Durham Research Online Deposited in DRO: 09 June 2009 Version of attached le: Published Version Peer-review status of attached le: Peer-reviewed Citation for published item: Jaques, S. D. and Dobney, K. and Van Neer, W. (2003) 'Diet, economy and status : evidence from the animal bones.', in Excavations at Tell Brak vol 4 : exploring an Upper Mesopotamian Regional Centre, 1994-1996. Cambridge: McDonald Institute for Archaeological Research / British School of Archaeology, pp. 417-430. McDonald Institute monographs. (4). Further information on publisher's website: http://www.mcdonald.cam.ac.uk/publications/mcd-books/mcd-booksbydate2003 Publisher's copyright statement: Additional information: Use policy The full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that: • a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders. Please consult the full DRO policy for further details. Durham University Library, Stockton Road, Durham DH1 3LY, United Kingdom Tel : +44 (0)191 334 3042 | Fax : +44 (0)191 334 2971 https://dro.dur.ac.uk Chapter 12 Diet, Economy and Status: Evidence from the Animal Bones Keith Dabney, Deborah Jaques & Wim Van Neer Excavations at Tell Beak during 1994 to 1996 pro­ tebrate assemblage recovered from Tell Brak. duced a small but significant collection of vertebrate In this paper, and the accompanying tables and remains. These were recovered both by hand collec­ figures, phase numbers correlate with cultural peri­ tion and through the implementation of a s}'stematic ods and trenches as follows: sampling and recovery programme using both dry­ and wet-sieving techniques. The vast size of the Ph~ Cullur;ll pt'riod Trr:nch mound of Tell Beak,. and the inevitably limited physi­ I Early Uruk H56 II Middle Uruk H51 cal scope of archaeological investigations, brings into III Ninevite 5 H52,. H54, HF, HL question just how representative any of the exca­ IV Later third millennium HS3, H54, HS5 vated material really is of the changing aspects of V Early second millennium HN daily life and economy in the past. This concern is further compounded by the often small size of the Methods recovered assemblages. In terms of identifiable frag­ ments, the numbers ofbones recovered and recorded The vast majority of the hand-collected and dry­ from the three seasons at Tell Beak. were relatively sieved material was identified and recorded on-sile small: less than 1500 fragments from dry sieving and during the field seasons, using a small portable com­ less than 1000 from wet sieving. Once these quanti· parative collection. Those remains which could not ties are divided between the different excavated be confidently assigned to species, in addition to all trenches, and the broad range of phases present in the vertebrate fragments sorted from the wet-sieved each trench, individual assemblage size falls well residues, were exported to laboratories in Britain below 500 fragments (see Table 12.1). As a result, and Belgium where more detailed identification and much care, and indeed scepticism, must be employed analysis was undertaken. The mammal and avian when attempting to draw statistically valid and remains were identified using the comparative col­ meaningful interpretations from such small data­ lection of the Environmental ArchaeOlogy Unit, sets. Nevertheless, with these caveats in mind, some University of York, whilst the fish remains were com­ interesting observations can be made about the ver- pared to the extensive collection housed at the Royal Museum of Central Africa, Tervuren, Belgium. T;lble 12.1. Numbt, 0/ id~lllifillbl~frllgm~nls rtrordnl from hllnd­ co"~cltd lind dry-sWvt flSMmblllgN by tu'nch lind by phll~. Due to the short time available for the record­ ing of the hand-eollected and dry·sieved material II III IV V Toul during the field season, a limited suite of elements HF 3 3 was recorded for the most common species: sheep/ HFI 21 HF2 "6 6 goat, pig. cattle, gazelle and equid_ For these taxa. HL 18 18 the skeletal elements that ,..'ere included in the final HN 317 317 recording protocol were selected on the basis of their H51 371 371 archaeological visibility and usefulness, namely ease H52 183 183 H53 ,.7 267 of identification to species, representation of differ· H54 39 40 79 ent parts of the skeleton, and the availability of use­ H55 12 12 ful biometrical and age-at-death data. The protocol H56 48 48 for recording is available upon request from the au­ Tol;ll 48 371 270 37' 317 1385 thors. 