1Manuscript received on 30 April 2019, accepted on 9 October 2019

2Over the ages, Mesopotamia has represented a unique convergence point wherein urbanism was born, states first arose and fell, long-distance trade networks developed and disintegrated, and patterns of human mobility catalysed cultural change at differing rates.

5Despite a certain degree of homogeneity in terms of material culture characterising the entire Northern Fertile Crescent, each single “marginal” area may be seen to have acted and reacted differently one from another, as is indicated by its own local character in settlement patterns (nucleated versus sparse), and in territorial organisation, as well as, most vitally, in pottery style, the lattermost laying the foundation for any chronological assessment of prehistoric periods. In this scenario of cultural variability, the “lived environment”, i.e. the landscape inhabited, acted both as trigger and constraint.

7The combined data offer a range of forms of information capable of addressing some of the questions raised above both in terms of quality (updated investigations and pertinence of the achieved information) and quantity of data. Settlement patterns and land use, and stratigraphic sequences and assessment of pottery will all here be considered as useful indicators of the variation occurring within the landscape throughout the Late Chalcolithic Age.

The survey region and excavated sites

9Partially defined by natural as well as artificial boundaries, the survey area covers a representative sector of a broad ecological zone including the Northern Zagros foothills. The region is characterised by a strikingly diverse physical environment, featuring river valleys and wadis, springs and streams, rolling plains, rocky gorges, and narrow canyons, as well as steep passes cutting up into rugged mountain chains (fig. 1).

10Both the Tigris River and its Ḫabur tributary were vital arteries for land use and communication, playing a paramount role in shaping human settlement over the millennia. This is a zone of both rainfed and irrigation agriculture, as well as of seasonal pasturage for sheep and goats. Such variety in landscape provided room for various lifestyles and subsistence niches, comprising of settled farmers, and nomadic and semi-nomadic groups. In the past, as today, all these people were closely interwoven, symbiotically interacting in a manner best exploiting such a diversified physical environment. “Transitional” and “permeable” are therefore the terms best describing the entire landscape at both a local and super-regional level.

13Geographically, the latter is not equally distributed among the various aforementioned zones; as is further discussed, most sites strongly cluster in the Tigris alluvium and the plains of its immediate hinterland, while the Ǧebel Biḫair (i.e. the preliminary mountain range of the Zagros) may be shown to have played a vital role, acting as something of a barrier and constraint.

15The bulk of material hails, however, from the stratified sequences of the other two sites investigated in 2015, namely Muqable I (fig. 2) and III. These are both located 5 km to the south of Bassetki and belong to an unusually thick cluster of settlements. Both were chosen not only because they display a continuous interlinked sequence of occupation, but also more crucially as they were menaced by heavy construction works in connection with the current building of a new airport at Duhok (Pfälzner and Sconzo 2015, 2016a, 2016b; Pfälzner et al. 2017).

16While Muqable I (site B11) dates almost exclusively to an early phase of the LC (4500-3800 BC), Muqable III (site B13) is a multi-period mound wherein LC marks the beginning of a long-lasting superimposed occupation yielding the formation of a true tell. Both sites have already provided well-stratified material and may become highly valuable benchmarks for the Chalcolithic and Early Bronze Age archaeology of Upper Mesopotamia (fig. 3). Muqable I, moreover, feasibly acted in the early 4^th millennium BC as pottery production centre, as is suggested by the retrieval of a pottery kiln associated with a bunch of wasters, and a reasonable amount of locally produced ceramic material in situ (Pfälzner et al. 2017).

17The Muqable cluster lies along the Wadi Saru Kani, one of the larger streams in the region, which was historically watered by several productive springs. The wadi is nearly dry today in the summer, but, until recently, irrigated fertile gardens and fields along the valley floor, and supplied water to villages further downstream. The Muqable area represents an ideal setting for self-sufficient agricultural settlements, thus forming a small and peculiar ecological niche. The cluster consists of seven hills altogether, lying in a radius of 350 m, and at a lesser distance apart from one other (fig. 3a).

