Chapter VI HERDING IN PERIOD VIA. DEVELOPMENT AND CHANGES FROM PERIOD VII Ldszld Bartosiewicz Introduction This chapter is aimed at identifying changes in the exploitation of and attitudes toward animals as societies became increasingly complex during Late Chalcolithic. The settlement of Arslantepe is located in the SE section of the Malatya plain, Eastern Anatolia, approximately 15 km from the Euphrates on the right bank of the river. The focus of research is a transitional period between the presumably more egalitarian, subsistence-based systems of Neolithic villages and the development of hierarchical societies during the Late Chalcolithic and Early Bronze Age in the region. Signatures of social differentiation are shown in the variability of archaeozoological evidence for meat consumption and, indirectly, production. Issues of socioeconomic complexity will be examined on the basis of meat provisioning. The degree of redistributing meat supplies will be generally assessed as a reflection of centralized husbandry. Mass herding of sheep and goat have stereotypically been identified as characteristics of centralized urban economies throughout Southwest Asia. The same trend has been identified at Arslantepe1. The question posed in this paper is how we can revise and refine details of this gross tendency in various parts of the site. The picture drafted will inevitably be general, synthesizing trends in animal exploitation between" the mid 4,h to the early 3rd millennium BCE, widely thought to parallel events in the Late Uruk civilization of Mesopotamia. Animals are indicators of change in subsistence that must have been interrelated with, or in fact, may have triggered shifts in social organization. Material and Methods Archaeozoology at Arslantepe was started by Sandor Bokonyi (1926-1994) in 1979. He visited the site regularly and accumulated a major body of data. This work has been continued by this author. The analysis is based on animal remains from the final Late Chalcolithic Period VI A (Late Uruk. 3350-3000 BCE), in comparison with the Late Chalcolithic Period VII Bokonyi 1983. 119 Economic Centralisation in Formative States. Chapter VI Studí di Preistoria Orientale 3 (3800-3350 BCE) immediately preceding it. Aside from diachronic differences in meat consumption, various contemporaneous areas have been distinguished by settlement function. The material was grouped by rooms within the major areas/chronological units investigated. Early materials (Period VII) In his 1983 article, Bokonyi published animal remains from Period VII uncovered in the northeastern sector of the tell. Materials from common households (A 11, A12, A21, A22, A23, A24. A25) are known only from this part of the site. Moreover, the Period VII sequence in the western elite residential area included a monumental building with columns (the earliest feature so far excavated) and elite residences (A582, A617, A646,A647, A648, A682, A684). Intermediate levels were distinguished between layers of the column building and the levels of subsequent Temple C (and related rooms) within the western elite area. Features in these levels (A580, A581. A618) were partially preserved domestic structures from. The end of Period VII is represented by Temple C and long rooms (A563, A571, A842, A848, A850, A858, and pit K680) behind it. possibly used for storage or in crafts related to activities in the temple itself. Late materials (Period VI A) For the main period under discussion here, animal remains were analyzed from high-status residential quarters and public spaces on the SW side of the mound. In Late Chalcolithic 5 Period VI A. research focused on the "palace complex", including Temples A and B. Bokonyi first analyzed bones from Temple A (A42. A36, A46. A77), then from store rooms A340, A364 and A365 on either side of a reception niche adjacent to the temples. Of these, A340 is of special interest as it yielded a concentration of cretulae (clay used for sealing), closely associated with large storage vessels which they had once officially closed. A smaller sample of animal remains was analyzed from room A206 in the Period VI A palace complex, where thousands ot cretulae used as storage jar sealings were likewise discarded. The significance of these deposits, as well as additional provenances with concentrations of fragmented cretulae is that they may have been related to redistributing provisions. Additional deposits of discarded broken seal imprints from storage jars kept in the palace provisioning areas (A77. A206 and A4j0) were also studied. Animal bones from two rooms (Al 13 and A127) in a representative minding (111) have also been analyzed. No Period VI A common households were discovered: the Late Uruk Period was represented only by elite residences (A650.A651. A727. A728,A734,A747, A933.A946) in the investigated area. Material rom these dwellings may be compared to animal remains from the palace complex as well as w.th those from households of more ordinary Period VII folk on the other side ot the tell. rhe aim was to detect differences diagnostic of or influenced by a central elite, intervening in, nealing and provisioning practices. Thus, an enclave of utmost administrative importance could be delineated ,n the Late Uruk town, dated to the time when Arslantepe became a major u-m /h'f m a" emerging ^ntralized system2. The spatial patterning of faunal remains across architectura features and public spaces will be used in illustrating diachronic shifts in Bmn fw iT f,at'rcation- Duri"g the discussion references will be made to the Early Bronze Age I (Period VIB. 3000-2750 BCE). when pastoral economy dominated the picture. i-ran 'gipane. Palmieri 1988-89. Recording and quantification All materials were hand-collected. Sporadic finds of bird and fish could not be identified due to a lack of collections, but offer evidence for fowling and fishing. The finds were recorded in terms of the number of identifiable specimens (NISP) as the least manipulated form of information. Given the complex depositional history of the site, minimum numbers of individuals (MNI) were not calculated, as the disarticulation and redeposition of carcasses could not be followed even on a macro-level. Therefore aggregation effects might emerge: single individuals may be represented in several provenances, contributing redundant information to MM estimates. Given sufficiently large sample sizes, taxonomic compositions tend to display comparable proportions regardless of method. Appendices 1 to 6 show NISP values assigned to each taxonomic group. Due to osteological similarities between certain species and fragmentation that left many pieces without diagnostic-features, identifications were sometimes possible only on higher taxonomic levels, usually families or genera. In such cases a species name based on uncertain identification would have been misleading, resulting in the "over-identification" of non-diagnostic bone fragments. The best example of this standard is the subfamily name Caprinae, commonly used in archaeozoology to describe non-distinguishable bone fragments of sheep and goat. Based purely on size, the wild forms of these two species also occurred at Arslantepe, compounding problems of identification. One of the problems in this study has been different sample sizes. Percentua! proportions popular in faunal studies are misleading when not supported by sufficient numbers of cases. Differences in percentages were thus tested for heterogeneity using Chi2 tests. Bokonyi (1983: 581) cited only the small number of complete long bones (20 of 3880) as a proxy to fragmentation. Non-identifiable bones were also counted in recent analyses as indicators of preservation. Chances of identification radically decline with fragmentation. Non-identifiable bones were classified by size as large (cattle size) and small ungulates (small ruminants and pig). Previous calculations at the site have shown that the mean length of non-identifiable large ungulate bones equaled that of identifiable caprine remains. Naturally, the latter produced even smaller non-identifiable specimens3. In the material under discussion. 20% of the remains could only be identified by size category. Sheep and goat tend to dominate NISP values. However, the bones of large ungulates (more damaged by intentional and natural fragmentation) occurred in greater numbers among the non-identifiable (table VLI). In spite of different sample sizes, the heterogeneous distribution (Chr=l 18.085 d.f.=8. P=0.000) of non-identifiable elements across provenances shows that as the combined effect of butchery, pot-sizing and post-depositional taphonomic events (trampling, leveling, etc.). the Table VI .1 . The proportion of identifiable and non-identifiab.e animal remains. Temple C Storage/Temple C levi Period Vll, intermediate levels Temple B Period VI A elite residences Total NISP 497 Large ungulate Small Total ungulate Non-identifiable % 562 1953 140 1327 506 4423 306 _36 136! 2395 ! 1.6 296 3 166 171 1794 26.« 38 j 692| 24 ; 370 568 5485 10.9 19.4 Bartosiewicz 1998: 227. table 3. 120 121 Economic Centralisation in Formative States. Chapter VI Studí di Preistoria Orientale 3 proportion of non-recognizable bone fragments strongly varied between areas of different functions. DlACHRONTC AND FUNCTIONAL VARIABILITY IN TAXONOMIC COMPOSITION Animal remains will be reviewed in light of their economic importance and cultural meanings. Faunal lists, detailed by provenances are listed in Appendices 1 to 6. Domestic animals In both Periods VII and VI A, the majority of meat was provided by domestic ruminants - cattle (Bos taunts L.. 1758), sheep (Oris aries L., 1758) and goat (Capra hircus L., 1758) -and to a lesser but diagnostic extent by domestic pig (Sits domesticus Erxleben, 1777). The contribution of these species ranged from 90% in Period VII, to 98.9% in Temple A in Period \ I A. Not even a higher value would be indicative of subsistence hunting, although the decreasing contribution of game to the diet is significant in statistical terms (Chi2= 159.964, d. f .=9, P=0.000). Fig. VI. 1 shows the ratios of bones from domestic ungulates by provenance. As has been pointed out before4, the main trend in this graph is the diachronic decline of pigs, concentrated in the common households and elite residences of Period VII. It was suggested that pigs in the Levant had been exploited mainly locally when no central authority interfered with herding practices and self-reliance was more prominent5. At Arslantepe pork consumption was rare in public areas: it seems to have been concentrated in homes. This confirms the stereotype of self-sufficient households: pork possibly came from animals kept around the house. Sporadic pig bones were also found in Period VIA Temple B and elite residences. They were, however, missing from Temple A and administrative areas. Cattle also contributed the smallest proportions to food refuse in cmiilae dumps and the representative building from this period. Due to the possibly differential fragmentation of these bones this difference is more difficult to interpret. It onlv shows that pork was least important in areas of administrative functions. Sheep and goat are not shown in fig. VI.1 separately. Due to the intensive fragmentation, only a fraction of caprine bones could be assigned to species: diagnostic features are easily lost when bones are poorly preserved. Table VI.2 shows the NISP values for sheep and goat by provenance. Table VI.2 - The ratios between identifiable bones of sheep, goat and caprines. ; VII elite i \ li common ' ^ ' A£litc residences LJjM£!e_C^ [Temple B Not Sheep 81 34 164 19 32 44 Goat 154 35 92 10 15 included in calculations due to small sampl Caprine 996 289 1739 290 310 Sheep/Goat 0.5 0.9 e size. Bokom i 198V Frati»imnP <• Hesse !'»!. á!pane- S,rdcusano "998: Bartos, levvicz 1998. 122 N of identifiable specimens, % 25 50 75 100 representative building III cretulae dumps Temple A VI A store rooms Temple B VIA elite residences storage Temple C level Temple C VII Intermediate levels VII elite residences VII common households Chi2 = 1559.037 d.f.=16 P=0.000 22 236 271 111 555 308 117 361 69 49 997 169 1995 319 0* J 31 1 r 1821 841 345 1339 □ 16 IE 33 ID 12 ID 83 384 64 TD 25 ED 1243 358 (NISP) in various samples. Hntries marked b> an asterisk could not be included in^C^^due to the small number .N.SP<5, of p,g Fig. VI.I - The percentages of domestic ungulates (? ii» bones. Goat remains decline through time and from residential to public areas in a statistically significant manner (Chi-156.296, d.f =9, P=0.000). It may be argued that goat, a species of superior milk production, are more typical of self-sufficient households than large scale herding operations. Reality, however, may have been more complex: here only the trend can be established. Various equids identified at the site were unlikely to have contributed to the meat diet. Period VII domestic horse (Equus caballus U 1758) must have been tar too precious tor everyday meat consumption, although horse remains began sporadically occurring during this period6. The wild or domestic status of horses is difficult to judge, however, due to the small number of finds, equally characteristic of rare game or a high status, non-meat purpose domesticate. The first convincing date for horse domestication has just been reported from the Eneolithic Botai culture of Kazakhstan, dating to about 3500 BCE that corresponds to Period VII7. r ,,--«,.,, Bokonyi himself pointed out that the presence of domestic ass (Equus annus L I />8:, he had identified was unexpectedly early for Arslantepe8. Subsequent discover.es of bones from ' Boessneck. Von den Driesch 1976: 81. Outram et al. 2009'. 1335. ! Bökönvi 1983: 589. 