and others within domestic areas. Unfortunately, no cemeteries have been found at
sites with late Early Dynastic private houses, and no private houses have been
uncovered at sites with cemeteries.
Smaller sitesThe late Early Dynastic period was a time of maximum urbanization within Sumer
(Adams and Nissen 1972 : 21 ). Small sites were therefore quite rare and no small
settlements dating to this time have been excavated. The overall organization of only
two non-urban late Early Dynastic sites^16 can be mapped with some precision with
high-resolution satellite imagery. One, Diyala Survey site 109 (Adams 1965 : 139 ) at 5. 5
hectares should be considered a small center and exhibits the same dense housing that
has been seen at the larger cities. The second, Nippur Survey site 1271 (Adams 1981 : 279 ;
see Figure 8. 5 ) is more interesting. Covering only one half hectare, it is surrounded by
a rectangular fortification wall, with its interior characterized by dense housing with
perhaps a central. It seems unlikely that this was typical of small late Early Dynastic
settlements in general. Other smaller late Early Dynastic sites, though their surface
traces are not particularly clear, nevertheless exhibit traces of dense architecture very
different from that seen on the West Mound of Abu Salabikh, suggesting that both
large and small late Early Dynastic settlements were comparable in their density of
domestic housing if not in their institutional complexity.
LATE EARLY DYNASTIC SETTLEMENTTo sum up the foregoing discussion, both large and small sites dating to the late Early
Dynastic period were characterized by high-density occupation with little in the way
of open space. The residential districts were characterized by houses built around
courtyards and sharing party walls strung along quite narrow streets and alleyways,
some with dog-leg bends in them. The differences in density between sites seem to have
resulted from the limitations in space imposed by the presence of city walls (Postgate
1994 ). Differences in crowding, however, are not manifested by the presence of empty
space; rather the less crowded settlements, like Abu Salabikh, had larger, more
sprawling houses whereas places like Khafajah had very small houses. In no case could
differences in crowding or domestic architecture be discerned between different areas
within even very large cities.
In addition to domestic areas, these late Early Dynastic cities included major
institutions of various sizes, mostly temples, many of which were embedded within
residential districts. Many of the cities were multi-mounded and, at least at Lagash and
Abu Salabikh, there is clear evidence that these were divided one from the other by
watercourses. In at least some instances, each mound seems to have been walled rather
than having a single wall surrounding the entire settlement.
It was during this time that secular rule evolved. In some instances, residential
structures were clearly connected to the main temple, such as the house associated with
the Temple Oval at Khafajah and the Palace A at Kish. But by the very end of the Early
Dynastic period–a time that might actually be contemporary with the earliest
Akkadian rulers (Gibson 1982 ; Gibson and McMahon 1995 )–palace complexes at Kish
and Eridu were built on separate mounds far from the main temple. In the case of
–– Elizabeth C. Stone ––