SAPIENZA UNIVERSITA DI ROMA Dipartimento di Scienze Storiche Archeologiche e Antropologiche dell'Antichitä
Studi di Preistoria Orientale (SPO) volume 3
Economic Centralisation in Formative States. The Archaeological Reconstruction of the Economic System in 4th Millennium Arslantepe
edited by Marcella Frangipane
M. Frangipane, C. Alvaro, F. Balossi Restelli, L. Bartosiewicz, M.B. D'Anna, GM. Di Nocera, P. Guarino, R. Laurito, C. Lemorini, G. Liberotti, A. Masi, G. Palumbi, P. Piccione, R. Quaresima, L. Sadori, F. Susanna
Roma - 2010
CONTENTS
Introduction
M. Frangipane......................................................................................................... pag. 5
Part One. Theoretical and Historical remarks
Chapter I - Politics, economy and political economy in early centralised societies. Theoretical debate and archaeological evidence
M. Frangipane........................................................................................... » 11
Chapter II - Arslantepe. Growth and collapse of an early centralised system: the archaeological evidence
M. Frangipane........................................................................................... » 23
Part Two. Architecture at Arslantepe: the expansion and change of public area. building features, techniques, materials and function
Chapter III - Architecture and the organisation of space....................................... » 45
III. l - Arslantepe in Periods VII, VI A, and VIB2: Changes in settlement
arrangements
C. Alvaro......................................................................................... » 45
III .2 - Public and private architecture at Arslantepe in Period VI A
C. Alvaro......................................................................................... » 54
Chapter IV - Building materials.............................................................................. » 73
IV. 1 - Building materials in the 4th and early 3rd millennium monumental
architecture at Arslantepe: Mudbricks and plaster
G. Liberotti, R. Quaresima............................................................. » 73
IV.2 - Timber use at the end of the 4th millennium BC at Arslantepe
C. Alvaro, L. Sadori, A. Masi, F. Susanna..................................... » 81
IV3 - Building techniques and materials in monumental architecture at
Arslantepe: a summary
C. Alvaro......................................................................................... » 94
Part Three. Food production and the management of primary economy
Chapter V - Agriculture at Arslantepe at the end of the IV millennium BC. Did the centralised political institutions have an influence on farming practices? F. Balossi Restelli, L. Sadori, A. Masi...................................................... » 103
3
!
I
Economic Centralisation in Formative States. Contents
Chapter VI - Herding in Period VI A. Development and changes from Period VII L. Bartosiewicz...........................................................................................
Chapter VII - Pastoral models and centralised animal husbandry. The case of Arslantepe
G. Palumbi.................................................................................................
Part Four. Food storage accumulation and consumption
Chapter VIII - The Ceramic containers of Period VI A. Food control at the time of centralisation
MB. D Anna..............................................................................................
Chapter IX - Continuity and changes in the elite food management during the 4th Millennium BC. Arslantepe Period VII and VI A: A comparison MB. D'Anna, P. Guarino..........................................................................
Chapter X - The Ceramic containers of Period VI B2: Food storage, processing and consumption in a village community of the early 3rd millennium P. Piccione..................................................................................................
Chapter XI - Food circulation and management in public and domestic spheres in Periods VIA and VI B2. A comparative perspective
MB. D'Anna, P. Piccione......................................................................... » 231
Chapter XII - Daily and ritual activities during the palatial phase. Wear traces analysis of chipped and ground stone industries
C. Lemorini................................................................................................ » 241
pag.   119 |
i
» 149
» 167 » 193 » 205
Part Five. Craft activities and wealth production
Chapter XIII - Metals and metallurgy. Their place in the Arslantepe society between the end of the 4th and beginning of the 3rd millennium BC GM. Di Nocera......................................................................................... » 255
Chapter XIV - Textile tools and textile production. The archaeological evidence of weaving at Arslantepe
R. Laurito........................................ » 275
Conclusions
Chapter XV - The political economy of the early central institutions at Arslantepe. Concluding remarks
M. Frangipane............................................
Bibliography..............................
» 289 »    309 ,
4
Chapter II
ARSLANTEPE. GROWTH AND COLLAPSE OF AN EARLY CENTRALISED SYSTEM: THE ARCHAEOLOGICAL EVIDENCE
Marcella Frangipane
The development of a 'palatial' centre at the end of the 4th millennium B.C. (Period VI A)
The period on which this paper focuses in particular is Arslantepe VI A (Late Chalcolithic 5, 3350-3000 BC), contemporary with the developments of the Late Uruk culture in Mesopotamia and at the height of Uruk 'colonisation' in the Syro-Anatolian regions. It was during this period that Arslantepe's public area underwent large-scale development, in terms of both its extension and the concentration of activities, accompanied by changes in its architectural and functional features1. The area used for public activities became a cluster of functionally differentiated buildings, attached to one another on several terraces hewn out of the south-western slope of the mound (fig. II.1). What has clearly emerged from recent architectural surveys of this monumental complex is that the various sectors were not built as a whole, but there was an expansion from the oldest northernmost area, which included 'Temple B' and the imposing Building XXXV (of which only one corner, A830, has been brought to light so far), towards the southernmost and lower areas on the slope2. Even though the construction work was probably phased-in across the years, we are fairly certain that all the buildings were contemporarily in use in the Final phase of the architectural complex; furthermore, the progressive extension of the structures seems to have been planned according to a precise design keeping pace with the new functional demands of public life, which also increased considerably. Whereas in its early phases the complex seems to have consisted mainly of ceremonial (Temple B) and official buildings (Building XXXV and perhaps shortly afterwards the structures on the western terrace - Building IV), in its later development it expanded to include buildings and spaces with a basically non-cultic character, which were specifically intended for economic and administrative activities, and acquired a greater capacity to admit larger numbers of people involved in the daily life of the public area: A group of storerooms and a courtyard were added, the access corridor was extended and widened, a monumental gate was built at the