123 Economic Centralisation in Formative States. Chapter VI Asiatic wild ass (Eqiuis hemiomts Pallas, 1775) at this site raise the possibility that he also encountered a metacarpal from an unusually gracile individual of this latter species. Dog (Canis familiaris L., 1758) occurred sporadically, the small numbers of their remains among the food refuse do not seem to indicate cynophagy. Marks of dog gnawing are more common in Period VII. a possible sign of dogs living in closer proximity to humans than in urbanized settlement. The presence of dogs may also have been related to hunting, probably more commonly practiced by the early inhabitants of the site. Non-domesticates and taxonomic richness By the Bronze Age, hunting lost economic importance at urban settlements in the Near East, in contrast to the Chalcolithic when it was still a major source of animal products at some sites". Various deer, gazelle, and wild ass were regularly preyed upon, but their remains occur in small numbers at Arslantepe. Wild animals, however, have contributed a very important qualitative aspect to the reconstruction of daily life. It would be erroneous to discuss all "wild" animals as a homogeneous unit: they include bones from large, meat purpose game, a variety of carnivores, small. commensal mammals as well as bird and fish remains. Structural differences between these bones as well as their different numbers in various animals affect their taphonomic histories10 thereby making direct comparisons to large mammals meaningless if not misleading. The abundance of mammals exploited was studied as a function of sample size. The number of taxa identified increases with the number of identifiable bone specimens (NISP). Therefore comparisons between samples of different sizes can be strongly biased, as the repertoire of species would reflect the number of bones available for study rather than the culturally idiosyncratic number of taxa sought by the analyst11. Although there is a usually high correlation between these two variables, the relationship is not linear: the number of taxa increases in a degressive manner and is exhausted when new species are no longer encountered. Therefore this relationship is shown in decimal logarithms in fig. VI.2 in which relatively rich and poor samples may be identified regardless of size, on the basis of their position relative to the trend line fitted onto all data points. In fig. VI.2. provenances from Period VII do not deviate from the main trend, in spite of the tact that, hypothetically. hunting must still have been important. Elite residences, Temple C, as well as the small intermediate sample in-between look ordinary. The number of taxa (showing a major contnbution of wild animals) is only slightly higher in common residences and the storage area near Temple C. This suggests that most wild animals had no special status value, their remains equally occurred in elite areas and mundane households. The situation became more polarized by Period VI A. While Temple B, storage areas and cretulae dumps fit the trendline (i.e. their numbers of taxa are in proportion with sample size), elite residences as well as the representative building seem to be rich in species. Meanwhile Temple A falls short of the fauna! diversity expected. An increasing differentiation in Period VI A space use may be related to increasing social complexity reflected in meat consumption. Some, eating m e.te residences and using the representative building had access to a variety of animal ST' rJ th°Seeating in TemPle A especially, consumed a limited variety of animals whether by economic standing or as a means of self-expression Studi di Preistoria Orientale 3 815t O © §■10 7- Temple C VII VIA VII stora9e elite repr. elite ^common bdg.lll^ '0 intermediate Temple C A Temple A " VIA storage and Temple B cretula dumps 0.171 y = 3.869 x R2= 0.430 100 1000 N of identifiable specimens 10000* Fis. VI.2 - The position of various samples in terms of taxonomic richness relative to the trendline (both axes were converted into decimal logarithms to produce a linear trend). Black symbols: Period VII. Light svmbols: Penod \ I A. Herbivorous game When discussing the Chalcolithic at Arslantepe. Bokonyi emphasized the possibility of local domestication'2, as he identified the ancestors of all main domesticates including aurochs (Bos primigenius Bojanus, 1827), Asiatic moufflon (Ovis orientalis Gmelin, 1774). bezoar goat (Capra aegagrus L., 1758), wild pig {Sits scrofa L.. 1758) and even wolt (Cams lupus L., 1758) among the carnivores. Chalcolithic animal husbandry, however, was firmly established. Tedious domestication efforts would have been unnecessary at this site. While bones of s.zes transitional between the wild and domestic forms are present'1, some may result from spontaneous crossings between the two forms. Over a decade ago. Bokonyi (1993: 337) saw the complexity of local cattle domestication in the absence of large aurochs horn cores at Arslantepe. Meanwhile phenotypic size alone turned out to have been deceiving in light ot recent DNA research. In spite of "transitional forms"', the local domest.cation ot aurochs during the Late Neolithic of Central Europe was refuted: most cattle tested were ot Near Eastern origins, unrelated to the local aurochs14. Bokonyi noted the high ratio of horn cores from adult rams and bocks (in the case ot wild sheep the male to female ratio is 6:1 and among wild goats only 9 male horn cores occur) . This secondary evidence for local domestication can be interpreted differently Should hunting have been a form of self-representation for the elites, trophy hunting may have b,ased the Oason. Buitenhuis 1998. Barttisien-ie/. Gái 200". (iravson 19X4: 136-137. ,: Bokonyi 1993a: 128. 13 Bokonyi 1993b: 357, figs. 3 and 4. 14 Edwards et al. 2007. 15 Bokonyi 1993b: 355. 124 125 Economic Centralisation in Formative States. Chapter VI Studí di Preistoria Orientale 3 sample, as capital males would have been selectively targeted. E.g. the dense, sharply-pointed horns of bezoar bucks make desirable trophies, a strong motive behind poaching even today. Red deer (Cervus elaphus L.. 1758) contributed almost half of the wild animal remains at Arslantepe16. although this ration is somewhat biased by the inclusion of fragments from potentially shed antler. However, the majority were skeletal bone, showing the popularity of venison. Red deer was also the most common game at comparable site of Korucutepe17. Remains of Persian fallow deer (Dama mesopotamica Brooke, 1875) were identified in small numbers and those of roe deer (Capreolus capreolus L.. 1758) are outright rare. Even during period VI A. there were coniferous woods around Arslantepe18. However, as forested red deer habitat were lost to pasturelands, sporadic occurrences of probably goitered gazelle (Gazella : 0.159 i Temple A 0.394 0.616j_ -0.1 S6: -0.179 1 VII elite residences 0.255 0.032 1 -0.093 i -0.095 j Temple C 0.188 0.006 ! 