18The earliest evidence for occupation (only traces) at Muqable dates back to the Neolithic (Halaf) period, albeit the major occupation is attested during the Late Chalcolithic (fig. 3c-d) and the Bronze Age. The cluster also continued to be settled (though to a lesser extent) in the Parthian, Sassanian, and Islamic periods. During each period, only some sites appear to have been settled contemporaneously, the settlement location(s) moving over time from site to site within the cluster in a circular pattern. Spatially, this topographic settlement system can thus be defined as a “crown cluster”.

History of research and chronological issues

19Prior to the beginning of the Tübingen project, very little was known of the early settlement history of the entire area. Practically no archaeological fieldwork had occurred within this region, and the area often appeared as a blurred or empty patch within distribution maps depicting prehistoric and early historical periods. This was due to various reasons, including its remote geographic location, the mountainous rugged terrain hampering any assessment on the sole basis of remote sensing (Pfälzner and Sconzo 2016a: 16), and the unstable political situation that characterised the recent history of Iraqi Kurdistan in different decades of last century.

22While, as previously mentioned, our zone was not directly affected by fieldwork, the investigations undertaken along the Eski Mosul lake provided a large array of new data which have also become important to the present understanding of the settlement history of Northern Iraqi Kurdistan as a whole (State Organization of Antiquities and Heritage 1987). The sites mapped by the SOAH were revisited by the EHAS team; nevertheless, the chronological assessment was only confirmed for three of them, two of which date to the Late Chalcolithic (table 1). One, Gire Baqal (site B193), mentioned as Tell Kamuna on the map, was excavated for a short season in 2010 by the DGA of Duhok under the direction of H.A. Qasim, but the results remain presently unpublished.

23Thereafter, apart from sporadic local activities, the whole region was inaccessible for over two decades due to extensive military operations connected with the former political unrest.

24Turning to chronology, it must here be noted that an assessment of a fine-tuned occupational history of the EHAS region is, for the LC, as for many other periods, still hampered by various factors. After almost 30 years, the overall chronology of the Late Chalcolithic of Upper Mesopotamia and the resulting finer fivefold periodisation established by the advanced seminar of Santa Fe in 1991 (Rothman 2001; Wright and Rupley 2001), still remain hardly applicable in dealing with survey material; but, as regards ceramic changes from one phase to the following, they result in being more quantitative than qualitative in nature, and also as being geographically sensitive.

25The new KUGAMID excavations have not presently brought to light any striking enduring local sequence spanning the entire LC period, but rather only snapshots of specific and chronologically distinct stages in the life of the settlements at hand.

Landscape and settlements in the Late Chalcolithic 1-2

27As a general pattern, the overall Chalcolithic occupation in the Upper Zagros Piedmont is still characterised by a rather dispersed form of settlement and land use (Iamoni 2014). Although a general increase from the Northern Ubaid to the Late Chalcolithic in site number (about 25%) and in overall occupied area could be sketched, such increase cannot be considered neither linear nor drastic, as the continuity between the two broad periods concerns less than 40% of sites, while the remaining ones are either new foundations or resettling of earlier sites occupied in the Neolithic.

28At the cultural baseline of the LC1-2, according to the results of our survey work, just a small network of villages and towns still dotted the Northern Zagros foothills, representing about 15% of the settlement sites presently detected. With the only exception of the so-called “Muqable trajectory” (to be treated later), these spread out across the region without according to any remarkable linear patterns: while, on the one hand, the latter may be understood as an indicator of the lack of emerging contact routes, on the other, this could possibly testify an ancillary means of exploiting the different potentials provided by the surrounding environment, and of the synergising of the land (fig. 5a).

30No visible occupation has been instead detected in the eastern stretch of the Ḫabur and Rugarm rivers and their tributaries, where, however, the agriculturalists of the Pottery Neolithic period had already established their first seasonal camps (Sconzo and Pfälzner 2016a: 24-26).