1 Frangipane 1997; Frangipane (ed.), 2004.
2 See Alvaro, chapter HI2, present volume.
23
Economic Centralisation in Formative States. Chapter II
Studi di Preistoria Orientale 3
Trmcea Schat-ricr
Fig. II.2 - Arslantepe. Plan of the palatial complex with the indication of successive expansion phases: dark grey indicates the oldest nucleus; medium grey refers to the second phase of expansion; light grey is used for the last construction phases. * Cretulae dumps; ♦ Cretulae in situ; ■* Wall paintings.
V.' -
a.i -
Arslantepe, Period VI A (LC5). views
«fAepaJatial building compiex.
entrance, new open spaces, probably intended for receiving people and giving access to the various buildings, made the complex wider and more articulated (fig. 112).
This compound and imposing architectural complex was the visible, and hence ideological, expression of the unity of the buildings - and then of the institutional functions they represented - which, while clearly distinct from each other (virtually non-intercommunicating, and even standing on different levels), made a compact whole meeting very clear-cut criteria, as already pointed out; at the same time the complex hosted a variety of different activities, which were prevailingly economic and administrative, while there was a marked contraction in the dimensions, role, and importance of the temple buildings, represented by Temple A and Temple B, in comparison with the large isolated ceremonial building (Temple C) of the final phase of the previous period VII (Late Chalcolithic 4, 3500-3400 BC)3. In other words, the huge expansion of the architectural complex and the activities connected with it does not seem to have been correlated to a parallel growth in the importance of the cultic or ceremonial places, as was the case, for example, in the large area of the Eanna at Uruk. On the contrary, the cult buildings of period VIA actually declined, at least in terms of the loss of their original impressive architectural appearance, physical separation, access capacity (the public no longer
Frangipane 2000c; 2003; Alvaro, chapter III.2, present volume.
24
25
Economic Centralisation in Formative States. Chapter II
Studí di Preistoria Orientale 3
entered the main cult room), and the variety of the functions performed there (for example, the scale of redistribution practices seems to have decreased).
For all these reasons, I believe that the complex of public buildings at Arslantepe can be defined as 'palatial' or 'proto-palatial', not in the sense of having been the residence of a sovereign, but rather as a unitary, or at least internally correlated (also in architectural terms) seat of the public institutions and their multiple functions, which gradually tended to take on a more explicitly 'secular' character. Although the agglutinated pattern of the whole complex can be linked to what were more specifically Anatolian architectural traditions, this local choice was nevertheless designed to meet the changed, and more complex, needs of the public institutions, which seem to have acquired wider-ranging prerogatives and tasks than they had previously had, partly detached from the ideological-religious and ceremonial mediation which was typical of the elite management of power in the first half of the fourth millennium. In the Arslantepe palace, in other words, we gain a glimpse of a kind of initial division between a more specifically managerial-administrative sphere, through which centrally managed goods and labour flowed and which occupied their own spaces (stores and courtyard), and an ideological-symbolic sphere connected with cult and ceremonial practices (Temples A and B).
A similar functional separation can perhaps also be detected at Uruk-Warka between the various isolated buildings in the large public area of the Eanna, or between it and the White Temple on the Anu ziggurat. And it is certainly an indication of the development towards more complex, solid and organised forms of central power. But the Warka temples, characterised by gigantic dimensions and a great openness towards the outside world, show no differences in the scale and dimension of the flows of individuals from any other buildings in the public area, and remain architecturally dominant as well as spatially separated. At Arslantepe, conversely, the ceremonial practices, though we assume they must have continued to play an important role in the formation and maintenance of consensus, took place in two buildings which, besides being strictly part of the architectural whole, were able to hold only a limited number of people and must hence have been more exclusive and 'separate' actions in the sense of being performed in a more restricted, and probably elitist, environment.