0.004 j 0.102 ' Storage Temple C level -0.069 -0.011 i 0.196 : 0.026 i VI A elite residences -0.229 •i>.0~i>; 0.208 i Store rooms VI A -0.464 -0.412! -0.148 [ 0.206 j VII intermediate levels -0.497 ^^767j_ 0.595; ---f -0.530 i Cretulae dumps -0.542 -0.327! -0.149: 0.00" ' _.-1 Representative Building III -0771_ -0.592 _ 0.495 \ ■0.23S ; Temple B __JK79XX 0.03S: -0.010 i With the exception of the intermediate levels. Period VII samples f^^^ fig. VI.3: B (marked by shading), determined by the co-occurrences of domestic p,g and w,ld Bartosiewicz 2005a: 53-55. -I! Frangipane. Siracusano 1998: 242. 21 Helmer 2008: 173. - Izbirak 1976: 131. 127 Economic Centralisation in Formative States. Chapter VI Studí di Preistoria Orientale 3 0.6 CO CM IN to < 0.0- -0.6 H hare -1.2 ff 9°sfO °w capnne -0.6 0.0 Axis 1: 56.5% pig O red deer ff "• Tri o.s-i 0.0- -0.64 Temple B cretula , dumps' Temple C storage • VIAÄ eilte» K VIA 9 storage Temple A VII elite VII common Temple C repr.bdg. Ill# intermediate # levels 0.6 B -1.2 -0.6 0.0 Axis 1: 56.5% 0.6 I .j!. VI.3 - Results of correspondence analvsis showing relationships between diagnostic animal species (Graph A) and sample ongmutmg lr«n different provenances. Graph B shows two strong polanzations: from pig to a sheep/aoat ccoiio . i ----- ^uuiig pv^itti i/.aiiuii.l. 1 tUUl pi li IVJ a Slit ■m> ftrom nght to left) and a less marked '"upward" shift from woodland to grassland habitats. ruminants. This coherent picture -'explodes" as a likely result of socioeconomic developments impacting on meat consumption in Period VI A, paralleling functional differentiation in the studs area. While elite residences of this later period remain closest to the norm represented bv Period VII provenances, others (especially representative building III as well as Period VIA dumps and storages) are scattered toward the "goat" direction. The absence of pig and red deer in Temple A and an increase in hare bones in Temple B sets these samples apart from mundane assemblages. Period VII Temple C was remarkably different. In the main cult room (A42) in Temple A there was a wild boar mandible near the altar/basin on the floor23. This as well as the remains of wild animals recovered from Temple C are indicators of the roles wild animals played in ceremonial activities. Carnivores As a consequence of the degressive relationship between assemblage size and taxonomic richness, rare bones of brown bear (Ursus arctos L., 1758) and lion {Panthera leo U 1758) have a greater probability to occur in large samples such as the "temple" assemblage of room A4M) and the general pooled sample-^. Although one may hypothesize that some bear and lion teeth and metapodium fragments were attached to prestigious skins worn or displayed otherwise in public spaces, their discovery is also a product of large sample size. Bear bones appear in the less distinct "general" areas as well during Period VII (such as elite residence AM > a reminder that their meat was consumed* During this period meat bearing bones 83. ' Bokern Í 1983: 590. at Arslantepe but throughout the Upper Euphrates Valley, e.g. at Norsuntepe2h. Tepecikr and Korucutepe28. Brown bear, a top predator reflects the state of ecosystems as a whole. Nowadays, it mostly occupies mountain forests, but also occurs in plains woodlands. It is there where bear is most threatened by habitat loss due to deforestation. The reduction of its quantitative contribution from a food species to probable trophies by Period VI A mirrors the process as lions went extinct in the Balkans between the Copper and Bronze Ages29. That time, lions must have been around in Anatolia, although their remains occur scarcely as they may have been seldom hunted. Wild cat (Felis silvestris, Schreber 1777), red fox (Vulpes vulpes L.. 1758) and various mustelids, including weasel (Mustela nivalis L., 1766) may have been of smaller symbolic significance. In fact, the bones of burrowing foxes may form secondary deposits in archaeological strata. The weasel skeleton recovered from room A850 associated with Temple C may also be considered an intrusion. Weasel remains occur regularly at archaeological sites in Anatolia30. Commensal animals A small group of partially identified remains seem to originate from animals sharing human habitats but not introduced intentionally into the archaeological record. Due to the lack of water-sieving, rodent (Rodentia) bones were few and far between. They could not be identified in relation to the archaeological record. A far more peculiar situation of archaeological relevance was observed in the case of remains that may be tentatively identified as bones from commonly occurring greater horseshoe bat (Rhinolophus cf. ferrumequinum Schreber, 1774) widely distributed both in Anatolia and SE Europe31. In situ skeletal remains of these animals were found on the floor of Temple C. The largely intact bodies of these delicate animals may have been deposited at a time when the roof of the building was at least partially still in place, but human trampling in the (presumably abandoned) building was minimal. This simple taphonomic phenomenon offered a glimpse at the building's own taphonomy. Among the poikilotherm animals identified at the site, the sporadic bones of frogs toads (Anura) and Greek tortoise (Testudo graeca L., 1758) may also be considered natural deposits, although according to Bokonyi some tortoise finds were contemporaneous with the archaeogical levels as some carapax fragments were cut and sometimes charred32. Birds and aquatic animals In the absence of water sieving, the bones of large species dominated the small assemblage of avian remains. The bones of great bustard {Otis tarda L.. 1758), white pelican (Pelecanm onocrotalus L., 1758) and some eagle (Falconiformes) represent different habitats, largely in the plain/floodplain area north of the site. The remains of perching birds (Passeriformes.) 26 Boessneck, Von den Driesch 1976: 98. 27 Boessneck, Von den Driesch 1973: 114. 28 Boessneck, Von den Driesch 1975: 142-143. 29 Bartosiewiez 2008: 768. 30 Boessneck 1973. 31 Bilgin et al. 2009. 32 Bökönyi 1993b: 355. 