417 Chapter 12 In the case of less common taxa such as wild It is clear from Table 12.2 that the most com­ mammals and birds, although few in number, all mon species recovered from the site are not surpris­ identifiable fragments were recorded where possi­ ingly the bones of domesticated mammals, primarily ble. As a result of this inherent bias in the data set, sheep and goat, followed by domestic pig and cattle. any comparative statistical analysis between species Although the majority of wild taxa, both birds and is restricted to the most common taxa mentioned mammals, arc present in much lower frequencies, above. the remains ofgazelle are more numerous than those of even equid and cattle. The major mammal species Figure 12.1 shows the frequency of the main mammals by phase, based upon the number of iden­ Frequency oftaxa lifiable specimens ( ISP). As can be clearly seen, the The total assemblage was recovered from deposits importance of remains of sheep and goat is highest which, in tenns of ceramicall)'-based and calibrated throughout all phases. In phase 11, however, their radiocarbon dating. span man)' centuries from late remains represent >90 per cent of all species recov­ fifth to earl)' second millennia Be. Once the verte­ ered (but of a tolal of only 371 identifiable frag­ brate assemblage is classified in terms ofchronologi­ ments), whilst in phases III-V, their proportion is cal date, dear differences are apparent between the reduced to between 50 per cent and 40 per cent. The relative frequencies of the most common mamma­ remains of larger domestic bovids never rise above lian taxa. 10 per cent for all periods (their highest frequency being in phase V), whilst the Tolblr 12_2. Numbt.,. ojid,·ulijiaIJI,· fragmrtrt~ r«ordcd from JrQlld-rollt'CtrJ and dry-s,rw IIsSfmh/agrs importance of domestic pigs It!! SI,,-ri.·s lIud by pl'II5/.'. is at its highest during pe­ riod IV, where their remains Spttin IV V TotiJl " '" are almost as frequent as Apodnnu!'/,\Ius sp. ~~ 2 , Mu~ muscutus houS(' moust' 1 sheep and goat. In terms of Lrpus c~rrsis CiJpe h4tw 1 wild game. the only species d. ,\ldlllvrQ C~T/StS ?honey badger/utel 1 present in substantial num­ Canidae dog famil)" 3 9 3 16 bers is the gazelle which, on d. VulpN sp. sm",11 fo:. 1 , 1 , d. Vul~ wlp6 ?fo:. 1 , the basis of overall size and rl"lrssp. c., 1 on hom-core morphology of cf.I'lllllh~rll Ifo ?hon 1 1 the few fragments recovered, Equidae horse/wild ass/donh')' , , 18 96 is likely to be the gnitred ga­ 5,/s f. domestic pig 7 18 149 55 230 Ccn'id;l" d<.'Cr , , zelle, Gazella sllbglltt/lrosa. ?Cl'r"id,,(' ?dL'('r Figure 12.1 shows lhat tht· Crn'lIs tlOpJrIlS roo deer , , remains of gazelle arc the &Ssp. CiJltll' , 13 9 69 second most frequent mam­ Gou-I/Il sp. S<17.elll' 5 <0 n "7 '0" 139 d. Gou-I/Il sp. ?ga7.e1l1' 3 3 mal recovered from phase III CQl'ril f. dornt'Stir goo' 3 19 22 11 79 deposits, representing >30 d. Cllprll f. domest;c ?goal , "20 9 <0 11 52 per cent of all major mam­ Ol'fS f. dOml'Stic sheep 9 65 <9 31' mal fragments, and also d. Ol"s f. domeslll:­ ?sha-p I , , <0 3 '''' " similar in frequency to cat­ CiJprine sht.--ep/goat 133 69 52 "339 sheeplgoat Iga7.('lh:­ " I "3 , 13 lie and pig in period V. Larg(' mammal I "1 A separate and argu­ d. HrllT/11l 1t'IlCOPS'S ?bamac1(' goose 1 I ably more representative d. Antl1ropoll!rs ",rgo ?dcmoisdl., crane 1 1 method of calculating the P/rroc/(s sp. sand grouse , <0 relative importance of spe~ Columbidae Piscon Idove famil}' 1 1 d. CoIl/mbll livjo ?rock dove 3 , des through time is by us­ COrl'll§ roTllX L raven 1 ing minimum number of individual (MNI) counts. Te:studo sp. lorlo;S(' This method reduces the bias Amphibi'ln of using fragment counts by TotiJl 37S 290 391 1426 only including bones, or " fragments of bone, which 418 Diet, Economy and Status NISP (t-bin m~m.millls) cannot be counted more than once. Figure J2.2 shows '00 ............ the percentage frequency of the three most common .5«,1._ taxa as represented by their MNI counts. Rather com­ 00 A ....... .~." fortingly. this method of calculation shows an al­ 00 ~ X e.,- most identical pattern of the relative importance of ". • ~ each species at Tell Brak through time to that noted 00 " using raw identifiable specimen counts (N1SP: Fig. M 12.1). Caprines (i.e. sheep and goats) arc once again , very common from phase U deposits.