A third function (more specifically social and political) may possibly be recognised in other buildings inside the palace, which were apparently used for official events and representative tasks, such as hosting people in a non-ritualised form (Building III and IV, probably Building XXXV) (fig. H2). One of these is the well-known Building HI - the 'Weapons Building' -which, even though it is only partially preserved, does not seem to possess any of the features recognisable as belonging to a cult building4. The best preserved room there contained two groups of arsenical copper weapons, spearheads and swords, which, on the basis of their stratigraphical and spatial location, one presumes must have been attached to the walls, emphasising their symbolic and prestige value5. Weapons of this kind have no equal so far attested in any other sites from this period, but they did become more common later in the same region of the Upper/Middle Euphrates6, consolidating a metallurgical and fighting tradition that may have originated precisely at Arslantepe; the swords in particular were the very first, and, for a very long time afterwards, the only evidence of the use of this weapon
* Frangipane, Palmieri 1983a: 307-315.
3 Di Nocera present volume.
6 Hassek Hoyuk (Behm-Blancke 1984) B' t Kara Hassan burials (Wooiley 1914; WooIIeyTl/ 75?V^JT        SqUa<W 2<*»), Carchem.sh and a. Arslantepe ..self (Frangipane et al. 2001; Di Nocera peLm *° 'm ** EB 1 *»* Tomb
m
A significant place in the ideological and symbolic communication system was occupied by wall paintings, which, in the buildings of the palatial complex, seem to have had a function that was much more than purely decorative. For they depicted figurative characters laden with meaning, and, in the innermost part of the long access corridor, a full 'narrative' scene was painted on the walls (fig. II.3). The positioning of the paintings, too, which were always by the sides of the doors or in places through which people passed, emphasises the fact that the intention behind it was to send out specific messages to the people entering the areas, and certainly referred to a shared ideological and perhaps also mythological baggage, which must have been intended to confer legitimacy on the power of the elites and consequently on the actions that were performed in the buildings under their authority. Among the best preserved paintings are representations of two stylised and almost identical anthropomorphic figures, on the two sides of what had originally been a doorway, subsequently bricked up, leading from the main access corridor to the internal courtyard of the palace by passing through the central room in the stores (fig. 113: a). The two figures, decorated with stylised elements, were standing behind a kind of small table or altar in an attitude of prayer and perhaps depicted mythical figures or divinities, judging from their extremely standardised traits and the way their features were obsessively repeated7. More elaborated and interesting still is the large scene painted on one of the walls of the innermost corridor, depicting a kind of cart (or plough?) being drawn by one, or more possibly two, oxen and driven by a coachman, which seems to be moving towards the outside of the building (fig. II.3: c-d). This scene is very closely paralleled by a cylindrical seal impressed on the cretulae from Arslantepe and a similar depiction on a Uruk seal, interpreted as scenes of ritual threshing8. The symbolic reference to agriculture and the role of the leader in activities connected with agriculture are very consistent with the redistribution model and the large flows of food commodities in the palace areas, clearly evidenced from the stores and the large quantities of cretulae and mass-produced bowls.
The most remarkable phenomenon documented by our archaeological investigations in the 4th millennium palace at Arslantepe is the extraordinary growth of the administrative organisation, which attained levels of sophistication and complexity wholly similar to those of the major Late Uruk Mesopotamian centres, even though some of the innovative elements found on those sites (hollow spherical bullae, often associated with tokens, numerical tablets, and, only in the city of Uruk-Warka, pictographic tablets) were absent. The study of the vast amount of administrative materials from Arslantepe9 has revealed that the cretulae by themselves were a most effective means of controlling the movement of goods, practised in northern Mesopotamia since the Neolithic10 and fully developed by the early central authorities in the fourth millennium. The cretulae, in other words, were used as 'documents' of the carried out transactions, by setting them aside and subsequently accounting them, as part of a system that worked perfectly well even without the support of any other accounting or recording tools"; this, however, was obviously possible only up to a certain quantity of operations, so that it was possible to account for them year by year (or from one administrative period to the next), without needing to memorise them over a long time. A few examples of a particular type of counting tablets, which however have nothing in common with the Mesopotamian numerical tablets, were
• 2£££        *»■■<-* »°*°~
9 Frangipane et al. 2007.
111 Akkermans (ed.) 19%; Akkermans, Duistermaat 1996. 11 Ferioli, Fiandra 1983; 1994; Frangipane et al. 2007.
26
27
Economic Centralisation in Formative States. Chapter II
I*- Wall painty i„ the ^
C ",d * ™e «™ depicted in thetS0rPhÍC ^ Ín the Storc
28
Studi di Preistoria Orientale 3
indeed also found at Arslantepe, and alongside this kind of accounting devices there was one single small tablet bearing an engraved sign12. Things were perhaps already moving at Arslantepe as well in the direction of finding additional tools to be used for managing an increasing mass of transactions.