128 129 Economic Centralisation in Formative States. Cliapter VI Studi di Preistoria Orientale 3 including crows were also found. The majority of bird remains, however, could not be identified on location. The few bones of carp-like fish (Cyprinidae) point toward the Euphrates as a source of aquatic food as well as the non-worked piece of riverine shell (Unio sp.) that had to be brought onto the mound by people. Human mediation is even more evident in the case of a purple snail (Murex sp.) fragment. This snail is the source of costly and labor-intensive dyes that had to be imported to the Malatya plain from the Mediterranean. The single find originates from room A954 in the Temple C area. Food remains from the VI A storage area Along with increasing centralization during Phase VI A, the general concept of redistribution became eminently important in the form of reinvesting into a labour force that performed communal tasks, possibly through the distribution of meals, including meat. During the earlier, formative periods of the settlement this may have taken place within the context of religious ceremonies, as in Temple C or possibly in temples A and B. Subsequently, a more secular form of redistribution seems to have been administered in the storerooms of the palace. Based on field observations a comparison between the animal bone assemblages from two of the Period VI A store rooms - A340 and A365 - looked especially promising (fig. VI.4= figl 1 |.l.04-storerooms-x-web). Both were located to the south of Temple B, among a series of three rooms found towards the Southwest of this public building complex. Over a thousand (NISP=1546) animal bones came to light from room A340. This room is the smaller of the two and was designated the food "redistribution room" on the basis of over a hundred mass-produced bowls and some 160 cretulae, most tossed in its western corner, possibly used in controlling and recording withdrawals. Both artefact types were nearly absent from room A365, which was larger and full of storage vessels, and possibly serving as an actual deposit of meals provisions33, but it contained an entire order of magnitude fewer bones (NISP=104). The overwhelming majority of animal remains in both rooms came from domestic ruminants. A few wild ungulate remains was also recovered, however, the two features could be best compared in terms of the anatomical distribution of cattle and caprine bones. Due to intensive fragmentation, less than 20% of the latter could be identified to species, sheep remains consistently outnumbering those of goats in both rooms (table VI.5). Table VI.5 - The number of bones from domestic ruminants. Room ,_ Cattle Caprine Sheep Goat Total A340 434 919 123 12 1488 A365 35 55 4 3 97 Fig. VI.5 shows the frequencies of individual skeletal elements for cattle (top) and caprines (bottom) in the two rooms of interest (NB: NISP values are shown in decimal logarithmic scales improving the visibility of smaller numbers). A notable feature of both cattle and caprine remains is that given sufficiently large sample sizes (A340) their anatomical distribution is relatively even in terms of the classification by three gross meat quality categories as defined Frangipane. Palmieri 1988-89. ■ - f nd in Period VI A storage rooms Fig. VI.4 - The different quantities of domestic animal remains ou _ ^ ^ (expressed in actual number of bones). The find spots of cretulae are indiuut . A340 and 2i)07b. fig 130 131 Economic Centralisation in Formative States. Chapter VI Studí di Preistoria Orientale 3 Best quality Medium quality Poor quality me concentration of poor qualify cattle bone in room A365. by Uerpmann34. This relatively even representation r«f ,11 u , -e,ati„„s be„ee„ ,e freq Jncies ^^^^^^t" Table VI.6 - Coefficients of Spearman rank correlation between skeletal elements. i A365 cattle ___A3j40_cattle___ rs=0.367 _±^0^052_I A365 caprine rs=0.562 P=0.003 i A340 caprine rs=0.594 P=0.0(P rs=0.635 ____p^aooi__^ The only discrepancy (shown by a near lack of ™rr.ut- v compositions of cattle bone in rooms A340 an^A365 rt } 7$ « is dominated by bones from the feet (dr/lilwh / ^ samPle f«™ ™™ A365 evenly represented skeletal r.™ I • }' therefore * »s dissimilar to the more in A340. " ^ remamS °f Capnnes and ev™ the larger sample of cattle bones J I'erpmann 1973. The exact relation between these bones - mostly representing the animals' entire body -and the activities that once took place in this area was further clarified with regard to the taphonomy of animal remains. Could the bones be leftovers of primary butchering (especially of caprines represented in very similar ways in both rooms) that preceded handing out food provisions? Or do these bones represent simple "fast food" consumed on location by officials who handled the seals and products in the storage jars that were not necessarily of animal origins? These questions were studied in additional detail using the microstratigraphy of the A340 bone assemblage that was, in fact, formed by the collapse of a two storey building. It was hypothesized that, among other goods, meat was also redistributed in this area, given the major difference observed in the number of bones in storeroom A340 - which is more likely to have been the scene of redistributing food to people working in the palace - and A365, which during its excavation seems to have been simply a storage facility for food to be sent to other distribution rooms. It was of utmost interest, whether the animal remains in A340 were mainly concentrated in the upper part of the fill or in the lower portion, near the floor (layers lab represent post destruction fill in this area). As has been ascertained through the analysis of the distribution of cretulae and pottery within the layers of this room, however, it seems certain that material from layers 2a to 2f belonged to the upper storey, whereas those found in layers 2g-h and 3 were associated with activity in the lower storey of the building. It was hoped that differences between the strata assigned to the lower and upper deposits (fig. VI.6v4) would reflect possible functional differences between the two levels of the building. For one thing, the majority (102) of cretulae found had been deposited in the strata of the lower storey where less than a quarter (220) of all animal bones were found. The rest of the bones (773) came to light from the ruins of the upper level accompanied by only one third of all cretulae (63) found in the building (table VI.7). Table VI.7 - The microstratigraphic distribution of animal remains and cretulae in the layers assigned to the two levels of room A340. A340 Upper storev Lower sto rev j Layer 2a 2b 2c 2d 2e 2f 2g 2h 3 ! Cattle 28 31 57 62 60 27 28 10 -j Caprine 87 52 104 75 122 39 55 38 54 Dog 1 1 i Aurochs 1 1 3 1 f 7 1 Wild caprine "> 1 1 Red deer -> 3 i ! j ! ■ ,-1-i Fallow deer i Wild pig 1 1 3 1 1 1 ! Hare 1 3 Bird (small) 1 1 I Total 117 85 171 142 188 70 88 51 81 ! Cretulae 3 1 3 8 25 23 21 18 63 j 132 133 Economic Centralisation in Formative States. Chapter VI a cl z> 1a 1b 2a 2b Lä-AJLs..