while sheep/goat, cattle and pigs were the main species consumed. A few ‘hold-out’
sites such as Tell Bderi (Becker 1988 ) and Tell Beydar (Pruβand Sallaberger 2003 / 04 )
maintained a higher degree of hunting in EJ I through III that may be related to their
locations near basalt desert or migration routes. Within domesticates, there was also a
shift from a wide range of species to greater specialisation on sheep and goats across the
third millennium BC. But there was, as in botanical material, sub-regional variation.
Pigs made up a fifth to a quarter of the faunal assemblage at EJ III–IV Brak and Leilan
but 1 per cent or less at contemporary Beydar and Chuera (Emberling et al. 1999 ; Pruβ
and Sallaberger 2003 / 04 ; Zeder 1998 ). At Raqa’i and Atij, pigs were 20 – 25 % of the
assemblage in the earlier third but declined to 1 – 2 % by the mid-third millennium
(Zeder 1998 ). Other animals varied accordingly: Brak had a relatively high percentage
of equids, Beydar had more frequent cattle, Chuera had more sheep/goats (Pruβand
Sallaberger 2003 / 04 ). These variations and temporal shifts may be related to different
contexts (pigs are particularly suited to households) or adaptation to micro-environ-
ments. Texts from Ebla indicate that specially bred equids from the Upper Khabur were
a valuable commodity (Sallaberger 1999 ). The frequency of equid bones at Brak may
reflect breeding there, while the Beydar glyptic’s images of equids and carts offers
oblique support, as does the more concrete evidence from administrative texts, which
record fodder for donkeys and rations for cart makers.
CLIMATE CHANGE
Past Mesopotamian climate changes are currently a topic of intense debate. Some
climatic and geomorphological data indicate an aridity increase in the terminal third
millennium BC(Courty and Weiss 1997 ; Cullen et al. 2000 ; De Menocal 2001 ; Weiss
1997 , 2000 a, 2000 b), but the data are not straightforward (Bottema 1997 ). The dating
of this aridity increase and its chronological relationship to archaeological levels and
political history remain blurred. The lake or deep water cores, ice cores and cave
speleothems in which this climate change registered come from a distance; climatic
data currently known from within northern Mesopotamia are minimal.
Human reaction to climate change is also debated, and particularly the degree of
political collapse and settlement pattern reorganisation in Subartu in EJ V is not clear.
Some sites’ occupation persisted while other sites shrank and some were abandoned;
strategies such as nomadism are not collapse but instead are successful adaptations; the
concept of collapse itself and its political, economic and social implications can be
deconstructed (see papers in Kuzucuoglu and Marro 2007 ). Indeed, close analyses of
data reveal that there were earlier strong fluctuations of aridity and temperature within
the third millennium, that is, 2600 – 2400 BC(Kuzucuoglu 2007 ), and that the Upper
Khabur in the third millennium BCwas already drier than in the fourth and under-
going changes in vegetation and river regimes (Courty 1994 ). The Upper Khabur’s
original ‘oak park woodland’ was already thinning and receding at the beginning of the
third millennium, perhaps linked to herd animal grazing and human over-exploitation
of timber for construction and fuel. Settlement patterns around Beydar and Tell al-
Hawa were disrupted in the early Akkadian period (EJ IV; Ball and Wilkinson 2003 ;
Wilkinson 2000 a), rather than the post-Akkadian EJ V. The impacts of human
occupation, agriculture, ever-expanding animal herds and urbanism are difficult to
disentangle from larger-scale climate change.–– North Mesopotamia ––