for knowledge or resources, some the emigration of people from the south, and some
emulation by local people of a desirable commodity, rather than close encounters with
the originators of the goods.
By the middle of the Uruk period, it is possible to identify a radical change in these
varied relationships with the founding of the first so-called colony site at Sheikh
Hassan on the Middle Euphrates (Boese 1995 ). Much has been written about the
elaborate system of so-called colonies which was in place by the late Uruk period
(Algaze 1986 : passim), of which Habuba Kabira and Jebel Aruda are perhaps the best
known examples. These sites were apparently intended to ensure a constant flow of
goods, especially metals and timber, southwards from the highlands of Anatolia into
the southern plain. Here, in the colonies, the material culture is identical with that on
the southern plain and seems to owe little to the area in which the settlements were
planted. In addition to the colony sites, there were also smaller way stations where
presumably caravans could rest and recuperate. Such sites have all the conveniences of
home, but show more evidence for local interaction. At Hassek Huyuk these facilities
include a small public building decorated with splendid mosaics. In the pottery corpus
there is also a ‘substantial component of indigenous late Chalcolithic chaff-tempered
forms’, suggesting a closer relationship with local people than at the colonies (Algaze
1993 : 86 ). Furthest from home there were small groups of people from the plain living
in enclaves in local settlements, presumably to facilitate the flow of goods to their
homeland. The best known of these enclaves is found at Hacinebi, while another
almost certainly existed at Brak in North-East Syria (Stein 2002 : 149 – 172 ).
Further north, where there is no evidence for enclaves, stylistic contacts with the
south can be recognised, sometimes in the glyptic and sometimes in the pottery. The
magnificent urban site of Arslantepe in eastern Turkey with its public buildings, wall-
paintings, sophisticated metalwork and elaborate accounting system has also produced
a limited amount of late Uruk pottery (e.g., Frangipane 2004 : fig. 37 ), and Uruk style
seals, suggesting a limited contact with the south, but the architecture and the
metalwork owe nothing to their southern counterparts and are equally sophisticated.
Even further north on the coast of the Caspian sea, recent studies have detected traces
of Uruk influence on the pottery (Munchaev and Amirov forthcoming).
Looking eastwards, there were also close ties between Susa and the south in the Uruk
period. By the late Uruk much of the material culture, perhaps most striking in the
iconography of the seals, is almost identical, but the exact nature of the relationship is
unclear and much debated. Differences in the shape of the proto-cuneiform tablets
from Susa and in the way signs are arranged both argue against the presence of
Sumerian bureaucrats (Potts D. 1999 : 52 ff.) and so of the actual conquest of southwest
Iran by south Mesopotamia. On the other hand, the pottery is Uruk in style. It also
seems unlikely that Uruk could maintain a full-blown military occupation of another
polity at this early time, but relations were undoubtedly close. The end of the Uruk
period, which coincides with the end of Susa II, saw a complete change in the material
culture of that city, marking an abrupt change in the relationship.
By contrast, relations with the Gulf seem to have declined later in the Ubaid period
and this region does not seem to have had a significant relationship with Uruk
Mesopotamia. A few pieces of Uruk-related pottery have been found in the Eastern
province of Arabia (Potts D. 1986 ). There are also scattered references in the texts to
Dilmun, which also appears as an element in a few personal names (Potts D. 1990 :
–– Harriet Crawford ––