__ 2d 1m 0) best medium TO poor best O c o medium CD Q poor © best £j medium CO Ü poor best c O medium TO I i...... O poor I— Upper N=908 Lower N=245 1 B 0 10 20 30 N of identifiable specimens, % h<. \ 1.6 - The strangraphic sect,on of A340 showing the mierostratigraphy of the upper and lower storey (A) and the perceptual d.stnbuUon of domestic ruminant bones (100%) by meat quality categories (B). Differences between the compositions of the upper and lower deposits are indicated by arrows. In addition to the aforementioned inverse pattern of bone and cretulae deposition in A340, more wild animal bones seem to originate from the upper storey. This in itself, however, may also be related to the significantly greater number of bones available from layers 2a through 2f that would by definition represent greater taxonomic richness (c.f. fig. VI.2). The overall composition of this sub-assemblage, however, largely conforms of other, well-defined concentrations of cretulae in the area (c.f. fig. VI.3A): the zoological material is dominated by the rather monotonous presence of bones from sheep and goat, complemented by a smaller portion of cattle remains. Cretulae were carefully discarded once they had served their function, some were even found in situ in the proximity of commodities they had once sealed. Their relation to animal bones, however, is somewhat ambiguous. As remains of domestic ruminants dominated in this sub-assemblage (notably, no bone from domestic pig could be identified!), anatomical distributions of their bones were analyzed in an effort to further elucidating functional differences within this feature of crucial importance. Table \ 1.8 shows the anatomical distribution of cattle remains within the mierostratigraphy summarized by categories of "best", "medium" and "poor" quality cuts. In general, remarkably few bones represent the axial skeleton: vertebrae carrying valuable meat (especially in the lumbar region) and even ribs, usually represented in great numbers as a combined result of butchering and fragmentation, seem to be missing Other large, meat bearing bones, such as those of the stylopodia, humerus and femur, are also rare. Remains of possibly highly fragmented mandibulae and probably associated teeth, however, occur commonly in the material. It is also surprising that dry limb bones, metapodia in particular, are represented by many fragments. Breaking up these robust bones in cattle would take a special effort resulting in very small yields of marrow. There is practically no edible meat around these bones of the dry limb. The distribution of sheep and goat bones across the stratigraphy is summarized in table V 1 .y. Studi di Preistoria Orientale 3 Table VI.8 - The anatomical distribution of cattle bones within the mierostratigraphy of room A340. A 340 Fill Upper storey Lower storey Sum , ! Cattle la lb 2a 2b 2c 2d 2e 2f 2g 2h 3a Atlas 2 1 1 ! 4 1 Axis 1 1 1 vertebra cervicalis 1 1 1 i 5 i vertebra thoracalis vertebra lumbalis 1 Scapula 1 1 1 ■> j 1 1 3 li ,-\ _L- Humerus 0 2 i Pelvis 2 2 3 1 1 9 -1 Femur 1 Best quality 5 4 4 1 4 4 1 2 1 6 33; __i_*- Cranium 5 i 1 2 4 6 6 2 s 1 1 35 Mandibula 6 12 2 12 16 14 16 11 5 1 1 96 j -—i Costa 1 1 2 1 1 1 9! -1 radius/ulna 1 1 1 3 1 7 ! tibia/fibula 1 1 1 1 1 5 j Medium quality 12 16 5 15 23 23 25 14 12 2 5 152 i processus cornualis I 2 1 1 7 I Maxilla 1 1 i H-! Dentes 4 -> j 9 10 15 15 4 5 5 i 70 j Carpalia 1 1 2 1 i 7; Metacarpalia 7 5 9 3 12 6 4 4 3 5 i 58 i nhalanx nroximalis 1 1 1 i li 5j l-'llCliLlll^ LJL \J*\. Ill 144 11 O ohalanx media 2 1 1 1 4 h*11 tilt* 11^ Hlvulu nhalanx dtstalis 2 1 n 4 J-'lltllCllljV UlJlUl 13 Calcaneus 1 -i- C* 1 ^- til 1V- LI A ^tra oik 111 c 1 i i il3U distil Ll?> Centrotarsale* 1 1 1 i i i 5 IVI e tata rsa 11 a 6 8 10 I 8 10 6 3 5 ~% 2 i 62 1 * *. W LUUil JUilU Poor quality- 19 22 19 15 30 35 34 11 15 8 16 j 224 Similarly to cattle remains, meat bearing bones from the axial skeleton are almost entire y missing, although the aforementioned stylopodium (esp. femur) fragments occur m relative^-high numbers. The dominant fractions of mandibula and tibia remains must be. at least in part, resulting from intensive fragmentation. Tooth and skull remains occurred ,n great numbers as well. Similarly to cattle, metapodium remains of negligible nutnt.ve value occur in relatively great numbers as well. 134 135 Economic Centralisation in Formative States. Chapter VI Table VI.9 - The anatomical distribution of caprine bones within the microstratigraphy of room A340. Cattle and caprine remains from the two storeys of A340 were summarized by meat value categories in table VI. 10. The percentual distribution of these remains is summarized by the two discussed deposits in fig. VI.6B. Revisiting our raw data in tables VI.8 and VI.9, however, is a reminder that the pattern thus emerging should be interpreted cautiously. In the case of cattle, it is mandible fragments that contribute this difference, while among caprine remains numerous fragments representing medium quality meat originate from the notoriously fragile zygopod.um bones, especially the tibia. Similarly to mandibles, the effect of post depositions Studi di Preistoria Orientale 3 fragmentation cannot be ruled out. Although it is difficult to tell to what extent qualitative differences in the anatomical composition of the upper and lower storey assemblages reflect nuances in diet or are a product of taphonomic bias, it can be said with certainty that animal heads and limbs dominated the repertoire of meat available and that far more meat must have been dealt with on the upper level of the building. This becomes patently clear in table VI.10 and fig. VI.8, a percentual summary of cattle and caprine remains from this two storey building. Table VI.10 - A comparison between the upper and lower sections of A340 by meat value categories as published by Uerpmann (1973). Fill (la-b) Upper 2a-f Lower 2g-3 Sum Cattle, best 9 16 7 32 Cattle, medium 28 105 19 152 Cattle, poor 41 144 F 39 224 Caprine, best 29 123 50 202 Caprine, medium 108 287 65 460 Caprine, poor 81 233 65 379 Sum 296 908 245 1449 Over three quarters of bones in the two discussed deposits in A340 originated from the upper storey. Given the evidently great difference between sample sizes and the basic similarity between the anatomical profiles of the upper and lower storey deposits (fig. VI.6S), the heterogeneity of distributions by meat quality categories was confirmed using a Chi2 test. The test showed that two trends visible in the percentual distribution of bones are significant in formal statistical terms: bone fragments representing medium quality beef and mutton were slightly more common in the larger deposit originating from the upper storey (Chi2=l 1.503: df=5; P=0.042). Bones of very high or very low quality did not show such differences. These results seem to suggest that food remains in this building were strongly associated with central administration (the absence of pig remains and the presence of bones from poss.bly high status game in the upper storey deposit). The presence of bones, however, is apparently not directly associated with that of cretulae. Considering the type of meat present, the almost complete lack of bones bearing valuable meat from the axial skeleton is remarkable: vertebrae and ribs make up one third of total skeletal weight both in cattle (18.8+13.9%) and sheep (21.7+12.8%)* Even if they tend to be relatively prone to natural fragmentation, they represent a mass of bone that would have been unlikely to disappear almost without trace in this deposit through even the toughest taphonomic process. It is likely that although the heads and feet of animals represent low quality meat by modern "western" tastes*, these body parts (especially in the case of cattle) were accumulated in room A340 selectively after hav.ng been separated 35 Reichstein 1974. 