cullen, Coode 1965‑1988: P.H. Davis, J. Cullen, M.J.E. Coode (dir.), Flora of Turkey and the East Aegean Islands, Edinburgh, Edinburgh University Press, 1965‑1988.

Di Cosmo 1994: N. Di Cosmo, “Ancient Inner Asian nomads. Their economic basis and its significance in Chinese history”, Journal of Asian Studies 53, 1994, pp. 1093‑1126.

Dutrieux et al. 2016: L. Dutrieux, C. Jakovac, S. Latifah, L. Kooistra, “Reconstructing land use history from Landsat time-series. Case study of a swidden agriculture system in Brazil”, International Journal of Applied Earth Observation and Geoinformation 47, 2016, pp. 112‑124.

Franchetti 2008: M. Franchetti, “Variability and dynamic landscapes of mobile pastoralism in ethnography and prehistory”, in W. Wendrich, H. Barnard (dir.), The archaeology of mobility. Old world and new world nomadism, Los Angeles, Cotsen Institute of Archaeology, 2008, pp. 567‑596.

Franchetti 2012: M. Franchetti, “Multiregional emergence of mobile pastoralism and nonuniform institutional complexity across Eurasia”, Current Anthroplogy 53, 2012, pp. 1‑38.

Franchetti et al. 2010: M. Franchetti, R. Spengler, G. Fritz, A. Mar’yashev, “Earliest direct evidence for broomcorn millet and wheat in the central Eurasian steppe region”, Antiquity 84, 2010, pp. 993‑1010.

Frangipane 2014: M. Frangipane, “After collapse: continuity and distruption in the settlement by Kura-Araxes-linked pastoral groups at Arslantepe-Malatya (Turkey). New data”, Paléorient 40/2, 2014, pp. 169‑182.

Frangipane, Palumbi 2007: M. Frangipane, G. Palumbi, “Red-black ware, pastoralism, trade, and Anatolian-Transcaucasian interactions in the 4th‑3rd millennium BC”, in B. Lyonnet (dir.), Les cultures du Caucase (VIe‑IIIe millénaires avant notre ère). Leurs relations avec le Proche-Orient, Paris, CNRS Éditions, 2007, pp. 233‑255.

French 1971: D.H. French, “An experiment in water-sieving”, Anatolian Studies 21, 1971, pp. 59‑64.

Froehle, Kellner, Schoeninger 2012: A. Froehle, C. Kellner, M. Schoeninger, “Multivariate carbon and nitrogen stable isotope model for the reconstruction of prehistoric human diet”, American Journal of Physical Anthropology 147, 2012, pp. 352‑369.

Geddes 1983: D. Geddes, “Neolithic transhumance in the Mediterranean Pyrenees”, World Archaeology 15/1, 1983, pp. 51‑66.

Goldschmidt 1979: W. Goldschmidt, “A general model for pastoral social systems”, in Équipe Écologie (dir.), Pastoral production and society, Cambridge, Cambridge University Press, 1979, pp. 15‑27.

Grillo 2014: K. Grillo, “Pastoralism and pottery use: an ethnoarchaeological study in Samburu, Kenya”, African Archaeological Review 31, 2014, pp. 105‑130.

Hakgören 1972: F. Hakgören, “Research on irrigation problems and their solutions at the soil-water cooperative area of Upper Pasinler Valley”, Ataturk Üniversitesi Ziraat Fakültesi Dergisi 3, 1972, pp. 41‑54.

Halstead 1987: P. Halstead, “Traditional and ancient rural economy in Mediterranean Europe. Plus ça change?”, The Journal of Helenic Studies 107, 1987, pp. 77‑87.

Halstead 1989: P. Halstead, “The economy has a normal surplus: economic stability and social change among early farming communities of Thessaly, Greece”, in P. Halstead, J. O’Shea (dir.), Bad year economics: cultural responses to risk and uncertainty, Cambridge, Cambridge University Press, 1989, pp. 68‑80.

Halstead 1996: P. Halstead, “Pastoralism of household herding? Problems of scale and specialization in Early Greek animal husbandry”, World Archaeology 28, 1996, pp. 20‑42.

Hillman 1981: G. Hillman, “Reconstructing crop husbandry practices from charred remains of crops”, in R. Mercer (dir.), Farming practice in British Prehistory, Edinburgh, Edinburgh University Press, 1981, pp. 123‑162.

