The Development of Pre-state Communities in the Ancient Near East: Studies in Honour of Edgar Peltenburg (Banea Publication)

An offprint from

DevelOpment Of pre-StAte COmmunItIeS In the AnCIent neAr eASt

edited by Diane Bolger and Louise C. Maguire

© Oxbow Books 2010 ISBn 978-1-84217-407-4

CONTENTS Editors’ Preface List of Contributors

vii ix

INTRODUCTION 1

The development of pre-state communities in the ancient Near East Diane Bolger and Louise C. Maguire

PART 1: SOCIAL 2 3

4 5 6

ORGANISATION AND COMPLEXITY IN PRE-STATE COMMUNITIES

Social complexity and archaeology: A contextual approach Marc Verhoeven Late Neolithic architectural renewal: The emergence of round houses in the northern Levant, c. 6500–6000 BC Peter M. M. G. Akkermans Abandonment processes and closure ceremonies in prehistoric Cyprus: In search of ritual Demetra Papaconstantinou A different Chalcolithic: A central Cypriot scene David Frankel Thoughts on the function of ‘public buildings’ in the Early Bronze Age southern Levant Hermann Genz

PART 2: EARLY

1

11

22 29 38 46

URBAN COMMUNITIES AND THE EMERGENCE OF THE STATE

The Tell: Social archaeology and territorial space Tony Wilkinson 8 Rethinking Kalopsidha: From specialisation to state marginalisation Lindy Crewe 9 From kin to class – and back again! Changing paradigms of the early polity Anne Porter 10 Different models of power structuring at the rise of hierarchical societies in the Near East: Primary economy versus luxury and defence management Marcella Frangipane 11 States of hegemony: Early forms of political control in Syria during the 3rd millennium BC Lisa Cooper

7

PART 3: TECHNOLOGY,

55 63 72

79 87

ECONOMY AND SOCIETY

12 A household affair? Pottery production in the Burnt Village at Late Neolithic Tell Sabi Abyad Olivier Nieuwenhuyse 13 Late Cypriot ceramic production: Heterarchy or hierarchy? Louise Steel 14 The domestication of stone: Early lime plaster technology in the Levant Gordon Thomas

97 106 117

Contents

vi

15 Domestication of plants and animals, domestication of symbols? Danielle Stordeur 16 Herds lost in time: Animal remains from the 1969–1970 excavation seasons at the Ceramic Neolithic settlement of Philia-Drakos Site A, Cyprus Paul Croft

PART 4: AGENCY,

123

131

IDENTITY AND GENDER

17 Agency in the Pre-Pottery Neolithic A Bill Finlayson 18 Understanding symbols: Putting meaning into the painted pottery of prehistoric northern Mesopotamia Stuart Campbell 19 Gender and social complexity in prehistoric and protohistoric Cyprus Diane Bolger 20 The painting process of White Painted and White Slip wares: Communities of practice Louise C. Maguire 21 The ceramic industry of Deneia: Crafting community and place in Middle Bronze Age Cyprus Jennifer M. Webb

PART 5: INSULARITY,

141 147 156 165 174

ETHNICITY AND CULTURAL INTERACTION

22 Outside the corridor? The Neolithisation of Cyprus Carole McCartney 23 Contextualising Neolithic Cyprus: Preliminary investigations into connections between Cyprus and the Near East in the later Neolithic Joanne Clarke 24 Was Çatalhöyük a centre? The implications of a late Aceramic Neolithic assemblage from the neighbourhood of Çatalhöyük Douglas Baird 25 The birth of ethnicity in Iran: Mesopotamian-Elamite cross-cultural relations in late prehistory Andrew McCarthy

185

197 207

217

1 INTRODUCTION: THE DEVELOPMENT OF PRE-STATE COMMUNITIES IN THE ANCIENT NEAR EAST Diane Bolger and Louise C. Maguire

Archaeological research on the small-scale societies which preceded the emergence of the earliest states in the Near East has a long and complex history which we cannot hope to review in detail in this brief introduction. However, it is important to situate the papers in this volume within a broad theoretical and methodological framework that distinguishes them from earlier approaches based on generic typologies of social organisation and unilinear trajectories of socioeconomic development. In the first section of this chapter we summarise those earlier approaches and review the challenges to unilinear models of social change which have continued to emerge over the last 20–30 years. In the second section we consider some of the more recent approaches to the study of pre-state societies in the Near East which have effectively replaced the earlier models and which provide a backdrop for the subsequent chapters of the book; these are introduced in the third and final section.

Challenging Traditional Models of Social Transformation in Pre-state Societies Neo-evolutionary categories of social organisation, as developed by anthropologists such as Service (1962, 1971) and Fried (1967) and further elaborated by Harris (1979) and Johnson and Earle (1987), were based on a broad social typology (bands-tribes-chiefdoms-states) that became highly influential in archaeological research during the 1970s and 1980s. While this tendency continued to some degree during the 1990s (e.g. Maisels 1990; Earle 1997), it began to fall out of use during the late 1980s as doubts arose among archaeologists concerning the ability of a limited number of abstract societal types to account for the great degree of variability and heterogeneity in past societies

(for general accounts of these developments, see Trigger 1989, chap. 9; and Renfrew and Bahn 2000, chap. 5; for more specific critiques, see McGuire 1983; Shennan 1993; Yoffee 1979; 1993; 2005; and Verhoeven, this volume). An increasingly widespread attitude to neo-evolutionism among archaeologists today is perhaps best expressed by Yoffee, who has called it “an illusion of history” (2005, 231). Critics of neo-evolutionism objected not only to the abstract or even fictive categories of analysis that comprise the band-tribe-chiefdom-state model, but also to the unilinear direction of social change inherent in neo-evolutionary thought. The assumption that cultures inevitably pass through a sequence of stages or steps from simple to complex, prestate to state, and that this occurs in a progressive stadial fashion, is no longer accepted by archaeologists today other than in the most general terms. In addition to the fact that evolutionary models cannot be sustained in the face of the detailed evidence that has accumulated over the last quarter of a century, neo-evolutionism can be regarded as a ‘meta-narrative’ that reflects the ethnocentric bias of Western thought (Rowlands 1989, 36). As Yoffee, Feinman and others argued, archaeologists needed to find alternative trajectories to social inequality and complexity and to understand transitions between the two within specific historical contexts (Yoffee 1993; Feinman 1995, 273–294; see also Bender 1989, 87). It is now widely acknowledged that social change among early societies such as those of the ancient Near East is likely to have been recursive and disruptive rather than unilinear, and that for a variety of reasons change occurred at different rates and in different ways from region to region and even from locality to locality (e.g. Renfrew 1984, 358–359; Shanks and Tilley 1987, 175–185; Price and Feinman 1989; Peltenburg 1993; Miller, Rowlands and Tilley 1995).

2

Diane Bolger and Louise C. Maguire

In response to these and other like-minded criticisms, many archaeologists have chosen to avoid the word ‘evolution’ in their research vocabulary and to replace it with other terms, such as ‘change’ or ‘transformation’ (e.g. Gledhill et al. 1995; Shanks and Tilley 1987), while others have adopted multilinear models of social change which highlight the variable paths to complexity demonstrated by the archaeological record (e.g. Sanders and Webster 1978; Trigger 1985). A relatively small number of archaeologists, however, continue to use ‘evolution’ cautiously in the belief that it still provides a useful framework for discussion if carefully defined (e.g. Stein and Rothman 1994; Yoffee 2005). In the present volume the word ‘development’ has been adopted deliberately to circumvent the more problematical connotations of ‘evolution’. ‘Development’, however, is not without its own connotations of progress, so it should be stated from the outset that the use of that term in this book is not intended to imply a uni-directional trajectory of social change. The view that ‘simple’ societies develop inevitably into increasingly complex forms of social organisation needs to be investigated rather than assumed; current approaches in archaeology acknowledge that there were variable pathways to social complexity and that more often than not these were circuitous. The gradual move in archaeological interpretation away from broad evolutionary models of social complexity has been accompanied by a shift in focus from over-arching systems or processes of social behaviour that emphasise similarities between synchronous cultures, to smaller scale research that demonstrates diversity both within and between them. At the same time, there has been a tendency in recent years to question the notion that external forces are primary causes of social change; while factors such as environment, population, climate and technology are important ingredients in social complexity it is now widely felt that they cannot, in and of themselves, explain social change (see Bender 1978 for an early example of this view). A further widespread criticism of ecosystems-based research is the contention that it provides a deterministic view of social development that tends to leave people out of the equation (see, for example, Bender 1978; 1989; Shanks and Tilley 1987; Brumfiel 1992; Hodder 2004). Taken together, the challenges to neo-evolutionary models briefly summarised above have resulted in the emergence of new approaches to the study of past societies that are not based exclusively on theories borrowed from other social sciences (particularly social anthropology), a phenomenon which frequently “condemned the past to resemble some aspect of the present” (Yoffee and Sherratt 1993, 8). While it is generally agreed that ethnographic evidence can provide valuable insights into the behaviour and experiences of prehistoric communities, it cannot and should not be applied uncritically. During the last 20 to 30 years, increasing numbers of archaeologists have underscored the

need to develop theories and methodologies that are more in keeping with the nature of their own concerns, such as the study of material culture and an understanding of longterm changes in human history (e.g. Plog 1974, preface and chap. 1; Renfrew 1984, 13; Yoffee 1993, 74; Gledhill et al. 1995, 27); at the same time, greater emphasis has been placed on interpreting the evidence within social, rather than exclusively environmental or economic frameworks. We address some of these new perspectives in greater detail in the following section.

Current Approaches to Pre-state Societies: Diversity, Scale and Context Over the last few decades the study of pre-state communities in the Near East has moved significantly beyond the abstract, stadial models of the 1970s and ’80s to adopt more nuanced approaches based on the detailed evidence that has emerged from the numerous surveys and excavations in the region. These efforts have revealed the inability of broad typological categories (such as band, tribe, chiefdom, state) to explain the variability of social organisation and social relations present in the archaeological record and have encouraged the formulation of new, socially-oriented research agendas centred on issues of diversity, scale and context.

Diversity A renewed appreciation by Near Eastern archaeologists of the diversity of cultures, patterns of social interaction, and trajectories of socio-economic development has emerged from intensive fieldwork in the region during the last 25 years. This has resulted from a number of factors, not the least of which has been the extensive rescue work carried out by local and foreign teams in association with several large dam projects in south-eastern Turkey and northern Syria, which have led to the discovery of a multitude of new sites whose developmental patterns appear to have differed considerably from those further to the south. It has also resulted from the finer-grained methods of modern archaeological fieldwork with its sophisticated means of recovering, recording and analysing data, and the multitude of specialists capable of generating detailed interpretations of the landscape, environment, technology, diet, material culture, and many other aspects of social life. Dating methods have also been refined, so that the temporal relationships between sites have become more clearly visible. The result of all of these developments has been a greater emphasis on difference and diversity that has made the traditional classification of societies into a limited number of ‘types’ appear to be overly simplistic. Similarly, the diverse nature of the archaeological evidence for the development of complex society has encouraged archaeologists investigating

1. Introduction: The development of pre-state communities in the ancient Near East social change among early communities in the Near East to move beyond the broad, programmatic schemes of research based on a single evolutionary model. As we shall discuss later in this section, the ability to achieve a more nuanced understanding of early societies is most successfully achieved by interpreting evidence within particular historical contexts (Hodder 1986). As Gledhill et al. have observed, “In the end there is no way to resolve the issue of state origins without attempting a detailed reconstruction, via archaeological materials, of the critical transitions which gave birth to the first manifestations of ‘civilization’ ”(1995, 25).

Scale The shift in focus from general to specific and from similarity to difference has underscored the importance of scale in archaeological research. As numerous studies over the last few decades have shown, greater attention to the micro-scale is crucial for looking at society from the bottom up rather than the top down (e.g. Renfrew 1984; Renfrew and Bahn 2000, chap. 5); for constructing alternative pathways to the accumulation of wealth, power and social inequality (e.g. McGuire 1983; Bender 1989; Price and Feinman 1995); and for understanding the processes by which complex society emerged and developed (e.g. Rowlands 1989; Stein and Rothman 1994). Consequently, archaeological research has become more specific by investigating particular groups, sub-groups and individuals, and the relationships between them, rather than focussing on abstract categories such as ‘society’ or ‘culture’. Research on the micro-scale can be approached by a variety of methods, such as examining the ways in which various sectors or groups within society functioned or changed, a process Stein and Rothman refer to as ‘organizational dynamics’ (1994, 1); focussing on the social practices of day-to-day existence (Bourdieu’s habitus; see Bourdieu 1977); looking for evidence of individuals (e.g. Knapp and Meskell 1997; Meskell 2002; Meskell and Joyce 2003); or investigating changes in the human life course (Gilchrist 1999, 88–100; 2004; Bolger 2004; 2008). All of these approaches regard the seemingly mundane activities of daily life as essential for understanding the relationships between people and the material world and demonstrate the need to develop theories of social change based on individuals and groups as active agents, rather than passive adapters to extrinsic environmental and economic forces. Temporal dimensions of scale are equally important. According to Annalist models, social change should ideally be investigated at multiple scales (short, medium and long term trajectories) that operate simultaneously. The coarse chronologies that most prehistorians are compelled to work with seem better suited to the investigation of long-term change, yet as several archaeologists have noted (e.g. Hodder 1986, 93; Feinman 1994; Bolger 2008), few archaeologists have specifically addressed particular social

3

issues over considerable expanses of time. Exclusive focus on short or medium range scales fails to take full advantage of the longer term trajectories that archaeological evidence can illuminate and that are essential for formulating theories of social change on a broader scale. Moreover, while long-term developments cannot directly shape the perceptions and behaviours of individuals over the short term span of a human life, they can serve as “constraining structures” that influence the range of possible behaviours and choices available to individuals and communities in the processes of daily life (Bintliff 1991, 7). These and other innovative applications of Annalist thinking to archaeological interpretation emphasise the multidimensional nature of time and its variability within and between cultures. This view of time effectively serves to ‘humanise’ the more abstract, objective approaches to the past by providing links between long-term trajectories of social change and the shorter, more subjective time scales of individual experience (Gosden 1994, 9).

Context The interpretation of evidence at different scales is closely related to questions of archaeological context. The view that it is essential to interpret the past in particular historical contexts (Hodder (1986, chap. 7) is now widely accepted, and contextual approaches have proved to be one of the most successful means of overcoming the limitations of broad evolutionary models of social change discussed above. While context in archaeology occurs in a wide range of dimensions, including temporal, spatial, typological and depositional (Hodder 1986, 125), it can also be understood to include the broader cultural and theoretical frameworks in which archaeologists interpret the past. The recognition that archaeological interpretation is a process involving an interactive or dialectical relationship between the archaeologist and the evidence lies at the heart of current research programmes and marks a radical departure with traditional methods of archaeological inference. As Tilley has stated, “contemporary archaeology is not a tabula rasa on which the context of the archaeological record simply inscribes itself awaiting its meanings to be captured” (1993, 9). Interpreting evidence of past societies within broader cultural and theoretical contexts places demands on archaeologists for a “comprehensive internal study of archaeological cultures” (Trigger 1989, 350) and for greater emphasis on the social context of cultural change (Price and Feinman 1995, 9). The recognition that groups and individuals can bring about transformations in social organisation, for example, reveals the limitations of processual models in which the environment was regarded as a prime mover that motivated people to change in order to more successfully adapt. While it is, of course, important

4

Diane Bolger and Louise C. Maguire

to acknowledge the constraints placed on individuals and groups by climatic, environmental and demographic forces, the greater degree of engagement with ‘the social’ in current archaeological interpretation helps us to appreciate the complexity and indivisibility of human experience in past societies and to recognise the multiplicity of meanings that can result by interpreting evidence within various contextual frameworks. It is fair to assert that a different kind of archaeology has emerged from the focus on diversity, scale and context outlined in the previous pages. This new way of looking at the past (which includes but is not limited to post-processualism) includes research on individuals, personal relations, kinship relations, social interactions, and individual and social identities; it considers questions of status, age, gender, cognition, habitus, performance, the body and social memory; it adopts a bottom up rather than a top down perspective; and it advocates a phenomenological approach centred on the active engagement of people with their environment (Renfrew and Bahn 2000, chap. 5). All of these areas of research fall within the rubric of social archaeology since to a large degree they are the result of human agency, inter-personal relationships and social networks. As shall become evident in the following section, the various papers in this book can be included in this new socially-oriented research agenda since collectively they reflect a growing concern within the field of ancient Near Eastern archaeology for issues of agency, difference, identity, community, materiality, ritual practice and cultural interaction, as well as their variations through time and space.

The Development of Pre-state Communities in the Near East The essays in this volume cover a variety of themes related to the development of pre-state communities in Anatolia, Cyprus, the Levant and Mesopotamia. They have been grouped, however, not by chronological period or geographical setting but according to the particular issues they address. In this way it is hoped that both the similarities and differences in theoretical and methodological approaches adopted by authors carrying out research in different regions of the ancient Near East can be appreciated more readily. The papers in Part 1 consider aspects of social organisation in pre-state communities while those in Part 2 investigate the structure of early urban communities and the development of the state. The contributions in Part 3 examine the interrelationships between technology, economy and society and those in Part 4 address various constructions of agency, identity and gender. Papers in the final section, Part 5, deal with issues of insularity, ethnicity and cultural interaction. All of the papers are introduced briefly in the pages that follow.

Part One: Social Organisation and Complexity in Pre-state Communities The first paper in this section, by Marc Verhoeven (Chapter 2), examines the concept of social complexity, a term which is still widely used by archaeologists, including many of the contributors to this volume, to indicate differences in the socio-economic structures of prehistoric communities. He argues that words such as ‘simple’ and ‘complex’ are value-laden terms which serve to promote the continuation of neo-evolutionary thinking; they can also distort our understanding of pre-state communities since various aspects of ‘complexity’, such as ritual behaviour, can be identified in some ‘simple’ Neolithic societies. The questions raised in this paper form a useful framework for the more specific treatments of social organisation that follow. The other papers in this section consider various aspects of social organisation and social practice among particular prehistoric communities in Cyprus, the Levant, and Mesopotamia. In Chapter 3, Peter Akkermans examines the shift from rectangular to circular buildings in SyroMesopotamia, which began around the middle of the 7th millennium or shortly afterwards and increased in importance during the 6th millennium. While a number of theories have been proposed to explain this change, Akkermans suggests that social characteristics, such as mobility and temporary habitation in small autonomous groups dispersed over the landscape, are of central importance to this discussion. In the following paper (Chapter 4) Demetra Papaconstantinou examines the life histories of houses in prehistory, focussing in particular on evidence from the Late Neolithic site of Ayios Epiktitos-Vrysi in Cyprus. After addressing issues of materiality, identity and ritual in the archaeological record, she looks at the relationships between ritual and domestic life and underscores the need for archaeologists to look for different ways of ‘narrating’ the past by drawing upon multiple strands of archaeological data. In Chapter 5, David Frankel presents evidence from the recently excavated site of Politiko-Kokkinorotsos in central Cyprus during the Middle-Late Chalcolithic periods. After summarising the major features of the excavations, and comparing the results with better-known material from the south-west of the island, he offers an alternative perspective on local economic patterns, cultural regionalism and chronology on the island during the first half of the 3rd millennium BC. The final paper in this section, by Hermann Genz (Chapter 6), examines the interpretation of EB II–III settlements in the Southern Levant as city-states, urban communities, or corporate villages by questioning the function(s) of nondomestic structures which have traditionally been regarded as public buildings and temples. Since a number of sites in the region have buildings that do not fit neatly into these categories, the socio-political organisation of these

1. Introduction: The development of pre-state communities in the ancient Near East communities is likely to have been more complex than has generally been acknowledged.

Part Two: Early Urban Communities and the Development of the State The papers in this section investigate the structure and development of early urban communities in the Near East, as well as the social processes by which particular communities became, or failed to become, more complex states. Chapter 7, by Tony Wilkinson, is concerned with the social archaeology of the Middle Eastern tell. By examining multiple strands of evidence (i.e. stratigraphic sequences, landscape archaeology, ethnographic evidence and historical documents) Wilkinson considers the ways in which the settlement community as a corporate group may have functioned and developed through time. He concludes that tells and the fields that surrounded them formed an indivisible social and landscape unit in which the social and economic processes of everyday life were carried out. In Chapter 8, Lindy Crewe discusses the social and economic forces that led to a concentration of power within the coastal centres of Cyprus during the Late Cypriot II period by considering evidence from the site of Kalopsidha. This small urban community in eastern Cyprus, which was located inland, played an important role in international relations during the transitional Middle Cypriot III–Late Cypriot I period. Crewe suggests that this may be due to its possible specialisation as a cult or agricultural production centre, and that its eventual marginalisation and abandonment was closely linked to the emergence of nearby Enkomi as a dominant urban coastal emporium. The paper by Anne Porter (Chapter 9) demonstrates the limitations of traditional models of the early state for explaining the emergence of early polities in the ancient Near East. On the basis of evidence concerning settlement hierarchies, settlement morphologies and burial practices in the Euphrates and Jazirah regions of Syria, she argues for kinship as a central attribute of social organisation and explores the complex and seemingly contradictory relationships between kinship and class that lay at the heart of the transformation from pre-state to state level political organisation in these regions. In Chapter 10, Marcella Frangipane examines the dialectical interplay of social and environmental factors that led to the development of different economic and political systems in the prehistoric Near East. By considering evidence for the growth of social complexity and political hierarchy in three different areas (northern Mesopotamia, southern Mesopotamia and western Anatolia) she demonstrates the inability of unilinear models of social change to account for the variable trajectories of social and economic development adopted in different regions. In the final paper (Chapter 11) Lisa Cooper looks at

5

archaeological and textual evidence from Ebla and Mari in Syria to investigate the emergence and development of early Near Eastern polities. By examining the social, economic and political developments at these sites both before and after they exerted hegemonic authority over their surrounding regions, she highlights the importance of considering long-term changes in material culture in order to effectively understand patterns of behaviour associated with increasing levels of socio-economic complexity.

Part Three: Technology, Economy and Society These chapters investigate technological and economic aspects of pre-state communities in Near Eastern society within particular social and temporal frameworks. The first paper, by Olivier Nieuwenhuyse (Chapter 12), discusses innovations in ceramic technology and style that transformed ceramic assemblages across Upper Mesopotamia in the late 7th millennium BC by focussing on pottery production at the Burnt Village of Tell Sabi Abyad in northern Syria. The detailed evidence for the production and use of pottery at this site indicates a complex network of intra- and inter-communal relationships involving the manufacture, consumption and exchange of raw materials and finished products, and furnishes insights into the ways in which material culture was used actively by pre-state communities to create and maintain social identities. In Chapter 13, Louise Steel considers some of the wider issues of socio-political organisation associated with developments in ceramic production and craft specialisation in Late Bronze Age Cyprus. After examining evidence for regionalism in ceramic traditions and tracing changes in ceramic production through time, she concludes that prior to the 13th century BC the island was divided into small regional groups that may reflect heterarchical rather than hierarchical forms of social organisation. Only in the 13th and 12th centuries did ceramic production on the island cut across regional boundaries, perhaps in connection with the increasingly centralised organisation of ceramic production and political control. In the following chapter (Chapter 14), Gordon Thomas considers the social practices and beliefs associated with the creation and development of lime plaster technology in the Neolithic period of the Levant. Following ideas developed by Pierre Lemonnier and others, Thomas focusses on the social variables of lime plaster technology rather than the functional aspects of its manufacture and use, and considers the roles it is likely to have played in ritual and cognitive dimensions of social behaviour at this time. Drawing upon Jacques Cauvin’s concept of the ‘revolution of symbols’, by which he characterised the origins of the Neolithic, Danielle Stordeur (Chapter 15) investigates the relationships between the use of symbols and the domestication of plants and animals in the Levant during

6

Diane Bolger and Louise C. Maguire

the Pre-pottery Neolithic A and B. Differences in the symbolic repertoires of these two periods, which involved the replacement of animal figures during the Pre-pottery Nelithic B by representations of human figures, are used as metaphors for shifting patterns of social and economic development as agriculture and herding became part of everyday life. In the final chapter of this section (Chapter 16), Paul Croft considers patterns of animal exploitation in the Late Neolithic period of Cyprus by analysing faunal evidence from the site of Philia-Drakos A, which he has examined recently for the first time. While Croft’s study of this material is not yet complete, he postulates on the basis of an initial sample that the faunal patterns observed at Philia conform to the standard pattern of very heavily deer dominated animal economies observed at other Late Neolithic sites on the island. It is hoped that continued study of the animal remains from this important site will establish in greater detail the ways in which various communities in Late Neolithic Cyprus functioned and interacted.

Part Four: Agency, Identity and Gender The papers in this section, which concern themes of agency, identity and gender, address some of the current issues in contemporary social archaeology. The first paper, by Bill Finlayson (Chapter 17), adopts a ‘bottom up’ approach to the growth of sedentism, agriculture and materiality during the Pre-pottery Neolithic A period. By focussing on internal social processes, such as agency, and considering detailed archaeological evidence at the local scale, he demonstrates how our understanding of the important transformations that occurred at this time can be enhanced by looking at various ways that Neolithic people inhabited their world. In the following paper (Chapter 18) Stuart Campbell investigates the symbolic aspects of prehistoric painted pottery from northern Mesopotamia. While painted decoration of the Samarran, Halaf and Ubaid periods has traditionally been used to define cultural divisions or chronological periods, little research has been carried out to interpret the meaning of the motifs and designs used in this pottery. Campbell explores various methods of extracting meaning from abstract geometric motifs and considers the ways in which pottery can play an important role in the development of social interaction and integration among pre-urban communities. The paper by Diane Bolger (Chapter 19) considers longterm changes in gender and social identity in Cyprus from the Neolithic period to the Early Iron Age. After discussing the ways in which genderless narratives of the past distort our understanding of ancient societies, she argues for a contextual approach that draws upon multiple strands of evidence (mortuary, figurative, material, etc.) in order to shed greater light on the changes in gender constructs during

the prehistoric and protohistoric periods on the island. Her results suggest that gender constructs changed considerably over time but did not necessarily follow a straightforward path. In the following chapter (Chapter 20), Louise Maguire moves beyond the standard typological study of prehistoric pottery in Bronze Age Cyprus by presenting a systematic reconstruction of the painting processes of White Painted ware. On the basis of a detailed analysis of brushstroke sequences on more than 200 vessels, she shows that major differences existed in various regions of the island that were the result of the positioning of the pot during the painting process, and that these practices changed over time. The complete transformation of communities of practice that occurred at the end of the Middle Bronze Age reflects a wider picture of social transformation which took place on the island at this time. The final chapter in this section (Chapter 21), by Jennifer Webb, investigates the ceramic industry at the Middle Bronze Age site of Deneia in north-western Cyprus. Drawing on the extensive body of published material from the cemeteries at this site, she explores the use of form and decoration to produce the assertive ceramic style which is a marked feature at the site and considers the ways in which community-specific styles, forms and motifs were used to craft social identities and to assert and maintain community affiliations.

Part Five: Insularity, Ethnicity and Cultural Interaction Like the papers in the previous section, those in this final section of the book (Part 5) address issues that are of current concern in social archaeology. The paper by McCartney (Chapter 22) looks at processes of Neolithisation in Cyprus and challenges traditional concepts of early island communities based on a priori assumptions about ‘insularity’, ‘island colonisation’ and ‘isolation’. Using multiple strands of evidence (e.g. economic practices, stone tool technology and the built environment), she argues for an historical approach that places Neolithic Cyprus within a wider Near Eastern context, and concludes that patterns of increasing sedentism, materialism and village development on the island follow similar trajectories to those observed in other regions of the Near East. The paper by Joanne Clarke (Chapter 23) addresses many of the same issues raised by McCartney by questioning traditional assumptions about the ‘insularity’ of Cypriot communities during the later Neolithic period. Recent research in the central Levant, she argues, warrants a reconsideration of the island’s insularity and isolation after the Pre-pottery Neolithic B. By looking at evidence for chipped and ground stone technologies, pottery manufacture, settlement organisation, economy and environment, she

1. Introduction: The development of pre-state communities in the ancient Near East maintains that a complete cessation of interaction between island and mainland is a highly unlikely scenario and that cultural connections with the Levant probably continued at this time. In the following chapter (Chapter 24), Douglas Baird investigates the social and economic relationships between the Neolithic community of Çatalhöyük and contemporary settlements in the surrounding region. Through a comparative study of flint and obsidian technology, he tests the proposal made by some archaeologists that Çatalhöyük served as a center of economic production. Since there is little evidence to suggest that Çatalhöyük played a distinctive role in terms of the distribution of chipped stone, he proposes that its centrality may have had more to do with the social networks through which material flowed around the landscape and between source area and consumption zones than with centralised production. In the final chapter of the volume (Chapter 25), Andrew McCarthy considers evidence for ethnicity in late prehistoric Iran through a stylistic and statistical analysis of 4th and 3rd millennium seals from Susa. His results suggest that two distinctive craft traditions at this site, a Mesopotamian and a proto-Elamite style, co-existed during the 4th millennium and became more sharply divergent during the 3rd millennium. He concludes that these differences were based on ethnic distinctions and that ethnicity may have been an important aspect, or even a defining quality, of the expression of state-level identity in Iran at this time. As this body of papers demonstrates, unilinear models of social change, which for many years were entrenched in the archaeological literature, have undergone a profound transformation. They are being replaced by more sociallyoriented approaches that cannot be reduced to a single explanatory model, but which embrace a wide variety of theories and methods. Through their joint concern with issues of diversity, scale and context, they are beginning to shed new light on the ways in which pre-state communities in the ancient Near East functioned, interacted and changed.

References Bender, B. 1989. The roots of inequality. In D. Miller, M. Rowlands and C. Tilley (eds) Domination and Resistance. London, Unwin Hyman. Bintliff, J. 1991. The Annales School and Archaeology. London and New York, Leicester University Press. Bolger, D. 2004. Gender in Ancient Cyprus: Narratives of Social Change on a Mediterranean Island. Walnut Creek (CA), AltaMira. Bolger, D. 2008. Introduction: Temporal dimensions of gender in ancient Near Eastern archaeology. In D. Bolger (ed.) Gender through Time in the Ancient Near East, 1–20. Lanham (MD), AltaMira.

7

Bourdieu, P. 1977. Outline of a Theory of Practice. Cambridge, Cambridge University Press. Brumfiel, E. 1992. Distinguished lecture in archaeology: Breaking and entering the ecosystem − gender, class and faction steal the show. American Anthropologist 94(3), 551–567. Earle, T. 1997. How Chiefs Come To Power: The Political Economy in Prehistory. Stanford (CA), Stanford University Press. Feinman, G. M. 1994. Social boundaries and political change: A comparative perspective. In G. Stein and M. Rothman (eds) Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity, 225–236. Madison, Prehistory Press. Feinman, G. M. 1995. The emergence of inequality: A focus on strategies and processes. In T. D. Price and G. Feinman (eds) Foundations of Social Inequality, 255–279. London, Plenum. Fried, M. 1967. The Evolution of Political Society. New York, Random House. Gilchrist, R. 1999. Gender and Archaeology: Contesting the Past. London and New York, Routledge. Gilchrist, R. 2004. Archaeology and the life course: A time and age for gender. In L. Meskell and R. W. Preucel (eds) A Companion to Social Archeology. Oxford, Blackwell. Gledhill, J., B. Bender and M. T. Larsen (eds) 1995. State and Society: The Emergence and Development of Social Hierarchy and Political Centralization (1st paperback ed.; orig. hardcover published by Unwin Hyman in 1988.) London, Routledge. Gosden, C. 1994. Social Being and Time. Oxford, Blackwell. Harris, M. 1979. Cultural Materialism. New York, Random House. Hodder, I. 1986. Reading the Past: Current Approaches to Interpretation in Archaeology. Cambridge, Cambridge University Press. Hodder, I. 2004. The “social” in archaeological theory: An historical and contemporary perspective. In L. Meskell and R. Preucel (eds) A Companion to Social Archaeology, 23–42. Oxford, Blackwell. Johnson, A. W. and T. Earle 1987. The Evolution of Human Societies: From Foraging Group to Agrarian State. Stanford (CA), Stanford University Press. Knapp, A. B. and L. Meskell 1997. Bodies of evidence on prehistoric Cyprus. Cambridge Archaeological Journal 7(2), 183–204. Maisels, C. K. 1990. The Emergence of Civilization: From Hunting and Gathering to Agriculture, Cities and the State in the Near East. London and New York, Routledge. McGuire, R. H. 1983. Breaking down cultural complexity: Inequality and heterogeneity. In M. Schiffer (ed.) Advances in Archaeological Method and Theory 6, 91–142. New York, Academic Press. Meskell, L. 2002. Private Life in New Kingdom Egypt. Princeton, Princeton University Press. Meskell, L. M. and R. A. Joyce 2003. Embodied Lives: Figuring Ancient Maya and Egyptian Experience. London, Routledge. Peltenburg, E. 1993. Settlement continuity and resistance to complexity in Cyprus, ca. 4500–2500 B.C.E. Bulletin of the American Schools of Oriental Research 292, 9–23. Plog, F. 1974. The Study of Prehistoric Change. New York, Academic Press.

