Archaeologists have noted a number of recurrent characteristics in the overall physical coniguration of urban settlements across northern Mesopotamia in the mid-late third millennium. Stone (2000:243244) describes the general patterns evident in the spatial layout of cities in this region. She states that a typical northern city of this period contains a high mound from which the lower town spreads in different directions. As the highest point of the settlement, the high mounD is usually the location for the main temple, while an extensive lower town is largely given over to residential sections. The lower town tends to be circumvallated with defensive walls, at one point along which the high mound can be found. Palatial structures may also be found along the fortiication wall, but such structures are probably within the lower town at some distance away from the religious precinct on the high mound.
This pattern is generally repeated among the contemporary large settlements in northern Mesopotamia and the region to the west of the Upper Euphrates, particularly Taya (Reade 1973, 1982, 1997), Leilan (Weiss et al. 1990, 1993), Hamoukar (Colantoni and Ur 2011; Gibson et al. 2002), Beydar (Lebeau 2012; Lebeau and Suleiman
2011) , Chuera (Dohmann-Pfalzner and Pfalzner 1996; Pruss 2000), Mozan (Buccellati 2005; Buccellati and Kelly-Buccellati 1998; Pfalzner
2012) , Brak (Emberling and McDonald 2003; D. Oates, J. Oates, and McDonaLd 2001; Ur, Karsgaard, and Oates 2011), Kazane (Creekmore 2010; see also Creekmore, Chapter 2 in this volume), Sweyhat (Danti and Zettler 1998, 2002, 2006), and Mardikh (Matthiae 1981, 2010) (Figure 3.1). All of these cities, except Brak, are walled, and the settlement size ranges between approximately 30-160 ha during this time, with the average site size of ca. 80 ha.2 Some of these cities are roughly oval and oblong (Leilan, Kazane, Mardikh, Brak, Hamoukar, and Sweyhat), while some others form a circular shape (Chuera, Beydar, and Mozan).3 These circular sites, known as Kranzhugeln ("wreath-mounds"), are also a common type, particularly seen between the Upper Khabur and Balikh regions as well as in northeast Syria (Akkermans and Schwartz 2003:256; Wilkinson 2000:239). Kranzhugel sites hoLd the citadel mound at the center, and the lower town surrounding the citadel mound is in turn circled wIth the outer fortification wall.
Although a single high mound is a consistent element of these northern urban centers,4 its location can be either near the center of the mounds (Beydar, Chuera, Kazane, MarDikh, Mozan, Sweyhat, and Taya) or at one end of their larger settlements (Brak, Hamoukar, and Leilan). The high mound was frequently fortified with an inner wall (e. g., Beydar, Chuera, Leilan, Mozan, Sweyhat, and Taya) demarcating the high mound from the rest of the settlement during this period. The size of the high mound as a proportion of the total community size varies from one site to another. At Mardikh and Taya, the high mound occupies only 3-5 percent of the total occupied area, whereas more than half of the site is comprised of the high mound at Brak and Chuera. The size of these high mounds ranges from 3 ha to as large as 43 ha with an average size of approximately 17 ha.5
The high mound was the primary location for not only main temples, but also royal and governmental residences, administrative buildings with storage facilities, and elite houses that were likely
Built around these public buildings. Whereas main temples tend to have been incorporated within a multiroom, 'palace-temple' complex in the southern Mesopotamian tradition, temples in northern Mesopotamia frequently stood independently. Temples excavated at these northern sites (e. g., Beydar, Brak, Chuera, Mozan, and Taya) are located at the highest point of the settlement, on the high mound. As the main temple was recently discovered at the southeastern edge of the lower town at Mardikh, it is possible, however, that major temples also existed in other parts of the community, but have not been identibed because the excavations at these large settlements have focused on the areas on or near the high mound. The elevated location must have meant that the main temple was the most prominent landmark of their landscape, for both the inhabitants and populations outside the city proper. At the same time, the availability of space on the high mound seems to have dictated the physical scale of the main temple, as well as the locations for secondary temples, if any existed. For example, the temple at Mardikh was constructed away from the small high mound (ca. 3 ha) which was already dominated by the presence of the Royal Palace G. In contrast, the large (ca. 43 ha) high mound at Chuera provided enough space for the construction of the royal palace as well as of the main temples, allowing some distance between these massive structures.
Secular public buildings - including palaces, governors' houses, and administrative structures - are increasingly seen at these urban settlements. A mid-late third millennium palatial structure has been securely identified at four northern sites (Beydar, Chuera, Mardikh, and Mozan), anD Akkadian administrative buildings have been found at two other cities (Brak and Leilan). Many of the palatial buildings found at these sites were rectilinear with large courtyards embedded among smaller rooms. As with main temples, these structures are all located on the high mound, at its periphery (in many cases, the western periphery), rather than at the central summit, keeping some distance from the main temples.6 It is also possible, or even likely, that more than one secular public structure existed in some cities, and that not all of these buildings would have been accommodated within the limited space of the high mound (Creekmore 2010; also see Creekmore, Chapter 2 in this volume). When secondary and tertiary secular buildings were to be constructed, there may have been a spatiaL limit on the high mound, which was already crowded with preexisting ceremonial and royal structures. This
May be precisely the reason why, by the first half of the second millennium BC, palaces and other major public buildings are increasingly seen within the lower towns at north Mesopotamian cities (e. g., Leilan and Mardikh). At these sites, many other secular public buildings or possible public buildings with unclear functions have been excavated from third-millennium contexts. Many of these possible public structures are administrative in nature, and they appear to have been clustered also on the high mound (e. g., the "Unfinished Building" at Leilan, the "U-Shaped Complex" and "Bi Building" at Beydar, and the "Kitchen Building" at Sweyhat).7
All of these regional centers contained vast residential quarters. These areas tend to have been concentrated in the lower town and filled with domestic structures, streets, and side alleys. Most structures stood contiguously on both sides of streets, often sharing party walls (e. g., Beydar, Chuera, Hamoukar, and Leilan). Even though some houses were built away from each other, leaving open spaces between them that were largely devoid of other structures (e. g., Taya), most of the examples - extensively excavated settlements, in particular - show that the densely crowded environment was the more common phenomenon. Streets were usually straight and sherd-paved, ca. 2-5 m in width. These streets provided direct access to domestic buildings (e. g., Chuera and Taya), whereas many houses could only be accessed via side alleys (e. g., Hamoukar and Leilan). Many of the dwelling structures in these urban settlements seem to have been the central-courtyard house type with varying sizes and numbers of rooms. The average house sizes are estimated for the Chuera houses as 63-147 m2, and for the Taya houses as 130-178 m2 (Pfalzner 2001:399), whereas the majority of the houses in these cities fall between 127-205 m2 (Colantoni and Ur 2011:36). Thus, we have steadily been building the large picture of the general city layout in the north Mesopotamian sites, but our ability to reconstruct socio-spatial patterns for the bulk of the population in the extensive residential quarters is still highly limited. The analysis and interpretation of the rich data from Titris help ill this gap and illuminate Hitherto underrepresented aspects of city life.
World History