Nearly 300 different seals, evidenced from their impressions, and thousands of cretulae (over 2,100 well preserved fragments) show that a large number of individuals were operating in the public area with administrative responsibilities, who, we believe, were primarily responsible for withdrawals of goods in a very intense and probably regular redistribution circuit, in the sense it was not necessarily linked to particular occasions or special events. This has been confirmed by the discovery of 175 cretulae in one of the three store-rooms, A340, which has been interpreted as a redistribution storeroom partly because of their presence and their association with about a hundred mass-produced bowls (fig. 11.4). Here, the cretulae were not only present in large numbers, but they seem to have been removed (and consequently attached) very frequently13, and then set aside temporarily, first in a corner of the room and then perhaps on an upper storey14. The thorough analysis of thousands of cretulae discarded as administrative materials in the main dump A206 inside the palace, however, made it possible to identify, besides the daily, or at all events regular and frequent withdrawals, the existence of particular transactions, probably carried out by senior-ranking officials, who withdrew and redistributed goods in large quantities, concentrated in a short space of time, probably for special, perhaps ceremonial, occasions15.
We cannot go once more into details about the administration in the palace complex, which has already been exhaustively shown in a recent book devoted specifically to that subject, to which the reader is referred for further details16. But I would like to draw attention to a few points which are very important for the subject we are dealing with in the present volume: 1) the huge quantities of transactions performed in the public area; 2) the fact that these transactions (at least the ones documented by the cretulae found) were exclusively internal movements of goods17 and almost entirely (perhaps exclusively) foodstuffs, as is indicated by the vast number of cretulae attached to pots; 3) the likely coexistence of forms of both regular redistribution and redistribution on special occasions; 4) and the fact that in these circuits there were individuals or officials with different responsibilities and probably of differing ranks in the hierarchy18. The existence of a fully-fledged bureaucracy with distinct tasks and ranks indicates that a class of specialists who only supplied services may have been emerging. The large numbers of people doing transactions in the stores, almost certainly people withdrawing foodstuffs, may have been palace outsiders who had links to the central institutions as workers or people giving their services or labour corvee; but officials operating intensely and continuously, perhaps above all on the occasion of special events, may have been of a socially higher rank, either linked or not to the paramount chiefs by kinship or lineage bonds.
Everything points to the central position that the administrative activity occupied in public life, and shows the magnitude of the redistribution practices, around which the life of the
12 Frangipane et al. 2007: 148-157.
13 Some were removed shortly after being attached, when the clay was still soft.
14 Frangipane et al. 2007: 157-173; 415-425.
15 Frangipane et al. 2007: 425-*59.
16 Frangipane et al. 2007.
17 Analyses of the clay used on a vast number of cretulae samples ail agree in showing it to be local in origin (Blackman 2007; Ralmieri, MorWdelli 2007).
m Frangipane et al. 2007: 468-J75.
29
Economic Centralisation in Formative States. Chapter II
Studi di Preistoria Orientale 3
palace seems to have revolved. Thousands of mass-produced bowls are the most frequent and the most widespread findings in the public area, and they were the main container for meals19, being associated in large quantities with cretulae. Not only did a very large number of individuals with administrative responsibilities withdraw the food, and in some cases probably regulate its supply, but even larger numbers of individuals used to take their meals in the public area20, and not exclusively or even mainly in connection with ceremonial or cultic events performed in the temples.
This is another major difference with what has been found in the final phase of period VII, from which time hundreds upon hundreds of mass-produced bowls have been found in the large Temple C21, while no similar quantities and concentrations of them have been found in any other building from the same period, even in those belonging to the elites.
What happened, then, in the transition between period VII and period VI A? What transformations took place, and what the aspects of continuity?
The background in Period VII
In this period, which developed homogeneously over a long period of time (covering Late Chalcolithic 3 and 4, without any particular cultural breaks, 3800-3400 BC), the settlement must have been fairly large, covering the whole area of the mound as it is today, showing a clear differentiation in functionally and symbolically separate areas. On the north-eastern edge of the site the excavations conducted in the 1960s and 1970s brought to light common mud brick houses with evidence of domestic activities and burials under the floors or close to the houses, which did not seem to show any particular differences in terms of funerary gifts22. Fairly imposing buildings had been conversely built in the central-western area of the tell, on what must have been the top of the early mound: they were larger in size, with walls over one metre thick, and, though the area underwent modifications in the floor plans and functions of the individual structures across time, the main buildings always maintained a fairly monumental, and what was certainly an elite, character. Outstanding among these was a large building (Building XXV) with a main hall decorated with paintings and mud brick columns (subsequently subdivided into four smaller rooms), whose ground plan and functional features did not seem to have any kind of public character and was more likely a residence of the elite23 (fig. 115). This does not exclude that the occupants, precisely because of their assumed high-ranking, might have had hosting and commensality functions, as the same transformation of part of the large room with columns into a store for foodstuffs with large pots may indicate. But there is no evidence of any large-scale redistribution practices (only a few bowls) or administrative activities.
This evidence was, conversely, substantially present in a large tripartite and almost certainly ceremonial building (Temple C), built next to the elite residences (immediately to the south/south west of them, towards the margins of the mound), standing apart from the other buildings and resting on a platform made of huge stone slabs and mud pise2* (fig. II.6). The cult, or more
S.159 J23
Arslantepe Period VI A Th
-    or —AM0
19 Frangipane 1989; Frangipane, Palmieri 1988-89; Frangipane (ed.), 2004; D'Anna present volume.
20 The huge numbers of bowls found in the palace area, often unbroken, are evidence of this. 11 Frangipane 2000c; Guarino 2008; D'Anna, Guarino present volume.