32The highest concentration of sites, as in any other period, is to be found in the southern sector of the Selevani plain (Daṣt-e Selevani). This plain is the only area of the Zagros Piedmont in the survey region featuring enough space for pastureland and farming. Crossed by several seasonal watercourses, this water-rich area was good enough for largescale rainfed agriculture and, consequently, for the establishment of a staple economy and eventually of a surplus gain. Both of these events will be basic requisite for the emergence and consolidation of large centres.

36Another interesting site nearby must have been Bassetki 3 (site B3), where the LC1-2 horizon marks the end of a long-lasting occupation originating in the Halaf period (Pfälzner and Sconzo 2015).

37To summarise, both artificial mounds and flat sites are attested, lying at an altitude between 329 m (Tigris alluvium) and 607 m. asl (Ǧebel Biḫair foothills). For some of them, an ancillary phase at the very beginning of the period is suggested by the presence of “Sprig Ware” and the abundance of other painted pottery of post-Ubaid tradition (Ball 1997, 2003: 11, 154).

38Site size spans from a few hundred square metres up to about 4 ha, the maximum size ultimately reaching 5 ha if the three LC1-2 sites of the Muqable cluster are considered as a single unit. The system of villages is thus resultantly fairly homogeneous in size and devoid of spatial hierarchies. To provide a comparative frame of reference, the overall occupied area in this period presently amounts to 57 ha, corresponding more or less to 1/7^th of what it was later attained in the Middle Bronze and Iron Ages. Thus, this is hardly similar to what simultaneously occurred to the South at Nineveh (Stronach 1994), and to the west of the Tigris, in the large sweep of the Jazirah (Wilkinson and Tucker 1995: 78; Ur et al. 2011), where LC1-2 spatial agglomeration and nucleation are evident at and around very large settlements such as Tell Hawa (Ball et al. 1989; Ball 1990), Hamoukar/Khirbet al Fakhar (Ur 2010: 96-98; Al Quntar et al. 2011) or Tell Brak (Oates et al. 2007; Ur et al. 2011: 2; McMahon and Crawford 2014).

39Despite the sparse occupation, village-size sites, and the absence of a true site hierarchy, the area eventually became embedded in a far-flung exchange network embracing the entirety of Northern Mesopotamia. The circulation of raw materials and craft products, as well as knowledge and modes of socio-economic organisation, are perceptible in some specialised activities related to the occurrence of both imported and local artefacts such as: lithic industry in exotic materials (mainly obsidian blades, attested in the survey collection at different sites; fig. 6a); serial-produced vessels (see below); and, above all, traces of an administrative system in the ancillary form of impressed stamp seals.

40Worth mentioning in this regard is a unique small clay seal impression from the Muqable cluster displaying the “quartered circle” (fig. 6b), a motif typical of the Gawra XII-IX style (Tobler 1950: 179-180, pl. CLVIII, fig. 12-15 and pl. CLIX, fig. 16-17; also attested at Hamoukar: Reichel 2009: fig. 12).

44Outstanding within the entire assemblage is the so-called “Red-Burnished Ware”, a specialised ware perhaps related to kitchen activities. The latter is characterised by a brittle, medium- to coarse-textured fabric, tempered with shell, large limestone grits, and even larger quartz crystals. In this instance, surface treatment is characterised by a thick coating of red slip and hand-burnished finishing. Incomplete oxidation occurs on many examples, as well as the double surface colour (especially in the area of the rim), tending towards a grey hue, and on the inner side of vessel surfaces. This ware occurs in a limited, specialised set of forms, among which may be found hole-mouthed pots, sometimes spouted (fig. 7.2-3).

45Apart from the latter ware and from the flat-based conical bowls, which are quite homogeneous in terms of recipes, size, and related set of forms, the rest of the production still displays interchanging rules in the association of not only shapes and wares, but also surface treatment and decorative patterns. On the whole, the repertoire does not seem to have completely abandoned Ubaid traditions.

49Although the 2015 excavation at Muqable I unearthed no conspicuous architectural remains (Pfälzner et al. 2017), the assemblage retrieved around and inside a pottery kiln (fig. 8), consisting of straight-sided bowls (fig. 9.1-2) and other closed vessels such as, for example, the pithoi with a hollowed rim (fig. 9.3-4) side by side with numerous wasters suggest that the area under analysis may well have served as a pottery production centre during LC2 (if not earlier).

