of the same professional group, as Susan Pollock argues was the case for Early Dynastic
I Khafajeh (Pollock 1999 : 123 – 134 ).
Although the size of houses within the settlements varied considerably, none were
very large, ranging from around 50 to 250 square metres. Exceptionally large houses at
Khafajeh with areas of 356 and 528 square metres had unusual plans and were perhaps
not ‘normal’ homes (Delougaz et al. 1967 : 277 ). Wealthier citizens presumably were in
a position to build larger houses, while the poor were restricted by available space and
thus their dwellings were more tightly packed together. Dividing the floor areas of the
houses in Tell Asmar into three groups, Henrickson ( 1981 ) suggests that the smallest
units may have represented shops, the next size up housed single nuclear families and
the largest units were home to wealthy nuclear families or extended family groups.
Regardless of available space, the plans of houses would have changed constantly as, for
example, when homes were divided among brothers at the father’s death. Thus a large
building might ultimately house a greater number of inhabitants rather than reflecting
a family’s prosperity.
What might one have seen when visiting a home at Eshnunna in the middle of the
third millennium BC? The building would have been approached along a narrow lane,
in some cases no more than a metre wide. These tiny alleys were lined with blank
facades formed by the thick mud brick walls of adjoining houses; outside windows
were rare and public space was clearly unimportant. The visual privacy of domestic
dwellings was a theme repeated in most dwellings, apparently regardless of whether
they housed the rich or poor. The majority of homes had only one entrance and the
doorways were small with a width of between 45 and 150 cm (Delougaz et al. 1967 : 153 ).
Pivots of stone or baked bricks were placed at both entrance and internal doors, so
wooden doors could be swung from an upright pole, the bottom of which rested in the
pivot, usually below floor level; only a few pivots have survived in the archaeological
record as they were presumably reused.
In layout there was much continuity with the Uruk period; although pure tripartite
houses disappeared, the arrangement of a central room or courtyard flanked by rooms
persisted. The entrance to the central courtyard was usually through a vestibule and not
directly from the street. Homes were thus completely introverted, they looked inwards
towards the courtyard, making the house an entirely private space; it had no public role
to play. The courtyard was literally the heart of the building, a communal space and
an area where the occupants could work without being subjected to the gaze of
outsiders. Harold Hill, the excavator of Tell Asmar, argued that these spaces were
roofed (Delougaz et al. 1967 : 148 – 149 ) but his evidence was questioned by Delougaz
(Delougaz et al. 1967 : 275 ) and it is generally accepted that the courtyard was open to
the sky. They were small spaces, with the largest courtyard only some 7 metres square,
but they nonetheless provided the principle source of light and air to the rooms around
its sides. The open courtyard was part of the adaptation of the building to the climate,
designed to protect the inhabitants from the daytime heat and cold at night. At night
the courtyard, and the rooms leading off it, were filled with cool air, which during the
day was heated by the sun. The rising hot air provided cooling breezes. At the hottest
time of the day, shadow helped to produce similar breezes.
The courtyard allowed the rooms around it to be subdivided into smaller dwelling
units for different members of a family by providing a common equal route of access.
The layout of rooms does not indicate a strong gendered separation of space within the
–– Everyday life in Sumer ––