36 Bartosievvicz 1997. 136 137 Economic Centralisation in Formative States. Chapter VI Studí di Preistoria Orientale 3 Upper storey Lower storey 14.3 \ V N=265 13.9-^^, 5.6 7:2 "V1 6.7 j \ / KBSS Jí9.8 i [8.2 Caprine Fig. v 1.7 - The anatomical distribution of bone finds from cattle (top) and caprines (bottom) in the upper and lower store) deposits of a340. Skeletal parts represented by over 5% of each species/deposit are labelled with the respective percentage values. These add up to around 80% in each case, clearly illustrating the relative absence of bones from the axial skeleton (vertebrae and ribs). For raw data see tables 7 and 8. the dressed carcass of greatest nutritive value. Primary butchery must have taken place elsewhere and these less precious cuts were taken to room A340 in the area of redistribution. Since the majority came to light from the upper storey, including some cuts of large game, these bones may represent meals by the personnel or guards in charge of dealing with the ware accounted for with the cretulae in the lower storey. Alternatively, they may be regarded as evidence of cheap cuts handed out as part of food distribution to those dependent on provisions by the central power that ruled Arslantepe during Phase VI A. The absence of pork from the menu (whoever consumed the meat once handled here) is remarkable: it seems to illustrate the centralized economic control mechanism that ultimately may have resulted in banning pork. - a staple in independent, rural household economies - by religious means37. The age distribution of cattle in selected areas within the Period VIA palace complex Patterns of meat consumption in different public and private contexts are crucial to our understanding of social relations at this settlement. Although the detailed analysis of age groups in various domesticates is beyond the scope of this study, in light of previous results the age distribution of cattle bones seemed worth considering. As one of the critical points in this paper is demonstrating the difference in animal remains between the various functional areas within the palace complex, this animal of great individual value seemed the best indicator ot such hypothetical differences. Diener. Robkin 1978. Therefore, ageable cattle bones found in temple areas A and B were compared to the storage area designated A340 and a selected set of rooms in the elite residences of Period VI A that contained major numbers of cattle bones. As caprine remains dominated in all these areas, cattle was represented by relatively small numbers of identifiable fragments. Although in store room A340 almost one third of the bone fragments originated from this animal, cattle was represented consistently by one quarter of the identifiable bones in the rest of the studied provenances (table VI. 11). Table VI.ll - The number and percentual contribution of cattle bones in the selected provenances. Provenance Cattle NISP % of total NISP 1 Store room A340 434 29.0 1 Period VI A elite residences total 117 24.4 Temple A total 112 25.9 _l.___---1 Temple B total 308 25.4 Unfortunately, however, only a fraction of these remains could be precisely aged, largely due to the high degree of artificial and natural fragmentation of the bones. The proportion of ageable cattle bones has varied considerably throughout the sample. The age compos.t.on of cattle remains in this sub-sample is summarized in table VI. 12. Table VI.12 - A comparison of cattle ages between the provenances selected for study. Fraction numbers show the proportion of ageable Room neonatus juvenilis ubadultus 3 "3 ■u maturus senilis specimens to all cattle bone s Store room 59/434 A340 1 10 9 37 ■> 0 Period VI A elite residences 6/45 A650 0 0 *> 4 0 0 1/1 A651 0 0 0 0 1 0 1/1 A727 0 0 0 0 1 0 22/46 _| Period VI A public _, A747 0 0 1 1 19 0 Temple A 1/3 A36 0 0 0 0 1 0 15/85 A42 0 3 4 8 0 0 5/9 A46 0 0 ~> -» j 0 0 1/1 A47 0 0 0 0 1 0 9/9 A84 0 0 0 9* 0 0 I t/f/f/tC u_______-_'—" 139/223 A450 0 0 1 9 129 0 10/24 A800 0 0 0 0 10 0 6/9 A809 _0_ 0 0 1 5 0 1/1 32/51 A810 A812 0 0 0 0 0 L 1 0 7 1 29 0 0 * Sinsle skull. 138 139 Economic Centralisation in Formative States. Chapter Ví Studí di Preistoria Orientale 3 The age distribution of cattle bone in some of these rooms is shown in absolute terms in fig. V1.8 (percentages would have been meaningless, given the small size of some sub-samples). Evidently, storage room A340 stands out with a reasonably varied age composition for the cattle slaughtered. We already know from figure VI.5, that this room had access to higher quality beef therefore eating the meat of relatively younger cattle (in fact of all age groups) fits a pattern of possibly high-status food. The opposite, a large quantity of presumably poorer quality, tougher meat from mature individuals was consumed in room A450 within the complex of Temple B. It is noteworthy that Temple B also stood out with its relatively great contribution of hare bones to the faunal sample (fig. VI.3S). Evidently, meaningful conclusions can be drawn only from these two, larger sub-samples that indeed seem different in terms of age composition. Although the number of bones is far too small to have this difference tested statistically, the hypothesis that age diversity may have been an artefact of sample size was considered in fig. VI.9. As may be seen, the overwhelming majority of sub-samples outline a strongly degressive trendline. Of the two major sub-samples, cattle bones from store room A340 basically follow this pattern: the better representation of a variety of age groups is, in part, the result of larger sample size. On the other hand, the greatest number of ageable cattle bones, recovered from room A450 of Temple B is dominated by bones from mature individuals. By modern standards, beef originating from these animals must have been far tougher, even if very old cattle or oxen possibly used in draught work could not be identified. It seems, however, that the meat consumed in this public area was of lesser gastronomic value. This places the interpretation of hare bones in a different light: could this small game have been a luxury food, or a complementary source of meat procured by common people opportunistically? In any case beef, even of poor quality, would not have been worth providing only to a few people. It must have been an important source of animal protein for "the masses". However, a differential distribution by quality (also related to the age of the animal at the time of slaughtering) seems apparent at Arslantepe. Discission and conclusions