8

Diane Bolger and Louise C. Maguire

Renfrew, C. 1984. Approaches to Social Archaeology. Edinburgh, Edinburgh University Press. Renfrew, C. and P. G. Bahn 2000. Archaeology: Theories, Methods and Practice (3rd ed.). London, Thames & Hudson. Rowlands, M. A. 1989. A question of complexity. In D. Miller, M. A. Rowlands and C. Tilley (eds) Domination and Resistance, 29–40. London, Unwin Hyman. Sanders, W. T. and D. Webster 1978. Unilinealism, multilinealism, and the evolution of complex societies. In C. Redman, M. J. Berman, E. V. Curtin, W. T. Langhorne, Jr., N. M. Versaggi and J. C. Wanser (eds) Social Archeology: Beyond Subsistence and Dating, 249–302. New York, Academic Press. Service, E. 1962. Primitive Social Organization: An Evolutionary Perspective. New York, Random House. Service, E. 1975. Origins of the State and Civilization: The Process of Cultural Evolution. New York, Norton. Shanks, M. and C. Tilley 1987. Social Theory and Archaeology. Cambridge, Polity Press in association with Basil Blackwell. Stein, G. and M. Rothman (eds) 1994. Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity. Madison, Prehistory Press.

Trigger, B. 1985. The evolution of pre-industrial cities: A multilinear perspective. In F. Geus and F. Thill (eds) Mélanges offerts à Jean Vercoutter, 343–353. Paris, Éditions Recherche sur les Civilisations. Trigger, B. 1989. A History of Archaeological Thought. Cambridge, Cambridge University Press. Yoffee, N. 1979. The decline and rise of Mesopotamian civilization: An ethnoarchaeological perspective on the evolution of social complexity. American Antiquity 44, 1–35. Yoffee, N. 1993. Too many chiefs? (or, safe texts for the ’90s). In N. Yoffee and A. Sherratt (eds) Archaeological Theory: Who Sets the Agenda?, 60–78. New Directions in Archaeology. Cambridge, Cambridge University Press. Yoffee, N. 2005. Myths of the Archaic State. Cambridge, Cambridge University Press. Yoffee, N. and A. Sherratt 1993. Introduction: The sources of archaeological theory. In N. Yoffee and A. Sherratt (eds) Archaeological Theory: Who Sets the Agenda?, 1–9. New Directions in Archaeology. Cambridge, Cambridge University Press.

PART 1

SOCIAL ORGANISATION AND COMPLEXITY IN PRE-STATE COMMUNITIES

2 SOCIAL COMPLEXITY AND ARCHAEOLOGY: A CONTEXTUAL APPROACH Marc Verhoeven

Introduction The notion of complexity is commonly used by archaeologists to indicate differences in the socio-economic structure of ancient societies. The use of the term implies that there are non-complex (simple) and complex societies, with the latter having evolved out of the former (e.g. ‘the rise of socioeconomic complexity’). In many cases, therefore, the use of the concept of complexity is a form of (neo-)evolutionary thinking. More particularly, in Near Eastern archaeology the term ‘complexity’ surfaces time and again in relation to societies dated to the Bronze Age and later, i.e. to (incipient) states and ‘civilisation’. Complexity, however, is a very subjective term which immediately leads one’s thoughts in specific directions. But what exactly is it? Is there an alternative to using the notion of complexity in terms of successive types of society? Are there perhaps different forms of complexity? In this contribution these issues will be explored. First, to set the stage, a critical account of what has been termed the ‘rise of civilisation in the Near East’ is given. This is followed by a discussion of social complexity and a critique of the common use of the notion of social evolution. In the second part of the paper two examples are presented of complex (ritual) contexts in communities that are generally regarded as (socio-economically) non-complex.

The Rise of Civilisation? For many years, Near Eastern archaeology has been preoccupied with the rise of socio-economic complexity and the development of states and civilisation. It is one of the most popular themes in introductory books as well as in academic teaching. Frequently one reads about the evolution

or development of complex society (e.g. Rothman 2004) or the evolution of early states (e.g. Stein and Rothman 1994; Yoffee 2005). The term ‘civilisation’ is less common in academic circles nowadays, but major publications have used it to structure their narratives. Oates and Oates (1976) and Redman (1978), for instance, speak of the “rise of civilization,” Maisels of the “emergence” (1990) and “cradle” (1993) of civilisation, and Algaze (1993) of the “dynamics and expansion of early Mesopotamian civilization.” In fact, these publications are not only structured by this notion; they are largely founded on it. For example, in the first chapter of his book Redman (1978, 1) writes: The Agricultural and Urban Transformations as they occurred in the ancient Near East are among the truly significant milestones in the history of humankind. The social changes that these processes fostered influenced all aspects of society and formed the structure out of which today’s world has emerged. The Near East has been selected as the geographical setting in which to examine the rise of civilization because changes took place there at a very early date, perhaps earlier than anywhere else in the world. In addition to this temporal priority, the history and prehistory of the Near East directly affected the emergence and growth of Western civilization.

Although this was written more than 30 years ago and, as already indicated, the term ‘civilisation’ seems to be less common nowadays in academic contexts, it is still regularly used in popular media, and I suspect that Redman’s ideas are still shared by a large number of Near Eastern archaeologists, especially those dealing with the historic periods. Moreover, many prehistorians dealing with the so-called Neolithic Revolution explicitly or implicitly use the idea of a rise of civilisation as well (e.g. Özdoğan and Başgelen 1999; Hauptmann and Özdoğan 2007). There are three main problems with this idea. Firstly, like the notion

12

Marc Verhoeven

of complexity, the term ‘civilisation’ is very subjective and involves a value judgement. For most people it means to be civilised, but what, then, about people who did or do not live in civilisations? Were and are they uncivilised? Not many would probably dare to say this nowadays, but the notion of civilisation, consciously or unconsciously, sets up a qualitative difference. Secondly, the concept of a ‘rise of civilisation’ is teleological, i.e. it supposes that ‘pre-civilised’, pre-state, and pre-historic communities and cultures were designed for or directed toward a final result: (modern Western) civilisation. In other words, in an explicitly evolutionary fashion, prehistoric societies are treated as stages towards more developed conditions and not as societies in their own right. Thirdly, the use of the notion of ‘civilisation’ often goes hand in hand with that of complexity, generally without mentioning what exactly is meant by it (but see Rothman 1994). In fact, ‘complex society’ is nowadays frequently used as a synonym for civilisation. Just like the term ‘civilisation’, the terms ‘emerging complexity’, ‘complex society’ and so on implicitly make a distinction between non-complex (simple) and complex societies, with the former being regarded as somehow inferior to the latter. In the following pages I wish to address some of the problems related to this dichotomous way of thinking and to argue for a more nuanced use of the notion of complexity.

Social Complexity Complexity is an intricate issue which is studied in many different disciplines outside the social sciences, e.g. biology, computer science, physics, mathematics. In this contribution I shall obviously be concerned with social complexity, i.e. complexity in past human societies (e.g. Randsborg 1981; van der Leeuw 1981; Price and Feinman 1995). This is a huge field of study (there is even a Journal of Social Complexity), but for the purposes of this paper I shall limit myself to terminology and the problems of the concept of social complexity. There are many different definitions of complexity, but recurrently it is stated that it denotes a whole made up of differentiated and interrelated parts: the more parts and the more connections between parts, the more complex the system or society. Flannery (1972) was one of the first archaeologists to explicitly use the concept of social complexity. He argued that there were two main dimensions to be distinguished: segregation, the degree of differentiation and specialisation within a system; and centralisation, the degree to which the parts of the system were interrelated. In an analysis of chiefdoms and early states in the Near East, Rothman (1994, 4) used Flannery’s ideas to define complexity as: “... the degree of functional differentiation among societal units or sub-systems.” Such

units can be households, political associations or villages. Moreover, Rothman distinguishes two axes of complexity: a horizontal axis made up of the individual parts of the system (segregation); and a vertical axis consisting of hierarchical levels (centralisation or integration). These horizontal and vertical dimensions of complexity are still widely used in analyses. However, because it has been realised that complexity consists of many potentially independent variables, further divisions have been made, which have enabled the analysis and distinction of different forms and degrees of social complexity (e.g. McGuire 1983; Blanton et al. 1993). However, as R. Chapman (2003, 84) notes, the most basic distinction is that between ‘surface’ traits (e.g. vertical social differentiation) and ‘deep’ traits (e.g. segregation and integration). I argue that the ‘surface’ traits are dominant in the popular conception of complex societies, more particularly the surface traits of stratified societies (chiefdoms and states). Thus, among other things, urbanisation, large populations, craft specialisation, social differentiation, hereditary ranking, centralised production, long-distance trade, writing, bureaucracy, formalisation of law and the monopoly on the right to use force (i.e. characteristics of chiefdoms and/or states: see e.g. Nelson 1995; R. Chapman 2003, 7, 91) are regarded as typical for complex societies and social complexity. However, as we shall see, when ‘deep’, structural traits are used, and when it is accepted that, according to context, different degrees of complexity can exist within the same society, almost any society can be regarded as complex in some or many respects. In fact, some have even argued that social processes are too complex and particular to be rigorously modelled in terms of complexity (e.g. Stewart 2001).

Social Evolution It is clear that the notion of social complexity is firmly embedded in that of social (or cultural) evolution, characterised by subsequent stages from simple to complex with increasing degrees of technological, social and economic complexity (e.g. R. Chapman 2003, 4–8). In fact, the ‘rise of civilisation’ discussed above is wholly based on an evolutionary framework in which societies are progressively ranked according to their socio-economic complexity (e.g. Redman 1978, fig. 6.13). Such evolutionary schemes go back to anthropological thinking in the 19th century (e.g. Tylor 1871; Morgan 1877; Spencer 1967 [1897]; see also Childe 1951), but the most famous models are those of Service (1962) and Fried (1967), both of which are mainly based on the ethnographic record. Service paid particular attention to the social structure of societies whereas Fried emphasised the role of political factors in social evolution. Table 2.1 presents an overview of the various stages

2. Social complexity and archaeology: A contextual approach Service

Fried

organisation

society

Social complexity

STATE

STATE

high

STRATIFIED

increasing

CHIEFDOM RANKED TRIBE EGALITARIAN

low

BAND

Table 2.1 The social evolutionary frameworks of Service (1962) and Fried (1967). Stage STATE (Service and Fried)

Social structure - specialised,

proposed while Table 2.2 lists the main characteristics of these stages (see R. Chapman 2003, 34–38 for an overview). Although the models of Service and Fried have been severely criticised, the succession of band-tribe-chiefdomstate (Service) or egalitarian, ranked, stratified, and statesocieties (Fried) is still widely used in archaeology (e.g. Earle 1991; Rowlands and Kristiansen 1998). As Shanks and Tilley (1987, 144) have noted, from the 19th century onwards social-evolutionary schemes have had seven main characteristics: (1) a totalising holism, a focus on the entire history of humanity; (2) gradualism, social change as a continuous and cumulative process; (3) universality, change as a generic and natural process shaping humanity and social institutions; (4) potentiality, change as endogenous and inherent in human societies: (5) directionality, change as a unified process (not cyclical or random), leading to optimal situations; (6) determinism, change as irreversibly and inevitably leading from the simple to the complex or

Subsistence and economy -

bureaucratic

13

Social hierarchy

Demography

no equal access to

use of legitimised force to establish and

-

resources

maintain authority of leadership;

-

government

-

- class society

complex of institutions by means

urbanisation; rise in population density

of which power of society is organised on a basis superior to kinship -

STRATIFIED

no equal access to resources

SOCIETY (Fried) CHIEFDOM (Service)

-

increase in size of

-

greater productivity;

residence groups;

-

mobilisation of human

- complex and

labour

status differences grounded in

-

economic differences

-

-

increased warfare

-

central control of social, economic

-

chiefs with ascribed statuses and

-

increase in hierarchy and inequality

rise in population density

-

and religious activities;

organised society

urbanisation;

rise in population density

rules of succession;

RANKED SOCIETY

-

equal access to resources;

-

low degree of social ranking;

- increased sedentism;

(Fried)

-

division of labour by age

-

competitive feasting

- rise in population

and sex; -

TRIBE (Service)

- larger number of

-

redistribution by chiefs;

-

agriculture;

-

kinship segments,

density

limited specialisation;

hunter-gatherers &

-

egalitarian;

- increased sedentism;

pastoralists;

-

situational leadership

-

each composed of

-

self-sufficient;

families;

-

agriculture

rise in population density

- non-residential groups (clans, etc.) - rituals - based on kinship

-

hunter-gatherers

-

egalitarian;

EGALITARIAN (Service

(nuclear family);

-

self-sufficient;

-

situational leadership

and Fried)

- absence of political,

-

BAND &

legal or religious groups

labour division by age

-

low population density;

-

mobile

and sex; -

communal access to sources

Table 2.2 Main characteristics of the social evolutionary frameworks of Service (1962) and Fried (1967).

14

Marc Verhoeven

from the homogeneous to the heterogeneous; and (7) causal reductionism, the notion that everywhere and at every time change was subject to the same causal laws. These seven characteristics can all be turned into criticisms on the basis of the fact that they are overgeneralising, ethnocentric (modern Western), essentialist and acontextual. In fact, there are many examples of societies that do not fit evolutionist typologies (see e.g. R. Chapman 2003, 41–44). Moreover, social evolutionary models hardly leave any room for contingency. Such models, then, are implicitly teleological, based on the idea that humanity is moving towards a predefined and predetermined goal. Moreover, diffusion and cultural contact between societies are largely ignored. How does such contact effect change? Finally, there is no real place for agency, with humans being regarded as subjects in an evolutionary process over which they have no control (Johnson 1999, 142; see also Smith 1973). Social evolution, then, does not necessarily lead towards more social complexity, however the latter is defined. Furthermore, complex societies are not always marked by social and spatial hierarchies or by social, political and economic centralisation. Multiple hierarchies can exist as well as hierarchies that have a limited power only. In the lowland Maya society, for instance, the political system was highly stratified and centrally controlled, while the economic organisation was horizontally structured (Potter and King 1995). The latter is an example of what has been termed a ‘heterarchy’. This term denotes social situations where everybody shares the same horizontal position of power and authority. More specifically, Crumley (1979, 144) has defined heterarchy as “... the relation of elements to one another when they are unranked ... when they possess the potential for being ranked in a number of different ways.” A heterarchy may be parallel to a hierarchy, subsumed by it, or contain hierarchies, but the two kinds of structure are not mutually exclusive. In fact, each level in a hierachical system is composed of a potentially heterarchical unit (Crumley 1979; 1995). The concept of heterarchy makes us aware that within the same society different types of social, economic, political, or any other complexity may exist. Complexity, then, is a multi-dimensional issue that needs to be contextualised and problematised in order to make sense of social situations and social change. In the remaining part of this paper I shall give two brief examples, one ethnographic and one archaeological, that indicate that social complexity is not or not only the outcome of social evolution or the prerogative of stratified or ‘complex societies’, but that different forms of complexity (the ‘deep traits’ mentioned above) can occur in any society. The focus shall be on ritual, used here to indicate performances which are distinguished in both space and time, marked by explicit material and immaterial symbolism, and often (but not always) related to the supernatural in which behaviour

is guided and restricted by tradition, rules and repetition (Firth 1951, 222; Verhoeven forthcoming).

Complex Hunter-gatherers In traditional evolutionary schemes, such as those of Service and Fried, hunter-gatherers are representatives of the simplest social systems; they are at the very base of human evolution. As indicated in Table 2.2, according to such schemes, their social structure is based on kinship; they are egalitarian and lack political, legal or religious groups; they are self-sufficient; they have a low population density; and they are mobile. However, as is well known, there are many examples of hunter-gatherer communities that are far from simple and that in fact have many characteristics of so-called complex societies. Hence they are often called ‘complex hunter-gatherers’ (Price and Brown 1985). Perhaps one of the most famous examples are the historically documented Indian tribes of the Northwest Coast of America, marked by hereditary social ranking, sedentary villages, dense populations, craft specialisation, warfare, private ownership, wealth differences and even slavery (e.g. Lightfoot 1993 cited in R. Chapman 2003, 85). A prime example from prehistory are the Jomon of Japan (c. 13000–2200 BP), hunter-gatherers who lived in large sedentary or semi-sedentary settlements, made elaborate pottery and probably cultivated nut trees. Moreover, circles of standing stones, thousands of human figurines, dental mutilation and elaborate decorative ‘art’ styles, among other things, are indicative of complex ritual and symbolic activities and beliefs (Habu 2004). These symbolic and ritual contexts are rarely mentioned in studies of hunter-gatherer complexity. I shall give an ethnographic example of complex ritual practice in what has been regarded as one of the most simple communities on earth, the (now virtually extinct) Selk’nam Indians of Tierra del Fuego in southern Patagonia.

The Selk’nam The Austrian anthropologist Martin Gusinde was the first to systematically describe and analyse the indians of Tierra del Fuego, which he did in three volumes published from 1931 to 1939 (Gusinde 1931). Like other hunting-gathering peoples of Patagonia (e.g. the Yamana, Haush and Alakaluf), the Selk’nam (formerly also known as the Ona) were marked by an extremely limited (simple) material culture. In 1928 Lothrop wrote, “Perhaps no feature of Ona life is more striking to us than their apparent unpreparedness to face the rigors of the Fuegian climate, especially as regards their house and their clothes” (Lothrop 1928, 51). Indeed, instead of dwellings they originally used plain windbreaks made of Guanaco hides. These skins also served as their

2. Social complexity and archaeology: A contextual approach

Main elements

Element composition

15

Element detail

Cosmos (1)

4 skies (4x1A)

Páhuil, Shenu, Sheit, Télil (4x1B)

Hut (2)

7 hutposts (7x2A)

Páhuil, Shenu, Shéit, Télil, Keyáishk, Wechúsh, Jóichik, (7x2B)

Lineages (3)

3 lineages (3x3A)

3 sho'on (3x3B)

Initiates (4)

young men (4A)*

kloketen (4B)*

Spirits (5)

2 main spirits (2x5A)

at least 30 Shoorts (men), Xalpen (effigy) (31x5B)

Shaman (6)

1 shaman (1x6A)

? (6B)

Body decoration (7)

painting, masks, dress (3x7A)

at least 30 forms of abstract paintings on bodies on kloketen and Shoorts, application of feathers, 2 types of masks (tolon, asl) (33x7B)

Connections main elements

1-2 / 1-3 / 2-3 / 4-5 / 4-6 / 4-7 / 5-6 / 5-7 (n=8)

Connections element composition

1A-2A-1A-2A-1A-2A-1A-2A / 1A-3A-1A-3A-1A-3A / 2A-3A-2A-3A-2A-3A / 4A-5A-4A-5A / 4A-6A / 4A-7A4A-7A-4A-7A / 5A-6A-5A-6A / 5A-7A-5A-7A-5A-7A-5A-7A-5A-7A-5A-7A (n=40)

Connections element detail

1B-2B-1B-2B-1B-2B-1B-2B / 1B-3B-1B-3B-1B-3B / 2B-3B-2B-3B-2B-3B / 4B-5B-4B-5B-4B-5B-4B-5B-4B-5B4B-5B-4B-5B-4B-5B-4B-5B-4B-5B** / 4B-6B / 4B-7B-4B-7B-4B-7B-4B-7B-4B-7B*** / 5B-6B-5B-6B-5B-6B5B-6B-5B-6B-5B-6B-5B-6B-5B-6B-5B-6B-5B** / 5B-7B-5B-7B-5B-7B-5B-7B-5B-7B-5B-7B-5B-7B-5B-7B-5B7B-5B-7B**+*** (n=83)

Table 2.3 The structure of the Selk’nam Hain ritual (based on A. Chapman 1997). * = group of initiates counted as 1; ** = based on 10 spirits; *** = based on 5 types of body decoration.

clothing, worn loosely over their otherwise naked bodies, with the fur on the outside. They had only few domestic tools, such as simple storage bags, baskets, scrapers and awls and their weapons were bows and arrows. On encountering Fuegians aboard Captain Fitzroy’s Beagle in 1831, Darwin, the godfather of evolutionism, exclaimed: “I could not have believed how wide was the difference between savage and civilized man. It is greater than between a wild and domesticated animal ...” (Darwin 1839, 228 quoted in Beer 1997, 149). Notwithstanding their indeed poor material culture, the ritual life of the Selk’nam was very complex. In fact, it is too complicated to even properly summarise here. Therefore, mainly by means of some citations from an analysis of the so-called Hain ritual (A. Chapman 1982, 1997) and a scheme (Table 2.3), I shall provide a taste of this complexity. THE HAIN RITUAL The Hain ritual was a rite de passage, a coming-of-age ceremony for young men during which the otherwise wandering family groups congregated. Both the ceremony and the sacred hut that was the focus of the rituals were called Hain. The objective of the extended ritual was to instruct the male initiates (kloketen) in the moral behaviour, oral tradition and subsistence activities of Selk’nam society. The two principal spirits of the Hain ritual were Shoort and Xalpen. Xalpen was the only (female) spirit represented by an effigy. Shoort was represented by fantastically and abstractly painted and masked male human actors (Fig. 2.1), who actively terrorised the initiates and women (except the wives and in-laws of the shamans). Among the audience

Fig. 2.1 Main participants of the Selk’nam Hain ritual: the two Shoorts of the north and left skies. Typically the men representing the spirits (which are primarily manifestations of the sun) are naked and decorated by painted abstract symbols. The masks are guanco-leather hoods stuffed with vegetal matter (after photo taken by Gusinde 1931, re-published by A. Chapman, 1997, fig. 71).

16

Marc Verhoeven

the mothers of the kloketens played prominent ceremonial roles. Shamans seem to have been the principal directors of the ritual. The Hain ceremony could last an entire year or even longer. The Hain hut symbolised both the cosmos and the social system. Its four principal posts were aligned in a circle at the cardinal points or ‘skies’ (sho’on). These posts represented the ‘centres of the skies’, the places of creation in the universe (‘wombs’). Three additional posts were set up between these main features. Each of the seven posts represented a number of territories and lineages (A. Chapman 1997, 88–89): The four skies were thought of as ‘invisible cordilleras of infinity’. The east sky, whose post was named Páhuil (a Haush word) was the most magnificent yet the most treacherous of all. Its great slippery cordillera was surrounded by a sea of boiling water. The magnificent cordillera of the west sky was the centre, or womb, of the wind (Shenu), for whom the west post of the Hain was named. The sun (Krren) was associated with the west sky, and Shenu was his brother. Shenu had assisted his brother Krren, then a powerful shaman, when they attacked the Hain of the matriarchy and its dominant leader, the female shaman Moon (Kreeh). In the beautiful cordillera of the south sky lived Owl (Sheit), for whom the south post was named. His mighty brother Snow (Hosh) also lived there with his sister Moon, once sovereign of the famed matriarchy, who had been defeated by her husband Sun. Finally, the cordillera of the north sky was the home of Sea (Kox) and his sister Rain (Chalu). Here the mystical Flamingo (Télil) was honoured as the north post of the Hain. The souls (kaspi) of humans returned at death to these wombs, the skies, with which each person had been identified during his or her lifetime. In the wombs of the vast outer space the souls were reunited with the eternal forces of the universe.

The Selk’nam cosmology was not only materialised in the Hain hut but also played an important role in the power and symbolism of Shoorts (A. Chapman 1997, 100): While all appear to be manifestations of the sun symbol, some have different functions and status. But there is a prototype Shoort who is represented by the k’tétu ..., a small white owl ... As a mythological being K’tétu had been a powerful shaman. He belonged to the west sky, the sky of the sun, was ‘very perfect’, handsome and muscular, and played the part of Shoort particularly well during the first men’s (mythological) Hain. Later he was transformed into this owl. The Shoort personage of the real Hain was decorated to depict the patterns of the owl’s feathers: white bands around the eyes of his mask, white knees and white splotches of paint on his body.

These examples suffice to show that the Hain ritual and associated Selk’nam cosmology are very complex. This is indicated in Table 2.3, which presents a general scheme of the complex structure of the Hain ceremony, especially of the connections (physical and/or non-material) between the various constituents. A conservative estimate indicates that there were up to 83 connections between the various

elements. Clearly, as already indicated by the citations, we have here a very complex cosmology, religious system and ritual practice in a politically egalitarian hunter-gatherer society that would be deemed ‘simple’ according to standard social evolutionary schemes. Our second example of ritually complex contexts in a presumed socially/politically unstratified society comes from the prehistory of the Near East, the early Neolithic site of Göbekli Tepe in Turkey.

Göbekli Tepe Göbekli Tepe (‘navel mountain’) is located near the city of Urfa in south-eastern Turkey in a rather desolated area at the top of a large limestone ridge. The site consists of several mounds, the total area of occupation measuring about 300 m in diameter. Since 1995, excavations covering an area of approximately 5000 km2 have been mainly carried out on the south-eastern mound. Göbekli Tepe dates from between c. 9100–8500 cal BC, i.e. from the Pre-Pottery Neolithic A (PPNA) to the Middle/Late Pre-Pottery Neolithic B (PPNB) periods. As is well known, Göbekli Tepe has yielded spectacular evidence of what were undoubtedly ritual practices, indicated by several stone buildings with megalithic decorated Tshaped pillars and a large stone sculpture depicting various humans and animals. Given that some of the T-shaped pillars (like those found in a ‘cult building’ at the nearby EarlyMiddle PPNB site Nevali Çori: e.g. Hauptmann 1993, 1999; Hauptmann and Schmidt 2007) were marked by bent human arms at the sides and hands and possible indications of dress at the front (all in bas-relief), it is highly likely that they depict humans, the body being represented by the vertical part and the head by the horizontal part of the ‘T’. Rich samples of wild plants and animals and the absence of houses and evidence for animal or plant domestication indicate that Göbekli Tepe was occupied by hunter-gatherers (e.g. Schmidt 2001; 2005; 2006; 2007; Hauptmann and Schmidt 2007). In standard social-evolutionary classifications, the people using this site would probably be expected to be part of a band or a tribe. ANIMALS THAT MATTER: RITUAL PRACTICE The functions and meanings of the buildings and sculptures is still puzzling and far from clear, but for the moment the excavator, Klaus Schmidt, suggests that the site was a central place for hunter-gatherer communities; a place where, first and foremost, rituals related to death were carried out (though so far, no burials have been found), possibly under the direction of shamans. The T-shaped pillars would have depicted supernatural beings or ancestors, who were protected by the dangerous animals depicted on the pillars. At the end of their use-life the buildings were deliberately filled and buried, possibly reinforcing their role in death

2. Social complexity and archaeology: A contextual approach

17

cults (Schmidt 2006; 2007). Whatever the precise meaning of the buildings and sculpture, it seems to be clear that here we have unmistakable and spectacular evidence for ritual activities (see e.g. Verhoeven 2002). In the following paragraphs, based mainly on the work of Schmidt (2006; 2007), I shall deal with the rich and complex iconography on the large T-shaped pillars (Fig. 2.2). I shall also focus on PPNA level III in which the oval stone buildings A, B, C and D have been excavated (on the basis of geophysical research, it is estimated that at least 20 additional buildings are present at the site). The largest building (C) measures approximately 30 × 25 m, while the smallest building (B) measures about 15 × 15 m. To date, a total of 38 stone pillars have been recovered from these structures. The pillars are situated within or against the stone walls and benches where they surround two larger free-standing central pillars (except in building A: see fig. 76 in Schmidt 2006). On the basis of the designs on the pillars, buildings A and B are known as the Snake-pillar Building and the Foxpillar Building respectively (a smaller rectangular building of the later PPNB level IIA has been termed the Lion-pillar Building). The pillars in buildings A–D have been listed in Tables 2.4 and 2.5); on the basis of published information (no detailed analyses are available yet), I attempt to present the decorations of each pillar in order to assess the degrees of iconographic and ritual complexity. As indicated in Tables 2.4–2.5, the maximum number of different symbols on a pillar is seven, and symbols are present in 55 cases. At least 48 different symbols are present on the pillars in 20 buildings. Building B was particularly rich in connections between different elements. As the majority of the pillars are partly hidden in walls, and as not all of them have been excavated down to their bases, many more symbols and connections are probably present. In fact, many other objects, features and contexts have been found with undoubtedly highly symbolic and, most probably, ritual functions and meanings, mainly large animal, human and human-animal statues. The ongoing excavations will undoubtedly reveal more such features. While it is as yet not possible to reliably estimate the number of connections between the different elements of the ritual ‘system’, the presence of at least 20 buildings and at least 38 (probably anthropomorphic) pillars with at least 48 different (animal and abstract) symbols in varying combinations, surely indicates that the buildings at Göbekli Tepe were quite complex symbolic structures. In fact, we can say that they were part of a complex symbolic, ritual and religious ‘universe’.

Conclusions It is undeniable that on a global scale there has been a general evolution from small-scale, politically largely

Fig 2.2 View of the front part of one of the large T-shaped stone pillars in a ritual building at Early Neolithic Göbekli Tepe (pillar 33 in Building D): symbolic connections between different animals: an insect and snakes (after Schmidt 2006, fig. 91).

Marc Verhoeven

18

Symbol 1

Building A

Symbol 2

Symbol 3

Symbol 4

Symbol 5

Symbol 6

Symbol 7

Total

1

snakes

wild sheep

2

2

bull

fox

3

?

?

4

?

?

5

snake

1

17

'head'

1

crane bird?

bucranium

snake

5

Building B 6

reptile,

snake

2

leopard? 7

?

?

8

?

?

9

fox

?

10

fox

?

14

fox?

?

15

?

?

16

?

?

34

?

?

11

bear, lion,

1

12

wild boar

Building C

leopard? fox

birds

abstract net

4

motif 13

?

23

wild boar

? 1

24

?

?

25

?

?

26

wild boar

1

27

?

?

28

wild boar

29

?

half moon

'baulk'

?

1

35

?

?

Table 2.4 The iconography of the pillars (indicated by numbers) in Buildings A–C of Level III (PPNA) at Göbekli Tepe. ? = no symbols recovered so far. Based on Schmidt 2006; 2007.

undifferentiated and principally egalitarian societies to our current (western) highly complex way of living. As heuristic devices, therefore, terms like band, tribe, chiefdom, and state are useful for giving an idea about the general nature of societies. However, such social-evolutionary schemes are over-generalising, ethnocentric (modern Western), essentialist and acontextual. Moreover, depending on contexts, complexity has many dimensions. As we have seen, societies that are materially and politically perhaps not complex (i.e. simple), such as the sub-recent Selk’nam in Patagonia or the hunter-gatherers at early Neolithic Göbekli Tepe in south-eastern Turkey, can nevertheless be very complex in other respects. In fact, many other

examples can be presented in the Near East, such as the complex hunter-gatherers of the Epi-Palaeolithic Natufian culture and early farmers and pastoralists at Late Neolithic Çatalhöyük and Tell Sabi Abyad I (see e.g. Verhoeven 2000; 2004; Hodder 2006). In assessing degrees of complexity, then, we should look at ‘deep’, structural traits (e.g. by counting the connections between different elements) instead of using the preestablished surface traits such as economic and political characteristics of types of societies. The alternative to using complexity to refer to a sequence of societal types is to acknowledge that according to context (material, technological, symbolical, ritual, etc.) there can be different

2. Social complexity and archaeology: A contextual approach Symbol 1

Building C

Symbol 2

Symbol 3

Symbol 4

Symbol 5

Symbol 6

19

Symbol 7

Total

36

?

?

37

?

?

39

?

?

Building D 18

human arms

fox

H-symbol

circle

half-moon

5

19

snake

20

snake

bull

fox

1 3

21

gazelle

onager?

lion,

3

22

fox

snake

hare?

3

30

onager?

snakes

H-symbol

3

bucranium

leopard?

31

human arms

32

?

2

33

fox

snakes

cranes?

birds

spider

insect

H-symbol

38

bull

fox

wild boar

cranes?

Ibis?

bird

bucranium

39

?

?

40

?

?

41

?

?

42

?

?

43

crane birds?

?

H-symbol

7 7

2

Total

55

Table 2.5 The iconography of the pillars (indicated by numbers) in Buildings C–D of Level III (PPNA) at Göbekli Tepe. ? = no symbols recovered so far. Based on Schmidt 2006; 2007.

levels of complexity in any community or society. In other words, there are different forms of complexity. Communities or societies are not complex or simple; rather, they have both complex and non-complex dimensions. In such a contextual approach it is explicitly acknowledged that social complexity is not just the pre-ordained result of an evolutionary process, or the benchmark of a ‘civilised’ society. There must have been many different forms of complexity in the past, and analysis of these variable forms may generate a better understanding of human psyche, ingenuity and achievement, rather than of ‘progress’.