Hillman 1984: G. Hillman, “Interpretation of archaeological plant remains: the application of ethnographic models from Turkey”, in W. Van Zeist, W. Casparie (dir.), Plants and ancient man, Rotterdam, Balkema, 1984, pp. 1‑41.

Hillman 2001: G. Hillman, “Archaeology, Percival, and the problems of identifying wheat remains”, in P. Caligari, P. Brandham (dir.), Wheat taxonomy: the legacy of John Percival, London, The Linnean Society, 2001, pp. 27‑36.

Höhn, Neumann 2012: A. Höhn, K. Neumann, “Shifting cultivation and the development of a cultural landscape during the Iron Age (0-1500 AD) in the northern Sahel of Burkina Faso, West Africa. Insights from archaeological charcoal”, Quaternary International 249, 2012, pp. 72‑83.

Honeychurch, Makarewicz 2016: W. Honeychurch, C. Makarewicz, “The archaeology of pastoral nomadism”, The Annual Review of Anthropology 45, 2016, pp. 341‑359.

Hopkins 2003: L. Hopkins, An ethnoarchaeological study of Sos Höyük and Yiğittaşi village, Archaeology at the North-East Anatolian Frontier VI, Leuven, Peeters, 2003.

Hopkins 2004: L. Hopkins, “Agriculture and economy in a highland village: an ethnoarchaeological approach”, in T. Takaoğlu (dir.), Ethnoarchaeological investigations in rural Anatolia, Istanbul, Ege Yayinlari, 2004, pp. 39‑53.

Hovsepyan 2010: R. Hovsepyan, “New data on agriculture of Aparan-III early Bronze Age settlement (Armenia)”, Biological Journal of Armenia 4, 2010, pp. 31‑37.

Hovsepyan 2011: R. Hovsepyan, “Palaeoethnobotanical data from the high mountainous Early Bronze Age settlement of Tsaghkasar‑1 (Mt Aragats, Armenia)”, Journal of Ethnobiology 2, 2011, pp. 58‑62.

Hovsepyan 2014: R. Hovsepyan, “Pulses cultivated in prehistoric periods at the territory of Armenia: short review of present archaeobotanical data”, Legume Perspectives 5, 2014, pp. 7‑8.

Hovsepyan 2015: R. Hovsepyan, “On the agriculture and vegetal food economy of Kura-Araxes culture in the South Caucasus”, Paléorient 41, 2015, pp. 69‑82.

Howell-Meurs 2001a: S. Howell-Meurs, “Archaeozoological evidence for pastoral systems and herd mobility: the remains from Sos Höyük and Buyuktepe Höyük”, International Journal of Osteoarchaeology 11, 2001, pp. 321‑328.

Howell-Meurs 2001b: S. Howell-Meurs, Early Bronze and Iron Age animal exploitation in northeastern Anatolia: the faunal remains from Sos Höyük and Büyüktepe Höyük, BAR International Series 945, Oxford, Archaeopress, 2001.

Işıklı 2012: M. Işıklı, “The subsistence economy of the Kura-Araxian folks in eastern Anatolia: animal husbandry and pastoralism”, in Azerbaycanin Erken Ekincilik Dovru Abideleri, Baku, Azerbaycan Milli İlimler Akademiası Arxeologiya ve Etnoqrafiya İnstitutu, 2012, pp. 103‑112.

Jacomet 2006: S. Jacomet, Identification of cereal remains from archaeological sites, Basel, IPAS Basel University, 2006, (accessed 15/08/2014).

Jones 1984: G. Jones, “Interpretation of archaeological plant remains: ethnographic models from Greece”, in W. Van Zeist, W. Casparie (dir.), Plants and ancient man, Rotterdam, Balkema, 1984, pp. 43‑61.

Jones 1987: G. Jones, “A statistical approach to the archaeological identification of crop processing”, Journal of Archaeological Science 14, 1987, pp. 311‑323.

Jones 1990: G. Jones, “The application of present-day cereal processing studies to charred archaeobotanical remains”, Circaea 6/2, 1990, pp. 91‑96.

Jones 1992: G. Jones, “Weed phytosociology and crop husbandry: identifying a contrast between ancient and modern practice”, Review of Palaeobotany and Palynology 73, 1992, pp. 133‑143.