22 Palmieri 1978.
23 Frangipane 1993a.
24 Frangipane 2000c; 2003; Alvaro, chapter III-1, present volume.
30
3i
Economic Centralisation in Formative States. Chapter II
Fig. H.5-Arslamepe, Period VII Th -v
(LL3-4). The ehte residential Buildin:
'g XXV (column building).
Studi di Preistorja Orientale 3
32
Fig. II.6 - Arslantepe, Period VII, final phase (lc4). Temple C: plan and reconstruction (a): the niched NE corner in the main room (by, examples of mass-produced bowls (c) and cretulae (d).
33
Economic Centralisation in Formative States. Chapter 11
generally ceremonial, function can be seen, in my opinion, in the architectural features which resemble the standard layout of 'Temple' buildings in Mesopotamia: as in the case of our Temple C, they are characterised by a huge central room, disproportionately large in comparison with the very small side rooms, with a podium-platform in its centre facing an 'altar' leaning against the rear wall25 (the latter is probably evidenced in Temple C from a fragmentary small base, being that part of the wall greatly damaged). The tripartite ground plan, which so far has only been found in this one building in Arslantepe, and the presence of multiple recessed niches decorating the corners of the central room (fig. II.6: b) are wholly reminiscent of Mesopotamian architecture, even though the structure also exhibits features belonging to local, or at least northern traditions, such as the presence of wall paintings in the north-eastern corner of the large central room26 and the use of wooden beams arranged horizontally underneath the building, perhaps to distribute the weight and increase its stability27 (fig. 11.6: a). Temple C, which belongs to the final building phase of period VII, stood imposingly on the platform lifting it above the surrounding ground, probably ensuring that it was also clearly visible from the surrounding plain.
To the north-east of it, a series of long rooms with thick walls (A850-848-858-842)28 might have been linked to the temple by hosting activities and services somehow related to its functions, but these are not clearly proven by the archaeological record; at all events they had been, at least partly, used for craft activities: on the floors, pieces of semiprecious green stone waste have been found together with ochre fragments and numerous unused obsidian arrowheads29. In the westernmost room (the one nearest to the temple), 87 cretulae, the majority still bearing stamp seal impressions, had been interestingly discarded in small groups, and hence obliterated, under the floor.
Cretulae from period VII have only been found in this final level, of which, unfortunately, the only extant remains are, besides some pits, the temple and the long rooms mentioned above, which might have been somehow linked to it (fig. II.6: d). All we can therefore say for the moment is that the administrative activity so far documented in period VII developed at the end of the period and was exclusively linked to the elite area, particularly to ceremonies and practices bound up with cult or ritualised feasts. A total of 334 cretulae, most still bearing stamp seal impressions, have been found in this final level of period VII, partly discarded (under the floor of room A850 and in refuse pits in the same elite area), but the majority (179) indeed concentrated in Temple C. Most of them (162), moreover, were grouped in the southernmost of the eastern side rooms (A932), where they must either have been piled up, or had perhaps fallen from an upper storey.
Together with the cretulae, numerous mass-produced bowls have also been found in this room, thickly scattered on the floor, upside down, and in some cases piled up (fig. II.6: c); an extremely large number of these bowls were also found in the other northernmost side room (A950), always upside down, partly piled up on the floors and partly in the collapse layers
25 Margueron 2009.
26 Wall paintings seem to belong to a well-rooted Chalcolithic tradition in the Upper Euphrates regions from at least the Late Ubaid period. They have been found not only in the houses at Arslantepe VII but also in the Late Ubaid site of Degirmentepe, in the same Malatya plain (Esin 1983) and in the Late Chalcolithic levels at Norsuntepe, in the Bazig region (Hauptmann 1976).
27 A similar technique, which is poorly known, has been found in a similarly monumental building from the end of the fifth - beginning of the fourth millennium at Tell Brak (Oates et al. 2007, p. 589).
28 Alvaro, chapter III.l, fig. m.l 3 present volume.
29 Lemorini present volume.
Studi di Preistoria Orientale 3
after probably falling from an upper storey. This means that the bowls were collected and kept in huge quantities inside the building, probably ready to be used in ceremonials for the distribution of meals, which took place in the large central room30. For in this central room, to the south of the platform with hearth, dozens and dozens of mass-produced bowls have been found scattered all over the floor, practically the only type of object present in situ in the room, and indeed in the building as a whole, leaving no doubt that the main activity performed there was distribution of meals at celebrations, or on ritual and ceremonial occasions31. No similar concentration of bowls has been found in any other period VII building or in any other part of the settlement. The Temple was therefore the place where food was distributed and, as in most Mesopotamian contexts, including the temple areas atTepe Gawra32, these redistribution practices at all events entailed some form of administrative control, evidenced from the constant association of bowls and cretulae. An initial process of centralising foodstuffs, and almost certainly labour corvees, therefore took place, in period VII, within a 'religious' or cult environment, with the paramount social custom of distributing food in a ceremonial context.