Acknowledgements By working in different regions and time periods, Eddie Peltenburg has made important contributions to our understanding of the complexities of ancient communities, be it of the Neolithic colonisation of Cyprus or secondary state formation in Early Bronze Age Syria. Therefore he would probably be one of the first to acknowledge the contextual nature of complexity, and it is to him that this paper is dedicated. I am grateful to Louise Maguire and Diane Bolger for the invitation to write this paper, as well as for

their editorial assistance. Mikko Kriek made the drawings and Ans Bulles corrected the English text.

References Algaze, G. 1993. The Uruk World System: The Dynamics of Expansion of Early Mesopotamian Civilization. Chicago, University of Chicago Press. Beer, G. 1997. Travelling the other way: Travel narratives and truth claims. In C. McEwan, L. A. Borrero and A. Prieto (eds) Patagonia: Natural History, Prehistory and Ethnography at the Uttermost End of the Earth, 140–152. Princeton, Princeton University Press. Blanton, R., S. Kowalewski, G. Feinman and J. Appel 1993. Ancient Mesoamerica: A Comparison of Change in Three Regions. Cambridge, Cambridge University Press. Chapman, A. 1982. Drama and Power in a Hunting Society: The Selk’nam of Tierra del Fuego. Cambridge, Cambridge University Press. Chapman, A. 1997. The great ceremonies of the Selk’nam and the Yámana: A comparative analysis. In C. McEwan, L. A. Borrero and A. Prieto (eds) Patagonia: Natural History, Prehistory and Ethnography at the Uttermost End of the Earth, 82–109. Princeton, Princeton University Press.

20

Marc Verhoeven

Chapman, R. 2003. Archaeologies of Complexity. London and New York, Routledge. Childe, V. G. 1951. Social Evolution. London, Watts. Crumley, C. L. 1979. Three locational models: An epistomological assessment for anthropology and archaeology. In M. B. Schiffer (ed.) Advances in Archaeological Method and Theory 2, 141–173. New York, Academic Press. Crumley, C. L. 1995. Heterarchy and the analysis of complex societies. In R. M. Ehrenreich, C. L. Crumley and J. E. Levy (eds) Heterarchy and the Analysis of Complex Societies, 1–5. Archaeological Papers of the American Anthropological Association 6. Washington DC, American Anthropological Association. Darwin, C. R. 1839. Narrative of the Surveying Voyages of His Majesty’s Ships Adventure and Beagle between the Years 1826 and 1836, Describing their Examination of the Southern Shores of South America, and the Beagle’s Circumnavigation of the Globe. Journal and Remarks, 1832–1836, vol. 3. London, Henry Colburn. Earle, T. 1991. Chiefdoms: Power, Economy and Ideology. Cambridge, Cambridge University Press. Firth, R. 1951. Elements of Social Organization. London, Watts. Flannery, K. V. 1972. The cultural evolution of civilizations. Annual Review of Ecology and Systematics 3, 399–426. Fried, M. H. 1967. The Evolution of Political Society. New York, Random House. Gusinde, M. 1931. Die Feuerland-Indianer, Band I: Die Selk’nam. Mödling (Vienna), Verlag der Internationalen Zeitschrift Anthropos. Habu, J. 2004. Ancient Jomon of Japan. Cambridge, Cambridge University Press. Hauptmann, H. 1993. Ein Kultgebäude in Nevali Çori. In M. Frangipane, H. Hauptmann, M. Liverani, P. Matthiae, and M. Mellink (eds) Between the Rivers and over the Mountains, 37–69. Rome, University of Rome. Hauptmann, H. 1999. The Urfa region. In M. Özdoğan and N. Başgelen (eds) Neolithic in Turkey: The Cradle of Civilization: New Discoveries, 65–86. Istanbul: Arkeoloji ve Sanat Yayinlari. Hauptmann, H. and M. Özdoğan 2007. Die neolitische Revolution in Anatolien. In C. Lichter (ed.) Vor 12.000 Jahren in Anatolien: Die ältesten Monumente der Menschheit, 26–36. Stuttgart, Theiss. Hauptmann, H. and K. Schmidt 2007. Anatolien vor 12,000 Jahren: die Skulpturen des Frühneolithikums. In C. Lichter (ed.) Vor 12.000 Jahren in Anatolien: die ältesten Monumente der Menschheit, 67–82. Stuttgart, Theiss. Hodder, I. 2006. Çatalhöyük: The Leopard’s Tale. Revealing the Mysteries of Turkey’s Ancient ‘Town’. London, Thames & Hudson. Johnson, M. 1999. Archaeological Theory: An Introduction. Oxford, Blackwell. Lightfoot, K. G. 1993. Long-term developments in complex hunter-gatherer societies: Recent perspectives from the Pacific coast of North America. Journal of Archaeological Research 1, 167–201. Lothrop, S. K. 1928. The Indians of Tierra Del Fuego: An Account of the Ona, Alacaluf and Haush Natives of the Fuegian Archipelago. New York, Museum of the American Indian, Heye Foundation.

Maisels, C. K. 1990. The Emergence of Civilization: From Hunting and Gathering to Agriculture, Cities and the State in the Near East. London, Routledge. Maisels, C. K. 1993. The Near East: Archaeology in the ‘Cradle of Civilization’. London and New York, Routledge. McGuire, R. 1983. Breaking down cultural complexity: Inequality and heterogeneity. In M. B. Schiffer (ed.) Advances in Archaeological Method and Theory 6, 91–142. New York, Academic Press. Morgan, L. H. 1877. Ancient Society, or Researches in the Line of Human Progress from Savagery through Barbarism to Civilization. London, MacMillan. Nelson, B. A. 1995. Complexity, hierarchy and scale: A controlled comparison between Chaco Canyon, New Mexico and La Quemada, Zacatecas. American Antiquity 60, 597–618. Oates, D. and J. Oates 1976. The Rise of Civilization. Oxford, Elsevier-Phaidon. Özdoğan, M. and N. Başgelen (eds) 1999. Neolithic in Turkey: The Cradle of Civilization: New Discoveries, 65–86. Istanbul, Arkeoloji ve Sanat Yayinlari. Potter, D. R. and E. M. King 1995. A heterarchical approach to lowland Maya socioeconomics. In R. M. Ehrenreich, C. L. Crumley and J. E. Levy (eds) Heterarchy and the Analysis of Complex Societies. Archaeological Papers of the American Anthropological Association 6, 17–32. Washington DC, American Anthropological Association. Price, T. D. and J. A. Brown (eds) 1985. Prehistoric Huntergatherers: The Emergence of Cultural Complexity. New York, Academic Press. Price, T. D. and G. M. Feinman (eds) 1995. Foundations of Social Inequality. New York and London, Plenum. Randsborg, K. 1981. Complexity, archaeological data, social equalities and cultural diversity. In S. E. van der Leeuw (ed.) Archaeological Approaches to the Study of Complexity, 39–53. Amsterdam, Universiteit van Amsterdam. Redman, C. L. 1978. The Rise of Civilization: From Early Farmers to Urban Society in the Ancient Near East. San Francisco, W. H. Freeman. Rothman, M. S. 1994. Evolutionary typologies and cultural complexity. In G. Stein and M. S. Rothman (eds) Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity, 1–10. Madison, Prehistory Press. Rothman, M. S. 2004. Studying the development of complex society: Mesopotamia in the late fifth and fourth millennia BC. Journal of Archaeological Research 12(1), 75–119. Rowlands, M. and K. Kristiansen (eds) 1998. Social Transformations in Archaeology: Global and Local Perspectives. London, Routledge. Schmidt, K. 2001. Göbekli Tepe, southeastern Turkey: A preliminary report on the 1995–1999 excavations. Paléorient 26(1), 45–54. Schmidt, K. 2005. Ritual centres and the neolithisation of Upper Mesopotamia. Neolithics 2(05), 13–21. Schmidt, K. 2006. Sie bauten die ersten Tempel: Das rätselhafte Heiligtum der Steinzeitjäger. Munich, C. H. Beck. Schmidt, K. 2007. Die Steinkreise und die Reliefs des Göbekli Tepe. In C. Lichter (ed.) Vor 12.000 Jahren in Anatolien: Die ältesten Monumente der Menschheit, 83–96. Stuttgart, Theiss.

2. Social complexity and archaeology: A contextual approach Service, E. R. 1962. Primitive Social Organization: An Evolutionary Perspective. New York, Random House. Shanks, M. and C. Tilley 1987. Social Theory and Archaeology. Albuquerque, University of New Mexico Press. Smith, A. 1973. The Concept of Social Change: A Critique of the Functionalist Theory of Social Change. Boston and London, Routledge & Kegan Paul. Spencer, H. 1967. The Evolution of Society: Selections from Herbert Spencer’s Principles of Sociology (ed. R. L. Carneiro). Chicago, University of Chicago Press. Stein, G. and M. S. Rothman (eds) 1994. Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity. Madison, Prehistory Press. Stewart, P. 2001. Complexity theories, social theory, and the question of social complexity. Philosophy of the Social Sciences 31(3), 323–360. Tylor, E. B. 1871. Primitive Culture: Researches into the Development of Mythology, Philosophy, Religion, Art, and Custom. London, John Murray.

21

Van der Leeuw, S. E. 1981. Archaeological Approaches to the Study of Complexity. Amsterdam, Universiteit van Amsterdam. Verhoeven, M. 2000. Death, fire and abandonment: Ritual practice at later Neolithic Tell Sabi Abyad, Syria. Archaeological Dialogues 7(1), 46–83. Verhoeven, M. 2002. Ritual and ideology in the Pre-Pottery Neolithic B of the Levant and south-east Anatolia. Cambridge Archaeological Journal 12(2), 233–258. Verhoeven, M. 2004. Beyond boundaries: Nature, culture and a holistic approach to domestication in the Levant. Journal of World Prehistory 18(3), 179–282. Verhoeven, M. forthcoming. The many dimensions of ritual. In T. Insoll (ed.) The Oxford Handbook of the Archaeology of Ritual and Religion. Oxford, Oxford University Press. Yoffee, N. 2005. Myths of the Archaic State: Evolution of the Earliest Cities, States, and Civilizations. Cambridge, Cambridge University Press.

3 LATE NEOLITHIC ARCHITECTURAL RENEWAL: THE EMERGENCE OF ROUND HOUSES IN THE NORTHERN LEVANT, c. 6500–6000 BC Peter M. M. G. Akkermans

Round houses are a characteristic feature of the 6th millennium BC in Syro-Mesopotamia (all dates in this paper are calibrated dates BC). Substantial evidence for this kind of architecture first came to light at Yunus in south-eastern Anatolia and, in particular, at Tell Arpachiyah in Northern Iraq in the early 1930s. Ever since then, the circular structures have usually been referred to as tholoi (singular tholos) on the basis of some broad but misleading parallels in layout with Mycenaean tombs of a much later date (Mallowan and Rose 1935, 25). Two basic types are found: buildings simply made of one circular room; and more complex, keyhole-shaped buildings consisting of a circular room enlarged by a rectangular antechamber. Generally they have a diameter between 3 and 5 m although both smaller and much larger ones occur as well. The walls were made of mud brick or pisé, either laid on stone foundations or simply set on the surface without any support whatsoever. The walls carried a mud plaster often covered by a thin white gypsum coating, giving the buildings a bright white appearance. The tholoi are usually thought to have had beehive-shaped superstructures made wholly of mud bricks, but flat or pitched roofs made of reeds and other organic materials were probably used as well. By the mid-6th millennium, circular structures with or without a rectangular antechamber were immensely popular in the hamlets and villages of the Halaf culture over much of northern Syria, south-eastern Anatolia and Iraq. Sites of this period tended to consist almost exclusively of round houses although rectangular structures still occurred in (very) modest numbers. A few such rectangular buildings made of very small cellular rooms, for example, were located amidst a mass of round houses at Çavi Tarlası in south-eastern Anatolia, at Tell Sabi Abyad in Syria and at the mounds of

Yarim Tepe II–III in Iraq. These rectangular buildings, it has been suggested, were special-purpose installations, serving as the communities’ communal granaries or storehouses (Akkermans and Schwartz 2003, 117). Because of their extreme abundance at the 6th millennium Halaf sites, the tholoi have always been considered archetypical of this culture and period. It is important to realise, however, that the distribution of these round buildings cuts across the traditional cultural boundaries created by archaeologists in every possible way: not only do the circular dwellings appear to have had a much longer history, going back as far as the mid-7th millennium BC, but they also occur roughly contemporarily in places alien to the Halaf culture, such as the Yarmukian sites of the southern Levant (e.g. Gopher 1995). The Late Neolithic tholoi, it appears, were not simply part of the ‘checklist’ of a specific archaeological construct (culture, period and region) but served in many different circumstances. Reviews of the round buildings of the 6th millennium Halaf culture are relatively numerous (e.g. Akkermans 1989; Breniquet 1996; Tsuneki and Miyake 1998), but very little has been said until now about the tholoi of the 7th millennium BC. The latter structures are astounding finds if only because of their great and unexpected antiquity. Hence the purpose of this article is to briefly consider the nature and origins of the earliest round dwellings in Late Neolithic Syria, c. 6500–6000 BC, primarily on the basis of the excavations at the site of Tell Sabi Abyad on the Balikh. The paper is in honour of Eddie Peltenburg, who has been a friend for many years as well as an outstanding researcher in the field of Near Eastern archaeology, and who happens to share an interest in prehistoric round houses, particularly on the island of Cyprus (see, for example, Peltenburg 2004).

3. Late Neolithic architectural renewal: The emergence of round houses in the northern Levant

23

Fig. 3.1 Map of Syria and the location of Tell Sabi Abyad.

Seventh Millennium Origins The long held belief that the northern Levantine tholoi originated in the Halaf epoch (c. 5900–5300 BC), and not before, derived principally from the absence of any such structures in occupations of an earlier, pre-Halaf, date (although it must be added that the absence of evidence was primarily negative in the sense that the number of excavated sites of this earlier period were minimal). However, excavations since the start of the 1990s at several sites in Syria, such as Tell Sabi Abyad, Tell Halula, Tell el-Kerkh and Chagar Bazar, have begun to explore settlement strata of the late 7th millennium BC which, significantly, contained circular architecture. Most relevant in this respect is the extensive fieldwork at Tell Sabi Abyad in the Balikh region of northern Syria (Fig. 3.1), which has revealed a great number of round dwellings in association with rectangular buildings in occupation levels predating the onset of the Halaf culture by many hundreds of years. The site has provided the earliest Late Neolithic circular architecture known in the Levant so far. The earliest round structures at Tell Sabi Abyad (Fig. 3.2) occurred in the extensive exposure termed Operation III on the north-western part of the mound where work succeeded in establishing a long sequence of Late Neolithic

Fig. 3.2 This building with the hearth opposite the entrance is one of the earliest circular structures found at Tell Sabi Abyad so far (Operation III, Level 3A, c. 6400 BC). The man in it is sitting next to the hearth. Note the relatively wide wall and the restricted interior diameter (3 m).

(or Pottery Neolithic) occupations, radiocarbon dated to between c. 7000 and 6200 BC (cf Akkermans et al. 2006). Although the stratigraphic analysis is still in progress, at least 13 main levels of settlement (often divided into

24

Peter M. M. G. Akkermans

Fig. 3.3 The wall of the round house in Operation III, Level 2C, c. 6300 BC. A typical tholos wall, showing crumbly, orange, irregularly-shaped clay slabs joined by a solid, gray mortar. Different sources of clay must have been in use.

sublevels) were identified, some of them occurring in the broad horizontal exposures high on the mound, others in the deep but relatively narrow soundings on the slope. Circular buildings occur at Tell Sabi Abyad at an astonishingly early date, i.e. in Operation III from Level 5 onwards, c. 6500/6450 BC according to a number of radiocarbon dates. These earliest tholoi were between 3 and 5 m in diameter with walls about 40–50 cm thick, made of both sizeable, greyish to reddish-brown, irregularly shaped clay slabs and what seem to have been small, handmade mud bricks joined by a distinct grey mortar (Fig. 3.3). Different sources of clay must have been exploited for building purposes, each with its own qualities. There were no particular foundations other than thick layers of mortar into which the first layer of bricks of different sizes was pressed. Entrance to these buildings was through narrow doorways about 45 cm wide, which were preferably located in the south wall. Low clay thresholds were also present. Although it cannot be proved due to matters of preservation, it is tempting to regard these doorways as low portholes as such features were the usual kind of passage in the local settlements of this period (Akkermans et al. 2006). The buildings were covered inside and outside with a red or grey, often renewed mud plaster, which in once instance carried a thick white gypsum coating. Two tholoi had a mud brick platform about 30 cm high, one covering about one fourth, the other about one third of the building’s interior. Two other tholoi contained a horseshoe-shaped hearth close to the wall opposite the doorway, each almost 1 m long and 0.5 m wide (cf Fig. 3.1). Their floors consisted of a smoothed layer of burnt clay on a foundation of large sherds. One hearth had a partly intact, beehive-shaped roof made of clay slabs, 55 cm high. In these early levels of settlement, c. 6500–6200 BC,

the round structures occurred in very small numbers (the paucity of round buildings has nothing to do with sample size but is a reality when taking into account the very large area of excavation, i.e. over 1800 m2). A total of six tholoi have been found so far, in Operation III Levels 5C to 2B, respectively. In each of these strata only a single circular building stood amidst a range of large and small, single or multi-roomed, rectangular features. Moreover, their occurrence was not continuous in the sense that there were intermediate sub-levels without round buildings. Depositional and radiocarbon evidence suggests that the tholoi were relatively short-lived and used over perhaps one generation at most, then abandoned and left to the elements (in view of the collapsed wall debris in them). Only in one case did the round buildings show evidence of remodelling or reconstruction in the same place: the Level 2B tholos was (partially) founded upon the remains of the lower, Level 2C tholos. Significantly, the rectangular, multiroomed architecture situated on either side of the tholos had also been rebuilt in roughly the same place and on the same alignment, suggesting a continuous use of architectural space over an extended span of time, perhaps by one family. In this case, it seems, there was a straightforward, mutual relationship between both the round and rectangular architecture on the spot; both types of building may have served the needs of one and the same group of people although probably for different purposes when taking into account the difference in layout. The link is less clear in the case of the other tholoi, which all stood isolated in the yards at some distance from the rectangular buildings.

Round Houses in the late Seventh Millennium, 6200–6000 BC The long-lived settlement in the north-western part of Tell Sabi Abyad (Operation III) came to an end at about 6200 BC. The desertion did not involve a total abandonment of the site, however. Two new foci of occupation were founded, partly on virgin soil, partly on slope wash at the foot of the original mound (Operations I and II). Late 7th millennium settlement also included parts of the slope of the original site to the east and south although not always permanently. Interestingly, the shift in the area of habitation seems to coincide with many innovations and changes in the local organisation of settlement, architecture, material culture, economy, etc. (cf Akkermans et al. 2006). One of those changes involved a substantial increase in the use of round architecture although always in association with rectangular buildings. While, as we have seen, only a handful of circular structures were employed in the early levels, there were dozens of such buildings in the layers of the late 7th millennium, c. 6200–6000 BC. The round houses of this period were between 2.5 and

3. Late Neolithic architectural renewal: The emergence of round houses in the northern Levant

Fig. 3.4 A relatively large tholos, about 4.6 m across (interior), with a repeatedly renewed, horseshoe-shaped hearth east of the entrance. Tell Sabi Abyad, Operation I, Level 7B, c. 6100/6050 BC.

5.5 m in exterior diameter (Fig. 3.4). Their mud-brick or pisé walls, about 25–45 cm wide, carried a mud plaster both inside and outside, occasionally coated with a thin, white gypsum layer. The floors consisted of tramped earth, sometimes with an additional clay plaster or a layer of pottery sherds pressed into the mud for additional strength. Remarkably, in a few cases the plaster on the wall and floor had been heavily burnt. This reddish, burnt plaster was a characteristic trait of many circular buildings at Tell Sabi Abyad in the early 6th millennium Halaf period, but it was apparently used in the late 7th millennium as well. I assume that it served as a solid, hard coating, keeping out moisture and vermin (Akkermans 1989, 59). The tholoi walls stood directly on the slope of the mound without any foundation, and as a consequence the bases of the walls sometimes differed by as much as half a metre in elevation. In order to create a flat surface for living and working, the tholoi interiors were planed with fresh clay or wall fragments from other (collapsed) structures. The entrances to the tholoi were between 40 and 45 cm wide and were usually provided with a low clay threshold. In several instances the threshold contained a perforated limestone or a re-used fragment from a basalt grinding slab with signs of hollowing and polishing, which probably served as a pivot stone. Occasionally, a sherd pavement lay in front of the doorway. The passages were usually to the south or to the west, occasionally to the north but never to the east. Set against their wall, some tholoi had a rounded or horseshoe shaped hearth as large as 1.8 m across with a smoothed, burnt clay floor on small limestones packed in mud (cf Fig. 3.4). The hearths had been renewed often, up to 12 times in one case, indicative of long use. In a few cases, shallow pits entirely filled with ashes were sunk into the floor next to the hearth. One building had a small, round

25

Fig. 3.5 Spatial segmentation is a characteristic of the tholoi from the late 7th millennium onwards. This round building, 4 m in interior diameter, has been divided in roughly two halves. The division wall was supported by a small buttress. A small circular, pedestal-shaped hearth stood in the eastern half. Tell Sabi Abyad, Operation I, Level 7A, c. 6050 BC.

pedestal shaped fireplace in its centre, slightly raised above the floor. It was covered with a layer of white plaster with a shallow, red-burnt hollow in the centre. Fireplaces also stood outside the tholoi, right next to their entrance or set against their wall. Sometimes half of a large, broken pottery jar laid on its side was used as a fireplace. Installations such as platforms, seats or benches were virtually absent. In one tholos there was a low, whiteplastered bench or dais about 30 cm high and 40 cm wide set against the wall between the hearth and the doorway. A large jar neck had been sunk into it upside down, perhaps for use as a stove (ashes and charcoal were in it; cf Nieuwenhuyse 2007, 59). Compartmentalisation was a main characteristic of many (albeit not all) round houses in the late 7th millennium. The interior space was divided by one or more walls into smaller compartments for living and working (Fig. 3.5), some of which were provided with doorways while others were not (access to them must have been high up in the wall). The partition walls may have extended up to the roof level, creating small rooms and alcoves; alternatively, they may have been in the form of low barriers demarcating grain bins or other special-purpose installations. Very often the dividers were little more than two short walls perpendicular to the tholos wall, delimiting a roughly triangular compartment of such restricted size that it cannot have served any purpose other than storage. Also new was the occasional addition of a rectangular antechamber to the circular room. Such keyhole shaped buildings were a common feature at many sites of the 6th millennium Halaf culture, but their occurrence in 7th millennium deposits is most unusual. In the case of Tell Sabi Abyad, the few examples known so far seem to have been initially free-standing circular structures, which

26

Peter M. M. G. Akkermans

Fig. 3.6 A unique find at Tell Sabi Abyad: the small, white-plastered tholos about 2.3 m across (interior), flanked on either side by a rectangular room. Operation I, Level 5B, c. 5960 BC.

were extended at a later time by means of a rectangular room or incorporated into a much larger, multi-roomed rectilinear building. One small tholos in the south-eastern area (Operation I, Level 5) had a circular, white-plastered room about 3 m in diameter flanked on either side by a rectangular room, each measuring roughly 1 by 1.5 m (Fig. 3.6). The lack of bondage between the walls suggested that the side rooms were added to the tholos at a somewhat later time. Access to the building was from the south through the main circular room. One rectangular side room was almost completely occupied by a large horseshoe shaped oven with a domed cover. The tholoi are usually thought to have had a vaulted roof made entirely of mud bricks, primarily on the basis of assumed parallels with the ‘beehive villages’ which until recently dotted the countryside of northern Syria and southeastern Anatolia (Fig. 3.7; see also Copeland 1955; Aurenche 1981). However, it is doubtful whether this perspective is correct. Not only is there no historical relationship whatsoever between the two forms of architecture, but the archaeological evidence itself suggests either a flat or a pitched roof made of timber and reeds rather than a beehive shaped mud brick cover. Sunk into the floor in the centre of several tholoi at Tell Sabi Abyad was a concentration of limestone boulders, which probably served as a foundation for a wooden post upholding the roof (cf Akkermans 1989, 32; 1993, 63). Other round buildings at the site contained clay fragments with impressions of reeds and wooden poles in their room fill, clearly showing that their roofs were made of wooden rafters covered by reed mats, which in turn were probably covered by a thick mud layer (a kind of construction also used in the rectangular structures at the site). Both stratigraphic and radiocarbon evidence at Tell Sabi Abyad suggest that the late 7th millennium tholoi had a restricted lifespan. Within a single building phase lasting for

Fig. 3.7 Postcard from the late 1930s showing a beehive village in the countryside northeast of Aleppo in Syria. Reconstructions of the prehistoric round houses and the settlements in which they stood are commonly based on analogies with the modern, local villages shown on pictures like this. However, caution in the use of such analogies is required because Neolithic communities were highly varied and their building strategies were the result of specific historical and environmental circumstances.

about 20–30 years (Akkermans and Nieuwenhuyse in press) the buildings were often used for a short time then simply left to decay. Quite a number of them appear to have been supplanted two or three times by new ones founded upon the lower, levelled building remains. Their construction required little investment in terms of time and energy; the work could probably be completed by five or six persons within a week or so (cf Akkermans 1993, 302).

Late Neolithic Round Houses in Perspective The earliest architecture of the Epipalaeolithic to Neolithic Near East consisted of round or oval, sometimes semisubterranean dwellings 3–6 m in diameter. By 9000 BC the round houses were slowly replaced by rectangular structures in many shapes and sizes, which remained predominant for several thousand years. The recurrence of circular monocellular buildings around the middle of the 7th millennium has no immediate cultural or technical roots; it implies both an innovation in architectural design and a fundamental departure from the long lived tradition of rectilinear construction. In the first 200 or 300 years of their existence, the round houses were little more than a sporadic supplement to the existing architectural repertoire in the villages of the Late Neolithic, increasing swiftly in importance only after c. 6200 BC. Apparently it took many generations for these buildings to gain widespread use and acceptance, perhaps because there was initially little need for them or because a social resistance of some sort accompanied their introduction.

3. Late Neolithic architectural renewal: The emergence of round houses in the northern Levant Alternatively, we may consider the scarcity of round buildings in the early levels as intentional and meaningful, related to the use of these structures in the communities. The tholoi were unique with respect to their layout and conspicuous appearance, as well as their highly restricted occurrence. Hence, they may have served specific purposes very different from the usual rectangular architecture which stood abundantly around them. The persistent presence of a single round building in each of the upper occupational phases at Tell Sabi Abyad is indicative of the limited yet steady demand for these special purpose installations and their success over a prolonged period of time. However, it is still very difficult to establish what the assumed special function of the earliest tholoi might have entailed. There were no artefacts or installations in the buildings other than those mentioned above to provide clues to their use. Tholoi were carefully raised architectural features, but this was the case for most if not all of the rectangular structures as well. Their occasional white plaster and hearths are other traits shared with the many rectilinear buildings in the vicinity, suggesting regular, domestic arrangements. Any activity in the round buildings must have been on a relatively small scale, involving only a few people, given their single room and restricted size. It is tempting to attribute ritual significance to these rare and exceptional structures although there is no proof to either confirm or reject this idea. Were the circular buildings simply another form of domestic architecture of this period? The early tholoi scarcely differed from the many circular structures so characteristic of the late 7th and, especially, 6th millennium BC. There is widespread agreement that the tholoi of this period principally served domestic purposes (eating, drinking, sleeping, socialising, etc.). The abundant occurrence of the tholoi, the hearths and other installations often found in them, and the nature and distribution of artefacts in and around them are all mentioned as indicators of their use in daily life, i.e. as ordinary dwellings for living and working, granaries, storerooms or animal pens (see, for example, Oates and Oates 1976; Breniquet 1996). Moreover, in the case of 7th millennium Tell Sabi Abyad it has been argued that the larger tholoi must have been houses because they contained the most extensive (single room) interior spaces and as such were the only local features suitable for constant habitation. The rectangular buildings at the site of this period are considered primarily to represent storehouses divided into many (very) small, cellular rooms (Verhoeven 1999). The shift from rectangular to circular buildings, which began around the middle of the 7th millennium or shortly afterwards and which steadily increased in importance in the course of time, has not yet been satisfactorily explained. A variety of perspectives have been proposed, ranging from a revival of much older building traditions that had still survived in remote hinterlands to adaptations resulting from

27

a degrading natural environment and differences in social structure and community organisation (cf Akkermans and Schwartz 2003, 103–131 and references therein). The late 7th millennium communities themselves remained as they were before, i.e. predominantly small and dispersed, in the order of 0.5–1 ha, with the number of inhabitants restricted to a few dozen. The architectural change, however, was slow but radical in the sense that it entailed the virtually complete replacement of the large, multicellular, rectilinear edifices by small, monocellular, round structures. Although the once dominant rectangular houses did not disappear altogether, they were given a new meaning as buildings primarily intended for storage rather than domestic living and working. In light of the above, an entirely new architectural rationale – either utilitarian, societal and/or ideological – came into existence in many parts of the Levant in the late 7th millennium, which focussed on small group size and corporate efforts: the round houses were meant to accommodate a single person or a small household of five or six individuals at the most while the storage buildings were for the benefit of the entire group, whether that community was a single household or several such units (Akkermans and Schwartz 2003, 151). The new architecture often seems to have had a transient character. Although the rectilinear storage buildings may have remained in use for 20 or 30 years with little or no noticeable modification, the circular structures were used briefly then replaced, often in the same place and alignment. Only the larger tholoi experienced a more sustained occupation, given the repeated renewal of their central hearths. It is tempting to assume that many of the round buildings were used seasonally and subsequently left to their fate, especially because their construction required little investment in terms of time and labour. In this respect, it seems that the circular architecture of the late 7th and 6th millennia cannot be detached from another trait typical of this period, i.e. the focus upon mobility and the temporary living in small and autonomous groups dispersed over the landscape. Tholoi, it seems, were easily established and easily deserted and so were the settlements in which they stood so prominently.

References Akkermans, P. M. M. G. 1989. Tell Sabi Abyad: Stratigraphy and architecture. In P. M. M. G. Akkermans (ed.) Excavations at Tell Sabi Abyad: Prehistoric Investigations in the Balikh Valley, Northern Syria, 17–75. British Archaeological Reports International Series 468. Oxford, British Archaeological Reports. Akkermans, P. M. M. G. 1993. Villages in the Steppe: Late Neolithic Settlement and Subsistence in the Balikh Valley, Northern Syria. Ann Arbor, International Monographs in Prehistory. Akkermans, P. M. M. G. and O. P. Nieuwenhuyse (eds) in press.

28

Peter M. M. G. Akkermans

Excavations at Late Neolithic Tell Sabi Abyad: Interim Report on the Archaeological Research (1994–1999) in the Balikh Valley, Syria. Turnhout, Brepols. Akkermans, P. M. M. G. and G. M. Schwartz 2003. The Archaeology of Syria: From Complex Hunter-Gatherers to Early Urban Society, ca. 16,000–300 BC. Cambridge, Cambridge University Press. Akkermans, P. M. M. G., R. Cappers, C. Cavallo, O. Nieuwenhuyse, B. Nilhamn, and I. Otte 2006. Investigating the early Pottery Neolithic of northern Syria: New evidence from Tell Sabi Abyad. American Journal of Archaeology 110, 123–156. Aurenche, O. 1981. La maison orientale: L’architecture du Proche Orient ancien des origines au milieu du quatriéme millenaire. Paris, Geuthner. Breniquet, C. 1996. La disparition de la culture de Halaf: Les origines de la culture d’Obeid dans le nord de la Mésopotamie. Paris, Éditions Recherche sur les Civilisations. Copeland, P.W. 1955. Beehive villages of north Syria. Antiquity 29, 21–24. Gopher, A. 1995. Early pottery-bearing groups in Israel: The Pottery Neolithic period. In T. E. Levy (ed.) The Archaeology

of Society in the Holy Land, 205–225. London, Leicester University Press. Mallowan, M. E. L. and J. C. Rose 1935. Excavations at Tell Arpachiyah, 1933. Iraq 2, 1–178. Nieuwenhuyse, O. P. 2007. Plain and Painted Pottery: The Rise of Neolithic Ceramic Styles on the Syrian and Northern Mesopotamian Plains. Turnhout, Brepols. Oates, D. and J. Oates 1976. The Rise of Civilization. Oxford, Elsevier-Phaidon. Peltenburg, E. 2004. Social space in early sedentary communities of southwest Asia and Cyprus. In E. Peltenburg and A. Wasse (eds) Neolithic Revolution: New Perspectives on Southwest Asia in Light of Recent Discoveries on Cyprus, 71–89. Oxford, Oxbow. Tsuneki, A. and Y. Miyake (eds) 1998. Excavations at Tell Umm Qseir in Middle Khabur Valley, North Syria: Report of the 1996 Season. Tsukuba, University of Tsukuba. Verhoeven, M. 1999. An Archaeological Ethnography of a Neolithic Community: Space, Place and Social Relations in the Burnt Village at Tell Sabi Abyad, Syria. Istanbul, Nederlands Historisch-Archaeologisch Instituut.