Jones 2005: G. Jones, “Garden cultivation of staple crops and its implications for settlement locations and continuity”, World Archaeology 37, 2005, pp. 164‑176.

Jones, Rowley-Conwy 2007: G. Jones, P. Rowley-Conwy, “On the importance of cereal cultivation in the British Neolithic”, in S. Colledge, J. Conolly (dir.), The origins and spread of domestic plants in Southwest Asia and Europe, Walnut Creek, Left Coast Press, 2007, pp. 391‑419.

Jones, Sadler 2012: G. Jones, P. Sadler, “Age at death in cattle: methods, older cattle and known-age reference material”, Environmental Archaeology 17, 2012, pp. 11‑28.

Jones 1981: M. Jones, “The development of crop husbandry”, in M. Jones, G. Dimbleby (dir.), The environment of man: the Iron Age to the Anglo-Saxon period, Oxford, British Archaeological Reports, 1981, pp. 95‑127.

Kakhiani et al. 2013: K. Kakhiani, A.G. Sagona, C. Sagona, E. Kvavadze, G. Bedianashvili, E. Massager, L. Martin, E. Herrscher, I. Martkoplishvili, J. Birkett-Rees, C. Longford, “Archaeological investigations at Chobareti in southern Georgia, the Caucasus”, Ancient Near Eastern Studies 50, 2013, pp. 1‑138.

Khazanov 1994: A. Khazanov, Nomads and the outside world (2nd ed.), Madison, University of Wisconsin Press, 1994.

Khazanov 2009: A. Khazanov, “Specific characteristics of Chalcolithic and Bronze Age pastoralism in the Near East”, in J. Szuchman (dir.), Nomads, tribes and the state in the ancient Near East. Cross-disciplinary perspectives, Chicago, Oriental Institute of the University of Chicago, 2009, pp. 119‑127.

Kiguradze, Sagona 2003: T.B. Kiguradze, A.G. Sagona, “On the origins of the Kura-Araxes cultural complex”, in A.T. Smith, K.S. Rubinson (dir.), Archaeology of the borderlands: investigations in Caucasia and beyond, Los Angeles, Cotsen Institute of Archaeology at UCLA, 2003, pp. 38‑94.

Knipper et al. 2008: C. Knipper, S. Paulus, M. Uerpmann, H.-P. Uerpmann, “Seasonality and land use in Bronze and Iron Age Kakhetia (Georgia). Oxygen and strontium isotope analyses on horse and cattle teeth”, Archäologische Mitteilungen aus Iran und Turan 40, 2008, pp. 149‑168.

Kohl 2007: P. Kohl, The making of Bronze Age Eurasia, Cambridge, Cambridge University Press, 2007.

Kohl 2009: P. Kohl, “Origins, homelands and migrations: situating the Kura-Araxes early Transcaucasian ‘culture’ within the history of Bronze Age Eurasia”, Tel Aviv 36, 2009, pp. 241‑265.

Kushnareva 1997: K.K. Kushnareva, The southern Caucasus in Prehistory: stages of cultural and socioeconomic development from the eighth to the second millennium BC, Philadelphia, University Museum Monograph, 1997.

Legge, Williams, Williams 1991: A. Legge, J. Williams, P. Williams, “The determination of season of death from the mandibles and bones of the domestic sheep (Ovis aries)”, Rivista di Studi Liguri 57, 1991, pp. 49‑65.

Little 1989: M. Little, “Human biology of African pastoralists”, Yearbook of Physical Anthropology 32, 1989, pp. 215‑247.

Longford 2015: C. Longford, Plant economy of the Kura-Araxes. A comparative analysis of agriculture in the Near East from the Chalcolithic to the Middle Bronze Age, PhD thesis, Department of Archaeology, University of Sheffield, 2015 (unpublished).

Longford, Drinnan, Sagona 2009: C. Longford, A. Drinnan, A.G. Sagona, “Archaeobotany of Sos Höyük, northeast Turkey”, in A. Fairbairn, S. O’Connor, B. Marwick (dir.), New directions in archaeological science, published in Terra Australis 28, 2009, pp. 121‑136.

Makarewicz 2013: C. Makarewicz, “A pastoralist manifesto: breaking stereotypes and re-conceptualising pastoralism inthe Near East Neolithic”, Levant 2013/2, 2013, pp. 159‑174.