It is interesting to note that the material culture of period VII at Arslantepe, particularly pottery, while reflecting the general Late Chalcolithic trend towards mass production, chaff pastes, a lack of decoration and poor firing, revealed a specifically local horizon, typical of the Malatya plain, in terms of shapes and repertoire, mainly connected with the regions to the west of the Euphrates and the Antioch plain ('Amuq F). Pots were mostly made on the slow wheel or handmade, and seem to have been produced in workshops using mass production techniques. Simple and coarse incised potter's marks appeared at this time on all categories of pots, but not on all individual pots (they are put on a low percentage of examples for each category), and are perhaps to be interpreted as distinctive signs applied by the potters on one sample from each batch of containers to make them recognisable subsequently, when they were taken to common areas for drying or firing33.
All this suggests that there may have been an increased demand for ceramic containers, particularly bowls for the distribution of meals, which must have driven to a growing mass production aimed at meeting the changed needs and the new demands from the central institutions. But this was a quantitative and not a qualitative development in production, led by craft workshops or individual independent craftsmen, as evidenced from the need for them to mark their products. Production was organised at the community level with different traditions from those of southern Mesopotamia, where the demand for ever-increasing numbers of containers for the distribution of meals or rations had led to the development of special products, such as the coarse bevelled rim bowls, manufactured in environments (domestic?) and using techniques (hand-made manufacture) that differed entirely from those of the other classes of pottery, which were always more sophisticated and specialised. At Arslantepe, as indeed on most of the northern Late Chalcolithic 3 sites, the increased demand for bowls was met by the same local potters who also made all the other classes of containers, and organised their work accordingly, modifying their technology and their procedures. And no trace has been found of any bevelled rim bowls in period VII layers.
The tripartite floor plan and the decoration with niches in the large Temple C reveal a long history of relations with the southern world, probably dating back to the Ubaid period, giving
30 Guarino 2008; D'Anna, Guarino present volume.
31 Helwing 2003; D'Anna, Guarino present volume.
32 Rothman 2002; 2009.
33 Palmieri 1985: 194.
35
Economic Centralisation in Formative States. Chapter II
rise to parallel and shared developments of similar organisational systems and cultural models. The archaeological material from period VII however reveals no sign of any direct Mesopotamian influence in the Malatya plain in the first half of the 4,h millennium.
Continuity and change in Period VI A
In period VI A there was certainly a more marked reference to the formal models of the Uruk world, even though, as already stated, there were a great many original aspects and strong ties with the previous period VII.
One of the elements that changed most was the pottery. For Period VIA pottery was mainly wheel-made, rather fine, well-fired and using pale colours, and now also exhibited shapes and profiles clearly influenced by the Uruk models, whose aesthetic features dominated the potters' taste34. The repertoire also changed radically and was more varied and specialised than in period VII, with specific shapes for different uses, which also drew their inspiration from the Mesopotamian production, which had succeeded in responding optimally to the new needs for an increasingly centralised management of foodstuffs35. The vessels with a proper Uruk typology are however extremely rare and the pottery retains distinctive and typically local features. Mass production was limited to truncated conical bowls which were now made on the fast wheel. The potter's marks disappear. Now, in addition to the Uruk-inspired pottery, the repertoire included a hand-made red-black burnished ware, whose shapes, aesthetic taste, and production techniques reveal relations with the Anatolian world, mainly Central-Northern Anatolia36; this ware, which had already emerged with a small number of examples at the end of period VII, now became one of the typical products of the period, and seems to have been used primarily for the consumption of food in elite environments and in public ceremonial contexts (it almost exclusively consisted of high-stemmed bowls, bowls and cups, small jars and jugs). The manufacture of cooking pots, kitchen ware, and pithoi, all hand-made and perhaps produced in a non-specialised environment, differed even more37.
Pottery production in period VI A therefore reflect changed demands and new needs on the part of the central institutions and the social elites, met by solutions that refer to Mesopotamian models, thereby indicating that the Uruk world, which was very close in this period thanks to the presence of colonial settlements in the Middle Euphrates, was without doubt a symbolic and real benchmark. On the other hand, the VIA pottery reveals an expansion of the site's external relations in many different directions, and the involvement in its sphere of action of many cultural and probably also ethnic components. Finally, the characteristics of this pottery manufacture show that the craftsmanship and organisation of production was now radically different from what it had been in period VII.
A marked continuity with the developments of the first half of the fourth millennium is conversely evidenced in the layout of the settlement, in the process of the formation of the elites, and in the way they exercised and consolidated their power, which, in my view, began exactly with the foundations laid down in LC 3-4 to considerably develop centralisation and new structures for governing the community. The organisation of the public and private elite
I
» Frangipane, fttafaj I983a; Frangipane ^ M   eu Anna Present volume. Frangipane, Palumbi 2007- Palumhi ?nn» x^,   ,.