4 ABANDONMENT PROCESSES AND CLOSURE CEREMONIES IN PREHISTORIC CYPRUS: IN SEARCH OF RITUAL Demetra Papaconstantinou

While it remains correct to state that the archaeology of neolithic Cyprus is mainly the archaeology of domestic units, the prolific record of that period allows us, through the evaluation of conditions of abandonment, to investigate elusive sociocultural levels of explanation. To accomplish that, we need to treat house evidence as an interactive part of the totality of human experience. (Peltenburg 2003b, 118)

Perhaps one of the most fascinating moments in research is when you realise that the questions you have been investigating for so long and for which you thought you had found convincing answers, viewed from a different perspective, lead to a completely different field of inquiry and an entirely new range of interpretations. This is undoubtedly the most fascinating but at the same time the most frustrating and disturbing moment of research, one that makes you wonder about your conceptual constraints at any given time and at the same time emphasises in the best possible way the significance of interacting and experimenting with new ideas. The archaeology of prehistoric Cyprus has many such moments to offer. Owing to the wealth of its material and presenting a really intensive and well practiced archaeological activity by several committed researchers, archaeological evidence from Cypriot prehistory offers a remarkable record for the application of different perspectives, one which often competes with its revelations and with the findings of new excavations. Recent investigations on the Aceramic phase of the island with reference to its colonisation, based on the recovery of new material (Guilaine and Briois 2001; Peltenburg et al. 2003) and the re-examination of material from older excavations (McCartney 2006), present one of the most remarkable examples of this phenomenon. And this is just one example. Depositional and abandonment processes constitute a

similar case since from the moment of their appearance they manage, quietly but persistently, to revolutionise archaeological research and change the way archaeology perceives its object of study. For decades archaeologists have reconstructed and narrated the history of the sites they excavated as if everything they were finding, apart from the obvious cases of eroded deposits, had a systemic coherence and value and was the direct reflection of the socio-political systems of the communities they examined. Today, however, with the work that has been done on depositional and abandonment processes in ethnoarchaeology, it is almost impossible to make such connections without acknowledging the fact that any kind of interpretation about the past is in one way or another filtered through these processes. Research on prehistoric Cyprus has been very quick to respond to this kind of concern, and a significant number of projects have been conducted that are sensitive to depositional and abandonment processes (Frankel and Webb 1996; Peltenburg et al. 1998; Daune-Le Brun 2008). Even more significant is the fact these attempts are progressing to the final stage of publication and are therefore being ‘materialised’ into well integrated practice. One of the most significant patterns to have emerged out of this interest seems to be the identification of divergent practices in relation to floor assemblages among the main phases in the prehistory of the island, with emphasis shifting over time between burials and houses. During the Chalcolithic period there was an abundance of usable items on abandoned floors, in contrast to the rather poor burial depositions, while in the Aceramic and Early Bronze Age periods these same artefacts were discarded away from the abandoned floors (Peltenburg 2003b, 114–115, table 2). The present paper attempts to contribute to the above by reviewing the depositional practices at a specific site and discussing possible ways of understanding and

30

Demetra Papaconstantinou

defining closure ceremonies in the prehistoric record. As a framework for that exercise, a brief comment on the role of ethnoarchaeological data in depositional processes is necessary.

Discard and Abandonment in Ethnoarchaeology: Putting Our Knowledge to Work The study of formation processes constitutes one of the most powerful influences processual thinking has had on archaeology, and since the 1960s it has generated extensive discussions between archaeologists and ethnoarchaeologists concerning the nature of the archaeological evidence and the relationship between human behaviour and material culture (David and Kramer 2001; Cameron 2006). Apart from the importance of taphonomy and the specialised studies that concentrate on the formation of sediments, ethnoarchaeology was recognised as a new field that would be able to help archaeologists address and find solutions to these questions. As such, it produced a new and useful vocabulary for the identification and description of the different behavioural patterns involved in formation process: de facto refuse, primary deposition, secondary deposition/refuse, curation behaviour, curate and discard strategies, planned/unplanned abandonment, artifact scavenging, collecting and recycling, and ritual depletion or ritual/assemblage enrichment. This vocabulary was created primarily to describe two of the most important behavioural patterns of deposition: discard and abandonment, patterns that according to ethnoarchaeological experience are significant not only because they reflect the life history of house floors, but because they refer to daily practices that frequently have a strong symbolic content and are “guided by the social and symbolic fabrics of society” (Cameron 2006, 27). In support of that perspective, Richard Bradley has gone so far as to state that “the archaeological record only retains the amount of order that it does because of the very conventions by which it was formed” (2005, 208–209), and therefore the formality that certain archaeological patterns often contain constitutes the reflection of ritualised actions. Ethnoarchaeology has indeed a long list of cautionary tales to present in relation to ritualised activities and abandonment processes, activities which are usually related to the death of an occupant of a house. LaMotta and Schiffer have presented an extensive list of these kinds of cases (1999, 23–25): •

•

cases where houses were burned upon abandonment, usually as a result of the death of one or more of the occupants, or were systematically depleted before being burned, with foreign objects occasionally deposited on house floors as part of the abandonment ritual. other cases where only a portion of the deceased’s

•

•

•

material possessions were burned, comprising personal possessions or broken and no longer usable objects. examples where, in order to avoid the mass destruction of household objects, a dying person was moved to an empty, makeshift structure, which was later burned. cases in which groups who were practicing differential disposal of the deceased’s possessions left some objects destroyed or abandoned on house floors while other objects were interred with the individual, dumped in specialised refuse areas or redistributed. cases where objects were deposited upon abandonment as ‘offerings’ for an individual buried beneath the floor.

Indicative of the interpretative limits that these models present for archaeological research is the phrase that follows LaMotta and Schiffer’s final assessment, that “much more comparative research needs to be conducted on specific processes of ritual deposition before we will be able to fully recognise the end-products of these behaviours in the archaeological record” (1999, 24). Perhaps, in order to gain maximum input, one ought to be looking at another set of cautionary tales that ethnoarchaeology has to offer, and one that has more practical implications. I refer to insights that have to do with the very nature of depositional processes and which are usually overlooked (LaMotta and Schiffer 1999, 24–25; see also Allison 1999, 11–12). These include the fact that: •

•

• •

•

•

primary deposition of objects at their locations of use is a fairly rare phenomenon and it is more likely that we are usually dealing with abandonment or postabandonment processes artefact assemblages contained within stratigraphically distinct proveniences are not qualitatively comparable in all cases, and that house assemblage formation processes are not created equally; a number of causes may generate similar abandonment processes; several ritual abandonment processes might actually resemble other forms of cultural deposition, especially provisional and de facto refuse deposition; instances of ritual accretion and depletion may have altered the content of an assemblage dramatically; and, finally both floor and fill deposits may be created by the same or similar depositional processes, and that these deposits may be stratigraphically indistinguishable.

These statements, if taken seriously at an interpretative level, would certainly be enough to rewrite the history of many of the archaeological sites that have been excavated so far. Currently, however, when they are addressed, they usually result in rigorous taphonomic analysis and, as Bradley observes, they are used in order to mark the limits

4. Abandonment processes and closure ceremonies in prehistoric Cyprus: In search of ritual of what archaeological interpretation can achieve (2005, 208). Furthermore, from an archaeological point of view, there seems to be great concern with the fact that house floor assemblages in archaeology present the aggregate of different actions of many individuals, i.e. a ‘palimpsest of activities’ and as such cannot be perceived with the conventional manner used in anthropology or history as particular occurrences (Allison 1999; Bradley 2005; Lucas 2008). In their attempt to put ethnoarchaeological research to work for archaeology, scholars often seek alternative modes of interaction. Schlanger, for example, has suggested that we probably need to “discuss kinds of prehistoric abandonment behaviours, instead of the effects of some abandonment practices on the assemblages we study” (1991, 460–473) while Lucas, more recently, has questioned the conventional manner in which we think of events in archaeology as particular occurrences and has suggested that “rather than thinking about how objects can be interpreted in terms of the event, we ought to be thinking about how an event could be interpreted in terms of objects” (2008, 62). Within that context, the following discussion will attempt to re-examine material from a Late Neolithic site in Cyprus, placing emphasis on the event of the abandonment of its structures and exploring the possibilities that this type of analysis can offer for future research.

Abandonment Processes and Material Engagement in Archaeology: The Case of Ayios Epiktitos-Vrysi Ayios Epiktitos-Vrysi (henceforth Vrysi) is a Late Neolithic settlement dating to the 5th millennium BC and located on a coastal promontory extending from the foothills of the Kyrenia Mountains on the north coast of the island (Peltenburg 1982). It is part of the Ceramic Neolithic tradition which appeared after a gap of 500 to 1000 years from its predecessor, the Aceramic Neolithic, and is characterised by a number of settlements (Vrysi, Sotira-Teppes, KlepiniTroulli, Kantou-Koufovounos, Paralimni-Nissia) with quite distinctive pottery styles but similar intra-site arrangements of clusters of freestanding houses usually identified as general purpose habitation units (Peltenburg and Spanou 1999; Bolger 2003, 26–29; Steel 2004, 63–74). Vrysi is somewhat unique because the area in which it is located is divided by a number of deep hollows and ridges, most probably of natural origin, and as a result habitation developed mostly within semi-subterranean hollows, forming a long succession of superimposed layers (Fig. 4.1). The part of the settlement revealed by the excavation is divided by a central ridge into two sectors, north and south, with habitation starting initially from the north and extending to the south. Occupation in the area consists of single-roomed

31

structures with irregular plans, a rich inventory of domestic installations (hearths, fireplaces, bins and benches), and on-floor tool assemblages. Owing to its exceptionally deep stratigraphy and the variety of material culture recovered, the site has been the subject of many studies for the spatial arrangements and the depositional behavioural patterns it presents, exploring patterns of intergenerational competition (Peltenburg 1982; 1993) and revealing the dynamic nature and the continuous alterations in the life cycles of buildings (Papaconstantinou 2002). In one of the most recent re-examinations (Peltenburg 2003b), the evidence from Vrysi has been compared to ethnographic cases of closure ceremonies upon the death of the occupant of a house based on repeated inventories of on-floor usable materials, which in many cases seem to diverge from the economising functionalistic principles of planned abandonment processually characterised in ethnoarchaeology by depleted floors (Cameron and Tomka 2003). The large numbers of intact objects, which were covered with thick, artifact-free, sterile deposits, are considered to indicate a deliberate act rather than the regular wall collapse of a deserted building that had become a dumping ground (Peltenburg 2003b, 108–113). As such they resemble closure ceremonies that “may have been following the death of an important occupant who was buried outside her/his building but whose objects were largely retained in what then became a memorial” (Peltenburg 2003b, 117), emphasising major concerns with renewal and continuity. By placing emphasis on the significance of the exploration of events in archaeology, as suggested by Lucas (2008), and by recognising in the abandonment of structures the manifestation of one of the most clear events revealed by the archaeological record, the following discussion will concentrate on the comparison of the layers between floors. This will be carried out according to the stratigraphic description of the site in the publication, and will explore the different types of depositional behaviour in each structure in an attempt to understand the conventions and formalities that may have characterised their life cycles. A brief description of the abandonment processes that seem to have taken place in the life history of the houses at Vrysi raises significant issues for the understanding of such events in the archaeological record (Tables 4.1 and 4.2). If we compare the vocabulary used in ethnoarchaeology with that formulated by archaeological observation in the field, there are clear distinctions concerning not only the degree of intentionality but also the variety of processes that might intervene between the use and re-use of buildings. Hence the archaeological list of events connected to the abandonment of a structure involves a variety of processes that are not evident in the ethnoarchaeological record. These include the immediate re-use and re-surfacing of a floor (House 1, floor 4b; House 2A, floor 3; House 3, floor 5); the various ways a superstructure can collapse (roof

32

Demetra Papaconstantinou

Fig. 4.1 Plan and aerial photograph of Ayios Epiktitos-Vrysi (after Peltenburg 1982).

and wall collapse: House 1, floor 5); removal of roof and demolition of the upper walls (House 1, floor 4a, floor 1); partial collapse (House 6, floor 2); and a number of different post-abandonment activities, such as floors which were left open as a dumping ground (House 7, floor 1), areas of continuous, ephemeral occupation (House 1, floor 5; House 6, floor 2; House 2b, floors 4a–b; House 3, floor 4), and

areas cut by later features and eroded (House 5, floor 1; House 4a, floor 1b; House 2b, floor 1). Furthermore, it is clear that the life histories of specific buildings are characterised by several types of ‘planned actions’ which are quite different from each other. House 1, for example, presents a history of readily identifiable floors with short periods of immediate re-use and re-surfacing and

4. Abandonment processes and closure ceremonies in prehistoric Cyprus: In search of ritual House/ floor House 1 Floor 5 Floor 4b Floor 4a Floor 3 Floor 2 Floor 1

House 5 Floor 2 Floor 1

House 6 Floor 3 Floor 2 Floor 1 House 7 Floor 2 Floor 1

House 12 Floor 2 Floor 1

Succession of abandonment episodes

Roof and wall collapse Fill-traces of temporary occupation Re-surfacing Sterile deposit Re-surfacing Sterile deposit Roof collapse – Sterile deposit – collapse of superstructure

Re-surfacing Collapse of superstructure Disturbance – pit cut in rabble

Rubbish deposit primarily of debris Collapsed partly Series of ephemeral floors

Re-surfacing - Continuous occupation Collapse of superstructure Depression above the collapse - overlying dump

House/floor House 2A Floor 4b Floor 4a Floor 3 Floor 2 Floor 1 House 2b Floor 5 Floor 4a-b Floor 3 Floor 2 Floor 1

House 3 Floor 5 Floor 4 Floor 3 Floor 2 Floor 1

House 4a Floor 3 Floor 2 Re-surfacing - Continuous occupation Disturbance – erosion Re-surfacing - Continuous occupation Disturbance - erosion

Table 4.1 Abandonment processes, Ayios Epiktitos-Vrysi, North Sector (interpretation based on Peltenburg 1982, 21–53). Shaded areas indicate the lack of distinguishable floors and the existence of a continuous build up of occupation. Bold in ‘Succession of abandonment episodes’ column indicates the characteristics of the layer immediately above the floor.

Floors 1b Floors 1a House 4b Floor 2 Floor 1 House 9 Floor 2 Floor 1

longer episodes of abandonment in which the floor was covered with sterile, artifact-free deposits. Other buildings, on the contrary, display different processes, with evidence of continuous, indistinguishable layers of occupation and the absence of sterile fills (Houses 6 and 7 in the north sector; Houses 2A and 3 in the south sector). If we add to those patterns the evidence from the type of discard processes observed in relation to floors, the list of intentional acts is further enriched (Tables 4.3 and 4.4). Evidence from usable objects left on floors, in many cases forming clusters, is related to a variety of abandonment processes, such as floors which were covered with sterile deposits (House 1, floor 4a), short resurfacing episodes (House 5, floor 2), and partial collapse (House 6, floor 2) while the best preserved case of an in situ artifact assemblage comes from a house in which there was no prior tradition of similar discard behaviour (House 2A, floor 3). This small exercise indicates that focussing on a few isolated cases of the best preserved floor assemblages in order to understand abandonment processes is quite a restrictive practice for archaeology, and that the aggregation

33

Succession of abandonment episodes

Re-surfacing Deep fill– not sterile Resurfacing Collapse of superstructure Ephemeral occupation – Deep fill Erosion

Re-surfacing Re-surfacing - Continuous occupation Deep sterile fill Collapsed superstructure? Re-surfacing Disturbed – pit cut Fill – not sterile Traces of occupation - erosion

Re-surfacing- floor make up Re-surfacing - Continuous occupation Deep fill- not sterile Re-surfacing - Continuous occupation Re-surfacing - Continuous occupation Eroded – cut by pit

Re-surfacing - Continuous occupation Partly collapsed Partly eroded Re-surfacing Disturbance – pit and ditch cut Re-surfacing - disturbance

Collapse of superstructure Occupational debris Erosion - Disturbance

Deep fill – not sterile - Roof collapse? Disturbance – pit cut Erosion

Table 4.2 Abandonment processes, Ayios Epiktitos-Vrysi, South Sector (interpretation based on Peltenburg 1982, 21–53). Shaded areas indicate the lack of distinguishable floors and the existence of a continuous build up of occupation. Bold in ‘Succession of abandonment episodes’ column indicates the characteristics of the layer immediately above the floor.

of activities in the layers between the floors has an equally significant role to play in discussions about abandonment. Refined methods of analysis of microstratigraphy could further enhance our knowledge of these processes, but even at the level of ‘crude’ stratigraphic analysis just attempted, it is clear that abandonment processes vary significantly from floor to floor. In fact, one can say that apart from the one case that has been identified by the excavator as de facto deposition, based on evidence of a sudden roof collapse (House 1, floor 5: Peltenburg 2003b, 108–109), most of the behavioural abandonment processes appear to be intentional. Therefore the challenge for archaeology is not to distinguish

Demetra Papaconstantinou

34

House/floor

Curate and discard strategies

Number of finds

Succession of abandonment episodes

House 1 Floor 5

scatter

25

scatter identifiable pattern scatter

9 42 15

Roof and wall collapse Fill-traces of temporary occupation Re-surfacing Sterile deposit Re-surfacing Sterile deposit roof collapse sterile deposit-collapse of superstructure

identifiable pattern scatter

40 14

identifiable pattern

30

Floor 4b Floor 4a Floor 3 Floor 2 Floor 1

House 5 Floor 2 Floor 1

House 6 Floor 3 Floor 2 Floor 1 House 7 Floor 2 Floor 1

House 12 Floor 2 Floor 1

Re-surfacing Collapse of superstructure Disturbance – pit cut in rabble

Rubbish deposit primarily of debris Collapsed partly Series of ephemeral floors

Re-surfacing - Continuous occupation Collapse of superstructure Depression above the collapse - overlying dump

Re-surfacing - Continuous occupation Disturbance - erosion Re-surfacing - Continuous occupation Disturbance - erosion

Table 4.3 Curate and discard strategies, Ayios Epiktitos-Vrysi, North Sector (data after Peltenburg 1982, 22–52). Shaded areas indicate the lack of distinguishable floors and the existence of a continuous build up of occupation. The number of finds is not calculated for these floors. The distinction between identifiable patterns and scattered objects refers to cases where patterns in the distribution of objects could be discerned, due to their concentration against walls around domestic installations (hearths, benches and bins). The finds are mostly bone and stone tools.

between planned and unplanned abandonment behaviour, as ethnoarchaeology does, but to identify the type of planned abandonment processes and its relationship to the deposition of artifacts. Having extended our knowledge of abandonment processes in the archaeological record, we realise that the distinction between abandonment and discard in closure deposits refers in practice to two kinds of behavioural patterns: behaviour that has to do with the abandonment of a building, i.e. how buildings ended their lives; and behaviour that has to do with the role of material culture in that process of abandonment, i.e. patterns that refer to the low or high contextual integrity of house-floor assemblages. Viewed from this perspective, the different types of abandonment in the life history of buildings provide archaeologists with a meaningful pattern against which floor assemblages can be examined, and on the basis of which a variety of stories can be told. This approach does not contradict our choice of highlighting specific occurrences in the archaeological record, but it sets them in a wider context and helps us realise that the occurrences reserved in special deposits constitute only a small component of the repertoire of artifacts and behaviours one can observe

(see also Renfrew 2004, 30). In this respect House 1 at Vrysi must indeed have been a significant building in the history of the settlement since it was sealed off twice (floors 4a, 2), in the first case together with two identifiable pillars of phallic shape. However, while its use, re-use and transformation from domestic alterations (re-surfacing) to more ritualised instances of abandonment (identified by the deliberate deposition of usable on-floor assemblages and sterile, object-free, fills) certainly indicate “renewing senses of identity” (Peltenburg 2003b, 117), these occur on a rather limited, domestic scale. The abandonment of a house is the clearest event of change that exists in the archaeological record. Since houses in the Neolithic period seem to indicate simultaneously material and symbolic roles (Peltenburg 1982; 2003a; Watkins 2004; Koutrafouri 2008), the termination of the life-cycle of a house does not only reflect practical needs but also the symbolic connection to its occupants and therefore to their death. If we accept that ritual is a communication system of ritualised actions by practitioners who “have acted in view of dealing emotionally with situations of the reality, of understanding and explaining themselves and their world, so that order, in the way they understand it, can be maintained” (Koutrafouri

4. Abandonment processes and closure ceremonies in prehistoric Cyprus: In search of ritual House/floor

Curate and discard strategies

Number of finds

Succession of abandonment episodes

House 2A Floor 4b Floor 4a Floor 3 Floor 2

scatter scatter identifiable pattern / in situ finds scatter

2 9 20 3

Re-surfacing Deep fill– not sterile Resurfacing Collapse of superstructure Ephemeral occupation - Deep fill Erosion

Floor 1 House 2b Floor 5 Floor 4a-b Floor 3 Floor 2 Floor 1

House 3 Floor 5 Floor 4 Floor 3 Floor 2 Floor 1

-

scatter

8

scatter

8 31

scatter

3

House 4a Floor 3 Floor 2

scatter

5

Floors 1b

scatter

3

Floor 1a

scatter

22

House 4b Floor 2

scatter

5

Floor 1

scatter

5

House 9 Floor 2 Floor 1

6

35

Re-surfacing Re-surfacing - Continuous occupation Deep sterile fill Collapsed superstructure? Re-surfacing Disturbed – pit cut Fill - not sterile fill Traces of occupation - erosion

Re-surfacing – floor make up Re-surfacing - Continuous occupation Deep fill- not sterile Re-surfacing - Continuous occupation Re-surfacing - Continuous occupation Eroded – cut by pit

Re-surfacing - Continuous occupation Partly collapsed Partly Eroded Re-surfacing Disturbance – pit and ditch cut Re-surfacing – disturbance

Collapse of superstructure Occupational debris Erosion - Disturbance

Deep fill – not sterile - Roof collapse? Disturbance – pit cut Erosion

Table 4.4 Curate and discard strategies, Ayios Epiktitos-Vrysi, South Sector (data after Peltenburg 1982, 22–52). Shaded areas indicate the lack of distinguishable floors and the existence of a continuous build up of occupation. The number of finds is not calculated for these floors. The distinction between identifiable patterns and scattered objects refers to cases where patterns in the distribution of objects could be discerned, due to their concentration against walls around domestic installations (hearths, benches and bins). The finds are mostly bone and stone tools.

2008, 95–96, 108), then the abandonment of a house as an event that created a new reality, and most probably had to be dealt with emotionally, can be seen as the ideal case for the exploration of such ritualised actions. The question about the role of material culture in abandonment processes, however, still remains, especially since patterns from Vrysi have no parallel in previous periods and because ethnoarchaeological evidence, while identifying the role of possessions in closure deposits as a common feature of the whole ritual, makes no reference to how that material engagement came about. We seem to have two questions here: 1) Why is material culture taking up a new role in the closure of some of the buildings at a

particular site? and 2) Why is it doing so at this specific time of the prehistory of the island, the Late Neolithic period? For the first question assessments are difficult. The only observations one can make are that the deposition of usable artifacts is related to specific ‘events’ in the life history of houses; and that they do not seem to follow a widespread pattern of community beliefs about the way houses end their lives. The characteristics of a ritual that seeks to formulate a sense of the interrelated nature of things and to reinforce values that assume coherent interrelations, and they do so by virtue of their symbols, activities, organization, timing and relationships to other activities (Bell 1997), do not seem to coincide with the restricted patterns observed to

36

Demetra Papaconstantinou

date at Vrysi. Hence the important archaeological question for the time being is not so much why certain houses were associated with specific depositional patterns (i.e. on-floor, usable artefacts) but why the renewal of identity and the termination of the symbolic and domestic use of that house had to be expressed through the deposition of usable items. This brings us to the second issue, i.e. the specific timing. To address this question we should probably look at an even wider archaeological context. According to Clarke, the evidence of stylistic behaviour, which characterises the decorative schemes in the pottery of the island during this period, is the manifestation of symbolic style as a form of “formal variation in material culture which is personally based and which carries information supporting individual identity” (Clarke 2001, 72, citing Wiessner 1983, 258). If this is the case, then one could suggest that this new environment of engagement with a certain type of material culture (pottery) through which individual identity is expressed triggered an entirely new relationship with movable artifacts and accelerated notions of property and possession in the construction of identity. This relationship, which was unattested previously, did not have to be manifested only through pottery, but precisely because pottery was so important for symbolic reasons, it could be applied in the depositional practices regarding other types of artifacts (for a similar discussion, see Hodder 2006). Consequently, if architecture in the Aceramic Neolithic period had been a means of external symbolic storage, as has been suggested by Watkins (2004), then what we witness in the Ceramic Neolithic phase is “a new range of artifacts itself undertaking a measure of external symbolic storage and facilitating the development of new value systems” (Renfrew 2004, 25). One could even go a step further by suggesting that during the Ceramic Neolithic period houses lost their primacy as external symbolic storage as portable artifacts began to take over, reinforcing individual identity. The abundance of artifacts in burials and the eventual placement of burials outside of the settlements, a phenomenon that is manifested in later periods, might therefore indicate two things: 1) the restrictive role that a simultaneously material and symbolic house can play in the construction of individual identity, especially when the latter needs to be manifested to the rest of the community; and 2) the fact that burial practices in cemeteries, by becoming an ‘institutional fact’ and assuming an institutional role, can more effectively serve larger sections of society and can provide a more coherent basis for rituals involving the whole of the community. If this is the case, then the deliberate deposition of usable artifacts on floors during the Late Neolithic is yet another example of a symbolic act that reinforces the role of movable artifacts at a time before cemeteries and graves goods become institutional facts (Renfrew 2001; 2004).

Abandonment Processes in Archaeology: The Value Lies in between To more fully comprehend major episodes of house demise and rebuilding, it is necessary to contextualize such activities. For example, the ethnographic record supplies instances of abandonment ritual that strongly influenced the character of assemblages. (Peltenburg 2003b, 113)

The above extract succinctly addresses the paradox that characterises archaeological research in relation to an understanding of abandonment processes. The closest context we are trained to look at for the interpretation of archaeological patterns is not the one we have in the field, but that provided to us by the ethnographic record. To a certain degree this is understandable. “We see what we have seen,” as Bradley once said (1997), and it is almost impossible to recognise patterns in the field without referring to something we already know (otherwise we would not have the words to describe them). Too much reliance on the ethnoarchaeological record, however, might simply shift emphasis from the identification of cult areas to that of closure deposits and neglect a significant body of evidence that is available in the field. In acknowledging this, we might need to act as historians or anthropologists when we get to the floor, but for the deposits between the floors we must realise that we are pretty much on our own. Exhausting this unique field of inquiry might be the only chance we have to let history surprise us. Alongside the long ethnographic index of patterns with ritual deposits, archaeology can contribute an equally significant index of the types of abandonment processes one can observe in the field. In this respect, archaeological depositional patterns have yet to be revealed.

Acknowledgements I would like to thank Professor Trevor Watkins and Dr. Vasiliki Koutrafouri as well as an anonymous reviewer for their helpful comments on earlier drafts of the paper. Part of this work was presented in the “Seminars for Interdisciplinary Communication” in Athens (April 2008) and in The Sixth World Archaeological Congress in Dublin (June 2008). I am indebted to the participants of those meetings for the feedback I gained there. Last but not least, I would like to express my gratitude to Eddie Peltenburg for being such a great source of inspiration and for the guidance and support he has so generously offered me all these years.

4. Abandonment processes and closure ceremonies in prehistoric Cyprus: In search of ritual

References Allison, P. M. 1999. Introduction. In P. M. Allison (ed.) The Archaeology of Household Activities, 1–18. New York and London, Routledge. Bell C. 1997. Ritual. Perspectives and Dimensions. Oxford, Oxford University Press. Bolger, D. 2003. Gender in Ancient Cyprus: Narratives of Social Change on a Mediterranean Island. Walnut Creek (CA), AltaMira. Bradley, R. 1997. To see is to have seen: Craft traditions in British field archaeology. In B. L. Molyneaux (ed.) The Cultural Life of Images: Visual Representation in Archaeology, 62–72. New York and London, Routledge. Bradley, R. 2005. Ritual and Domestic Life in Prehistoric Europe. New York and London, Routledge. Cameron, C. M. 2006. Ethnoarchaeology and contextual studies. In D. Papaconstantinou (ed.) Deconstructing Context: A Critical Approach to Archaeological Practice, 22–33. Oxford, Oxbow. Cameron, C. M. and S. A. Tomka (eds) 1993. Abandonment of Settlements and Regions: Ethnoarchaeological and Archaeological Approaches. Cambridge, Cambridge University Press. Clarke, J. 2001. Style and society in Ceramic Neolithic Cyprus. Levant 33, 65–80. Daune-Le Brun, O. n.d. Les techniques de construction au Néolithique pré-céramique Recent de Chypre. Unpublished paper delivered in the 4th International Congress of Cypriot Studies, Nicosia, 29 April–3 May 2008. David, N. and C. Kramer 2001. Ethnoarchaeology in Action. Cambridge, Cambridge University Press. Frankel, D. and J. M. Webb 1996. Marki Alonia: An Early and Middle Bronze Age Town in Cyprus. Excavations 1990–1994. Studies in Mediterranean Archaeology 123(1). Jonsered, Paul Åströms Förlag. Guilaine J. and F. Briois 2001. Parekklisha Shillourokambos: An Early Neolithic site in Cyprus. In S. Swiny (ed.) The Earliest Prehistory of Cyprus: From Colonization to Exploitation, 37–53. CAARI Monograph Series 2. Boston, American Schools of Oriental Research. Hodder, I. 2006. Çatalhöyuk: The Leopard’s Tale: Revealing the Mysteries of Turkey’s Ancient ‘Town’. London, Thames & Hudson. Koutrafouri V. G. 2008. Ritual in Prehistory: Definition and Identification. Religious Insights in Early Prehistoric Cyprus. Unpublished Ph.D. thesis, University of Edinburgh. LaMotta, V. M. and M. B. Schiffer 1999. Formation processes of house floor assemblages. In P. M. Allison (ed.) The Archaeology of Household Activities, 19–29. New York and London, Routledge. Lucas, G. 2008. Time and archaeological event. Cambridge Archaeological Journal 18, 59–65. McCartney, C. 2006. The meaning of context for changing interpretations of the Cypriot Aceramic Neolithic. In D. Papaconstantinou (ed.) Deconstructing Context: A Critical Approach to Archaeological Practice, 79–97. Oxford, Oxbow. Papaconstantinou, D. 2002. Life cycles and the dynamics of domestic architecture: Households in Neolithic Cyprus. In D. Bolger and N. Sewint (eds) Engendering Aphrodite: Women and

37

Society in Ancient Cyprus, 33–52. CAARI Monograph Series 3. Boston, American Schools of Oriental Research. Peltenburg E. J. 1982. Vrysi: A Subterranean Settlement in Cyprus. Excavations at Prehistoric Ayios Epiktitos Vrysi 1969–1973. Warminster, Aris & Phillips. Peltenburg, E. J. 1993. Settlement discontinuity and resistance to complexity in Cyprus, c. 4500–2500 B.C.E. Bulletin of the American Schools of Oriental Research 292, 9–23. Peltenburg E. 2003a. Identifying settlement of the Xth–IXth millennium B.P. in Cyprus from the contents of KissonergaMylouthkia wells. In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre: Actes du Colloque International Organisé par le Départment des Antiquités de Chypre et l’École Française d’Athènes, Nicosie 17–19 Mai 2001, 15–33. Bulletin de Correspondance Hellénique Supplément 43. École Française d’ Athènes. Peltenburg E. 2003b. Incorporated houses, memory and identity in prehistoric Cyprus: Inferences from Ayios Epiktitos-Vrysi. In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre: Actes du Colloque International Organisé par le Départment des Antiquités de Chypre et l’École Française d’Athènes, Nicosie 17–19 Mai 2001, 99–118. Bulletin de Correspondance Hellénique Supplément 43. École Française d’ Athènes. Peltenburg, E. and S. Spanou 1999. The painted pottery from Ayios Epiktitos-Vrysi. Report of the Department of Antiquities, Cyprus, 13–34. Peltenburg, E., D. Bolger, P. Croft, E. Goring, B. Irving, D. A. Lunt, S. W. Manning, M. A. Murray, C. McCartney, J. S. RidoutSharpe, G. Thomas, M. E. Watt and C. Elliott-Xenophontos 1998. Lemba Archaeological Project II(1A). Excavations at Kissonerga-Mosphilia, 1979–1992. Studies in Mediterranean Archaeology 70(2). Jonsered, Paul Åströms Förlag. Peltenburg, E., D. Bolger, S. Colledge, P. Croft, C. Fox, E. Goring, A. Jackson, D. A. Lunt, C. McCartney, M. A. Murray, J. S. Ridout-Sharpe, G. Thomas and M. E. Watt 2003. Lemba Archaeological Project, Cyprus III(1). The Colonisation and Settlement of Cyprus: Investigations at Kissonerga-Mylouthkia 1976–1996. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Renfrew, C. 2001. Symbol before concept: Material engagement and the early development of society. In I. Hodder (ed.) Archaeological Theory Today, 122–140. Cambridge: Polity. Renfrew C. 2004. Towards a theory of material engagement. In E. DeMarrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World, 23–31. McDonald Institute Monographs. Cambridge, McDonald Institute for Archaeological Research. Schlanger, S. 1991. On manos, metates and the history of site occupations. American Antiquity 53, 460–473. Steel, L. 2004. Cyprus Before History: From the Earliest Settlers to the End of the Bronze Age. London, Duckworth. Watkins, T. 2004. Architecture and “theatres of memory” in the Neolithic of southwest Asia. In E. DeMarrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World, 97–106. McDonald Institute Monographs. Cambridge, McDonald Institute for Archaeological Research. Wiessner, P. 1983. Style and information in Kalahari San projectile points. American Antiquity 48, 253–276.