50This evidence is further strengthened by the retrieval of a series of fine mineral-tempered ring-shaped scrapers (fig. 9.5-7); other surface finds of the same type have been collected at this and other sites. Scraping is an age-old technique, already appearing in Ubaid contexts as a device for reducing the time of unbaked vessel drying, as well as for improving rapidity and serial production. These tools from Muqable were as well feasibly used in pottery finishing; typologically they recall those widely attested in Mesopotamia from the Uruk period onward: therefore, they further suggest that the origin of this class of finds should be moved back to the late 5^th millennium BC, thus making a northern feature preceding and later embodied in the urban revolution of the south (Alden 1988; Alden and Minc 2016).

Landscape and settlements of the Late Chalcolithic 3-5

52Contrary to what was preliminarily suggested in previous contributions based upon survey results from the seasons 2013-2014 (Pfälzner and Sconzo 2015, 2016a), the EHAS region apparently provides incipient hints of advancement in the trend toward settlement nucleation, redefinition of modes of production and consumption as well as socio-political reorganisation during this phase.

53The latter phenomena cannot be understood without issuing a caveat concerning site attribution. As a whole, the LC3-5 ceramic assemblage retrieved during our survey presently includes over two thousand diagnostic sherds, gathering together both a local/indigenous chaff-faced and a Uruk-related mineral-tempered production. It needs scarcely be stated that the survey material, as category, does not help to overcome the dichotomy between the “local” and the “foreign” and problems of attribution and data interpretation occur whenever forms which are expected to be found in mineral-tempered wares feature rough chaff-tempered fabrics or the other way around (Vallet et al. 2017: 74-75). The repertoire of collected sherds displays a certain degree of borrowings or hybridisation between southern and local types, which remain difficult to interpret in the absence of any true context. This means that the material culture from this very corner of Northern Mesopotamia remains still nebulous and would deserve new adequate reference stratigraphies in order to give a proper response to the question of culture contact and interaction.

55In the LC3-4 (5) local repertoire, the fine mineral-tempered fabrics, quite widespread in the LC1-2, disappear, being replaced by much coarser, low-fired, chaff-tempered examples. This phenomenon is accompanied by the vanishing of the painted decoration which (as has been previously identified) still lingered in the prior assemblages. On the other hand, all those aspects of pottery technology considered the major diagnostics of increased cheap and large-scale pottery production in the period do occur (Al Quntar and Abu Jayyab 2014: 99-106): uniformity in vessel type, simplification of form and manufacture, new strategies for fuel consumption, and a decrease in (if not abandonment of) aesthetic attributes, thus possibly testifying to a process of progressive decline in quality in favour of a more standardised, cruder, but also faster process of manufacture, and a progressive loss of the symbolic and representative functions of ceramics. Should a phase of increasing standardisation be sought, then this would be the appropriate period in which to locate the beginnings of such a development.

56The Uruk-influenced assemblage remains even more elusive, if no longer in terms of quantity (table 2), then, at least, in terms of variability. Apart from the chaff-tempered, usually mould-made bevelled-rim bowls (quite rare within the survey region as a whole) and another set of grit and/or chaff-tempered hand-made vessels, the assemblage features medium-to-fine wheel-made fabrics tempered with sand, calcite, and especially mica. Ranging from yellow to deep orange, vessels are evenly fired, thus strongly differing from the indigenous ceramic tradition. The part of the form spectrum which might be recognised is repetitive, being limited to nose-lug jars with incised and impressed decoration, bowls with oblique/triangular rims, jars with undercut rims (all very frequent), spouts (very few), and conical bowls with string-cut bases. One single case of reserved slip ware has been detected, while no red slip has presently been recognised.

IANES, University of Tübingen, drawing A. al-Hashemi

58This assemblage is not at all new, as it is strictly comparable to that recorded at other excavated sites of the lower Eski Mosul district, such as Tell Karrana 3, strata 4-3a (Rova 1993, 2003), Mohammed ‘Arab, phase MA 1 (Roaf 1998), and Tell Jessary, level 5, ash pit (Numoto 1990). It testifies to the existence a regional facies spanning from the northern Nineveh hinterland all the way to the preliminary Zagros Mountain chain at the time, if not, indeed, after the collapse of the Uruk network system.