37        ™^ZT^^Ksmt vo,ume-
36
Studi di Preistoria Orientale 3
areas in the two periods reveals an evident continuity: in period VI A, new buildings, also quite imposing and also residential in character, were built above the period VII elite residential buildings38 in the same central-western area of the tell which was then the peak of the mound; and even though the public area in VI A was much more extensive, composite, and complex, it occupied the south-western slope beginning initially only a short away from the dismantled Temple C, being located, as the latter, in a more peripheral part of the mound in comparison with the elite houses. This seems to reflect the same idea that the higher-ranking lineages should reside in the upper, and more central, parts of the settlement, while the areas used for public functions, despite their much more monumental character, were arranged around them, overlooking the plain. This indicates not only close continuity in the use of the "prominent" areas, but probably also a linkage between the high status lineages of both periods.
Such a policy would also indicate a similar mental universe and similar needs in managing relations with the population and with the territory: one crucial factor that certainly gave the organisation of the elites at Arslantepe a particular character even at the time of its full development, was the lack of urbanisation, which made the site a powerful early state, political, economic, administrative and religious centre, without any urban environment to which to refer, interacting with a rural population and a rural territory with little internal differentiation in terms of their features and functions39. The residents on the site in period VI A were even less than in the previous period, and were perhaps only limited to high-ranking lineages, as is suggested by the fact that the only houses found so far are the prestigious structures located on the upper part of the mound, and that very little evidence from this period has so far been found in the NE part of the hoyuk, which had conversely been heavily occupied in period VII.
Even though the glyptic and the administrative system had grown and expanded considerably in period VI A, they exhibited strong continuity and close links with the previous period, also iconographically40. Likewise, the marked redistributive character of the elites organisation does not seem to have been the result of emulation or of any new external influences from the Uruk world - with which they would certainly have had relations - but the origins lie unmistakably in the ceremonial redistribution system evidenced in Temple C.
The great development of Arslantepe in the palatial period would therefore seem to have been the final phase of an internal process of growth in the authority of the elites and their ability to centralise and manage resources. The site probably acquired a new regional role, too, which brought it into close relations both with the expanding Syro-Mesopotamian cultures and with the mountainous areas of Central-Northern and North-Eastern Anatolia, with their plentiful raw materials, particularly metals, while always retaining a pronounced political and cultural autonomy.
While the area of Malatya thus played an important and active role in the process of the formation of the State, in close connection with the parallel phenomena in the Mesopotamian world, it also brought about its own, different form of development, not based on urbanisation, which made the system less stable and less well-established then what occurred in highly urbanised environments, and ultimately led to its rapid collapse.
In the final phase of the palace, as clearly shown by C. Alvaro in chapter HI.2 in this volume, the monumental character of the entrance was downsized, accompanied by the construction of a kind of fortification wall, with an evident reduction in the symbolic, ideological
~^~anelor^ Alvaro, chapter HU, presenrvdume. » Frangipane et at. 2007: 475-477; Frang.pane 2009. «° Human 2007b.
37
Economic Centralisation in Formative States. Chapter II
and prestige aspects: wall paintings were coated with plaster layers or mud cover, and Temple A was converted into a less sacred building in which more common activities and forms of redistribution were performed41. The discovery of the weapons in Building III might also suggest that the chiefs now played a new important role, due perhaps to a possible increase in strife, and the emergence of a new threat to the power of the elites and to their activities.
The collapse of the centralised system: instability at the beginning of the 3r and the 'devolution' of the Arslantepe centralised society
millennium
Around 3000 BC, a devastating fire completely destroyed the palace, bringing to an end the Late Chalcolithic early state organisation and its system of interregional relations. The central institutions underwent a radical crisis, overwhelming the centralised system and bringing about its final collapse, while new relations with the northeastern Anatolian world and the regions south of the Caucasus gradually gained the upper hand, powerfully conditioning developments at Arslantepe in the third millennium.
The Malatya region in the first few centuries of the third millennium (Early Bronze I, Arslantepe period VI B) was highly unstable, perhaps embroiled in conflicts. The site was partially abandoned, followed by what were probably seasonal occupations by groups of transhumant pastoralists only using red-black burnished ware, who were linked to the world of the so-called Kura-Araks culture42 (period VI B1).
This short period was followed by a phase in which the settlement was rebuilt (period VI B2) with characteristics that were wholly different from what they had been before43. It was now a village made up of small mud brick houses separated by narrow streets, with evidence of domestic activities and with the floors covered by kilos of charred seeds (mainly barley, and to a lesser degree, wheat and legumes)44 which had probably fallen from the upper floor, where they may have been stored after harvesting and before the fire destroyed the village. There were also courtyards and open areas for communal activities, such as slaughtering animals (A199-203, A310)45 (fig. II.7: b) or copper smelting (A645-671)46 (see fig. III. 1.6 in Alvaro, chapter III.l of the present volume).