5 A DIFFERENT CHALCOLITHIC: A CENTRAL CYPRIOT SCENE

David Frankel

Introduction The development of archaeology is the product of many factors; as new techniques emerge, explanatory fashions come and go and new sites expand our database. Within this mix individual scholars have a major influence through the new material they expose and the way they define issues and problem domains. This is certainly true for prehistoric Cyprus where for over 30 years Eddie Peltenburg has laid out for us a material world of sites and artefacts and a virtual world of ideas and explanatory models. The projects he has managed at Kissonerga, Lemba and Souskiou define much of what we now understand as Chalcolithic Cyprus. This, inevitably, has not only expanded but also structures our views of this cultural system and its evolution. Our current models of Chalcolithic sites and strategies are largely based on these substantial villages of the southwest: here the excavation of richer, more archaeologically productive sites forms a selective filter which can only be modified by the gradual expansion of archaeological research. Material culture, primarily pottery, still has a primary role to play in establishing temporal variation. Our standard procedures, of necessity, inherently assume uniform patterns of change. In the absence of independent dating this inevitably militates against explorations of synchronic differences. These may take many forms and be the product of varied influences. The scale and nature of stylistic and technological variation may reflect geographic distance and corresponding degrees of social interaction. Raw material usage may be a matter of cultural choice or of availability; so too may be alternative economic strategies. Here environmental factors come into play to influence or structure cultural choices. Alongside these are variations in types of sites and their associated arrays of material culture.

In this brief paper I will take recent work at a small Chalcolithic site in central Cyprus as a starting point to consider the question of diversity in the Chalcolithic, treading warily as befits a Bronze Age poacher in Eddie’s Chalcolithic preserves.

Politiko-Kokkinorotsos The summary that follows is largely based on excavations carried out in 2006 and 2007 by Jenny Webb and myself, and studied with the very able assistance of Paul Croft and Carole McCartney upon whose insights I have drawn extensively (Webb et al. in press; Webb in press).

The Site Politiko-Kokkinorotsos (henceforth Kokkinorotsos) is a small site in the northern foothills of the Troodos, two kilometres west of Politiko (Fig. 5.1). It is situated on the interface between the igneous formations of the Troodos and the sedimentary structures of the northern half of Cyprus, in the narrow valley of the Koufos River as it flows out of the foothills to join the Pedeios, one of the main rivers which flow across the centre of the island. The site is defined by a discrete surface scatter of pottery of about 500 m2 in area. Although originally thought on the basis of surface finds to have material from the later 3rd millennium (Given and Knapp 2003, 192–197, 265–266 as Politiko-Phournia), 12 AMS dates show that it was only used between about 2880 and 2670 cal BC. No extant traces of structures were located, but very large quantities of pottery, stone and bone come from the fill of three irregular pits and an extremely large natural hollow (Fig. 5.2). All appears to be secondary

5. A different Chalcolithic: A central Cypriot scene

Fig. 5.1 Map of Cyprus showing the location of PolitikoKokkinorotsos and other relevant sites.

39

to the predominant fabric in Middle Chalcolithic Period 3B at Kissonerga-Mosphilia and on other Middle Chalcolithic sites in the south and southwest. The unpainted pottery and the coarse ware used for low-fired flat-based pans or trays are also generally within the common Middle Chalcolithic tradition. In the field five general varieties of unpainted pottery were recognised, separated from one another largely on the basis of surface colour. In other respects, however, they do not form discrete groups. While there is some association of these fabrics with particular shapes, technologically they do not differ significantly. A Principal Components Analysis shows the broad overlap of the five unpainted varieties in attributes of fabric and surface treatment (Fig. 5.4). Of these, Fabric E is the most distinctive. This finer quality variety is distinguished by a lustrous all black surface and small vessel size. All the unpainted fabrics are, however, best viewed as varieties or diffuse sub-sets within a common range and so belong to a single tradition of ceramic production. The most common shapes are those seen at all other Middle Chalcolithic sites: deep spouted holemouth jars, tubular-spouted vessels, large storage jars and small and medium-sized bowls. These suggest that a wide array of activities, including the use of finer quality and decorated vessels, were carried out at the site, broadly similar to those seen elsewhere.

Flora and Fauna Fig. 5.2 General view of the site and excavations at PolitikoKokkinorotsos.

or even tertiary redeposition, perhaps a result of material from middens and other discard contexts being shovelled into holes and hollows to level the area during a short and deliberate episode of clearance. Several explanations can be advanced for the lack of structural evidence at Kokkinorotsos. One is that the Chalcolithic buildings were entirely, perhaps deliberately, demolished in the 3rd millennium. A second is that they were removed in later times by erosion and other landscape transformations. A third is that there never were any other than ephemeral, light, perhaps temporary structures at the site. These, too, could have been deliberately demolished.

Pottery Over one tonne of pottery was recovered from the excavated area (Fig. 5.3). The great majority is unpainted, but there is a significant component of Red on White ware, similar

Both wild and cultivated species are represented in the plant remains. Their quantification is difficult, given the exigencies of deposition, fragmentation and sampling, but nuts predominate. It is likely that these represent the harvesting of wild plants; the same may be true for fruits such as olive, fig and grape. Barley and wheat are well represented and are most likely to be cultivated crops although they may not have been grown near Kokkinorotsos. The fauna from the site are of particular importance. Paul Croft’s analysis of 150 kg of animal bone (including more than 7,000 identifiable pieces of large mammalian bone) shows that it was dominated by fallow deer (Dama mesopotamica). Deer make up three-quarters of identified fragments in the assemblage; almost all the remainder were sheep with a small number of goats and pigs (Fig. 5.5). Although deer are the major contributor to the diet at contemporary sites, they do not dominate to the same extent as at Kokkinorotsos. Equally, if not more importantly, however, is the total concentration on hunting: unlike other Middle Chalcolithic sites, none of the animals at Kokkinorotsos appear to have been herded. Two lines of evidence point toward this. One is based on estimates of the ages at which animals were killed. The age profiles of caprines and deer are very similar, matching those of hunted deer elsewhere (as at Bronze Age Marki-Alonia)

40

David Frankel

Fig. 5.3 A selection of sherds from Politiko-Kokkinorotsos.

5. A different Chalcolithic: A central Cypriot scene

Fig. 5.4 Principle Components Analysis of pottery attributes showing the general overlap of varieties of pottery.

41

Fig. 5.5 Relative importance of animal species at PolitikoKokkinorotsos compared with Middle Chalcolithic KissonergaMosphilia and Lemba-Lakkous (data from Croft 1985; 1998).

and differing significantly from those expected of herded animals. The other is the representation of body parts. Once again these fit with a pattern characteristic of hunting where those elements with less meat (such as the head and lower limbs) are normally left at the kill site and are therefore under-represented in site assemblages.

Stone Tools Carole McCartney’s analysis of the assemblage of 5,704 chipped stone artefacts demonstrates that the full range of activities associated with core reduction and tool production is represented. Although some types were probably brought ready-made to the site, most were manufactured and rejuvenated there. Three-quarters of the chipped stone artefacts are of Lefkara basal chert, a moderate to good quality raw material probably collected from the local river bed. This, and the other types of stone, can all be found in the vicinity of the site. Knappers at Kokkinorotsos employed a simple core technology that fits well within the Cypriot Chalcolithic. It is primarily a flake-based industry with few blades or bladelets. The tool assemblage is heavily dominated by three major classes: pièces esquillées, scrapers and burins, which together make up two-thirds of the tools (Fig. 5.6). All can be associated with butchery or with processing animals. Pièces esquillées are likely to have been used as wedges for splitting bone, the scrapers for cleaning hides, and the burins for working with antler or bone. Most other chipped stone tool types are found in smaller quantities, suggesting that a relatively limited range of activities were carried out at the site. There are, for example, very few pieces showing

Fig. 5.6 Proportional occurrence of the main chipped stone artefacts types at Politiko-Kokkinorotsos.

signs of silica gloss such as would result from harvesting or cutting plants. Most of the varied types of ground stone tools were made from readily available igneous rocks. The most common are rubbers or querns, accounting for over a quarter of the sample. Other heavy duty tools such as pounders are also well represented, but there are comparatively few cutting or wood-working tools (Fig. 5.7). While all tool types show high rates of breakage, this was particularly true of rubbers, querns and bowls, all of which were incomplete. These were clearly among the most intensively utilised as well as the most common artefacts in the assemblage. The high breakage rate of simple pounders also suggests heavy use.

42

David Frankel Ambeikou-Ayios Georghios in the Cyprus Museum. There are, in addition, a handful of Dentalium and other shells and a picrolite bead. Although the function of the dozens of modified sherds remains unclear, they, together with these other items, show that some general activities seen on other Chalcolithic sites also took place at Kokkinorotsos.

Activities and Function

Fig. 5.7 Proportional occurrence of main functional categories of ground stone tools from Politiko-Kokkinorotsos compared with Middle and Late Chalcolithic Kissonerga-Mosphilia and LembaLakkous (data from Elliott 1985; Elliott-Xenophontos 1998).

Other Items Alongside the more common artefact types noted above a small number of bone and antler tools were recovered. The former include spatulate implements and points showing evidence of the application of considerable pressure during use, and a needle. The latter are principally cut from antlers with evidence of concentric grooving and snapping to facilitate beam removal. While these clearly indicate on-site processing, the number of worked fragments is very small in comparison to the size of the faunal sample, suggesting that antler working was not a major activity at Kokkinorotsos. A few other items are of less immediately practical utility. These include fragments of two Red on White anthropomorphic figurines similar to examples from Middle Chalcolithic levels at Kissonerga-Mosphilia and SouskiouVathyrkakas (Goring 1991; 1998, 154–158; 2006) as well as other sites, including an unpublished fragment from

At first sight various elements at Kokkinorotsos tell different stories. The lack of structures, the exclusive presence of hunted animals, and the high proportion of chipped stone tools associated with butchery or processing animal products, together with the lack of agricultural tools or those appropriate for forest clearance or timber-working, all strongly point to a site used intermittently, perhaps seasonally, specifically for hunting. The timing of mouflon culling indicates that this was likely to have been in the spring and summer. Other artefacts – the storage jars, the spouted pottery vessels useful for handling liquids, the domesticated cereals and the grinding equipment – indicate that a variety of activities were carried out and other types of food were prepared and consumed. The finer, decorated pottery figurines and personal items such as beads also indicate a broader range of behaviour. These varied aspects can, however, be reconciled. The site is best understood as primarily a hunters’ camp; a base used, perhaps seasonally, for forays into the surrounding woodland and for preparing hides and other products from the resulting quarry. The pottery and other material suggest that these were not brief periods of use, but that the site was occupied for longer periods, perhaps weeks or even months at a time, possibly by family groups rather than by hunters alone. The pottery is likely, therefore, to have been brought to the site along with agricultural produce and other supplies from a more permanent larger settlement, perhaps situated some distance away, either closer to the coast or in the more open and productive floodplain of the Pedeios River, perhaps in the vicinity of Nicosia. In such a model Kokkinorotsos functioned as one, special purpose component of a Chalcolithic economic and settlement system built around a larger, more permanently occupied agro-pastoral village.

Kokkinorotsos in Context Limited material and uncertain dating inhibit other discussions of variability in the archaeological record of Neolithic and Chalcolithic Cyprus. For example, in the absence of independent dating differences between contemporary sites in the presence or proportional occurrence of diagnostic pottery are hard to assess (cf Stanley Price 1979, 43). This is made more difficult by the uneven representation of material from different parts of the island.

5. A different Chalcolithic: A central Cypriot scene Regional variation is a common theme in Cypriot prehistory (Frankel 2009), but we have still to come to terms with what spatial differences in material culture (generally pottery) mean in terms of social relationships at different scales of distance and intensity. In some cases divergence may be the result of inherent processes of cultural ‘drift’ as new types or varieties develop in separated areas and gradually become increasingly common. The rate and degree of these developments can vary considerably, confusing our ability to establish contemporaneity or levels of interaction. Where other mechanisms come into play varied degrees of material difference can signal degrees of social interaction, defining for us – and possibly for people of the time – social identity at a different scale (Webb, this volume). There is at present no reason to see the latter aspect as significant in the regional variations in ceramics during the Chalcolithic. While specific longer distance chains of connection linking different parts of the island are indicated by the presence of picrolite and marine shells at Kokkinorotsos, general relationships and an associated diffusion of techniques and styles account for the broad similarity of the pottery and other material culture. The ceramic assemblage at Kokkinorotsos is, not surprisingly, most similar to material from other sites in the central region of Cyprus, notably Ambelikou-Ayios Georghios, Philia-Drakos site B and Kyra-Alonia, which contain the same array of unpainted, painted and coarse fabrics (Dikaios 1962, 141–155). In these central lowland and foothill sites unpainted wares dominate. This contrasts with the southwest where a high proportion of painted pottery is seen as characteristic of the earlier third millennium (i.e. the Middle Chalcolithic) and only replaced by monochrome wares in the Late Chalcolithic. Many years ago, on the basis of very limited evidence, Stanley Price argued for a distinctly different trajectory of pottery change in different parts of the island. North of the Troodos an earlier emphasis on monochrome pottery preceded a greater use of painted wares – the reverse of the sequence in the south (Stanley Price 1979, 38, 46, figs 8–9). Such broad regional variations in pottery production provide one possible explanation for the low representation of painted pottery at Kokkinorotsos at a time when these wares dominated elsewhere. The consistency of the 12 radiocarbon dates place occupation between 2900 and 2700 cal BC. While the closest Chalcolithic sites (Ambelikou, Kyra and Philia) are likely to be broadly contemporary, its period of use therefore falls within an apparent hiatus between Middle and Late Chalcolithic in the sequence at Kissonerga-Mosphilia and elsewhere. Apart from the central Cypriot sites, only the recently excavated site of Souskiou-Laona fills the chronological gap (Peltenburg personal communication). This raises the question of whether the site represents a specific response to particular events or is part of a more deeply embedded, longer-lasting tradition.

43

One approach to understanding Chalcolithic settlement history emphasises periodic collapse with a local fissioning or restructuring of specific systems due to social forces; another explains broader-scale developments by reference to climatic deterioration or amelioration. Following the latter approach, the settlement hiatus between the Middle and Late Chalcolithic at Kissonerga-Mosphilia and other southwestern sites is attributed to a decline in resource availability (Peltenburg 1993). The occupation of Kokkinorotsos at this time could then be regarded as a symptom of a widespread reversion to low intensity occupation and an increased emphasis on mobile hunting strategies, such as that which may have characterised the early phases of the Early Chalcolithic (Clarke 2007; Wasse 2007, 62). This would account for the size, location and nature of the excavated remains and the concentration on hunting although not for the importation of cereals and other goods. In addition, any broad climatic deterioration would have had an even greater impact in the centre of the island than in the better favoured environments in the southern coastal plains. It is therefore unlikely that such a settlement collapse in the south would have led to an increase in activity in poorer areas. At a different scale, specific catastrophic events are also sometimes adduced to explain disruptions to established patterns (Peltenburg 1978; 1982, 65; Stanley Price 1979, 73–77; Todd and Croft 2004, 222, 225). Although it is possible that the localised impact of natural disasters such as earthquakes might have had a short-lived effect on particular settlements, these need not have had more than transient effects. Elsewhere in the world, even where more devastating events such as volcanic eruptions blanket land with lava or ash, the effects are not always negative and may be surprisingly short-lived, while the social context is a critical factor in determining responses (Sheets 1979, 555–58; Grattan and Torrence 2007). Seen in this way, neither more general nor very specific natural factors provide an adequate explanation for islandwide developments involving the emergence of sites like Kokkinorotsos. Rather than being seen primarily as a response to environmental events, it is more likely that small-scale, periodically inhabited hunting villages like Kokkinorotsos may always have been part of the Chalcolithic landscape and one of a number of site types in contemporary use. At other Chalcolithic sites a mixture of wild and domesticated animals are in evidence with a concomitant variation in the proportional representation of sets of stone tools associated with butchery (Fig. 5.8). The variability between assemblages in these two aspects is such that no simple chronological trend can be observed. Kokkinorotsos differs in demonstrating a total focus on hunted, wild animals. As discussed above, this is best explained by regarding Kokkinorotsos as a special purpose hunting station while other sites had a more varied function.

44

David Frankel

Fig. 5.8 Proportions of chipped stone artefact types associated with butchery compared with the estimated contribution of deer to the meat supply at Chalcolithic sites.

This explanation also fits with the lack of formal structures at Kokkinorotsos, in contrast to the ‘classic’ Middle Chalcolithic sites where large, well built circular houses are the norm both in the south-west and north of the Troodos. Some other sites, such as those near Kalavasos, also appear to lack evidence of substantial above ground buildings (Todd and Croft 2004; Clarke et al. 2007), which may either have been entirely destroyed, have yet to be found or never existed (cf Todd 1987, 29). There is, however, a wider variety of pits and tunnels in the Kalavasos sites, features also common in the Early Chalcolithic. These sometimes have internal features which indicate their possible use for storage or some specific activities even if they were not general dwellings (Todd and Croft 2004, 220–23; Clarke et al. 2007, 49). The irregular pits at Kokkinorotsos, therefore, also differ from these features, again suggesting a low intensity of occupation. Contemporary sets of neighbouring Chalcolithic settlements are often regarded as forming site clusters with a complex history of associated phases of expansion, contraction, abandonment and re-establishment (Peltenburg 1993; 1998; 2003, 272). The sites are, for the most part, seen as equivalent to one another, differing in size rather than kind, with the possible exception of the Souskiou complex which may have been a regional integrative centre (Peltenburg 2006). Kokkinorotsos (and perhaps sites such as Kalavasos-Ayious) can also be seen as an example of a functionally different site type, distinct from the ‘normative’ permanent settlements such as Lemba-Lakkous and Kissonerga-Mosphilia.

Variation in functional site types is more evident in some periods of Cypriot prehistory than in others. This is in part a product of site survival and identification together with the history of research. It may also be that some cultural systems had more clearly distinct settlement types than did others. So, for example, several functional site types are well recognised for the Late Bronze Age, including urban centres, mining sites and agricultural villages (Catling 1962; Keswani 1993; Webb and Frankel 1994; Knapp 1997), but as yet no such structured differences have been identified during most of the Early and Middle Cypriot periods. Kokkinorotsos, seen in this analysis as a special-purpose hunting station distinct from a primary, permanent agropastoral village, opens up the possibility that community territories may have been made up of a series of sites, each the focus of particular activities appropriate to their ecological setting. In this way a main contribution of Kokkinorotsos to our understanding of the Chalcolithic is to expose more of the complexity and variability in the cultural system. It expands the frameworks, providing, perhaps, more questions than answers: but that, surely, is the nature of all archaeological research.

Acknowledgements As on so many occasions I am especially grateful to Jenny Webb, co-director of the Kokkinorotsos project. I have also relied heavily on the work of Paul Croft and Carole McCartney, who provided the essential documentation, analysis and insights regarding the animal bones and chipped stone on which any understanding of the site depends. Funding was provided by the Australian Research Council with additional support from La Trobe University.

References Catling, H. W. 1962. Patterns of settlement in Bronze Age Cyprus. Opuscula Atheniensia 4, 129–169. Clarke, J. 2007. On the Margins of Southwest Asia: Cyprus during the 6th to 4th Millennia BC. Oxford, Oxbow. Clarke, J., P. Croft and C. McCartney 2007. The 1940s excavations at Kalavasos-Kokkinogia and Kalavasos-Pamboules. Report of the Department of Antiquities, Cyprus, 45–86. Croft, P. W. 1985. The mammalian fauna. In E. Peltenburg et al., Lemba Archaeological Project 1: Excavations at Lemba Lakkous 1976–1983, 202–208. Studies in Mediterranean Archaeology 70(1). Göteborg, Paul Åströms Förlag. Croft, P. W. 1998. Animal remains: Synopsis. In E. Peltenburg et al., Lemba Archaeological Project II(1A): Excavations at Kissonerga-Mosphilia 1979–1992, 205–214. Studies in Mediterranean Archaeology 70(2). Jonsered, Paul Åströms Förlag. Dikaios, P. 1962. The Stone Age. In P. Dikaios and J. R. B. Stewart,

5. A different Chalcolithic: A central Cypriot scene The Swedish Cyprus Expedition, Vol. IV Part 1A: The Stone Age and the Early Bronze Age in Cyprus, 1–204. Lund, The Swedish Cyprus Expedition. Elliott, C. 1985. The ground stone industry. In E. Peltenburg et al., Lemba Archaeological Project I: Excavations at LembaLakkous, 1976–1983, 70–93, 161–195. Studies in Mediterranean Archaeology 70(1). Göteborg, Paul Åströms Förlag. Elliott-Xenophontos, C. 1998. Ground stone tools. In E. Peltenburg et al., Lemba Archaeological Project Volume II(1B). Excavations at Kissonerga-Mosphilia 1979–1992, 193–232. Edinburgh, University of Edinburgh, Department of Archaeology Occasional Paper 19. Frankel, D. 2009. What do we mean by ‘regionalism’? In I. Hein (ed.) The Formation of Cyprus in the 2nd Millennium B.C. Studies in Regionalism during the Middle and Late Bronze Age, 15–25. Denkschriften der Österreichischen Akademie der Wissenschaften, Contributions to the Chronology of the Eastern Mediterranean. Vienna. Given, M. and A. B. Knapp 2003. The Sydney Cyprus Survey Project. Los Angeles, University of California Press. Goring, E. 1991. The anthropomorphic figures. In E. Peltenburg et al., Lemba Archaeological Project Volume II(2). A Ceremonial Area at Kissonerga, 39–60. Studies in Mediterranean Archaeology 70(3). Göteborg, Paul Åströms Förlag. Goring, E. 1998. Figurines, figurine fragments, phalli, possibly figurative worked and unworked stones, unidentifiable worked stone and pottery fragments. In E. Peltenburg (ed.) Lemba Archaeological Project Volume II(1A). Excavations at Kissonerga-Mosphilia 1979–1992, 148–167. Studies in Mediterranean Archaeology 70(2). Jonsered, Paul Åströms Förlag. Goring, E. 2006. Anthropomorphic and zoomorphic figurines and pendants. In E. Peltenburg (ed.) The Chalcolithic Cemetery of Souskiou-Vathyrkakas, Cyprus. Investigations of Four Missions from 1950 to 1997, 69–90. Nicosia, Department of Antiquities, Cyprus. Grattan, J. and R. Torrence 2007. Living Under the Shadow: Cultural Impacts of Volcanic Eruptions. Walnut Creek (CA), Left Coast. Keswani, P. S. 1993. Local exchange in Late Bronze Age Cyprus. Bulletin of the American Schools of Oriental Research 292, 73–83. Knapp, A. B. 1997. The Archaeology of Late Bronze Age Cypriot Society: The Study of Settlement, Survey and Landscape. Glasgow, Department of Archaeology, University of Glasgow. Peltenburg, E. J. 1978. The Sotira culture: Regional diversity and cultural unity in Late Neolithic Cyprus. Levant 10, 55–74.

45

Peltenburg, E. J. 1982. Recent Developments in the Later Prehistory of Cyprus. Studies in Mediterranean Archaeology Pocketbook 16. Göteborg, Paul Åströms Förlag. Peltenburg, E. 1993. Settlement discontinuity and resistance to complexity in Cyprus, c. 4500–2500 B.C.E. Bulletin of the American Schools of Oriental Research 292, 9–23. Peltenburg, E. (ed.) 1998. Excavations at Kissonerga-Mosphilia 1979–1992. Lemba Archaeological Project II(1A). Studies in Mediterranean Archaeology 70(2). Jonsered, Paul Åströms Förlag. Peltenburg, E. (ed.) 2003. Lemba Archaeological Project III(1). The Colonisation and Settlement of Cyprus: Investigations at Kissonerga-Mylouthkia, 1986–1996. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Peltenburg, E. (ed.) 2006. The Chalcolithic Cemetery of SouskiouVathyrkakas, Cyprus. Investigations of Four Missions from 1950 to 1997. Nicosia, Department of Antiquities, Cyprus. Sheets, P. D. 1979. Environmental and cultural effects of the Ilopango eruption in Central America. In P. D. Sheets and D. K. Grayson (eds), Volcanic Activity and Human Ecology, 525–564. New York, Academic Press. Stanley Price, N. P. 1979. Early Prehistoric Settlement in Cyprus 6500–3000 B.C. British Archaeological Reports International Series 65. Oxford, British Archaeological Reports. Todd, I. A. 1987. Vasilikos Valley Project 6: Excavations at Kalavasos-Tenta. Studies in Mediterranean Archaeology 71(6). Göteborg, Paul Åströms Förlag. Todd, I. A. and P. W. Croft 2004. Vasilikos Valley Project 8: Excavations at Kalavasos-Ayious. Studies in Mediterranean Archaeology 71(8). Sävedalen, Paul Åströms Förlag. Wasse, A. 2007. Climate, economy and change: Cyprus and the Levant during the late Pleistocene to mid Holocene. In J. Clarke, On the Margins of Southwest Asia: Cyprus during the 6th to 4th Millennia BC, 43–63. Oxford, Oxbow. Webb, J. M. in press. Central Cyprus in the early third millennium BCE: Problems and prospects. In A. Demetriou (ed.) Proceedings of the Fourth International Cyprological Congress, April 2008. Nicosia, Society of Cypriot Studies. Webb, J. M. and D. Frankel 1994. Making an impression: Storage and surplus finance in Late Bronze Age Cyprus. Journal of Mediterranean Archaeology 7(1), 5–27. Webb, J. M., D. Frankel, P. W. Croft and C. McCartney 2009. Excavations at Politiko-Kokkinorotsos: A Chalcolithic hunting station in Cyprus. Forthcoming in Proceedings of the Prehistoric Society 75, 189–238.

6 THOUGHTS ON THE FUNCTION OF ‘PUBLIC BUILDINGS’ IN THE EARLY BRONZE AGE SOUTHERN LEVANT Hermann Genz

Introduction With his excavations at Jerablus Tahtani Eddie Peltenburg has greatly enhanced our understanding of the emergence and development of complex societies in the Northern Levant. Although in contemporary societies of the Southern Levant no elite tombs comparable with Tomb 302 at Jerablus Tahtani (Peltenburg 1999) are attested, the presence of a number of public buildings suggests that elites there played a somewhat different, but equally decisive role in the emergence of complexity. In the traditional view, the transition from the Early Bronze Age I to the Early Bronze Age II, dated to c. 3100 cal BC according to recent radiocarbon dates, marks the beginning of urban settlements in the Southern Levant (Kempinski 1978; Kenyon 1979, 84–86; Esse 1989; Miroschedji 1989; Mazar 1990, 108). These urban settlements are characterised by a dense population concentration in an area restricted by fortification walls, and especially by the presence of public buildings, generally called ‘palaces’ and ‘temples’. It has only recently become clear that the urbanisation of the Southern Levant followed quite a different trajectory from that of Mesopotamia and that the urban sites of the Southern Levant are not simply down-scaled versions of Mesopotamian cities (Falconer and Savage 1995; Chesson and Philip 2003, 8). Thus the earlier idea of secondary state formation triggered by developments in Mesopotamia (Kempinski 1978; Kenyon 1979, 84–86) is no longer acceptable. Recently even the existence of truly urban settlements in the Southern Levant during the Early Bronze Age has been questioned (Philip 2001; Chesson and Philip 2003), and other models such as ‘corporate village’, ‘heterarchy’ and ‘middle range society’ have been suggested to describe the Early Bronze Age settlements of the Southern Levant.

In the light of this controversy it is somewhat surprising that little work has been done on the nature and function of the so-called public buildings that have been excavated at a number of Early Bronze Age II–III sites in the Southern Levant (Fig. 6.1). While the study of domestic households has made great progress in recent years (Chesson 1997; Ilan 2001; Chesson 2003), public buildings are still often only described according to their architectural layout (Kempinski 1992a; 1992b). This is certainly due to the fact that for earlier excavations contextual information for the contents of the buildings and individual rooms is mostly lacking. A few examples from recent excavations, however, offer new insights into the possible functions of these buildings.

Temples Although a multitude of papers deal with Early Bronze Age temples of the Southern Levant (Amiran 1981; Mazar 1990, 125–126; Kempinski 1992a; Miroschedji 1993; Philip 2001, 175–176), there is still insufficient evidence for a clear definition of their precise function. In recent years more systematic definitions of how to identify cult buildings in the archaeological record have greatly enhanced our understanding of religion and cult practice in pre-literate societies (Renfrew 1985, 11–26; Alon and Levy 1989). However, not all of these can be directly applied to the Early Bronze Age Southern Levant for a variety of reasons further discussed below. The majority of scholars have focussed on architectural plans as the major element for defining temples. There seems to be a general agreement that the layout of Early Bronze Age temples in the Southern Levant followed the broadroom model, but when it comes to the identification of specific

6. Thoughts on the function of ‘public buildings’ in the Early Bronze Age Southern Levant

Fig. 6.1 Early Bronze Age II–III sites with public buildings mentioned in the text.

buildings as temples, opinions still widely differ. Thus, for instance, the so-called sacred area at Arad (Amiran 1981, 49) is now generally regarded as a domestic structure (Mazar 1990, 126; Kempinski 1992a, 57). Also the ‘bâtiment blanc’ at Tel Yarmuth, first identified as a temple (Miroschedji 1988, 40–41; 1993, 208–211) has been stripped of that status by the excavator himself (Miroschedji 1999, 9–10). Lastly, the suggestion that the ‘Acropolis Building’ at cAi/et-Tell was converted from a temple to a palace at the transition from the Early Bronze Age II to the Early Bronze Age III (Callaway 1972, 292–293) rests more on speculations than on hard evidence. It has to be noted that no other case of the conversion of a religious to a secular building is attested for the Bronze Age Near East. Many problems connected with the identification of cult buildings and the interpretation of their function are generally based on incomplete excavation and publication as well as, for the earlier excavations, a lack of detailed recording of objects and other materials associated with these buildings.

47

Therefore the discussion of temples will be restricted to the most completely excavated examples: temples 4040, 5192, 5269 and the circular platform from stratum XV at Megiddo (Loud 1948, 78–84; Kempinski 1989, 28–35; Esse 1991, 87–89) and the sacred precinct in the Upper City of Khirbet ez-Zeraqon (Mittmann 1994, 13–14; Genz 2002, 94–96; see Fig. 6.2), both of which date to Early Bronze Age III. Both at Megiddo and at Khirbet ez-Zeraqon the cultic complexes are situated in the highest parts of the settlements. Although their layout is different, both complexes consist of three major broadroom structures and a large circular stone platform, and both are enclosed by a wall. The broadroom structures at Megiddo and Khirbet ezZeraqon are remarkably similar, showing a main rectangular room with the entrance in one of the long walls and benches or podia opposite the entrance. Two stone slabs placed on the axis of the room served as bases for wooden pillars supporting the roof. At both sites the broadrooms had anterooms. At Megiddo these were entirely open to the front and only had two pillar bases indicating that they also were roofed. At Khirbet ez-Zeraqon, however, the antae of buildings B0.4 and B0.5 showed a right-angle bend and thus partly closed the anterooms. Nevertheless, as at Megiddo two stone pillar bases were attested. Only building B0.1 in Khirbet ez-Zeraqon differs in its layout as it lacks an anteroom. Instead it has a narrow rectangular room attached to the back of the main room. There seems to be general agreement that the buildings just discussed indeed served as temples. Their specific ground plans, and the fact that they lack installations such as ovens, fireplaces and grinding installations, preclude their identification as domestic buildings. The podia opposite the entrances definitely served as focal points in the main rooms. While it is generally assumed that these podia served to display the images of deities, it needs to be pointed out that neither Megiddo, Khirbet ez-Zeraqon nor any other presumed Early Bronze Age temple from the Southern Levant has provided any evidence for such images. It may be assumed, of course, that such images consisted of perishable materials such as wood. However, even the (admittedly scanty) iconographic evidence of the Early Bronze Age Southern Levant, such as clay figurines or cylinder seal impressions (Ben-Tor 1977; Miroschedji 1993, 211–212), does not depict any images of divinities. The anthropomorphic figures which are occasionally depicted are more likely to be human worshippers. Thus we not only lack information on how Early Bronze Age divinities were depicted, but we do not even know whether they were depicted at all. Based on evidence mainly dating back to the Chalcolithic period, de Miroschedji (1993, 213–216) has reconstructed a divine couple for the Early Bronze Age comprising a female deity connected with water and a male deity represented by horned animals such as rams or ibexes. Definite proof for such a divine couple in the Early

48

Hermann Genz

Fig. 6.2 Plan of the upper city of Khirbet ez-Zeraqon, Jordan (after Genz 2002, fig. 2).