One of the aims of this paper has been detecting possible economic interference by the early state elites in the life of their population, seen through the Phases VII-V1 A faunal data from the Chalcolithic of Arslantepe. The increasing dominance of caprine-based herding practices at the site was linked to the evidently new function that the settlement acquired by Period VI A as a great regional centre38. The site contains tantalizing hints of increasing social stratification reflected in the variability of access to the meat of particular animal species. Arslantepe is an ideal site in which to identify evidence of the early development of social complexity with regard to differential and centralized provisioning characteristic of this phase of the urbanization process. The bones of sheep, goat and cattle occurred most commonly, followed by those of pig (usually contributing less than 20% of NISP to the repertoire of domestic animals). In line with the results of previous investigations, the contribution of pork and game to the diet decreased between Periods VII and VI A. It must be emphasized, however, that hunting had hardly any practical s.gn.f.cance in meat provisioning during the studied late Chalcolithic Frangipane. Slracusano 1998: 242. 150 (A ■4— ZJ .= r ^ *£ a. vi ■r, s. C & .% 'Si u VI W U 1 tern E ii C Bökönvi1988 1405 100 167 1114 554 16 1 80 3437 Period VII common houses A21 65 15 25 184 96 2 7 394 A22 85 15 10 92 46 1 2 251 A23 3 1 2 6 A24 2 2 1 5 A25 14 3 9 13 39 Common total 169 34 35 289 156 ■> 9 695 Period VII elite residences A682 24 4 1 26 9 6 70 A617 548 32 60 550 145 30 1365 A684 25 3 3 29 7 1 68 A582 365 34 89 365 196 3 1052 A646 5 2 1 5 2 15 A647 28 8 1 26 5 2 70 A648 2 1 3 1 7 Elite total 997 84 155 1004 365 42 2647 Period VII Intermediate levels 1 A580 30 10 1 50 13 104 A581 2 1 3 A618 17 1 1 19 Intermediate total 49 12 2 50 13 126 Period VII public Temple C A900 10 5 2 54 4 75 A932 2 5 1 12 1 21 A934 11 2 1 36 5 55 A950 46 17 7 242 15 327 Temple C total 69 29 11 344 25 478 Storage/craft, Temple C level 1 A563 22 7 2 15 2 48 A571 3 3 A842 104 39 17 313 ti J) 503 A848 62 25 1 164 17 9 278 A850 61 38 16 250 16 11 392 A858 19 17 5 74 3 118 A953 7 3 39 5 54 A954 61 12 8 120 8 211 K680 25 7 5 159 5 6 207 Storage/craft total 361 148 54 1137 83 I 3! 1814 Clason, Buitenhuis 1998: 236. fis. 3. Frangipane. Siracusano 1998: 238. Bartosieuicz 2005b: 98. Frangipane. Siracusano 1998: 242: Bartosieuicz 2005c: 155. table 1. 142 143 Economic Centralisation in Formative States. Chapter VI Studí di Preistoria Orientale 3 Appendix 2 - Period VI A, domestic animals. Cattle Sheep Coat Caprine Domestic pig Horse Domestic ass Dog Domestic total Period VI A elite residences A650 45 2 7 65 ~» 1 123 A651 1 13 14 A727 1 2 -» j A72H 3 1 4 8 A734 2 13 15 A747 46 22 4 132 2 4 210 A933 6 1 41 48 A946 13 4 2 32 7 58 Elite total 117 29 14 302 12 5 479 Period VI A public Temple A A36 3 2 5 A42 85 15 8 266 2 376 A44 5 1 2 7 15 A46 9 3 8 20 A47 1 1 A49 2 2 AS4 9 5 14 Temple A total 112 19 10 290 1 432 Temple B A450 223 65 23 509 32 27 879 A800 24 7 60 2 93 A809 9 6 4 29 2 50 AS 10 1 2 3 A812 51 16 4 116 1 188 Temple B total 308 94 31 716 33 31 1213 Period VI A administrative Cretulae dumps A77 (Temple A) 61 87 28 773 7 956 A206 131 71 60 859 24 1 1146 A430 79 6 4 107 196 Cretulae dumps total 271 164 92 1739 31 1 2298 Store rooms A340 434 123 12 919 8 1 2 1499 A364 49 18 12 146 2 227 A365 35 4 3 55 5 1 103 A369 37 72 21 436 3 1 570 Store rooms total 555 217 48 1556 16 2 5 2399 Representative Bdg. III. Al 13 ->-> 17 9 183 1 2 234 A127 27 5 27 Representative Bde total 39 9 188 1 2 261 Appendix 3 - Period VII, large game. Aurochs Wild sheep Wild goat Wild caprine "zZ >i 55 SI ■o ZJ i-Zi w — Z> z* ■o ■D Zi a Persian fallow deer Roe deer 5t Wild ass Brown bear Lion w 5 it 1- Bökönvi1988 68 16 28 29 1 154 24 40 10 39 1 1 411 Period VII common houses A21 5 1 1 1 13 6 A22 4 2 9 14 4 1 3d A23 1 1 A24 A25 4 1 1 6 Common total 9| 9 3 10 32 11 3 1 78 Period VII elite 1 1 1 1 1 1 Elite residences A682 4 1 1 3 9 A617 37 6 11 11 29 8 2 4 1 io| IIS A684 2 -> A582 22 1 1 1 i IT A646 1 1 *■» A647 1 1 A648 Elite total 63 9 12 11 31 8 2 io| 153 Period VII Intermediate levels A580 2 6 1 1 10 A581 A618 1 1 ■■*> Intermediate total 3 7 1 1| «1 12 Period VII public Temple C 1 1 1 A900 TC 1 1 1 1 4 A932 TC --1 A934 T 2| 4 A950 TC 1 2 1 6 Temple C total Storage/craft, Temple C level A563 A571 A842 A848 A850 A858 A953 A954 K680 Storage/craft total 2 2 23 9 12 1 1 48 3 1 4 2 8 2 4 1 22 2 2 5 4 7 1 1 2 3 19 1 1 1 1 -> 1 1 1 8 6 11 3 1 o 5 i 14 Q 36 19 26 8 10 s 116 144 145 Economic Centralisation in Formative States. Chapter VI Appendix 4 - Period VI A, large game. Aurochs Wild sheep Wild goat Wild caprine Coitered gazelle ĺ. z* ■a it Persian fallow deer i-z* Zi ■0 ä? Wild pig Wild ass Brown bear Lion Gray wolf Large game total Period VI A elite residences AdSO 3 1 1 2 9 A651 1 1 A727 A728 A 734 A 747 1 1 I 1 1 11 A933 Al>4d 2 2 4 Elite total 9 1 1 5 2 1 1 3 2 25 Period VI A public Temple A A 3 6 A42 1 1 1 1 6 4 1 15 A44 A46 A47 A49 AS4 Temple A total 1 1 1 1 6 7 1 18 Temple B A450 15 1 5-2 1 3 21 A800 t 1 2 3 8 AS(W AS 10 1 1 ASI2 3 3 3 9 Temple B total 21 5 1 5 1 3 3 39 Period VI A administrative Cretulae dumps A 7 7 (Temple A) 1 7 14 4 23 A 20d 5 3 1 3 2 2 17 A430 8 1 1 1 11 Cretulae dumps total 7 3 11 2 4 16 1 1 6 51 Store rooms A340 12 J 13 2 9 39 A364 1 1 2 A365 A369 i 2 4 Store rooms total 12 3 2 13 3 IC 1 2 45 Representative Bdg. III. Al 13 2 2 K ) 17 A127 2 Representative total 2 i 1 1 1 1( 1 19 Studi di Preistoria Orientale 3 Appendix 5 - Period VII, small game. Bökönvi1988 Period VII common A21 A22 v23 A24 A25 Common total Period Ml elite Elite residences A682 A684 A617 A582 A646 A647 A648 Elite total Period VII_ Intermediate levels A580 A581 A618 Intermediate total C3 Period VII public Temple C_ A900 A932 A934 A950 Temple C total Storage/craft, Temple C lev, A563 A571 A842 A848 A850 A858 A953 A954 K680 I Storage, craft total 1 sk 3 3 3 3 1 um II 61 1 4 7 1 20 7 146 147 Economic Centralisation in Formative States. Chapter VI Appendix 6 - Period VI A, small game. s Wild cat ■a U 55 > Rodent indet. L... Bat indet. Great bustard Pelican Eagle Passeriform bird Bird indet. Greek tortoise Frog/Toad Cyprinid fish Fish indet. Small game total Period vi a elite residences Ao50 Ao5 i A~27 a?:x AT?4 i 1 1 1 1 1 1 1 1 1 1 1 1 1 i 1 1 A747 A933 1 A94h 1 1 Lliie total 1 1 1 ~» 1 1 1 1 1 1 11 I 1 1 3 Period vi a public Temple a A36 \4: 1 2 2 5 A44 A46 A4"7 A4° AS4 Temple A 1 1 Temple B A450 10 5 8 4 27 AXOO 2 2 AS09 1 1 AS 10 1 1 AS 12 17 2 2 ? 1 34 58 Temple B total 31 5 7 1 2 10 5 34 90 Period vi A administrate (retulae dumps A7" (Temple AI II 2 13 A206 1 2 i 1 5 A430 •> 9 13 33 57 (."retulae dumps total i 1 11 i 13 33 3 64 Store rooms A340 4 4 8 A3M 1 1 2 A3f>5 A3e9 Store rooms total I 4 5 10 Representative Bdg. iii. Al 13 1 1 2 AI27 Represemame iota! 1 1 2 148