Makarewicz 2014: C. Makarewicz, “Winter pasturing practices and variable fodder provisioning detected in nitrogen (δ15N) and carbon (δ13C) isotopes in sheep dentinal collagen”, Journal of Archaeological Science 41, 2014, pp. 502‑510.

Makarewicz, Sealy 2015: C. Makarewicz, J. Sealy, “Dietary reconstruction, mobility, and the analysis of ancient skeletal tissues. Expanding the prospects of stable isotope research in archaeology”, Journal of Archaeological Science 56, 2015, pp. 146‑158.

McCorriston 2013: J. McCorriston, “The Neolithic in Arabia: a view from the South”, Arabian Archaeology and Epigraphy 24, 2013, pp. 68‑72.

McCorriston et al. 2012: J. McCorriston, M. Harrower, L. Martin, E. Oches, “Cattle cults of the Arabian Neolithic and early territorial societies”, American Anthropologist 114/1, 2012, pp. 45‑63.

Mertz et al. 2013: O. Mertz, K. Egay, T. Bech Bruun, T. Svan Coldig, “The last swiddens of Sarawak, Malaysia”, Human Ecology 41, 2013, pp. 109‑118.

Messager et al. 2015: E. Messager, E. Herrscher, L. Martin, E. Kvavadze, I. Martkoplishvili, C. Delhon, K. Kakhiani, G. Bedianashvili, A.G. Sagona, L. Bitadze, M. Poulmarc’h, A. Guy, D. Lordkipanidze, “Archaeobotanical and isotopic evidence of Early Bronze Age farming activities and diet in the mountainous environment of the South Caucasus: a pilot study of Chobareti site (Samtskhe-Javakheti region)”, Journal of Archaeological Science 53, 2015, pp. 214‑226.

Miller 1984: N.F. Miller, “The use of dung as fuel: an ethnographic example and an archaeological application”, Paléorient 10/2, 1984, pp. 71‑79.

Miller, Smart 1984: N.F. Miller, T. Smart, “Intentional burning of dung as fuel: a mechanism for the incorportation of charred seeds in the archaeological record”, Journal of Ethnobiology 4, 1984, pp. 15‑28.

Monahan 2007: B.H. Monahan, “Nomadism in the Early Bronze Age southern Transcaucasia: the faunal perspective”, in L.M. Popova, C. Hartley, A.T. Smith (dir.), Social orders and social landscapes. Second University of Chicago Conference on Eurasian Archaeology, Chicago, Cambridge Scholars Publishing, 2007, pp. 379‑392.

Motuzaite-Matuzeviciute, Telizhenko, Jones 2012: G. Motuzaite-Matuzeviciute, S. Telizhenko, M. Jones, “Archaeobotanical investigations of two Scythian-Samatian period pits in eastern Ukraine. Implications for floodplain cereal cultivation”, Journal of Field Archaeology 37, 2012, pp. 51‑61.

Muldner 2013: G. Muldner, “Stable isotopes and diet: their contribution to Romano-British research”, Antiquity 87, 2013, pp. 137‑149.

Neef, Cappers, Bekker 2012: R. Neef, R. Cappers, R. Bekker, Digital atlas of economic plants in archaeology, Groningen, Barkhuis, 2012.

Nesbitt 1991: M. Nesbitt, Early Bronze Age crops and dung from Dilkaya Höyük, Kew Gardens, 1991 (unpublished).

Nesbitt 1995: M. Nesbitt, “Recovery of archaeological plant remains at Kaman-Kalehoyuk”, in P.T. Mikasa (dir.), Essays on Ancient Anatolia and its surrounding civilizations, Wiesbaden, Harrassowitz, 1995, pp. 115‑130.

Nesbitt et al. 2017: M. Nesbitt, J. Bates, G. Hillman, S. Mitchell, The archaeobotany of Asvan. Environment & cultivation in eastern Anatolia from the Chalcolithic to the Medieval Period, London, British Institute at Ankara, 2017.

Newton 2004: J. Newton, Life in the mountains of Anatolia. A study of a Bronze Age settlement in its environmental context, PhD thesis, University of Melbourne, 2004 (unpublished).

Ocak 2016: S. Ocak, “Transhumance in Central Anatolia: a resilient interdependence between biological and cultural diversity”, Journal of Agriculture, Environment and Ethics 29, 2016, pp. 439‑453.