On the top of the mound they built an imposing mud-brick fortification wall, 4 metres thick, resting on a 5 m thick stone foundation, with internal buttresses, which surrounded the upper part of the settlement, separating it from the rest of the village, that spread out at its foot along the slope of the mound (fig. II.7: a). We do not yet know the features of the walled zone because all that has been unearthed so far is a nwd-paved wide open area, as a kind of open space or wide road running along the wall, but the huge dimensions of the fortification suggest that it must have surrounded a kind of citadel, or at least a socially separate, and hence prominent, area.
Attempts were made to establish new forms of power on the site, which may have been the legacy of the powers of the old local elites, or perhaps the expression of a new political class that had emerged from the radical reorganisation of the regional system. There are evident
See chapters HI2 VIII and XII by Alvaro, D'Anna and Lemorini respectively, present volume « Frang.pane, D, Nocera. Palumbi 2005; Frangipane, Palumbi 2007; Palumbi 2ob£
Frangipane, Palmieri 1983b: 523-574; Frangipane 1998b. m Sadori et at. 2006a; Susanna 2007. 45 Bokonyi 1983; Bartosiewicz 1998. * Frangipane, Palmieri 1994-95; Di Nocera present volume.
Studi di Preistoria Orientale 3
~. vi R? ŕ FBI) View of the toruncaum. Rg. II.7 - Arslantepe, Penod VI B2 (hBI).   ^ ^ ^ ^
of the fortification »all on the top of the mound ia) and houses
39
38
Studi di Preistoria Orientale 3
elements of continuity both in cultural features and in the production modes from the previous period VI A. In period VI B2 there was a revival at Arslantepe of pale wheel-made ware of Uruk derivation, which, though showing an overall change in the repertoire, retained many of the features of the previous period and could at all events be ascribed to the ancient ceramic tradition of Uruk origin47. Livestock, too, which changed radically to concentrate on sheep and goats starting from the end of period VII and then particularly in period VI A48, retained the same features49, thus indicating that a deep-seated structural change had taken place in the modes of production practised by the population of the plain in this sector around the mid fourth millennium.
But the way in which the new political authority was exercised, and the characteristics of that authority, changed completely: there were no longer any traces of centralisation and the redistribution of resources, or of administration, or of ideological and religious or ceremonial mediation. In this period there were no longer any mass-produced bowls, or cretulae, or temples. Whereas the common population returned to occupy the site, building an extensive rural village, power seemed above all to manifest itself in the construction of defence structures and in the warlike character of the chiefs50.
This becomes very clear from the discovery of an exceptional burial of an important personage, perhaps a chief, on the edge of the settlement, outside the central fortified area, which has become known as the 'Royal Tomb' (Tl): a carefully built stone cist excavated at the bottom of a large pit contained the remains of an adult man who had been buried with a wealth of grave goods, with four adolescents who had been sacrificed on the cist cover (fig. II.8)51. The person inside the cist was a male aged about 40, resting on a wooden 'bed', and probably wrapped in a shroud, accompanied by numerous pots, personal ornaments made of silver, gold, cornelian, and rock crystal, with a small hoard of metal objects placed behind his back, mainly comprising arsenical copper weapons (spearheads, a sword, and 4 daggers), working tools and other silver and copper-silver alloy ornaments. Whereas the weapons indicated that the person had been a warrior, the other objects emphasised the prestige of the chief probably also through the inherent value of the metal used.
The new political/military power, hypothesisable from the nature of the gifts in this extraordinary burial and the construction of the huge fortification wall on the top of the mound, was nevertheless unable to withstand the contradictions that had emerged in the territory of Malatya after the collapse of the fourth millennium centralised system; these, in my opinion, would have been internal contradictions, coupled with possible conflicts that may have broke out with new ethnic-social and economic groups (pastoralists) spreading throughout the plain.
The outcome was another collapse, and the breakup and splintering of the communities in the region between 2750 and 2500 BC (periods VI C and VI Dl), until a new resettling occurred in the second half of the third millennium, in the course of period VI D2.
This turbulent and well-documented historical development of the site, with its many splits and traumas, that also reveals the internal political, social and economic developments taking place there, is an .deal field for attempting the study we shall be addressing in this volume.
Frangtpane, Palmier* 1983b: 523-574; Piccione present volume. Bokony. 1983; Bartosiewicz and Palumbi present volume Bokony, 1983; Bartosiewicz 1998; Palumbi present volume. *' Frangipane 2001a.
Sl Frangipane et al. 2001; Frangipane 2007-08.
Fig. 11.8 - Arslantepe, Period VI B2. The
so-called "Royal Tomb" with some of the rich metal gifts.
41
40
Economic Centralisation in Formative States. Chapter II
Period VII, which was followed in such close continuity by the growth of the early state system documented in the Arslantepe palace, provides important elements for comparisons in the analysis of the formation process of a solid centralised economic system and its production and socio-political bases; period VI B2, conversely, enables us to make a comparative examination of a local community life which followed the early state phase in historical continuity to it, in which it is archaeologically documented that all the economic centralisation features had disappeared.
42