59Turning to settlement system, if we start approaching it from a neutral perspective (i.e. with no distinction between local and Uruk-oriented sites), it might seem as if the old pattern might remain more or less intact: in LC3-5, site number in fact grows by only a few units (34-39). Fundamentally, however, more than half of the LC1-2 settlements is temporarily or definitively abandoned, while various entirely new foundations are recorded (fig. 12). The aggregate area remains more or less of equal size (ca. 50 ha); site size ranges between 0.5 and 4 ha, while a size of 1-2 ha is now overwhelmingly the norm. While, on the Hezil Su and the Ḫabur valleys, occupation remains steady and continues mainly in the form of bluff-top sites set high up on the alluvial plain, a slight increase in number of tells can be noted within the Tigris hinterland, such as the Selevani Plain, testifying to the beginning of a rather long-lasting phase of settlement, at least in some areas.

60Moreover, spatial planning becomes more organised and settlements tend to cluster along rather visible routes.

61This becomes more apparent when, in light of the pottery evidence previously outlined, recorded data are split between sites in which elements of Southern Mesopotamian inspiration are evident and those with an exclusively local repertoire (fig. 13).

62First of all, sites displaying a mixed repertoire greatly outnumber those only featuring a purely “foreign” or “local” style. Southern Mesopotamian oriented settlements are all new foundations clustering along the Tigris and its immediate hinterland, while local LC3-4 (5) ones apparently lie at the piedmont of the Ǧebel Biḫair along an internal trajectory, and usually stem from an earlier LC1-2 nucleus. Among the latter is site B147, a flat larger settlement located at the confluence of two wadis to the north of Bassetki. Here, the initial LC2 occupation of about 4 ha may now be seen to have spread on the other side of one of the wadis, thus reaching an extension of ca. 6 ha. The extraordinary amount of flint blades and grinding stones retrieved identifies its status as a specialised extraction (?) and/or production centre within the region.

63By means of those sites with mixed assemblages, a deliberate choice of strengthening occupation along three main trajectories can be envisaged: two of them, riverine in character, included an E-W route along the Ḫabur River and a S-N example along the Tigris itself; the third was instead an inland route along the Wadi Amlik, directly linking the two rivers through a mountain pass.

64Worth mentioning is the easternmost settlement presently detected along the Ḫabur, Gundi Kusa (site C64), located close to a spring in a hilly landscape some 5 km south of the Ḫabur River. Characterised by a prominently Uruk-oriented ceramic repertoire, the site presents a comparable amount of local chaff-tempered pottery and hybrid artefacts (Pfälzner and Sconzo 2016a: 24, fig. 8b). Given the impossibility of ascertaining the relationship between the two parts of the assemblage (whether spatial or temporal), it remains unclear as to whether this might be interpreted as some manner of southern implant or the seat of an indigenous community fully integrated within the southern Mesopotamian network as is the case for Hacinebi on the Middle Euphrates (Stein 2001, 2002a, 2002b). Its perfectly isolated geographical location, off any beaten track, but not too far away from the Ḫabur River may suggest the existence of exchange routes linking the Tigris lowlands to the Iranian uplands further east.

65A second tiny hill-top site (B151) overlooking the Wadi Amlik is located on a natural hillock bordered by two wadis, immediately to the south of the modern highway leading to Peš Ḫabur. It belongs to a string of three sites similar in nature set at a certain distance between the Tigris and the Zaxo pass; its strategic hill-top setting was clearly intended to control both the mountain route, on the one hand, and the access to the Tigris alluvial plain, on the other. As previously mentioned, this prompts speculation that it might have been envisioned as a penetration route, in this instance connecting the eastern bank of the Tigris to the Ḫabur across the only existing north-south mountain pass: namely the “Zaxo connection”.