Öztürk, Çağlar, Bulut 2006: A. Öztürk, O. Çağlar, S. Bulut, “Growth and yield response of facultative wheat to winter sowing, freezing sowing and spring sowing at different seedling rates”, Journal of Agronomy and Crop Science 192, 2006, pp. 10‑16.

Palumbi 2003: G. Palumbi, “Red-black pottery, eastern Anatolian and Transcaucasian relationships around the mid-fourth millennium BC”, Ancient Near Eastern Studies 40, 2003, pp. 80‑134.

Palumbi 2009: G. Palumbi, “Collassi, trasformazioni e mondo Kura-Araks nell’Alta Valle dell’Eufrate tra IV e II millennio. Sviluppi locali o migrazioni?”, Scienze dell’antichità Storia Archeologia Antropologia 15, 2009, pp. 157‑171.

Palumbi 2010: G. Palumbi, “Pastoral models and centralised animal husbandry. The case of Arslantepe”, in M. Frangipane (dir.), Economic centralistion in formative states. The archaeological reconstruction of the economic system in 4th millennium Arslantepe, Rome, “Sapienza” Università di Roma, 2010, pp. 149‑165.

Palumbi 2012: G. Palumbi, “Bridging the frontiers. Pastoral groups in the Upper Euphrates region in the early third millennium BCE”, Origini 34, 2012, pp. 261‑278.

Payne 1973: S. Payne, “Kill-off patterns in sheep and goats: the mandibles from Aşvan Kale”, Anatolian Studies 23, 1973, pp. 281‑303.

Paz 2009: S. Paz, “A home away from home? The settlement of Early Transcaucasian migrants at Tel Bet Yerah”, Tel Aviv 36, 2009, pp. 196‑216.

Pearson et al. 2013: J. Pearson, M. Grove, M. Ozbek, H. Hongo, “Food and social complexity at Çayönü Tepesi, southeastern Anatolia: stable isotope evidence of differentiation in diet according to burial practice and sex in the early Neolithic”, Journal of Anthropological Archaeology 32, 2013, pp. 180‑189.

Piro 2009: J. Piro, Pastoralism in the Early Transcaucasian culture: the faunal remains from Sos Höyük, PhD thesis, New York University, 2009 (unpublished).

Piro, Crabtree 2017: J. Piro, P. Crabtree, “Zooarchaeological evidence for pastoralism in the Early Transcaucasian culture”, in M. Mashkour, M. Beech (dir.), Archaeozoology of the Near East 9, Oxford, Oxbow Books, 2017, pp. 273‑283.

Pollington 1840: V. Pollington, “Notes on a journey from Erz-Rum, by Mush, Diyar-Bekr, and Bireh-jik, to Aleppo, in June 1838”, Journal of the Royal Geographical Society of London 10, 1840, pp. 445‑454.

Potts 2014: D. Potts, Nomadism in Iran: from Antiquity to the Modern Era, New York, Oxford University Press, 2014.

Riehl 2006: S. Riehl, “Nomadism, pastoralism and transhumance in the archaeobotanical record. Examples and methodological problems”, in S. Hauser (dir.), Die Sichtbarkeit von Nomaden und saisonaler Besiedlung in der Archäologie Multidisziplinäre Annäherungen an ein methodisches Problem, Halle-Wittenberg, Orientwissenschaftliches Zentrum der Martin-Luther-Universität, 2006, pp. 105‑125.

Ristvet, Bakhchaliyev, Ashurov 2011: L. Ristvet, V. Bakhchaliyev, S. Ashurov, “Settlement and society in Nakhichevan: 2006 excavations and survey of the Nakhichevan Archaeological Project”, Iranica Antiqua 46, 2011, pp. 1‑53.

Rothman 2003: M. Rothman, “Ripples in the stream: Transcaucasia-Anatolian interaction in the Murat/Euphrates Basin at the beginning of the third millennium BC”, in A.T. Smith, K.S. Rubinson (dir.), Archaeology of the borderlands: investigations in Caucasia and Beyond, Los Angeles, Cotsen Institute of Archaeology, 2003, pp. 95‑110.

Rothman 2005: M. Rothman, “Transcaucasians: settlement, migration, and trade in the Kura-Araxes periods”, Archäologische Mitteilungen aus Iran und Turan 37, 2005, pp. 1‑10.

Rothman, Kozbe 1997: M. Rothman, G. Kozbe, “Muş in the Early Bronze Age”, Anatolian Studies 47, 1997, pp. 105‑129.