Bronze Age is still lacking, however. In addition, it has to be pointed out that in Stratum XV at Megiddo three temples existed simultaneously. Regardless of what happened inside the broadroom buildings, the restricted space inside the rooms suggests only a small number of participants at a time; thus access to the buildings clearly must have been limited to specific groups of the community. The round stone platforms attested at Megiddo and Khirbet ez-Zeraqon are generally interpreted as altars. Indeed, the remains of burnt animal bones were directly associated with this structure at Khirbet ez-Zeraqon (Mittmann 1994, 13; B. Dechert, personal communication). Unfortunately, no evidence is available for Megiddo, but it seems reasonable to assume that both structures served the same purpose, namely as altars for animal sacrifices. At Khirbet ez-Zeraqon the large courtyard around the platform allowed the participation of larger numbers of people in such events. The majority of the burnt bones consist of the lower leg bones of sheep

and goat. It remains unclear what happened to the rest of the meat, but it can be assumed that it was consumed either by the specialists who conducted or supervised the ceremonies, or even by all the participants. In this respect it is interesting to note that B0.2, very likely a service building situated directly north of the altar in Khirbet ez-Zeraqon, contained installations and pottery vessels connected to food preparation (Genz 2002, 95) although again for only a limited number of people due to its rather small size and the presence of only a few cooking pots. Except for the animal sacrifices there is little evidence for other kinds of offerings. Philip (2001, 176) has noted the general absence of offering pits in Early Bronze Age temples in contrast to their marked presence in Middle and Late Bronze Age cultic contexts. The only exception could be the deposit of 21 semi-complete or complete vessels deposited in a stone-lined pit (R0.11/i0.3) in the courtyard of the cultic complex at Khirbet ez-Zeraqon (Genz 2002, 95–96 and pl. 3–8).

6. Thoughts on the function of ‘public buildings’ in the Early Bronze Age Southern Levant Also noteworthy is the general absence of ritual objects. In addition to a lack of all kinds of cult images, other objects, such as incense burners or offering stands, which characterise later Middle and Late Bronze Age temples, are not attested in the Early Bronze Age. The only exception is again found at Khirbet ez-Zeraqon where the occurrence of small ceramic bowls with snake applications is restricted to the cultic complex in the Upper City (Genz 2002, 102). Thus they may be interpreted as vessels with a specific cultic function. Philip (2001, 176) has observed that Early Bronze Age temples in the Southern Levant are never associated with storage and administrative features. Thus it can be concluded that temples had an exclusively religious function. Among the rituals performed, animal sacrifices and food consumption, very likely for a limited number of participants, can be identified as the main activities in the cultic complexes. While some of the criteria established by Renfrew (1985, 11–26) for the identification of cult buildings are thus fulfilled, such as special location, special layout, the existence of boundary walls between sacred and profane areas, and the presence of focal points in the buildings, others are lacking, especially evidence for cultic paraphernalia, religious symbols and images of divinities.

Palaces Non-religious public buildings have traditionally been termed ‘palaces’ (Mazar 1990, 123; Kempinski 1992b, 78). This term generally implies that these buildings served as elite residences as indeed is attested for Near Eastern palaces from the Middle Bronze Age onwards. The Early Bronze Age in the Southern Levant, however, is characterised by a notable lack of features associated with elites, such as prestige objects made of precious or exotic materials or elaborate and richly furnished tombs (Genz 2000, 58; Philip 2001, 165). It is therefore advisable to use a more general term for these buildings such as ‘administrative structures’ as has been suggested by Philip (2001, 176–178). Such administrative structures are so far only known from four sites: ‘palace 3177’ at Megiddo (Loud 1948, 70–76; Esse 1991, 83–84; Kempinski 1992b, 78); ‘palace B’ in Tel Yarmuth (Miroschedji 1999, 10–12); the ‘palace’ in the Upper City of Khirbet ez-Zeraqon (Mittmann 1994, 14; Genz 2002, 96; see Fig. 6.2); and the so-called commercial building in Field 1 from stratum L2 on the Lower tell at Tell es-Saidiyeh (Tubb et al. 1997, 55–65). While the former three date to the Early Bronze Age III, the building at Tell es-Saidiyeh is firmly dated to Early Bronze Age II. Only ‘palace B’ at Tel Yarmuth has been completely excavated. The incomplete exposure of the other three buildings makes a study of their layout and an interpretation of their function rather difficult. In the case of the building at

49

Tell es-Saidiyeh it is not even clear whether we are dealing with one building (Tubb et al. 1996, 20) or two buildings separated by an open space as the most recent report suggests (Tubb et al. 1997, 64). All of these public buildings consist of a large number of rooms of varying sizes connected by courtyards and corridors. While the buildings at Tel Yarmuth and possibly at Megiddo seem to have been erected according to a preconceived plan (Miroschedji 1999, 10; 2001), at Khirbet ez-Zeraqon different building stages can clearly be recognised, as building B0.8 was added some time after the more solidly executed parts B0.7 and B0.10 had been erected (Fig. 6.2). Access to these buildings seems to have been restricted as demonstrated by the walls surrounding ‘palace B’ at Tel Yarmuth and building 3177 at Megiddo. Also, for the building at Khirbet ez-Zeraqon only one entrance has been identified. The careful and solid construction of these buildings, as well as their layout, clearly shows that they were at least partly constructed for representational purposes such as receptions. Indeed, evidence for food consumption on a larger scale is attested at Tell es-Saidiyeh (Tubb et al. 1997, 62). However, the most prominent function of these buildings was for storage. At Tel Yarmuth more than 150 large pithoi were discovered in the rooms of ‘palace B’ (Miroschedji 1999, 11–12) while at Khirbet ez-Zeraqon (Genz 2002, 96–97 and pl. 25–37) and at Tell es-Saidiyeh (Tubb et al. 1997, 59–62) large numbers of storage vessels were found. The quantity of goods stored in ‘palace B’ at Tel Yarmuth clearly must have exceeded the needs of the people possibly living in this building. It has to be kept in mind that the inhabitants of the domestic dwellings kept their own supplies as attested by numerous pithoi and storage jars discovered in such buildings (Genz 2002, 99–104). Thus in the administrative complexes the surplus not needed for immediate consumption seems to have been stored. The nature of the vessels discovered suggests mainly the storage of valuable liquids such as oil or wine (Genz 2003). It is not known whether these commodities were collected in the administrative buildings for trade or for other purposes, and the methods of recording and administrating these goods remain unclear. While marking systems such as cylinder seal impressions (Flender 2000) and incised signs on pots (Genz 2001) are widely attested during the Early Bronze Age in the Southern Levant, they cannot yet be conclusively linked to the administrative buildings as they also frequently occur in domestic dwellings.

Other Buildings of a Public Nature The so-called ‘granary’ from Early Bronze Age III levels at Beth Yerah is thus far unique in the Southern Levant, but its interpretation as a granary is generally accepted (Esse 1991, 53; Kempinski 1992b, 75–78; Mazar 2001). According

50

Hermann Genz

Fig. 6.3 Plan of the lower city of Khirbet ez-Zeraqon, Jordan (after Douglas 2007, fig. 4).

to Mazar’s calculations (2001, 458), its storage capacity considerably exceeded the needs of the city’s inhabitants, especially if additional storage in domestic dwellings is taken into consideration. Thus it is very likely that only the surplus not needed for immediate consumption was stored in this building, either for trading or as a reserve for times of famine. The broadroom in the centre of the structure has been interpreted as a sanctuary by some scholars (see Mazar 2001, 454–455) although this interpretation is by no means secure. Unlike temples in Mesopotamia there is so far no other evidence that temples in the Southern Levant had any control over agricultural production, as mentioned above. A completely different type of public building was excavated in the Lower City of Khirbet ez-Zeraqon, dating to the later parts of the Early Bronze Age II and the earlier parts of the Early Bronze Age III (Douglas 2007, 7). Situated directly east of the Lower City gate, building B1.5 consists of a large rectangular room with a row of column bases along its axis and a large courtyard between the room and

the gate house of the city gate (Douglas 2007, 46–47 and Abb. 3–4; see Fig. 6.3). The layout of the building, as well as the fact that it did not contain any installations for food storage and preparation, precludes its interpretation as a domestic building. Instead the building seems to have been linked to the city gate. This is suggested by the close proximity of the two structures, but even more so by the fact that B1.5 was abandoned and replaced by a large open court shortly before the city gate went out of use by being blocked (Douglas 2007, Abb. 5). Unfortunately, the lack of in situ finds in B1.5 prevents a secure determination of its specific function. The building may have had a military function to accommodate off-duty guards, or an administrative function for the registration of the flow of goods and people through the gate. An almost identical building is attested in Tel Yarmuth with the ‘bâtiment blanc’ (Miroschedji 1988, 35–41; 1999, 9–10), and there the building is also situated reasonably close to a city gate; thus a similar function as for B1.5 at Khirbet ez-Zeraqon may be suggested.

6. Thoughts on the function of ‘public buildings’ in the Early Bronze Age Southern Levant

Conclusions Although admittedly few public buildings from urban settlements in the Early Bronze Age Southern Levant have been discovered so far, it has to be kept in mind that extremely few sites have been excavated on a large scale. It can be assumed that public buildings formed a regular component of at least medium to large-sized urban settlements in the Southern Levant during the Early Bronze Age II–III. A surprisingly large variety of public buildings is attested, ranging from religious to administrative and storage structures. Even with the limited amount of public buildings available, some kind of standardised plan for buildings of different functions can be discerned in the Southern Levant and even beyond. The existence of a larger variety of presumably standardised public buildings seems to suggest the existence of more complex societies organised beyond the level of kinship ties in the Southern Levant than has been acknowledged in recent discussions (Philip 2001; Chesson 2003). While this does not imply a return to the city-state model prevalent before the 1990s (Esse 1989; Miroschedji 1989), it certainly suggests that elites had important functions in economic and administrative processes, as suggested by Joffe (1993). Elites in these societies do not seem to have invested their wealth in prestige objects and elaborate burial structures but rather in the economic and administrative sector. This suggestion is in accordance with the staple finance model, which has previously been proposed for the Early Bronze Age society of the Southern Levant (Philip 2001; Genz 2003). In order to fully understand the socio-political and economic developments of societies during the Early Bronze Age in the Southern Levant, public buildings should not be treated as exceptions but as integral parts of the social systems which produced them.

References Alon, D. and T. E. Levy 1989. The archaeology of cult and the Chalcolithic sanctuary at Gilat. Journal of Mediterranean Archaeology 2, 163–221. Amiran, R. 1981. Some observations on Chalcolithic and Early Bronze Age sanctuaries and religion. In A. Biran (ed.) Temples and High Places in Biblical Times, 47–53. Jerusalem, Hebrew Union College. Ben-Tor, A. 1977. Cult scenes on Early Bronze Age cylinder seal impressions from Palestine. Levant 9, 90–100. Callaway, J. A. 1972. The Early Bronze Age Sanctuary at cAi (etTell). London, Quaritch. Chesson, M. S. 1997. Urban Households in Early Bronze Age Communities of Syro-Palestine. Unpublished Ph.D. thesis, Harvard University. Chesson, M. S. 2003. Households, houses, neighborhoods and corporate villages: Modeling the Early Bronze Age as a house society. Journal of Mediterranean Archaeology 16, 79–102.

51

Chesson, M. S. and G. Philip 2003. Tales of the city? ‘Urbanism’ in the Early Bronze Age Levant from Mediterranean and Levantine perspectives. Journal of Mediterranean Archaeology 16, 3–16. Douglas, K. 2007. Die Befestigung der Unterstadt von Hirbet ez-Zeraqon (Deutsch-jordanische Ausgrabungen in Hirbet ezZeraqon 1984–1994. Endberichte Band III/1. Abhandlungen des Deutschen Palästina-Vereins 27(3). Wiesbaden, Harrassowitz. Esse, D. L. 1989. Secondary state formation and collapse in Early Bronze Age Palestine. In P. de Miroschedji (ed.) L’urbanisation de la Palestine à l’âge du Bronze ancien. Bilan et perspectives des recherches actuelles. Actes du Colloque d’Emmaüs (20–24 octobre 1986), 81–96. British Archaeological Reports International Series 527. Oxford, British Archaeological Reports. Esse, D. L. 1991. Subsistence, Trade and Social Change in Early Bronze Age Palestine. Studies in Ancient Oriental Civilizations 50. Chicago, Oriental Institute of the University of Chicago. Falconer, S. E. and S. Savage 1995. Heartlands and hinterlands: Alternative trajectories of early urbanization in Mesopotamia and the southern Levant. American Antiquity 60, 37–58. Flender, M. 2000. Cylinder seal impressed vessels of the Early Bronze Age III in northern Palestine. In G. Philip and D. Baird (eds) Ceramics and Change in the Early Bronze Age of the Southern Levant, 295–313. Levantine Archaeology 2. Sheffield, Sheffield Academic Press. Genz, H. 2000. The organization of Early Bronze Age metalworking in the southern Levant. Paléorient 26, 55–65. Genz, H. 2001. Early Bronze Age potmarks from Khirbat azZayraq n: Some aspects concerning their meaning. Studies in the History and Archaeology of Jordan 7, 217–228. Genz, H. 2002. Die frühbronzezeitliche Keramik von Hirbet ez-Zeraqon, Nordjordanien. Mit Studien zur Chronologie und funktionalen Deutung frühbronzezeitlicher Keramik in der südlichen Levante (Deutsch-jordanische Ausgrabungen in Hirbet ez-Zeraqon 1984–1994. Endberichte Band V. Abhandlungen des Deutschen Palästina-Vereins 27(2). Wiesbaden, Harrassowitz. Genz, H. 2003. Cash crop production and storage in the Early Bronze Age southern Levant. Journal of Mediterranean Archaeology 16, 59–78. Ilan, O. 2001. Household archaeology at Arad and Ai in the Early Bronze Age II. In S. R. Wolff (ed.) Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse, 317–354. Studies in Ancient Oriental Civilizations 59. Chicago, Oriental Institute of the University of Chicago. Joffe, A. H. 1993. Settlement and Society in the Early Bronze I and II Southern Levant: Complementarity and Contradiction in a Small-scale Complex Society. Monographs in Mediterranean Archaeology 4. Sheffield, Sheffield Academic Press. Kempinski, A. 1978. The Rise of an Urban Culture: The Urbanisation of Palestine in the Early Bronze Age 3000–2150 B.C. Israel Ethnographic Society Studies 4. Jerusalem, Israel Ethnographic Society. Kempinski, A. 1989. Megiddo: A City State and Royal Centre in North Israel. Materialien zur Allgemeinen und Vergleichenden Archäologie 40. Munich, Beck. Kempinski, A. 1992a. Chalcolithic and Early Bronze Age Temples. In A. Kempinski and R. Reich (eds) The Architecture of Ancient

52

Hermann Genz

Israel from the Prehistoric to the Persian Periods, 53–59. Jerusalem, Israel Exploration Society. Kempinski, A. 1992b. Fortifications, public buildings, and town planning in the Early Bronze Age. In A. Kempinski and R. Reich (eds) The Architecture of Ancient Israel from the Prehistoric to the Persian Periods, 68–80. Jerusalem, Israel Exploration Society. Kenyon, K. M. 1979. Archaeology in the Holy Land. 4th ed. London, Benn. Loud, G. 1948. Megiddo II, Seasons of 1935–39. Oriental Institute Publications 62. Chicago, University of Chicago Press. Mazar, A. 1990. Archaeology of the Land of the Bible 10,000–586 B.C.E. New York, Doubleday. Mazar, A. 2001. On the significance of the Early Bronze III granary building at Beit Yerah. In S. R. Wolff (ed.) Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse, 447–464. Studies in Ancient Oriental Civilizations 59. Chicago, Oriental Institute of the University of Chicago. Miroschedji, P. de 1988. Yarmouth 1. Rapport sur les trois premières campagnes de fouilles à Tel Yarmouth (Israel) (1980–1982). Paris, Éditions CNRS. Miroschedji, P. de 1989. Le processus d’urbanisation en Palestine au Bronze ancient: Chronologie et rythmes. In Miroschedji, P. de (ed.) L’urbanisation de la Palestine à l’âge du Bronze ancien. Bilan et perspectives des recherches actuelles. Actes du Colloque d’Emmaüs (20–24 octobre 1986), 63–79. British Archaeological Reports International Series 527. Oxford, British Archaeological Reports. Miroschedji, P. de 1993. Cult and religion in the Chalcolithic and Early Bronze Age. In A. Biran and J. Aviram (eds) Proceedings of the Second International Congress of Biblical Archaeology, 208–220. Jerusalem, Israel Exploration Society.

Miroschedji, P. de 1999. Yarmuth: The dawn of city states in southern Canaan. Near Eastern Archaeology 62, 2–19. Miroschedji, P. de 2001. Notes on the Early Bronze Age metrology and the birth of architecture in ancient Palestine. In S. R. Wolff (ed.) Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse, 465–491. Studies in Ancient Oriental Civilizations 59. Chicago, Oriental Institute of the University of Chicago. Mittmann, S. 1994. Hirbet ez-Zeraqon: Eine Stadt der frühen Bronzezeit in Nordjordanien. Archäologie in Deutschland 1994(2), 10–15. Peltenburg, E. 1999. The living and the ancestors: Early Bronze Age mortuary practices at Jerablus Tahtani. In G. del Olmo Lete and J.-L. Montero Fenollós (eds) Archaeology of the Upper Syrian Euphrates. The Tishreen Dam Area: Proceedings of the International Symposium held at Barcelona, January 28th – 30th, 1998, 427–442. Aula Orientalis Supplementa 15. Barcelona, Editorial Ausa. Philip, G. 2001. The Early Bronze I–III Ages. In B. MacDonald, R. Adams and P. Bienkowski (eds) The Archaeology of Jordan, 163–232. Levantine Archaeology 1. Sheffield, Sheffield Academic Press. Renfrew, C. 1985. The Archaeology of Cult: The Sanctuary at Phylakopi. London, Thames & Hudson. Tubb, J. N., P. G. Dorrell and F. J. Cobbing 1996. Interim report on the eighth (1995) season of excavations at Tell es-Sacidiyeh. Palestine Exploration Quarterly 128, 16–40. Tubb, J. N., P. G. Dorrell and F. J. Cobbing 1997. Interim report on the ninth season (1996) of excavations at Tell es-Sacidiyeh, Jordan. Palestine Exploration Quarterly 129, 54–77.

PART 2

EARLY URBAN COMMUNITIES AND THE EMERGENCE OF THE STATE

7 THE TELL: SOCIAL ARCHAEOLOGY AND TERRITORIAL SPACE Tony Wilkinson

Introduction The tell is the dominant feature of the archaeological landscape of the northern Fertile Crescent, and in recent years excavations have enabled archaeologists to make great strides in the understanding of the social archaeology of the tell. This paper is dedicated to Eddie Peltenburg, who among Near Eastern archaeologists has been a vigorous proponent of the need to extract social interpretations from archaeological data. Whereas the excavators of settlement and burial sites have progressed significantly in their quest for a new social archaeology, archaeological surveyors, on the other hand, have tended to focus upon the landscape and economic features associated with tells and related sites rather than on the social archaeology of their surrounding landscape. This paper therefore attempts to address this imbalance by examining the immediate landscape of tells in the northern Fertile Crescent, bringing together field evidence, ethnography and texts to distil some of the common factors in the social archaeology of the tell. One feature that has emerged from archaeological surveys in recent years is that settlement patterns in the Near East fluctuated through time between nucleated and dispersed modes (Wilkinson 1995). In a paper published in 2000, Edgar Peltenburg demonstrated that the tell, as a nucleated community, was a widespread feature across the Near East and eastern Mediterranean, and that alternations between nucleated tell-based communities and dispersed settlement were important factors in understanding cycles of settlement over long spans of time. Before embarking on this narrative, it is necessary to caution that tells were hardly likely to be socially or functionally uniform. As Postgate (1982, 310) has argued for Assyrian rural society, one should be careful about imposing a “false homogeneity” on ancient communities.

Tells and their landscapes were probably different from region to region, even within the same area. Whereas some will have been dominated by palaces and temples, others would mainly have comprised domestic houses, storage, or cult structures. For example, in the region of Tell Beydar the Beydar cuneiform texts implicitly assess the agricultural potential of the subordinate tells by referring to the numbers of draft animals (Widell 2004), but they also highlight that one of those sites, Sulum, was a centre of religion (Sallaberger and Ur 2004). Although it is not known which site corresponds to ancient Sulum, all tells in the Tell Beydar complex appear at least superficially similar, and from surface inspection it is not evident which could be a religious centre and which were devoted to agricultural production. Similarly, the site of Gre Virike in the Turkish middle Euphrates, although in appearance a conventional tell, also probably functioned as some form of cultic place near Carchemish (Ökse 2007, 100). This paper illustrates how land was probably configured around typical small to medium sized tells in Upper Mesopotamia, and how the inferred land tenure provides clues concerning the social structure of the community that occupied the tell. The essay also draws together threads from earlier arguments that point to supposedly ‘ancient’ forms of land holding, and attempts to bring these in alignment with results derived from ethnographic studies and archaeological surveys of the landscape of the Fertile Crescent. Fundamental to this discussion is land tenure. Many scholars have used cuneiform texts to analyse land tenure to describe the shapes of fields, as well as to show how land was allocated for private holdings, estates or communally organised land. Less effort, however, has been spent on analysing these topics within an overall landscape framework. Although such an ambitious treatment is beyond

56

Tony Wilkinson

the scope of this paper, preliminary analysis suggests how the archaeological landscape, land tenure and the social archaeology of the settlements themselves can be viewed within one overall framework. Land tenure is important to this narrative because of the role it plays in linking together field evidence, cuneiform texts and social archaeology, and also because land-use practices have the capacity to mediate between social groups and the environment. Initially, this paper summarises the layout of a typical tell-type settlement in the northern Fertile Crescent and outlines the field evidence for the surrounding agricultural landscape. This is followed by a discussion of some key features of land tenure interpreted from cuneiform texts. Finally, the ethnographic evidence from traditional agricultural communities in the Levant is harnessed to outline equivalent features or processes and to suggest how these flesh out the interpretations of the landscape of Bronze Age Mesopotamia. The perspective is from the outside looking in: no attempt is made to tackle the abundant evidence of tell communities derived from excavations.

Field Evidence The tell is one of the most familiar and conspicuous features of the Near Eastern landscape. In Upper Mesopotamia, many tells attained their maximum size and bulk in the 3rd millennium BC (Early Bronze Age). Most are of relatively limited size, being usually 1–10 ha in area, although some Upper Mesopotamian tells attain 50 ha in area with surrounding walled quarters that take the overall settlement up to some 120 or even 160 ha (Stein 2004, 67). Tells are usually topographically constrained, and even when they do have lower or outer towns, only in exceptional circumstances do small dispersed settlements occur immediately beyond the limits of the tell. One such unusual case is found at the site of Tell Brak where an extensive outer settlement as well as numerous small outliers of Late Chalcolithic 3–4 (3900–3400 cal BC) settlement occur a short distance beyond the limits of the main tell (Ur et al. 2007). Intensive survey away from many tells has demonstrated that although extensive pottery scatters may occur, recognisable sites are scarce within what appear to have been the fields surrounding the tell. Again, exceptions can be cited, specifically the small early 3rd millennium site of Hajji Ibrahim, which appears to have been a small fortified granary in the vicinity of Tell Sweyhat along the Syrian Euphrates (Danti 1997, 89–92). That many archaeological mounds were genuinely nucleated settlements is hinted at by textual evidence from Babylonia indicating that houses tended to be located adjacent to houses and fields were adjacent to fields (Leemans 1982, 248; but see Steinkeller 2006 for an alternative view of the region of Umma). For the Amuq plain, Magness-Gardiner

Fig. 7.1 The archaeological landscape to the NE of Tell Beydar showing the inferred cultivation, the limits of that cultivation (inferred from the fade-out zone of hollow ways) and what were probably the common grazing lands in between (redrawn from original unpublished map by Jason Ur).

similarly concludes that the population lived within nucleated settlements and walked out to their fields (1994, 40). The significance of this will be discussed below. In those parts of Upper Mesopotamia where landscape features are well preserved, settlement territories can be inferred from the configuration of hollow ways that radiated from the central tell. Such features, which appear to have functioned both to gain access to the fields and as droveways along which the flocks and draft animals moved, often fade out some 3–5 km from the central tell, thereby implying the existence of an approximate boundary to the cultivated land (Fig. 7.1; Wilkinson 1994; Ur 2003). Moreover, the distinctive radial pattern of such hollow ways demonstrates that the main locus of movement was out from the settlement towards the fields, outlying pastures, and in some cases to the neighbouring settlements and centres. Unfortunately, such hollow ways are only preserved in the less transformed landscapes of the Khabur basin and neighbouring parts of north-western Iraq. On the other hand, in western Syria and southern Turkey, intensive surveys demonstrate that the main area around the tell was apparently devoid of settlements (Philip et al. 2005; Casana 2007). Consequently, it is reasonable to infer that such areas within a walking distance of the tell (some 2–3 km) were probably also given over to fields. In favourable circumstances, intensive field survey has

7. The Tell: Social archaeology and territorial space also demonstrated that some major tells were surrounded by low-density scatters of potsherds that are inferred to have resulted to a significant degree from the spreading of midden material and domestic refuse on the fields as fertilizer. Such ‘field scatters’ are by no means universal, however, and many tells show little evidence for them. For example, sample transects laid out around sites in the Amuq plain (Turkey) and in the Homs region (Syria) show negligible off-site scatters of the Bronze Age, whereas those around Tell Hammam et-Turkmen in the Balikh and Kurban Höyük in Turkey show only very sparse scatters of Early Bronze Age ceramics (Casana and Wilkinson 2005; Philip et al. 2005; Wilkinson 1990; 1998). In these cases, if off-site scatters are present in significant amounts, they consist of later pottery, mainly of Hellenistic, Roman, or Late Roman/Byzantine date. Nevertheless, where field scatters are of Early Bronze Age date, they demonstrably focus upon the central tell from which they were presumably spread (Wilkinson 1989). Finally, estimates of the populations that could be supported by the inferred field areas suggest that where hollow ways and textual records occur in the same area, the estimated population of the central tells could be approximately supported by the inferred field catchments, as well as by the number of draft animals recorded on the texts (Wilkinson et al. 2007; Ur and Wilkinson 2008). However, the larger settlements, such as Tell Beydar and Brak, were surrounded by relatively smaller cultivated areas in proportion to their site area than the smaller tells, and these smaller tells were capable of generating surplus agricultural production. This suggests that the smaller tells produced a surplus of cereals that went to the main centres to sustain the larger populations and public institutions, either as tax or some equivalent measure of surplus. The Beydar texts suggest that the settlements in the region of Beydar/Nabada were assessed in the form of the number of plough animals from each site (Widell 2004, 721–723; Ur and Wilkinson 2008, 313–314) and that this may have related to the agricultural capacity of each village, as discussed below. In sum, the landscape evidence suggests that tells were nucleated communities that were surrounded by a definable area of cultivated fields accessed by radial pathways from the settlement with un-bounded lands (presumably pasture and scrub woodland) beyond. Such fields were sufficient to supply the inhabitants of the central communities with staple foods, and the entire assemblage – tell, outlying fields, and pasture – went together as a single, and perhaps indivisible, unit of the landscape.

Historical Evidence Unfortunately, cuneiform texts are not explicit in describing how the land associated with the smaller rural communities in rain-fed northern Mesopotamia was used. As stated by

57

both Adams (1982, 11) and Diakonoff (1975, 131), most texts relate to the state and temple sectors of the urban centres of power rather than to the rural communities that formed the matrix of Upper Mesopotamian society. In terms of land tenure Postgate has reminded us that: “…clearly most existing entitlement to land would be unwritten, though fixed by custom, and the need for documentary proof probably crept in with the gradual encroachment of urban landlords” (1982, 311–312). In other words, the societies that are often studied by archaeologists are not explicitly discussed in many texts. A similar silence may be observed in documentation from Ottoman Syria indicating that when land was re-distributed communally, peasants held no written title to their shares (Admiralty 1944, 266). When lands are discussed in the texts, as in the Nuzi texts, the discussion may be couched in the form of unusual practices, such as adoption, that are seemingly developed perhaps to avoid customary law (Zaccagnini 1999). Nevertheless, many archaeologists and ancient historians refer back, often rather vaguely, to an ‘early’ phase of rural settlement in which the village communities were organised along communal lines. For Lamberg-Karlovsky (1999, 179–180) this was during the ceramic Neolithic, that is the Hassuna and Samarra periods, when there is archaeological evidence for communal storage facilities. Steinkeller (1999), mainly discussing southern Mesopotamia, maintains that this phase continued until the Late Uruk period, whereas Diakonoff (1975, 132) sees such arrangements as continuing until at least the Old Babylonian period, ending by the early 1st millennium BC when Neo-Assyrian administrative practices resulted in a major transformation of land holding. Finally, Fales (1984, 8) argues that such a communal phase of village land holding continued until the last two centuries of the Neo-Assyrian empire. Referring to this broad span of time, Adams states: “Hence it serves here merely to underline that communal forms of de facto tenure, as distinguished from whatever forms of titular ownership of land may be formally recognised by law, are surely a very ancient, complex and widely occurring institution. They deserve to be taken fully into account in any reconstruction of early Mesopotamian agriculture” (1982, 10). The practice of communal ownership by the village community then begs the question as to who was the ultimate owner of the land. Whereas Diakonoff (1975) and Jankowska (1969) consider the lands of villages in the region of Assur and Nuzi as forming the communal property of the “village commune”, Postgate regards the “ultimate owner” as the crown or the Assyrian palace (1982, 310). Fortunately, hints concerning the nature of village land holdings come from Bronze Age texts excavated from Alalakh, Ebla and other sites in northern Syria and southern Turkey. Such texts refer to or hint that villages were transferred between rulers together with their population, their surrounding lands and the revenues derived from

58

Tony Wilkinson

Fig. 7.2 The village of Bar Elias in Lebanon, an example of a village that had been under musha’ cultivation in the 19th century AD (after Weulersse 1946, fig. 37).

them (Wiseman 1953, 47–48; Zaccagnini 1989; MagnessGardiner 1994, 40; Schloen 2001, 306–307). This practice was also attested around Early Bronze Age Ebla where land grants included entire villages and their populations. Nevertheless, ultimate ownership appears to have been in the hands of the king and his family within what may be termed a patrimonial state (Steinkeller 1999, 300; Schloen 2001). In other words, these ancient rural communities can be regarded as forming one unit that could be exchanged at will, without it being necessary to negotiate over details of land ownership. Although a compelling case can be made for equating the rural communities referred to in the Alalakh texts with those composite landscape units of tell, tracks, fields and surrounding pastures evident in the Khabur basin, it is

necessary to examine how such land holdings may have been operated in practice.