68At the same time, it should not be underestimated the possibility that such a regular spacing along the river terraces reflects—in turn—the establishment of a system of agricultural production for large urban centres not necessarily located within the limit of the survey area, rather located downstream, in the Nineveh plain or even southward.

70In summary, a complex picture emerges from that which has been sketched above, in which one and the same region features areas apparently rather segregated and, in turn, less affected by the international network emerging from the LC3 onward, and yet also other areas—such as those of the Tigris and Ḫabur rivers—which are far more embedded within this network. Within such a reconstruction, it might be argued that the local component played an active role in the exchange network, local sites possessing the upper hand within resource extraction.

Conclusion

72The results of the EHAS survey and KUGAMID excavations in the northernmost region of Iraqi Kurdistan doubtless bring new insights into the life of Chalcolithic communities in Upper Mesopotamia. While the continuation of investigations of the earlier occupation levels at Muqable, and consequently the further establishing a solid stratified pottery sequence, would ultimately lead to a better understanding of the paths toward urbanisation within this region, a few conclusive remarks may nevertheless now be sketched.

73As far as settlement is concerned, in a broad sense, the East-Tigridian plateau extending north of Nineveh and south of the first ridge of the Zagros Mountains, namely the Ǧebel Biḫair, seems apparently to have been continuously settled from the 6^th to the 3^rd millennium BC, while the Late Neolithic and Early Chalcolithic penetration further north seems to be minimal and strictly confined to the main waterway of the Ḫabur. If so, this first Zagros ridge may be seen to have acted as a barrier for at least the greater part of the Chalcolithic period.

74In particular, the Late Chalcolithic in the EHAS region does not seem to feature at any stage those forms of accelerated urbanisation envisaged for the Jazirah steppe to the west, or at Nineveh to the south, as testified by population nucleation, spatial agglomeration and expansion, and, moreover, by the reorganisation of political institutions.

75In turn, the area does not seem to be cut off completely from the urban formation process as a whole. First of all, the beginning of a long-lasting settlement strategy is suggested by the fact that the LC settlement gives preference to watercourses, especially to those fed by perennial springs, which would have assured water supply throughout the year.

78The LC1-2 settlement dynamics display the establishment of trajectories that would later develop and expand in the LC3-5: salient is the E-W example along the Ḫabur and a SW-NE axis connecting the Tigris to the Ǧebel Biḫair Piedmont. The latter consists of a chain of village-size sites set at a short distance along the same wadi (Saru Kani), thus suggesting a penetration route leading from the Tigris alluvium towards the hinterland, perhaps connected to the exploitation at a regional scale of natural resources set at the foot of the mountains. A similar pattern has been noticed, albeit at a larger and more organised scale, a few miles to the south, in the Jebel Zawa Piedmont, by the LONAP project.

79In this picture, the Muqable cluster, set midway between river and mountains, may have functioned as a small hub within the proposed local network.

80The broader eastern Tigridian plateau, moreover, already displays material evidence of an extensive cross-regional interaction with the surrounding societies of Upper Mesopotamia including the exchange of obsidian, other raw materials and crafted commodities, and technology and knowledge.

81Conversely, as far as pottery production is concerned, the abandonment of the Ubaid legacy, begun in the entire Northern Fertile Crescent right after the mid-5^th millennium, seems to have been a rather sluggish, gradual process at the foot of the Zagros Mountains, as is attested by some lingering elements in the pottery repertoire, such as the persistence of the painted tradition, mineral-tempering, and costly firing techniques down to the end of the LC2. These features seem to make up a local development of this very sector of the Zagros Piedmont.

82Regarding further development in pottery, it might be stated that if a change or break is to be envisaged, then this must apparently occur at the cultural baseline of the LC3. In fact, it is at this stage that a decisive step toward serial production is taken, with the result that a clear loss of quality is traded in favour of a gain in speed of manufacture and subsequent labour specialisation. At this time, the region seems once more to have ceramically been clearly oriented towards the steppe of the Jazirah, as hinted by the fact that certain typical features of the North Mesopotamian “chaff-faced koiné” (such as casseroles and hammerhead bowls) find their south-easternmost border in the upper EHAS region.