Rühl, Herbig, Stobbe 2015: L. Rühl, C. Herbig, A. Stobbe, “Archaeobotanical analysis of plant use at Kamennyi Ambar, a Bronze Age fortified settlement of the Sintashta culture in the southern Trans-Urals steppe, Russia”, Vegetation History and Archaeobotany 15, 2015, pp. 413‑426.

Sagona 1984: A.G. Sagona, The Caucasian Region in the Early Bronze Age, BAR International Series 214, Oxford, BAR Publishing, 1984.

Sagona 1993: A.G. Sagona, “Settlement and society in Late Prehistoric Trans-Caucasus”, in M. Frangipane (dir.), Between the rivers and over the mountains, Rome, “Sapienza” Università di Roma, 1993, pp. 453‑474.

Sagona 2000: A.G. Sagona, “Sos Höyük and the Erzurum Region in Late Prehistory: a provisional chronology for Northeast Anatolia”, in C. Marro, H. Hauptmann (dir.), Chronologies des pays du Caucase et de l’Euphrate aux IVe‑IIIe millénaires, Paris, De Boccard, 2000, pp. 329‑373.

Sagona 2004: A.G. Sagona, “Social boundaries and ritual landscapes in Late Prehistoric Trans-Caucasus and highland Anatolia”, in A.G. Sagona (dir.), A view from the highlands. Archaeological studies in honor of Charles Burney, Leuven, Peeters, 2004, pp. 475‑538.

Sagona 2011: A.G. Sagona, “Anatolia and the Transcaucasus themes and variations ca 6400-1500 BCE”, in S. Steadman, G. McMahon (dir.), The Oxford handbook of Ancient Anatolia 10,000-323 BCE, Oxford, Oxford University Press, 2011, pp. 683‑703.

Sagona 2014: A.G. Sagona, “Rethinking the Kura-Araxes genesis”, Paléorient 40/2, 2014, pp. 23‑46.

Sagona et al. 1996: A.G. Sagona, M. Erkmen, C. Sagona, I. Thomas, S. Howells, “Excavations at Sos Höyük, 1995: second preliminary report”, Anatolian Studies 46, 1996, pp. 27‑52.

Sagona et al. 1997: A.G. Sagona, M. Erkmen, C. Sagona, I. McNiven, S. Howells, “Excavations at Sos Höyük, 1996: third preliminary report”, Anatolica 23, 1997, pp. 181‑226.

Sagona et al. 1998: A.G. Sagona, M. Erkmen, C. Sagona, I. McNiven, S. Howell-Meurs, “Excavations at Sos Höyük, 1997: fourth preliminary report”, Anatolica 24, 1998, pp. 31‑64.

Sagona, Sagona 2000: A.G. Sagona, C. Sagona, “Excavations at Sos Höyük 1998‑2000: 5th preliminary report”, Ancient Near Eastern Studies 37, 2000, pp. 56‑127.

Sagona, Sagona 2004: A.G. Sagona, C. Sagona, “Frontiers and mountains: fifteen years of archaeology in north-east Anatolia”, in T. Murray (dir.), Archaeology from Australia, Melbourne, Australian Scholarly Publishing, 2004, pp. 361-373.

Sagona, Sagona, Özkorucuklu 1995: A.G. Sagona, C. Sagona, H. Özkorucuklu, “Excavations at Sos Höyük: first preliminary report 1994”, Anatolian Studies 45, 1995, pp. 193‑218.

Sagona, Zirmansky 2009: A.G. Sagona, P. Zirmansky, Ancient Turkey, London, Routledge, 2009.

Sagona 1999: C. Sagona, “An archaeological survey of the Erzurum Province 1999: the region of Pasinler”, Ancient Near Eastern Studies 36, 2009, pp. 108‑131.

Salzman 2002: P. Salzman, “Pastoral nomads: some general observations based on research in Iran”, Journal of Anthropological Research 58, 2002, pp. 245‑264.

Shahack-Gross 2011: R. Shahack-Gross, “Herbivorous livestock dung: formation, taphonomy, methods for identification and archaeologial significance”, Journal of Archaeological Science 38, 2011, pp. 205‑218.

Shimelmitz 2003: R. Shimelmitz, “A glance at the early Trans-Caucasian Culture through its nomadic components”, Tel Aviv 30, 2003, pp. 204‑221.