Traditional Patterns of Land Holding and the Community In the early 1980s significant attention was paid to the possibility that communal forms of land-holding, similar to those of open field agriculture in Europe, were also characteristic of rural communities in the Near East (Adams 1982, 8). Such systems, known as musha’, musha‘ or mesha’a, were a common form of agriculture in the Levant in the 19th century and were considered by some scholars to have stretched back to ‘ancient times’ (Warriner 1948, 20;

7. The Tell: Social archaeology and territorial space Granott 1952, 213; Seeden and Kaddour 1984). In musha’ agriculture, the cultivated lands of the village (or most of them) were held by the village community, that is by a group of elders, the sheikh, the heads of the main households, clans or even all the males of the village. The area of land was frequently assessed in terms of the faddan, the amount of land that could be ploughed in one day by a pair of draft animals (usually oxen). A particularly common mode of land distribution was for the individual parcels of land to be redistributed among the households by the casting of lots. According to Granott (1952, 229), although initially lots were drawn every year, they may later have been apportioned every two years and as much as nine year intervals, with the result that in some cases the interval was increased to permanent use. Without delving further into the rather extensive literature on musha’ agriculture, it is necessary to emphasise that because the land was held in common, it was not only very difficult to alienate or sell individual parcels of land to outsiders, but equally the tax burden from the land could be shared among the shareholders so that the entire village and its territory could be assessed as a single unit. Fortunately, maps of musha’ field systems, or post-musha’ landscapes that have been ‘frozen in time’ (Weulersse 1946) have been published, so that it is possible to view their basic elements. For example, the village of Bar Elias in Lebanon shows a tightly clustered nucleated village (with its tell) surrounded by a highly fragmented landscape of small field parcels, which were exclusively devoted to cultivation and were bereft of habitations or other structures (Fig. 7.2; Weulersse 1946, fig. 37). This lack of outlying settlement is congruent with the land tenure. This is because in musha’ systems individual parcels of land are rotated regularly, with the result that no individual ‘owns’ the land or can therefore settle on it. In addition to the dense matrix of fields, one or two radial features led away from the village of Bar Elias, and at the time of mapping were infilled with small field parcels (Fig. 7.2). Before they became infilled, these features appear to have originally been broad radial tracks similar to the hollow ways or drove-ways that radiate from Upper Mesopotamian tells. Although the regular reallocation of land, as well as ease of access to fields, provides a compelling reason for the clustering of settlements, it would be overly simplistic to ascribe settlement nucleation to just one factor, namely land tenure. Rather, geographers frequently posit a number of inter-related factors as contributing to settlement nucleation. According to B. K. Roberts (1996, 35–37) there are three social conditions of nucleation: 1) Communality of assent, namely those social aspects of life that stem from ties of kinship, whether assumed or created; often these are represented by the patrimonial household as recognised by Weber (1978) and more recently Schloen (2001). 2) Communality to economise, perhaps over-simplified as

59

‘many hands make light work’; in other words, there are tangible benefits to combining labour for certain tasks or for sharing labour-saving devices such as plough teams. 3) Communality of enforcement, which includes defence and the requirement of having an overall strong figure, such as a sheikh, a chief or the lord of the manor, as head of the community. In terms of the community of the tell, the inhabitants may have been linked by common ancestry, as well as being bound together by collaboration for the common good. Defence would have been in the hands of either a single figure, the head of the clan, or a group of elders, who were representative of the kinship or social groups of the village. Such communities would not only be defended by the community and its head, but would also be physically bounded by its wall, which would not only provide defence and a clear delimitation as to the limits of the community, but would also serve to constrain the sediments of redundant buildings, thereby maintaining the distinctive morphology of the tell. In addition to such constraints, monumentality would have been contributed by the house of the chief or local ruler, public buildings, and storage structures, the entire site being rendered immobile perhaps by the sanctity of cult buildings and/or water sources. At a general level, the agricultural landscape of a musha’ community approximates to that of a nucleated village in western Europe, as well as to reconstructions of the agricultural landscape surrounding the Bronze Age tell (compare Roberts 1996, fig. 2.5, with reconstructions in Wilkinson 1994 and 2003). Although one must be wary of uncritically extrapolating ethnographic practices back into the past, many studies of musha’ systems provide valuable hints as to how society may have functioned within a system of collective land management. First of all, it is evident that musha’ as a collective way of land distribution is not simply an economic system. Thus Fischbach (2000, 201) maintains that, “Especially in musha’ villages, property ownership was integrally linked with social relations. Persons owned abstract shares in the total village land, shares that had been originally determined as part of a collective village decision. The terms of reference for landed property were thus socially constructed.” In traditional collective villages in Egypt, not only did everyone know everyone else, but also there was a tendency for members of the community to create solidarity by working for a common interest (Gerber 1987, 144, citing the work of Ayrout). Not only did musha’ often function within a patriarchal social system, but it has also been seen as a way of preserving the integrity of the village community (Granott 1952, 215). The musha’ village was specifically structured so that it was a self-contained unit; it was difficult for non-villagers to acquire village land or to penetrate it socially (Schaebler 2000, 289). Marriage to someone within the village was also preferable to someone from without (Gerber 1987, 144), and

60

Tony Wilkinson

families combined into larger groups termed hamula (clan). This was conceived as a form of mutual protection, within which there was a desire to keep the family property intact (Granott 1952, 216). Such communities epitomised Roberts’ concepts of communality because “They lived together, they worked together, and when the need arose they were comrades in self defence”….“The chief force which held the community together and the shield of its integrity was the common ownership of the soil….” (Granott 1952, 216). Moreover, “Its essence was that social groups of different sizes held their land as an indivisible possession” (Granott 1952: 215). Such systems hint that Bronze Age rural settlements could also be interpreted as indivisible possessions that could therefore be exchanged together with their population and immediate territories. The above can be used to provide a linking argument, namely that communally organised landtenure systems represent social systems that were nucleated and indivisibly bound to their surrounding territory.

Conclusions The above narrative has discussed the role of the tell as a nucleated community in which collective action probably played a significant role in the agricultural systems around the tell. This echoes conclusions drawn by Peltenburg in his consideration of the mobilisation of labour in the erection of fortified structures along the Euphrates. These can be interpreted not just as service facilities: “Their construction, involving pooled labour in a joint enterprise that also was a unifying end in itself, exhibits settlement nucleation as a means to integration and creation of new identities” (Peltenburg 2007, 14). In some studies there has been a tendency for the communal village system to be perceived as a form of overly romanticised ethnographic analogy. For example, Weulersse described these musha’ type systems as “extremely robust, healthy and balanced (social) and economic organizations” (Seeden and Kaddour 1984, 504). On the other hand, such systems were prone to disputes over the allocation of shares, which led to local conflicts. In addition, depending upon the way the musha’ was administered, practices of divided inheritance could result in the land parcels becoming progressively reduced in size. However, this criticism could be levelled at private land holdings as well. Musha’ communities have also been regarded as uneconomical in their use of land because they tended to practice the system of biennial fallow, which although beneficial as a system for conserving soil moisture, left half the land fallow every year. Moreover, collective re-distribution of the land meant that there was no incentive to improve the soil because others would “reap the fruits of his initiative” (Granott 1952, 218). In other words, under collective ownership, the

share-holders were unlikely to either intensify agricultural production or invest in the land. Hence manuring, as well as the construction of field enclosure walls or buildings, was discouraged (Granott 1952, 218, 246; Fischbach 2000, 85). Because such features provide a physical signature, they are significant archaeologically, and are potentially recognisable in the landscape. In other words, when we start to see archaeological signatures of enclosure walls, buildings and manuring, it is likely that the landscape had switched from some form of communal administration to private or institutional land holdings that allowed investment to take place. As observed above, the area immediately surrounding many tells in upper Mesopotamia and the northern Levant is relatively bereft of archaeological sites or features. Field boundaries are not in evidence, nor are settlements or individual buildings very common. However, the latter do occur, and when they are recognised, as in the area of Tell Sweyhat, their presence may well be significant in terms of the pattern of land holding. As observed above, field scatters, which frequently result from ancient fertilisation and are indicators of investment in the land, are present around certain sites but are not present around all Bronze Age tells. The absence or presence of field scatters might therefore result from the practice of collective re-distribution of land (no field scatters) or a switch to institutional or private land holding (field scatters). Unfortunately, because field scatters will only register when settlement refuse or midden material is scattered upon fields, such an elementary equation is not entirely valid. Midden material would only have been used as fertilizer when the community was without animal dung as a result of the dung being burned as fuel because all the scrub or woodland in the region had been removed. Consequently, there are both palaeoecological and socioeconomic reasons behind the appearance of field scatters. Nevertheless, a compelling case can be made for viewing the smaller tells, specifically those that are without field scatters, as being representative of communities that practiced a form of communal agriculture. This would accord with the observation made by Adams (1982, 8) that in the Ottoman period musha’ villages occurred in transitional areas between fully sedentary intensively cultivated areas around the larger towns and regions further to the east that were given over to pastoral nomads. Thus Tell al-Hawa and Hamoukar in north-western Iraq and north-eastern Syria respectively, where 3rd millennium field scatters are dense and extensive (Wilkinson 1994; Ur 2002), could be viewed as the larger towns around which agriculture was intensive (and under the administration of institutions, estates or even given over to private land holdings). Conversely, smaller tells, such as Tell Judeidah, Kurban Höyük and others in the Homs and Tell Beydar areas around which field scatters are rare or absent, might represent a form of collective, lowintensity agriculture equivalent to the musha’ system.

7. The Tell: Social archaeology and territorial space Such communities may not have lived in an agricultural paradise, however, because it is likely that they were subject to taxation. As indicated in the Beydar texts, the presence of lists of draft animals suggests that some form of census was required. In the case of the Beydar area, the neighbouring smaller tells were apparently producing a surplus cereal production, which may have been a ‘tax’ that went to the main town or city, in this case Beydar and Brak, Nabada and Nagar respectively (Wilkinson et al. 2007; Ur and Wilkinson 2008). Whether indicative of taxation or not, the assessment of 19th and early 20th century land holdings as faddans (i.e. the amount of land that can be ploughed by a pair of oxen in one day) is indicative of the strength or wealth of the community. As Schaebler has stated, “The faddan was therefore not a metric measurement or, strictly speaking, even a standard indication of surface area; it was rather a social unit of measurement testifying to the productive strength or wealth of village or individual” (2000, 258); and “Land in musha‘ communities is clearly more than a means of production. It is rather the very expression of the community. A village is expressed in a given number of faddans” (2000, 288). In other words, the counts of plough animals in the Beydar texts, rather than relating to the direct administration of these subordinate places, might equally be an assessment of the productive strength of the outlying villages, or for taxation. In sum, evidence from landscape surveys, historical texts and ethnography converge on the conclusion that the tells and the fields that surrounded them formed an indivisible social and landscape unit. Moreover, the tell not only represented a nucleated community, but one that was a strong, probably patrimonial social group, which exercised some form of collective redistribution of land. The land was not simply an economic resource, but it was viewed as a collective good and was part of the social processes of everyday life for a community that clustered together for its own perceived common advantage.

Acknowledgements This paper is dedicated to Eddie Peltenburg, whose research has contributed significantly to my own thoughts on the development of the social archaeology of the ancient Near East. This paper also draws on research from the Modeling Ancient Settlement Systems (MASS) project, and I specifically wish to thank McGuire Gibson, David Schloen, Tate Paulette, Jesse Casana and Jason Ur for various discussions relating to this topic over the past ten years.

References Adams, R. McC. 1982. Property rights and functional tenure

61

in Mesopotamian rural communities. In J. N. Postgate, I. Dandamayev, H. Gershevitch, G. Klengel, G. Komoróczy and M. T. Larsen (eds) Societies and Languages of the Ancient Near East. Studies in Honour of I.M. Diakonoff, 1–14. Warminster, Aris & Phillips. Admiralty. 1944. Syria. British Admiralty Geographical Handbook Series, BR 513. British Naval Intelligence Division. Casana, J. J. 2007. Structural transformations in settlement systems of the northern Levant. American Journal of Archaeology 112, 195–222. Casana, J. and T. J. Wilkinson 2005. Settlement and landscapes in the Amuq Region. In K. Aslihan Yener (ed.) The Archaeology of the Amuq Plain, 25–65, 203–280. Oriental Institute Publications 131. Chicago, Oriental Institute of the University of Chicago. Danti, M. 1997. Regional surveys and excavations. In R. L. Zettler (ed.) Subsistence and Settlement in a Marginal Environment: Tell es-Sweyhat, 1989–1995 Preliminary Report, 85–94. Philadelphia, Museum Applied Science Center for Archaeology (MASCA), University of Pennsylvania Museum of Archaeology and Anthropology. Diakonoff, I. M. 1975. The rural community in the ancient Near East. Journal of Economic and Social History of the Orient 18(2), 121–133. Fales, M. 1984. A survey of Neo-Assyrian land sales. In T. Khalidi (ed.) Land Tenure and Social Transformation in the Middle East, 1–13. Beirut, American University of Beirut Press. Fischbach, M. R. 2000. State, Society and Land in Jordan. Leiden, Brill. Gerber, H. 1987. The Social Origins of the Modern Middle East. Boulder (CO), Lynne Rienner. Granott, A. 1952. The Land System of Palestine in History and Structure. London, Eyre and Spottiswood. Jankowska, N. B. 1969. Communal self-government and the king of the state of Arrapha. Journal of the Economic and Social History of the Orient 12, 233–82. Lamberg-Karlovsky, C. C. 1999. Households, land tenure and communication systems in the 6th–4th millennia of Greater Mesopotamia. In M. Hudson and B. A. Levine (eds) Urbanization and Land Ownership in the Ancient Near East, 167–201. Peabody Museum Bulletin 7. Cambridge (MA), Peabody Museum of Archaeology and Ethnology, Harvard University. Leemans, W. F. 1982. The pattern of settlement in the Babylonian countryside. In J. N. Postgate, I. Dandamayev, H. Gershevitch, H. Klengel, G. Komoróczy and M. T. Larsen (eds) Societies and Languages of the Ancient Near East: Studies in Honour of I. M. Diakonoff, 246–249. Warminster, Aris and Phillips. Magness-Gardiner, B. 1994. Urban-rural relations in Bronze Age Syria: Evidence from Alalah Level VII palace archives. In G. M. Schwartz and S. E. Falconer (eds) Archaeological Views from the Countryside: Village Communities in Early Complex Societies, 37–47. Washington DC, Smithsonian Institution Press. Ökse, A. Tuba 2007. A ‘high’ terrace at Gre Virike to the north of Carchemish: A power of local rulers as founders? In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 94–104. Levant Supplementary Series 5. Oxford, Oxbow.

62

Tony Wilkinson

Peltenburg, E. 2000. From nucleation to dispersal: Late third millennium BC settlement pattern transformations in the Near East and Aegean. In O. Rouault and M. Wäfler (eds) La Djéziré et l’Euphrate syriens: de la protohistoire à la fin du IIe millénaire av. J.-C.: tendances dans l’interprétation historique des données nouvelles, 183–206. Subartu 7. Turnhout, Brepols. Peltenburg, E. 2007. New perspectives on the Carchemish sector of the Middle Euphrates River Valley in the 3rd millennium BC. In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 3–24. Levant Supplementary Series 5. Oxford, Oxbow. Philip, G., M. Abdulkarim, P. Newson, A. Beck, D. Bridgland, M. Bshesh, A. Shaw, R. Westaway and K. Wilkinson 2005. Settlement and landscape development in the Homs region, Syria: Report on work undertaken during 2001–2003. Levant 17, 21–42. Postgate, J. N. 1982. Ilku and land tenure in the Middle Assyrian kingdom – a second attempt. In J. N. Postgate, I. Dandamayev, H. Gershevitch, H. Klengel, G. Komoróczy, and M. T. Larsen (eds) Societies and Languages of the Ancient Near East. Studies in Honour of I. M. Diakonoff, 304–313. Warminster, Aris & Phillips. Roberts, B. K. 1996. Landscapes of Settlement: Prehistory to the Present. London, Routledge. Sallaberger, W. and J. Ur. 2004. Tell Beydar/Nabada in its regional setting. In L. Milano, W. Sallaberger, P. Talon and K. Vanlerberghe (eds) Third Millennium Cuneiform Texts from Tell Beydar (Seasons 1996–2002), 51–71. Subartu 12. Turnhout, Brepols. Schaebler, B. 2000. Practicing musha‘: Common lands and the common good in southern Syria under the Ottomans and the French. In R. Owen (ed.) New Perspectives on Property and Land in the Middle East, 241–309. Cambridge (MA), Harvard University Press. Schloen, J. D. 2001. The House of the Father as Fact and Symbol: Patrimonialism in Ugarit and the Ancient Near East. Cambridge (MA), Harvard Semitic Museum Studies in the Archaeology and History of the Levant. Seeden, H. and M. Kaddour 1984. Space, structures and land in Shams ed D n Tann ra on the Euphrates: An ethnoarchaeological perspective. In T. Khalidi (ed.) Land Tenure and Social Transformation in the Middle East, 495–526. Beirut, American University of Beirut Press. Stein, G. 2004. Structural parameters and sociocultural factors in the economic organization of north Mesopotamian urbanism in the third millennium BC. In G. M. Feinman and L. M. Nichols (eds) Archaeological Perspectives on Political Economies, 61–78. Salt Lake City, University of Utah Press. Steinkeller, P. 1999. Land-tenure conditions in third-millennium Babylonia: The problem of regional variation. In M. Hudson and B. A. Levine (eds) Urbanization and Land Ownership in the Ancient Near East, 289–329. Peabody Museum Bulletin 7. Cambridge (MA), Peabody Museum of Archaeology and Ethnology, Harvard University. Steinkeller, P. 2006. City and countryside in third millennium BC Babylonia. In E. Stone (ed.) Settlement and Society: Essays Dedicated to Robert McCormick Adams, 185–211. Los Angeles, University of California, Cotsen Institute of Archaeology. Ur, J. A. 2002. Settlement and landscape in northern Mesopotamia:

The Tell Hamoukar survey 2000–2001. Akkadica 123, 57– 88. Ur, J. A. 2003. CORONA satellite photography and ancient road networks: A northern Mesopotamian case study. Antiquity 77, 102–115. Ur, J. A., P. Karsgaard and J. Oates 2007. Early urban development in the Near East. Science 317, 1188. Ur, J. A. and T. J. Wilkinson 2008. Settlement and economic landscapes of Tell Beydar and its hinterland. Subartu 21, 305–27. Tell Beydar Studies 1. Turnhout, Brepols. Warriner, D. 1948. Land and Poverty in the Middle East. London and New York, Royal Institute of International Affairs. Weber, M. 1978. Economy and Society: An Outline of Interpretive Sociology (eds G. Roth and C. Wittich). Translated from the German by E. Fischoff et al. Berkeley and London, University of California Press. Weulersse, J. 1946. Paysans de Syrie et du Proche Orient. Paris, Gallimard. Widell, M. 2004. Some observations on the administration, agriculture and land management of Tell Beydar. UgaritForschungen 35, 717–733. Wilkinson, T. J. 1989. Extensive sherd scatters and land use intensity: Some recent results. Journal of Field Archaeology 16(1), 31–46. Wilkinson, T. J. 1990. Town and Country in Southeastern Anatolia Vol.1: Settlement and Land Use at Kurban Höyük and Other Sites in the Lower Karababa Basin. Oriental Institute Publications 109. Chicago, Oriental Institute of the University of Chicago. Wilkinson, T. J. 1994. The structure and dynamics of dry farming states in upper Mesopotamia. Current Anthropology 35(1), 483–520. Wilkinson, T. J. 1995. Late Assyrian settlement geography in upper Mesopotamia. In Mario Liverani (ed.) Neo-Assyrian Geography, 139–159. Quaderni di geografia storica 5. Rome, Università di Roma ‘La Sapienza’. Wilkinson, T. J. 1998. Water and human settlement in the Balikh Valley, Syria: Investigations from 1992–1995. Journal of Field Archaeology 25, 63–87. Wilkinson, T. J. 2003. Archaeological Landscapes of the Near East. Tucson, University of Arizona Press. Wilkinson, T. J., J. H. Christiansen, J. Ur, M. Widell and M. Altaweel 2007. Urbanization within a dynamic environment: Modeling Bronze Age communities in upper Mesopotamia. American Anthropologist 109(1), 52–68. Wiseman, D. J. 1953. The Alalakh Tablets. Occasional Publications of the British Institute of Archaeology at Ankara 2. London, British Institute of Archaeology at Ankara. Zaccagnini, C. 1989. Asiatic mode of production and ancient Near East: Notes toward a discussion. In C. Zaccagnini (ed.) Production and Consumption in the Ancient Near East, 1–126. Budapest, University of Budapest. Zaccagnini, C. 1999. Economic aspects of land ownership and land use in Northern Mesopotamia and Syria from the late third millennium to the Neo-Assyrian period. In M. Hudson and B. A. Levine (eds) Urbanization and Land Ownership in the Ancient Near East, 331–352. Peabody Museum Bulletin 7. Cambridge (MA), Peabody Museum of Archaeology and Ethnology, Harvard University.

8 RETHINKING KALOPSIDHA: FROM SPECIALISATION TO STATE MARGINALISATION Lindy Crewe

Introduction This paper is offered to Eddie Peltenburg in thanks for all his generous support and valued advice over the years as doctoral supervisor and beyond. The title is a reference to Eddie’s paper, “From isolation to state formation in Cyprus, c. 3500–1500 BC” (1996), which was my entry point into the complexities of Late Cypriot social complexity. This paper examined the earliest manifestations of ‘statehood’ on Cyprus at the Middle–Late Cypriot transition, centred on the evidence from Enkomi and the construction of a series of apparently contemporary ‘fortifications’ located at key points around the island. These structures, and the majority of other sites yielding evidence for the beginning of the Late Cypriot period, remain enigmatic and inaccessible to archaeologists. Since the 1920s when a house of Bronze Age date was excavated and published by Einar Gjerstad at Kalopsidha (1926), the site has played a pivotal role in reconstructing Bronze Age Cypriot society. Whether this is truly due to the site’s exceptional nature or through a mere accident of excavation is a question that is unlikely to be resolved in the near future, but by asking different questions of the data and re-examining the material more recent research continues to allow new interpretations (e.g. Frankel 1974; Barlow 1985; Pilides 1996; Webb 1999; Maguire 2009). Further excavations at Kalopsidha in 1959 by Paul Åström (1966) increased the site’s status as a leading early participant in relations with the surrounding eastern Mediterranean region and provided additional evidence for the abandonment of at least part of the settlement contemporary with the foundation of a coastal trading emporium at nearby Enkomi during the Middle–Late Cypriot transition (MC III–LC IA, c. 1750–1550 cal BC). Kalopsidha’s decline from “the old capital of the east Mesaoria” (Catling 1973, 168) has been

explained as either a destruction of the site by people who then settled at Enkomi or a relocation by the inhabitants to Enkomi (Åström 1966, 140). Through the later twentieth century and beyond, excavations in the Levant and Egypt (recently summarised in Maguire 2009) have continued to produce evidence that the majority of the earliest pottery exports found outside of Cyprus during the mainland Middle Bronze horizon originated in the east and can be tied particularly to styles found at Kalopsidha and apparently produced there (Åström 1966, 8). Due to the limited resolution and scope of the data, discussed below, examination of the Kalopsidha evidence leaves us with more questions than answers. Indeed, why Kalopsidha? If copper is considered to be central to Cypriot social transformations (Muhly 1986; Knapp 1988 and Peltenburg 1996 amongst others) why does a site located well inland and far from the copper sources play such a leading role? Did it in fact play the primary role that pottery styles found abroad would seem to indicate? My aim here is not to speculate on how Kalopsidha gained the special position it apparently occupied by the end of the Middle Cypriot (henceforth MC) period but to consider how the waning fortunes of the site can inform on the wider picture of social transformation on Cyprus. I would like to reconsider the material culture evidence, particularly the styles, technologies and types of pottery that feature strongly at the site. One of the most intriguing features of the Kalopsidha assemblage is the apparent contradiction between the precocious adoption of a swathe of both foreign and locally inspired innovations in tandem with a lasting technological and stylistic conservatism in production of the distinctive White Painted pottery for which the site is renowned.

Lindy Crewe

64

Settlement

Korovia-Nitovikla

Fortification

Phlamoudhi-Vounari

MESAORIA

PLAIN

Enkomi Ayios Sozomenos

Athienou

Kalopsidha

Dhali-Kafkallia

Kissonerga -Skalia 0 N

Fig. 8.1 Map of Cyprus showing sites mentioned in the text.

Fig. 8.2 Aerial view of Kalopsidha showing locations of the excavated areas (image: Google Earth).

50km

8. Rethinking Kalopsidha: From specialisation to state marginalisation

65

The Evidence from Archaeological Investigations at Kalopsidha Archaeological evidence from the area of Kalopsidha is dispersed discontinuously over a 150 ha area, located in the fertile alluvial Mesaoria Plain in southeast Cyprus, 13 km from the coast and 11 km southwest of Enkomi (Fig. 8.1). Tombs located in four discrete cemeteries (Sites A, B, D and E) and the remains of a settlement (Site C) dating from EC I–MC III were first investigated by J. L. Myres in 1894 (Myres 1897) (Fig. 8.2). A house occupied from MC III–LC I (and overlying an earlier EC III–MC II structure) at Site C and a trial trench at the locality Koufos (Trenches 8–9 on Fig. 8.2) were excavated by Gjerstad in 1924 (Gjerstad 1926, 27–37). Further excavations (Trenches 1–9) were carried out in 1959 by Paul Åström, and the subsequent publication summarises both his and earlier excavations (1966). Despite Åström’s attempts to deal fully with the material from earlier excavations, the majority of the Kalopsidha tombs were poorly excavated and incompletely described (Myres 1897), and the material is now lost. Keswani (2005, 380, 392) suggests that the reported horse bones along with a few imported items and copper objects in MC tombs indicate that the inhabitants of the area were engaged in the same intensification of mortuary-centred prestige displays as the rest of the island towards the end of the MC. In comparison with the majority of Cypriot Bronze Age settlements, Kalopsidha shows a long period of occupation (primarily EC I–LC IIA). It is not possible to ascertain the extent of prehistoric settlement at Kalopsidha during any period of occupation, and areas with identified settlement evidence are dispersed over 14 ha. Åström suggests that the Late Cypriot (henceforth LC) settlement (exposed in excavation only in Trench 9) may have only been an area of 40–50 m in diameter (Åström 1966, 48). Gjerstad’s house apparently adjoined another to the east and a road ran along the southern face (Gjerstad 1926, 27; see Fig. 8.3, a), although it has been questioned whether Gjerstad’s house was in fact a single architectural unit (Frankel and Webb 1996, 54). Åström positioned Trenches 1–9 in an effort to locate stratified deposits to define the relative chronology of the LC, but Trenches 1, 2, and 4–8 contained no in situ material (Åström 1966, 37–38). Trench 3, Trench 9 and Gjerstad’s house provide the only glimpses into the nature of the settlement and will be discussed in further detail below.

Trench 3 The small but tantalising exposure of Trench 3 (8 m N–S x 1 m E–W) still remains the most useful for the purposes of understanding Kalopsidha’s settlement organisation during the MC III–LC IA transition. Three walls were partially exposed (Fig. 8.3, b) and only a single construction and occupation phase was attested. The most substantial, and presumably exterior, wall (Wall A), is 1.2 m wide

A DJO I NI NG B UI LDI NG Unexcavat ed

1 2 3 4

11 5 6 10 7

9

8 S

A.

0

T

R

5m

E

E

T

Wall C 2

Wall B 1 Wall A N

B.

Fig. 8.3 MC III–LC IA structures at Kalopsidha. A: Plan of Gjerstad’s house (after Gjerstad 1926, 28, fig. 3). B: Trench 3 (after Åström 1966, fig. 21).

and preserved to a height of 45 cm. Åström (1966, 41) speculated that it may have been a fortification wall, but more recent research has suggested that it is more likely these structures represent a social strategy of emulation of Levantine elite behaviour rather than possessing a purely militaristic function (Philip 1991; Crewe 2007b; Horowitz 2007; Peltenburg 2008). Parallels for roughly contemporary walls of this monumental (in Cypriot terms) building style are attested across the island at Enkomi in the Area III, Level A pre-fortress wall remnant and the Level IA fortress building (Dikaios 1969–1971, 15–16); in the external walls of the fortresses at Korovia-Nitovikla (Sjöqvist 1940, 64–73); at Dhali-Kafkallia (Swiny 1972, 28) and Ayios Sozomenos (Gjerstad 1926, 37–47); and in the mound of Phlamoudhi-Vounari (Horowitz

66

Lindy Crewe

2007, 140–41, table 6.6). An example from the southwest of the island may also now be attested at Kissonerga-Skalia (Crewe et al. 2008) as pottery evidence from the latest occupation surface associated with the 1.2 m wide rubble wall (Wall 68) in Trench G also dates to MC III–LC IA. Construction techniques for these structures are variable; some exhibit a casemate technique and others use a fairly haphazard rubble construction set in variable amounts of mud mortar. All were apparently constructed within MC III–LC IA. Walls B and C of Trench 3 were around 40 cm wide and ran more or less perpendicular to Wall A, suggesting they were internal walls. The structure was abandoned suddenly, and Åström (1966, 47) concluded that Room 2 was destroyed by fire. The remains, including three pithoi in Room 2 and two in Room 1, indicate that it may have been a storage area.

Gjerstad’s House (Site C) The house excavated by Gjerstad measures approximately 12 m × 15 m and consists of eleven rooms, including Room 5, which Gjerstad interpreted as an open inner courtyard (Fig. 8.3, a). Gjerstad excavated seven strata but 3–7 underlay the house and belonged to an earlier structure. Gjerstad and Åström dated strata 1–2 to MC III. The entire house was excavated only down to Stratum 2, and only one room was excavated to Stratum 7 (Barlow 1985, 49). Evidence of conflagration in Room 7 was interpreted as a destruction of the house in Stratum 2 (Åström 1966, 139). Stratum 1 represents the construction of a new floor level and brief reoccupation of the house before final abandonment. There are many problems with the stratigraphy, classification and subsequent storage of the material (for discussion of these problems, see Barlow 1985 and Crewe 2007a, 51). A number of wares of both Cypriot and nonCypriot origin were classed as ‘foreign’ (Gjerstad 1926, 269). It is unclear as to which strata belong the “great quantities of Syrian ware, large pithoi with pointed base and vertical shoulder-handles, also jugs and flasks” cited by Gjerstad (1926: 36). Åström (1972a, 170–71) summarised the wares present but noted that much of the material from Strata 1–2 was unfortunately mixed together at a later date. Åström (2001) has since noted White Painted VI, Monochrome and Bichrome Wheelmade ware in these strata, signalling that occupation continued into LC I.

Trench 9 No intact architecture was encountered in Trench 9 (5 m N–S × 1 m E–W), but an incredibly large number of sherds were excavated (223,000 of which 95,936 were classified), predominantly dating from MC III–LC IIA. The bottommost layers appeared to be undisturbed and were dated to MC III, but an early LC I date has since been suggested (Merrillees

1971, 66–67; Manning et al. 2002). This is considered likely here based on the presence of local wheelmade wares. There were smaller numbers of LC IIB–LC III sherds, and postBronze Age material was found throughout most levels, indicating much of the area was disturbed (Åström 1966, 48). In addition to ceramics, small numbers of copper and bronze fragments and artefacts, slag and a possible crucible suggest that metalworking was carried out in the vicinity (Watkins 1966, 113–115). Åström interpreted Trench 9 as a dump associated with an unlocated settlement, but Webb (1999, 113–116) has recently reinterpreted it as a likely sanctuary deposit due to parallels with the site of Athienou-Bamboulari tis Koukounninas (Dothan and Ben-Tor 1983). Similarities include the large numbers of low-fired miniature vessels, which may have served as votives, large amounts of immature goat/sheep bones, evidence of metal working at the site (including some scrap interpreted as votive in nature) and location on a possible overland route between copper mines and coastal centres. Further evidence of the likelihood of the deposit relating to a non-domestic function may be seen in the large numbers of painted wares (60% in the lower levels and 26.6% overall were of White Painted varieties) compared to proportions found associated with domestic architecture at Kalopsidha (10.7% in Trench 3) and also within the Enkomi Areas I and III buildings (up to 10%, Crewe 2007a, 133–143). Although use of Trench 9 lingered on into LC IIA, the majority of the pottery is of LC I date, and it is likely that after LC IB activity was much reduced. In sum, the latest occupation in the excavated buildings and the earliest deposition in Trench 9 are at least partially contemporary with the first occupation of Dikaios’ Area I building at Enkomi in LC IA1. Evidence for how Kalopsidha functioned prior to MC III is elusive, except for the presence of another (largely unexcavated) structure underlying Gjerstad’s house. In the remainder of this paper, I would like to explore the material indicators of Kalopsidha’s occupation and how they may tie in with the wider social processes occurring on the island.

Pottery and Kalopsidha’s International Relations Kalopsidha has been identified as a pottery production centre (Åström 1966, 8) based upon the presence of misfired sherds at both Site C and in Trench 9 and a preponderance of unique types of certain wares, particularly some of the White Painted varieties (Fig. 8.4), Red Slip, Black Slip and Plain White wares. Although, as noted above, the site is renowned for its reputation as the source of the earliest exported White Painted wares, the most common pottery style at Kalopsidha is Plain White Handmade ware, accounting for 37% of the Trench 3 assemblage (Åström 1966, 40–44). The significance of this may again be highlighted through

8. Rethinking Kalopsidha: From specialisation to state marginalisation

67

Fig. 8.4 Juglets of White Painted ware variants manufactured at Kalopsidha. A: White Painted III–IV Pendant Line Style (Åström 1972a, 28, fig. IX.4). B–C: White Painted IV–VI Cross Line Style (Åström 1972a, 65, figs IX.11–12; scale 1:4). D: White Painted V Eyelet Style (Åström 1972a, 69, fig. XVI.15). E: White Painted V Tangent Line Style (Åström 1972a, 71, fig. XVI.17). F: White Painted VI Soft Triglyphic Style (Åström 1972b, 58, fig. XLI.2).