83Since the beginning of the LC3, moreover, a further path toward the formation of an urbanised landscape might be traced. Recent investigations along the Zagros Piedmont have shown that, within the entire area, urbanisation does not necessarily seem to be connected to the formation of large or centralised sites, but rather to the establishment of a strong network of small centres interacting at different levels (on this issue, Boaz Bruun Skuldbøl and Colantoni 2016a).

84As sketched above, the EHAS region seems somewhat to follow this pattern, albeit to a lesser extent. Despite the absence of an explicit size-based settlement hierarchy, sites are scarcely randomly dispersed, but rather strictly aligned along old and new paths and trajectories. A certain degree of site interdependence is implied by the close proximity of some sites to one another.

85The E-W riverine penetration route along the Ḫabur is extended eastwards, connecting with the Tigris hinterland thanks to a mountain crossing over the Zagros, namely the Zaxo pass. The SW-NE trajectory along the Wadi Saru Kani clearly survives, but is now embedded in a wider network spanning the whole left bank of the Tigris. Here, between the Middle and Late Uruk period, a series of strongholds are newly founded. As of yet, twelve new sites can be counted, about 1/3 of the overall amount of settlements attested in the period. As suggested by pottery assemblages, some of these new sites were clearly inhabited by both local and Southern Mesopotamian oriented communities (living side by side in the form of “enclaves”, or perhaps succeeding each other), while others seem to display an exclusively Southern Mesopotamian oriented taste within their material culture.

86Given these premises, this new settlement dynamic along the Tigris seems to have been strongly affected and feasibly catalysed by the interaction between the expanding colonial trade network of the Southern Uruk city states and the polities of Upper Mesopotamia (according to models well explained in the past by Algaze 2008; Stein 1999, 2002b, 2012; also Rothman 2001) and, therefore, to be the outcome of direct contacts between lowland Mesopotamian communities and the populations set along the main routes between the Tigris Piedmont and Central Zagros.

87The location of most such foundations up above a riverine lower terrace suggests, in turn, a strong bond with an ancillary exploitation of the alluvium, which becomes from now on a prime agricultural resource, as well as the desire to connect with the hinterland, the latter being fulfilled by the enhancement of all those aforementioned wadi routes reaching the Zagros Piedmont. In the latter area, as already stated, a chain of sites with older LC1-2 roots probably involved in the exploitation of local resources and/or in the manufacture of lithic artefacts had still survived, albeit displaying instead throughout their life a purely local repertoire. Exemplary in this sense is site B147, which precisely at this time almost doubles its size, reaching over 6 ha in extension. It might be proposed that the pre-existing system of resource management acting at a regional level was enlarged and finally embedded in a supra-regional network of trade and exchange. All this reinforces that, during the LC3-5, sites of varying functions and roles co-existed: “local” sites deeply rooted in the territory; outposts set along a series of penetration routes providing a more complex material culture associating (spatially, temporally?) local and Uruk influenced material; and, finally, new foundations, mostly on the riverside, with a purely Uruk-related repertoire. All of these may have been in a structural and not discontinuous relationship, and therefore are potentially to be regarded as forming part of one and the same integrated complex spreading pervasively across the landscape.

88This multi-variate evidence of cultural encounter with Uruk Mesopotamia not only alters the picture achieved so far by current investigations in Upper Iraqi Kurdistan, but also engenders further unsolved questions which only future field research along the Upper Iraqi Tigris and Ḫabur rivers may bring to a solution.

89In understanding such a diversified panorama, the impact of temporality naturally remains a clue. A more precise chronological setting for the new foundations along the Tigris, in fact, becomes compelling in better distinguishing the role played by external forces in the above-sketched formation process. Whatever the initial impact of this might have been, worth stressing for the time being is that most of the LC3-5 sites, and especially the new riverine foundations and the Muqable cluster, provide intimations of a continuous occupation also in the terminal phase of the Uruk period and even later, in an age in which the Uruk trade network had collapsed and the true Uruk colonies of the Middle Euphrates and elsewhere had already perished.