Shishlina, Gak, Borisov 2008: N. Shishlina, E. Gak, A. Borisov, “Nomadic sites of the South Yergueni Hills on the Eurasian Steppe”, in H. Bernard, W. Wendrich (dir.), The archaeology of mobility. Old world and new world nomadism, Los Angeles, Cotsen Institute of Archaeology, 2008, pp. 230‑249.

Smith 2005: A.T. Smith, “Prometheus unbound: southern Caucasia in Prehistory”, Journal of World Prehistory 19, 2005, pp. 229‑279.

Smogorzewska 2004: A. Smogorzewska, “Andirons and their role in Early Trascaucasian Culture”, Anatolica 30, 2004, pp. 151‑177.

Spengler, Chang, Tourtellotte 2013: R. Spengler, C. Chang, P. Tourtellotte, “Agricultural production in the Central Asian mountains: Tuzusai, Kazakhstan (410-150 BC)”, Journal of Field Archaeology 38/1, 2013, pp. 68‑85.

Spengler et al. 2014: R. Spengler, B. Cerasetti, M. Tengberg, M. Cattani, L. Rouse, “Agriculturalists and pastoralists: Bronze Age economy of the Murghab alluvial fan, southern Central Asia”, Vegetation History and Archaeobotany 23, 2014, pp. 805‑820.

Spengler, Franchetti, Domani 2014: R. Spengler, M. Franchetti, P. Domani, “Late Bronze Age agriculture at Tasbas in the Dzhungar Mountains for eastern Kazakhstan”, Quaternary International 348, 2014, pp. 147‑157.

Spooner 1973: B. Spooner, “The cultural ecology of pastoral nomads”, Anthropological Quarterly 44, 1973, pp. 198‑210.

Stevens, Fuller 2012: C. Stevens, D.Q. Fuller, “Did Neolithic farming fail? The case for a Bronze Age agricultural revolution in the British Isles”, Antiquity 86, 2012, pp. 707‑722.

Styring et al. 2014: A. Styring, R. Fraser, R. Arbogast, P. Halstead, V. Isaakidou, J. Pearson, M. Schaefer, S. Triantaphyllou, T. Valamoti, M. Wallace, A. Bogaard, R. Evershed, “Refining human palaeodietary reconstruction using amino acid delta N-15 values of plants, animals and humans”, Journal of Archaeological Science 53, 2014, pp. 504‑515.

Summers 2004: G. Summers, “Yanik Tepe and the Early Transcaucasian culture: problems and perspectives”, in A.G. Sagona (dir.), A view from the highlands. Archaeological studies in honor of Charles Burney, Leuven, Peeters, 2004, pp. 617‑643.

Summers 2014: G. Summers, “The Early Trans-Caucasian culture in Iran: perspectives and problems”, Paléorient 40/2, 2014, pp. 155‑168.

Tapper 1977: N. Tapper, “Pashtun nomad women in Afghanistan”, Asian Affairs 8, 1977, pp. 163‑170.

Thevenin 2011: M. Thevenin, “Kurdish transhumance: pastoral practices in south-east Turkey”, Pastoralism: Research, Policy and Practice 1/23, 2011, pp. 1‑24.

Valamoti 2007: T. Valamoti, “Detecting seasonal movement from animal dung: an investigation in Neolithic northern Greece”, Antiquity 81, 2007, pp. 1053‑1064.

Van der Leeuw, McGlade 1997: S.E. Van der Leeuw, J. McGlade (dir.), Archaeology: time, process and structural transformations, London, Routledge, 1997.

Van der Veen, Jones 2006: M. Van der Veen, G. Jones, “A re-analysis of agricultural production and consumption: implications for understanding the British Iron Age”, Vegetation History and Archaeobotany 15, 2006, pp. 217‑228.

Van Zeist, Bakker-Heeres 1975: W. Van Zeist, J. Bakker-Heeres, “Prehistoric and early historic plant husbandry”, in M. Van Loon (dir.), Korucutepe, vol. 1, Amsterdam, North Holland Publishing Company, 1975, pp. 225‑257.

Van Zeist, Bakker-Heeres 1985: W. Van Zeist, J. Bakker-Heeres, “Archaeobotanical studies in the Levant 1. Neolithic sites in the Damascus basin: Aswad, Ghoraife, Ramad”, Palaeohistoria 26, 1985, pp. 151‑170.

Van Zeist, Bakker-Heeres 1986: W. Van Zeist, J. Bakker-Heeres, “Archaeobotanical studies in the Levant 2. Neolithic and Halaf levels at Ras Shamra”, Palaeohistoria 26, 1986, pp. 151‑170.

Van Zeist, Bakker-Heeres 1988: W. Van Zeist, J. Bakker-Heeres, “Archaeobotanical studies in the Levant 4. Bronze Age sites on the north Syrian Euphrates”, Palaeohistoria 27, 1988, pp. 247‑316.

Viner et al. 2010: S. Viner, J. Evans, U. Albarella, M. Parker Pearson, “Cattle mobility in prehistoric Britain: strontium isotope analysis of cattle teeth from Durrington Walls (Wiltshire, Britain)”, Journal of Archaeological Science 37, 2010, pp. 2812‑2820.

Wasylikowa 1981: K. Wasylikowa, “The role of fossil weeds for the study of former agriculture”, Zeitschrift für Archäologie 15, 1981, pp. 11‑32.

Wendrich, Barnard 2008: W. Wendrich, H. Barnard (dir.), The archaeology of mobility. Old world and new world nomadism, Los Angeles, Cotsen Institute of Archaeology, 2008.

Wilkinson 2003: T. Wilkinson, “Archaeological survey and long-term population trends in Upper Mesopotamia and Iran”, in N.F. Miller, K. Abdi (dir.), Yeki bud, yeki nabud: essays on the archaeology of Iran in honour of William M. Sumner, Los Angeles, Cotsen Institute of Archaeology, 2003, pp. 39‑51.

Wilkinson 2014: T. Wilkinson, “The Early Transcaucasian phenomenon in structural-systemic perspective: cuisine, craft and economy”, Paléorient 40/2, 2014, pp. 203‑229.

Yakar 2000: J. Yakar, Ethnoarchaeology of Anatolia. Rural socio-economy in the Bronze and Iron Ages, Tel Aviv, Emery and Claire Yass Publications in Archaeology, 2000.

Yakar 2006: J. Yakar, “Traits of nomadic people: ethnoarchaeolgical and archaeological research in Turkey”, in S. Hauser (dir.), Die Sichtbarkeit von Nomaden und saisonaler Besiedlung in der Archäologie Multidisziplinäre Annäherungen an ein methodisches Problem, Halle-Wittenberg, Orientwissenschaftliches Zentrum der Martin-Luther-Universität, 2006, pp. 45‑63.

Zohary 1973: M. Zohary, The geobotanical foundations of the Middle East, Amsterdam, Swets and Zeitlinger, 1973.

Zohary, Hopf, Weiss 2012: D. Zohary, M. Hopf, E. Weiss, Domestication of plants in the old world, Oxford, Oxford University Press, 2012.


1 Sagona has since revised his opinion and regards the Kura-Araxes as settled agro-pastoralists (Sagona 2011, 2014).

Table des illustrations

Titre Fig. 1 – The Pasinler Valley, Erzurum Province, Turkey, indicating the location of Sos Höyük and other places mentioned in the text; elevation intervals are every 100 m (C. Longford).
Fichier image/jpeg, 557k
Titre Fig. 2 – Graph of crop item percentages for each period at Sos Höyük.
Fichier image/jpeg, 288k
Titre Fig. 3 – The products and by-products of crop processing for glume wheats and free threshing cereal (free threshing wheat and barley); note that hulled barley is a free threshing cereal and is processed the same way as bread/durum wheat (for barley the “hulls” are the palea and lemna rather than the glumes) (based on Charles 1984; Van der Veen, Jones 2006).
Fichier image/jpeg, 388k
Titre Fig. 4 – Discriminant analysis plot comparing the Sos Höyük samples to ethnographic crop-processing groups from Amorgos, Greece, using physical characteristics of weeds seeds (Jones 1984).
Fichier image/jpeg, 323k
Titre Fig. 5a – Sample composition of ethnographic crop processing samples from Amorgos, Greece (Jones 1990) for comparison with fig. 5b.
Fichier image/jpeg, 230k
Titre Fig. 5b – Sample composition of Sos Höyük based on the proportions of grain, rachis internodes and weed seeds.
Fichier image/jpeg, 193k


The University of Sheffield, United Kingdom

The University of Melbourne, Australia


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search