Fig. 8.5 Plain White Handmade ware pithoi from Strata 1–2 at Gjerstad’s house. A: Selection of pithos sherds with relief bands. B: Plain White ware pithos of EC III–MC II type.

comparison with the contemporary occupation level in LC IA Area I at Enkomi where Plain White (including both handmade and wheelmade forms) forms only 8% of the assemblage (Crewe 2007a, 139, table 17.14). Gjerstad did not isolate the ware during his excavation, and it seems to have been lumped in with the foreign wares. Amounts are also lower as material from the earlier structure is present in the deposits. LC IA–IIA Trench 9 has comparable numbers to later levels at Enkomi at around 31% (Åström 1966, 49). Again, this figure includes all manufacturing technologies at Enkomi but reflects only the handmade variety at Kalopsidha. I will return to this point below. Kalopsidha also has the majority of the earliest examples of Plain White on the island (Crewe 2009). It is perhaps not difficult to see how Plain White ware may have developed at Kalopsidha as it is essentially White Painted ware without any painted decoration. The ware also owes stylistically and typologically to Middle Bronze plain wares from the Levant

(Crewe 2007b). In addition to the novelty seen in the surface treatment of Plain White ware, new vessel types also occur. I will briefly note the significance of one of these: the Plain White Handmade pithoi. Cypriot communities began producing large storage vessels as early as EC III–MC I, but they became more common across the island during MC II (Pilides 1996, 107). The earlier examples were manufactured in Red Polished ware, usually with a large round mouth, short neck with two handles from the neck to shoulder, piriform body and small flat base, ranging from 0.75–1 m in height (Pilides 2000, 4, fig. 1). Probably in MC III (Åström 1966, 205–206) a new, slightly smaller type appeared, ranging from 33–56 cm in height (Pilides 2000, 3), which was manufactured in a fine, hard variant of Plain White ware often with relief or incised bands at the neck or shoulder (Fig. 8.5, a) and usually without handles. The Levantine affinities of the Plain White pithoi with

68

Lindy Crewe

relief bands were noted by Åström (1972a, 230–231), and these vessels have since been identified as exports to Tell el-Dabca (Maguire 2009, 157–161) and Ugarit (noted in Pilides 2000, 51). I have also recently examined sherds from Tell el-‘Ajjul and located further examples, along with a number of similar local Levantine variants. This phenomenon of stylistic and technological borrowings along with true imports and exports occurred across the eastern Mediterranean during the Middle Bronze Age. Related to this is the circulation of juglets, discussed further below. The manufacture of this new style of vessel on Cyprus can be seen as an adaptation of a Levantine form but drawing on pre-existing Cypriot requirements for increased storage capacity. Syrian inspiration has also been seen in some of the White Painted V vessels imitating Syrian form and with painted eyes (Fig. 8.4, d; Åström 1972a, 223), and Kalopsidha also has the earliest locally manufactured Levantine-style saucer lamps with pinched rims (Åström 1966, 111, fig. 107). An example of the experimental and hybridised nature of Cypriot pottery at this time can be seen in a handle from a pithos of earlier Red Polished type but manufactured in Plain White ware (Fig. 8.5, b). In association with these locally produced Plain White pithoi must briefly be mentioned the significant quantities (a minimum of 26 vessels) of imported Canaanite jars found in both Trench 3 and Gjerstad’s house, and including the stratum underlying the structure. Despite reservations noted above on the attributions of this material to particular strata, sherds from a minimum of five different vessels are stored in a box marked “Room 9, Str. 3” and are thus attributable to MC II–III, predating Gjerstad’s house. Amongst other boxes from “Stratum 1 or 2” or only labelled “MC III” are sherds from at least a further 20 vessels. At least one vessel and probably more occur in the Trench 3 building. Kalopsidha therefore had exceptional storage capacity for the later MC and into early LC IA, both incorporating local and imported vessels, and on current evidence the site exhibits privileged access to imported goods and trade relations. In contrast, only one definite Canaanite jar fragment is mentioned from LC IA–IIA Trench 9 (Åström 1966, 76) although other unidentified sherds may have been present. Turning to the Trench 9 deposits, although the earliest deposition of material may be contemporary with occupation in the excavated structures, the area continued to be used for around another 200 years after their abandonment. The low numbers of wheelmade wares in Gjerstad’s house and Trench 3 are indicative of this very early LC IA date before wheelmade wares were commonly produced (Crewe 2007b), but in comparison with other parts of the island the continuing strong handmade tradition seen in the Trench 9 deposits is significant (less than 2% of the total material is of wheelmade types). This was explained as “conservatism” by Åström (1966, 64), but it is also here that we see the vast quantities of White Painted ware that correlate so well with

types found outside of Cyprus and in small numbers in the earliest Enkomi deposits (12% in the earliest occupation but decreasing to 2% by LC IB; Crewe 2007a, 113).

Discussion In both the MC III–LC IA structures in Trench 3 and Gjerstad’s house the pottery is characterised by a substantial proportion of innovatory forms and styles. This includes the earlier strata beneath Gjerstad’s house with significant quantities of White Painted ware, as White Painted itself was an innovation at the beginning of the MC. On the other hand, the LC IA–IIA assemblage of Trench 9 is characterised by technological and stylistic archaisms. By attempting to understand these conflicts between tradition and innovation, and perhaps between different sectors of the one community, it is possible that we can understand the processes of change on Cyprus. Peltenburg views the first monumental ‘fort’ buildings as “instrumental in securing for islanders a more active and independent role in the realisation of potential of local resources” (2008, 153). The destruction and abandonment of the Trench 3 building and Gjerstad’s house provide a glimpse into the probable conflict taking place between traditional ways and those who seemingly embraced a new world of foreign symbols and ideas at the MC–LC transition. The Trench 3 building was constructed on a grand scale, and both buildings contained imported and locally-produced stored surplus, which may well have contained the ingredients for the manufacture of the material to be transported in the White Painted vessels. Whether the building programme and the housing of the goods reflect the actions of an individual, group within the community or outsiders is unknown. If sectors of the community (rather than outsiders) were responsible for the destructions, then the subsequent adherence to traditional pottery manufacture and decorative techniques by those who continued to live at Kalopsidha may reflect a (temporary) return to traditional and tested methods. Maguire notes that of the significant quantity of Cypriot pottery vessels imported to Tell el-Dabca prior to the New Kingdom, the majority are White Painted handmade wares of styles most commonly found at Kalopsidha although she adds the caveat that Kalopsidha is the only site to have been excavated in the region (2009, 26). Maguire has also examined the phenomenon of the circulation of a range of small jugs and juglets as precious commodity containers (probably for perfume or oil), which were produced in Cyprus, Egypt and the Levant (1995; 2009, 67–68) and were apparently a crucial component of the Hyksos-period material culture complex. At the beginning of the Late Bronze Age, styles and production zones changed dramatically, apparently replaced almost entirely

8. Rethinking Kalopsidha: From specialisation to state marginalisation by Cypriot-produced Base Ring and Red Lustrous ware versions (Maguire 1995, 63). The lack of eastern-style White Painted wares post-dating LC IA at Enkomi or in New Kingdom/Late Bronze deposits (Maguire 2009, 18) suggests that Kalopsidha ceased to play a role in production and export at this time. Occupation at the site persisted for perhaps 100 years as evidenced by the Trench 9 deposit, but based on comparison with the Enkomi assemblage it is likely that the majority of Trench 9 use dates only just into LC IB with greatly reduced activity after this. It is reasonable to suppose that Kalopsidha and Enkomi had a symbiotic relationship upon the first foundation of the latter site, with Enkomi serving as a gateway for inland-produced goods and also taking advantage of the increased prosperity brought by interactions with the eastern Mediterranean. It is worth noting that White Painted is the only ware in the Enkomi settlement assemblage in which juglets form a significant component (around 25%, contrasted with around 3% in other wares). Based on their deposition contexts, juglets appear to have played a largely funerary-specific role during the Cypriot Bronze Age, and greater amounts in the settlement strata may therefore be interpreted as the traces of these vessels passing through the site for export (see Crewe 2007a, 132 for further discussion). Whilst the Trench 9 deposit may partially represent the debris of a ‘packaging’ centre, the crudely made, low-fired miniatures (painted and unpainted) found in abundance were certainly not exported and seem more likely to represent votives manufactured perhaps by the ‘worshippers’ as suggested for the cultic deposits at Athienou (Dothan and Ben-Tor 1983, 56–57). The transformation from international production centre to rural sanctuary may have been facilitated by the tradition surrounding the goods that were produced: perfumed oil containers that played a crucial role in mortuary ritual. Perhaps in an effort to avoid losing their advantageous position as Enkomi began to become a focal centre in its own right, Kalopsidhans increased production of the resource that they had long specialised in. The final style of the sequence, the White Painted VI dated exclusively to LC I (Fig. 8.4, f), has a simplified linear decoration but is still handmade. The adherence to handmade painted forms may be a combination of maintaining the ready identification of a long-established product combined with an ‘if it isn’t broken don’t fix it’ attitude of producers. It has been suggested that mortuary consumption was a “driver of economic intensification” for copper production (Keswani 2004, 153). This may also be applied to perfumed oils and the containers they were packaged in, with greater value attached to certain styles of known quality. Keswani (2004, 140) also notes that numbers of tombs within the LC coastal centres are insufficient for the entire population and

69

considers it likely that some inhabitants of the coastal towns were transported back to their ancestral inland villages for burial. Kalopsidha may therefore have been ideally placed as a centre for funerary and related ritual. Indications that the site may have remained active in the funerary industry attested by Kalopsidha-style White Painted juglets found in quantity not only at Enkomi but also at Ayios Iakovos and other sites in the Karpas as well as occasional examples farther afield (details in Åström 1972a, 17–78; 1972b, 53–68). Kalopsidha may have been supplanted by other communities located on more direct routes to the copper sources, such as Athienou, in LC IIA. By this time, the Kalopsidhans had lost their niche, which was transferred to other localities producing and exporting juglets of Base Ring and Red Lustrous Wheelmade wares (Merrillees 1971). The other possible reason for Kalopsidha’s decline may have been that one of the ingredients needed for their special product was no longer available. As Maguire has noted, there is no reason to assume that Cypriot pottery vessels contained Cypriot goods (1995, 54). From LC IA onwards Enkomi retained at least some of the Canaanite jars that were coming into the port. There are significant quantities of Canaanite jars found in LC IA levels (c. 10% of the pottery assemblage in both Areas I and III, Crewe 2007a, 124–125). The final demise of Kalopsidha’s international role may also be linked to the 50% reduction in Canaanite jars in LC IB deposits at Enkomi, potentially due to the disruption of trading routes after the expulsion of the Hyksos from Egypt (Maguire 1995, 63). Taking into account the transportation to the site of bulk goods contained in Canaanite jars, the evidence of Levantine-influenced, locally-produced vessels for additional bulk storage, which were in turn exported, luxury goods in tombs (including equid burials), and the local manufacture of quantities of fine, painted jugs and juglets that were also exported to the surrounding eastern Mediterranean, we can speculate that both bulk commodities and value-added goods (Sherratt 1999) were being produced at Kalopsidha. The oft-repeated MC II “trickle” of exports becoming a “flood” during MC III (Catling 1973, 174) may reflect new methods of intensifying production leading to the adoption of the stylistic and technological innovations seen in the Plain White ware. It is likely that the initial impetus for production was a local need for perfumed oils for funerary use as part of the increased importance of mortuary ritual in competitive display (Keswani 2005). From LC IA, goods were being transported through Enkomi with the likely result that Enkomi eventually restricted the movement further inland of many of the imports. In addition, if Kalopsidha’s early ascendancy was due to its ability to muster agricultural surplus, then once the eastern Mesaoria became more extensively occupied it lost its uniqueness as Enkomi’s own hinterland comprises fine agricultural land. We will probably never be able to establish the exact role

70

Lindy Crewe

of Kalopsidha in the initial stages of the move towards complex society on Cyprus, but it seems likely that the site’s later fortunes were linked closely to those of nearby Enkomi. When Enkomi was first founded we see perhaps the greatest period of prosperity at Kalopsidha, but eventually as Enkomi was transformed into a central place the site became marginalised and finally abandoned.

Acknowledgements I would like to thank Diane Bolger and Louise Maguire for the organisation of and invitation to participate in this volume. Thanks also to Karen Slej and staff at the Medelhavsmuseet, Stockholm for permission to examine material from Kalopsidha and to Rachel Sparks, Keeper of the UCL Institute of Archaeology collections for allowing me to examine the Tell el-‘Ajjul material from Petrie’s excavations. This research was conducted whilst I held the British Academy Reckitt Postdoctoral Fellowship in Archaeology at the University of Manchester, for which I would like to gratefully acknowledge the British Academy.

References Åström, P. 1966. Excavations at Kalopsidha and Ayios Iakovos in Cyprus. Studies in Mediterranean Archaeology 2. Lund, Paul Åströms Förlag. Åström, P. 1972a. The Middle Cypriot Bronze Age. The Swedish Cyprus Expedition IV(1B). Lund, Swedish Cyprus Expedition. Åström, P. 1972b. The Late Cypriot Bronze Age. Architecture and Pottery. The Swedish Cyprus Expedition IV(1C). Lund, Swedish Cyprus Expedition. Åström, P. 2001. Bichrome Hand-made Ware and Bichrome Wheelmade Ware on Cyprus. In P. Åström (ed.) The Chronology of Base-Ring Ware and Bichrome Wheel-made Ware. Proceedings of a Colloquium held in the Royal Academy of Letters, History and Antiquities, Stockholm, May 18–19 2000, 131– 142. Stockholm, Kungl. Vitterhets historie och antikvitets akademien. Barlow, J. A. 1985. Middle Cypriot settlement evidence: A perspective on the chronological foundations. Report of the Department of Antiquities, Cyprus, 47–54. Catling, H. W. 1973. Cyprus in the Middle Bronze Age. In I. E. S. Edwards, C. J. Gadd, N. G. L. Hammond and E. Sollberger (eds) The Cambridge Ancient History, vol. II(1). The History of the Middle East and the Aegean Region c. 1800–1380 BC, 165–175. Cambridge, Cambridge University Press. Crewe, L. 2007a. Early Enkomi: Regionalism, Trade and Society at the Beginning of the Late Bronze Age on Cyprus. British Archaeological Reports International Series 1706. Oxford, Archaeopress. Crewe, L. 2007b. Sophistication in simplicity: The first production

of wheelmade pottery on Late Bronze Age Cyprus. Journal of Mediterranean Archaeology 20(2), 209–238. Crewe, L. 2009. Regionalism and the first appearance of Plain White Handmade Ware in the Middle Cypriot Bronze Age. In I. Hein (ed.) The Formation of Cyprus in the 2nd Millennium B.C. Studies in Regionalism during the Middle and Late Bronze Age. Proceedings of a Workshop Held at the 4th Cyprological Congress, May 2nd, 2008, Nicosia, 79–90. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean. Vienna. Crewe, L., P. Croft, L. Graham and A. McCarthy 2008. First preliminary report of excavations at Kissonerga-Skalia, 2007. Report of the Department of Antiquities, Cyprus, 105–119. Dikaios, P. 1969–1971. Enkomi. Excavations 1948–1958, vols. I –IIIB. Mainz, Philipp von Zabern. Dothan, T. and A. Ben-Tor 1983. Excavations at Athienou, Cyprus. 1971–1972. Qedem 16. Jerusalem, Institute of Archaeology, The Hebrew University of Jerusalem. Frankel, D. 1974. Middle Cypriot White Painted Pottery: An Analytical Study of the Decoration. Studies in Mediterranean Archaeology 42. Göteborg, Paul Åströms Förlag. Frankel, D. and J. M. Webb 1996. Marki-Alonia: An Early and Middle Bronze Age Town in Cyprus. Excavations 1990–1994. Studies in Mediterranean Archaeology 123(1). Jonsered, Paul Åströms Förlag. Gjerstad, E. 1926. Studies on Prehistoric Cyprus. Uppsala, Uppsala Universitets Årsskrift. Horowitz, M. T. 2007. Monumentality and Social Transformation at Late Bronze I Phlamoudhi-Vounari, Cyprus. Unpublished Ph.D. thesis, Columbia University. Keswani, P. S. 2004. Mortuary Ritual and Society in Bronze Age Cyprus. London, Equinox. Keswani, P. S. 2005. Death, prestige and copper in Bronze Age Cyprus. American Journal of Archaeology 109, 341–401. Knapp, A. B. 1988. Copper production and eastern Mediterranean trade: The rise of complex society on Cyprus. In J. Gledhill, B. Bender, and M. T. Larson (eds) State and Society. The Emergence and Development of Social Hierarchy and Political Centralization, 149–169. One World Archaeology 4. London, Routledge. Maguire, L. C. 1995. Tell el-Dabca: The Cypriot connection. In W. V. Davies and L. Schofield (eds) Egypt, the Aegean and the Levant: Interconnections in the second Millennium BC, 54–65. London, British Museum. Maguire, L. C. 2009. Tell el-Dabca XXI: The Cypriot Pottery and its Circulation in the Levant. Ausgrabungen in Tell el-Dabca (series ed. M. Bietak). Untersuchungen der Zweigstelle Kairo des Österreichischen Archäologischen Instititutes 33. Vienna. Manning, S. W., D. A. Sewell and E. Herscher 2002. Late Cypriot IA maritime trade in action: Underwater survey at MaroniTsaroukkas and the contemporary east Mediterranean trading system. Annual of the British School at Athens 97, 97–162. Merrillees, R. S. 1971. The early history of Late Cypriote I. Levant 3, 56–79. Muhly, J. D. 1986. The role of Cyprus in the economy of the eastern Mediterranean during the second millennium B.C. In V. Karageorghis (ed.) Acts of the International Archaeological Symposium ‘Cyprus between the Orient and the Occident’,

8. Rethinking Kalopsidha: From specialisation to state marginalisation Nicosia, 8–14 September 1985, 45–60. Nicosia, Department of Antiquities, Cyprus. Myres, J. L. 1897. Excavations in Cyprus in 1894. Journal of Hellenic Studies 17,134–173. Peltenburg, E. J. 1996. From isolation to state formation in Cyprus, c. 3500–1500 BC. In V. Karageorghis and D. Michaelides (eds) The Development of the Cypriot Economy from the Prehistoric Period to the Present, 17–43. Nicosia, University of Cyprus and Bank of Cyprus. Peltenburg, E. J. 2008. Nitovikla and Tell el-Burak: Cypriot midsecond millennium BC forts in a Levantine context. Report of the Department of Antiquities, Cyprus, 145–157. Philip, G. 1991. Cypriot bronzework in the Levantine world: Conservatism, innovation and social change. Journal of Mediterranean Archaeology 4(1), 59–107. Pilides, D. 1996. Storage jars as evidence of the economy of Cyprus in the Late Bronze Age. In V. Karageorghis and D. Michaelides (eds) The Development of the Cypriot Economy from the Prehistoric Period to the Present, 107–124. Nicosia, University of Cyprus and Bank of Cyprus. Pilides, D. 2000. Pithoi of the Late Bronze Age in Cyprus: Types from the Major Sites of the Period. Nicosia, Department of Antiquities, Cyprus.

71

Sherratt, E. S. 1999. E pur si muove: Pots, markets and values in the second millennium Mediterranean. In J. P. Crielaard, V. Stissi and G. J. van Wijngaarden (eds) The Complex Past of Pottery Production, Circulation and Consumption of Mycenaean and Greek Pottery (Sixteenth to Early Fifth Centuries BC), 163–211. Proceedings of the ARCHON international conference held in Amsterdam 8–9 Novermber, 1996. Amsterdam, J. C. Gieben. Sjöqvist, E. 1940. Reports on Excavations in Cyprus. Revised Reprint from the Swedish Cyprus Expedition Finds and Results of the Excavations in Cyprus 1927–1931, vol 1 (text and plates). Stockholm, Swedish Cyprus Expedition. Swiny, S. 1972. Part II: The fortified settlement. In J. C. Overbeck and S. Swiny, Two Cypriot Bronze Age Sites at Kafkallia (Dhali), 25–31. Studies in Mediterranean Archaeology 33. Lund, Paul Åströms Förlag. Watkins, T. W. 1966. Metal finds. In P. Åström, Excavations at Kalopsidha and Ayios Iakovos in Cyprus, 113–115. Studies in Mediterranean Archaeology 2. Lund, Paul Åströms Förlag. Webb, J. M. 1999. Ritual Architecture, Iconography and Practice in the Late Cypriot Bronze Age. Studies in Mediterranean Archaeology Pocket-book 75. Jonsered, Paul Åströms Förlag.

9 FROM KIN TO CLASS – AND BACK AGAIN! CHANGING PARADIGMS OF THE EARLY POLITY Anne Porter

In this celebration of Eddie Peltenburg’s career it is more than fitting to revisit a topic to which he has made such a significant contribution: the genesis and organisation of the early polity in northern Syria. Since it is now widely recognised that the neo-evolutionary models of the early state, as characterised either by developmental stages or classificatory types, are no longer useful for understanding these issues (see Yoffee 2005 for a succinct treatment), there are a multitude of approaches and frameworks to consider. But whether we consider them in terms of the ‘archaic state’ (Feinman and Marcus 1998), ‘incipient’, ‘nascent’ or ‘transitional’ states (all terms employed by Trigger 2003) or ‘early complex polities’ (Smith 2003), and whether they are implicit or explicit, there is an understanding in any study of the early polity (my own terminological preference) that something qualitatively, if not quantitatively, different has occurred in political history – a new state of political being, not just in grand organisational terms but in terms of the way people conceive of themselves and their interactions with others, has come into existence. In the past one key aspect of this change in political being was characterised as a shift from kin-based societies – where people both understood their place in the world and their responsibilities to others to be arranged according to their blood relations, i.e. their birth into a certain group – to a class-based society where the affinities (or lack thereof) between much larger numbers of unrelated people who coexisted in the same space were derived from their position in socio-economic hierarchies (for a random sample see Adams 1966, 80; Yoffee 1993, 69; but cf Yoffee 1997, 261–262; see also Zagarell 1986, 416; Pollock 1991, 177; McCorriston 1997, 518; Trigger 2003, 152–153). The applicability of this change was universal. In some way or another all societies went through it – or didn’t. If a given

society did not, or had not yet, then that society did not qualify as a state. Various terms have been applied to this pre-state/state alternative entity in both the anthropology of the contemporary world and the archaeology of the ancient world: chiefdom (Earle 1997; Flannery 1999) and segmentary society (Stephen and Peltenburg 2002) are two of the most prominent. Both in some way harken back to a kin/tribal basis for these pre/non-state societies despite the problematic nature of the term ‘tribe’. But as the emphasis in American archaeology of the Near East shifted from the problem of state formation to a more diverse set of issues, as excavation results increasingly found not the transition to the state but the remains of functioning polities or complex societies (Stein 1994; 1998), and as political theory shifted at the same time as the practice of archaeology itself changed, evidence began to emerge that the polities of the north, and even of the south (Yoffee 2005, 110, 214), were not quite so ‘developed’ as to have entirely abandoned kin relations for class structures long after such changes were supposed to have been accomplished. This then seems to require a choice: are the early politics of northern Mesopotamia not far enough along the continuum to really qualify as states? Or should we think about the situation in entirely other frameworks? There is an interesting divergence here between highly local studies of the polity and global ones: those focussed on northern Mesopotamia tend to the former by employing the term ‘tribe’ (Steinkeller 1999; Stein 2004; Cooper 2006; 2007) or, like ‘segmentary society’, a label based on characteristics of the tribe (such as Peltenburg 2007a, 11) and deriving ultimately from a long history of usage in anthropology (Morgan 1877; Evans-Pritchard 1940; Service 1962; Sahlins 1968; see Peletz 1995, 353 for its different meanings). Those situated in a larger discourse opt for other

9. From kin to class – and back again! Changing paradigms of the early polity approaches altogether (e.g. Blanton 1998; Smith 2003; Yoffee 2005). I would like to bring both together in this paper by considering local northern Mesopotamian polities through a different framework, one which does not contrast one state of political being (tribe) with another (state) because practices of social and political interaction, which are entirely interdigitated but not necessarily correlative, vary from polity to polity in northern Mesopotamia. In each polity there are situations where kinship is the dominant mode of interaction and situations where socio-economic position (‘class’ is not really an apt term) and/or civic identity is operable and administrative structures are framed through and deploy all modes of interaction at different times. This variability characterises both early polities and much later ones and extends far beyond northern Syria as well. It is therefore not a matter of viewing the northern polities as less than fully developed states, but of thinking about the state differently, perhaps indeed dispensing with it (and associated terminologies) as an analytical frame of reference altogether. The only way to depict such variability is to employ a wide range of descriptive terms on a contingent basis because the production of monolithic paradigms driven by the need for succinct ways of expressing political organisation, as well as the ‘hows’ and ‘whys’ of its change, leads ultimately to a counter-productive reductionism (see Smith 2003, 40 and note 9). There are many ways of thinking about the constituent relationships contained within the early polity (cf Stein 1998), but there are at least four that top my list and consequently four sets of terminologies to be developed. One way is in terms of the spatial organisation of the polity, which I term polity morphology. It tends to be a given in most discussions of the Near Eastern state that a polity controls contiguous territory, either in a highly circumscribed manner such as the Mesopotamian city state or in the broader manner of the Syrian state, with the empire being the most attenuated, but still contiguous, form. Yet the very idea of colonialism, largely accepted as pertaining in some way in the 4th millennium (Surenhagen 1986; Algaze 1993) raises other possibilities, one of which is the ‘bifurcated polity’ in which two components of the same political entity exist at a distance from each other. A wide range of political relationships may be posited for this spatial situation, ranging from two essentially independent groups only nominally connected, to differential specialisations among the components of the bifurcated polity (Porter 2009). A ‘dispersed polity’ is one that has a significant component of its populace engaged in mobile enterprises, including, but not limited to, both localised and broad-range pastoralism, where territorial usage shifts, either seasonally or over the long term. On this issue I diverge from Smith (2003, 153), who argues that a polity and its catchment area are not coterminous. They may be, if the people who utilise the catchment area are in fact governed by the polity

73

or consider themselves to be members of, and therefore governed by, the polity. It does not matter if the polity does not have the means to enforce its desires over this catchment area, only that its members think it does. But a dispersed polity might also include scattered components integrated into a single entity through conquest, for example, where control is more direct than only the paying of tribute (Wattenmaker 1994, 197–198) or as when Samsi Addu appointed one son over Mari, the other over Ekallatum while he ruled at Shubat Enlil (van der Mieroop 2004, 102). An extended morphology, in contrast, posits a more regularised spatial connection, somewhat like an arm stretched out to hold in hand a distant object. The Assyrian trading colonies might provide such an example, where stable and frequently utilised routes of passage and communication between homeland and colony are maintained whether through treaty or tribute, so that it appears as if the homeland controls the space of the routes themselves. While these terms represent different points on a continuum, there are others that represent different ways of spatial distribution entirely, and so multiple terms to describe morphology might be used of any one polity. Variable polity morphologies are in evidence over a very long period in the Near East, starting at least in the 4th millennium. The relationship between Habuba Kabira and Uruk, for example, might represent a bifurcated polity or a dispersed model if integrated political relationships (as opposed to only economic ones coupled with social memory of chronologically distant origins) really can be demonstrated between Uruk and some of the other sites that share its material culture. In the 3rd millennium Ebla controlled Carchemish (Fronzaroli 2003) and perhaps Emar as well (Archi and Biga 2003, 10).While evidence for these relationships has prompted some to claim that Ebla was a massive empire (e.g. Astour 1992), there is no indication, archaeological or textual, that substantiates a claim for contiguous territory from Emar to Ebla to Carchemish; indeed, the evidence is to the contrary. There were clearly independent kingdoms in between. The dispersed polity of Samsi Addu has already been mentioned, but it may not have been the only non-contiguous polity of the 2nd millennium: the Emutbala kingdom of Larsa may have constituted some form of bifurcated polity with the Yamutbal of northern Syria (Porter 2009; see also Steinkeller 2004). Ideas about the effective limitations of direct political control over space (e.g. Johnson 1978; Giddens 1981) are no doubt part of the reason that divergent polity morphologies are not well considered, but there are multiple ways of stretching time and space, so that the distance that may be travelled in a day (Flannery 1999, 5) does not condition the degree to which, or the ways, a polity might extend itself. However, the spatial organisation of a polity and the territorial relationships within it do pose a set of problems for political organisation and operation, just as the way a society is organised may

74

Anne Porter

constrain the potential for different territorial configurations. As I have recently argued (Porter 2009), kinship practices might be part of the way a polity maintains integrity over space, but this should not be taken to mean that this is always the case nor that kinship may not be an equally critical component of contiguous or local polities. Nevertheless, differences in how kinship systems are structured and how kinship practices work between bifurcated, dispersed and contiguous polities would not be surprising. Kinship systems and practices form a key component of social configuration, the complex of structures, ideologies and practices that form the basis of social interaction, and the second way in which polities vary in their organisation and operation. Kinship, defined ego-centrically or intragenerationally (the living kin of an individual) and sociocentrically or multi-generationally (the relationships accrued through having a common ancestor or common descent; see Holy 1996; Fowles 2002) is in evidence in one way or another, and from a variety of sources, as a powerful component of the social worlds of a large number of sites across Syria. Since kinship is established by both horizontal and vertical relationships and can be socially constructed as opposed to only attributed by birth (Porter 2009, 215), genealogies, ancestor practices such as the kispu (Tsukimoto 1985), certain rituals of association (Durand 1992, 117; Bonechi 1997, 480) and relationship terminologies (e.g. Sasson 1998, 462) may all be considered evidence of it. Archaeological evidence is perhaps a little less direct, but it is nevertheless persuasive and can be found in burial practices (Peltenburg 2007b); architecture (Dohmann-Pfälzner and Pfälzner 1996); the spatial organisation of a settlement (Porter 2009); and even individual artifacts (Hempelmann n.d. a; n.d. b). Much of the data for some form of kinship, however, is interpreted as creating and perpetuating various kinds of ancestor systems (Peltenburg 1999; Porter 2000; 2002a; Schwartz 2007). There is, of course, the possibility that once having discovered ancestors we are inclined to see them everywhere (Whitley 2002), but the written evidence does support such claims made of burial practices. The relationship between ancestor traditions and political systems is not straightforward, however. Ancestors can be deployed to do different things in different ways at different times. Kinship systems and ancestor practices do not even have to work in tandem, and it is possible that ancestor practices exist without functioning kinship systems as relict of earlier practices that have since disappeared, leaving only a rhetoric of legitimation behind. Kinship rules may work to define who has membership of a network of living relations while the practices that create and perpetuate ancestors, ranging from burial customs to commemorative rituals, may have nothing to do with group membership at all but work to perpetuate an ideal vision of society or implement a socio-civic code of moral authority (Porter 2002a). In broadest terms, kin relations are “potentially boundless”

(Holy 1996, 40) and can be used to construct inclusionary societies where linkages are made between groups that might sometimes be far distant in time and space, and exclusionary societies where membership of the group is restricted through a narrow conception of kin relations (Porter 2002b). Kinship, moreover, may be used to create and substantiate differentiation within a kin-based system where members of one lineage are privileged over others (usually that closest to the founding ancestor), leading to authoritarian structures of power and/or segregated elites; or it may work to promote corporate political behaviour where social-economic differences are offset by kin connections across hierarchies. Kinship, despite long academic traditions of assumptions otherwise (again, summarised succinctly by Yoffee 2005, 23), implies nothing about political operation. Not all kinship systems are the same, and kinship systems do not in themselves give rise to one form of political organisation and operation or another. Kinship is a set of social rules and resources. The state, on the other hand, is a political structure in which different social configurations may pertain. The equation of kinship structures with a) a particular type of political form or b) pre- or non-state societies, especially when viewed through the lens of ancestor practices, is therefore misleading. Moreover, kinship and a high degree of social stratification are not mutually exclusive, for they may be functional in different arenas of life, so that individuals become embedded in multiple and cross-cutting social networks. This understanding stems from Service (1962) and is one aspect of neo-evolutionary frameworks that has been sadly neglected in archaeological discourse (although see Fowles 2002). Members of a single kin group may occupy different levels of a social hierarchy. A grouping based on socio-economic position may include members of multiple kin groups, and which set of allegiances is invoked will depend on the situation at hand. Nevertheless, systems of social standing, kinship and ancestor practices all form a critical part of the third arena in which polities vary, political ethos, by which I mean the way a group conceives of itself and its members’ relationship to it in terms of its character and principles (Porter 2009) and which can be described as, amongst other ways, communitarian, familial, hierarchical, authoritarian, heterarchical, heterogeneous and factional. An ethos, however, does not necessarily correspond to political practice, the fourth element of my framework for thinking about the early polity, for practice is situated in the shortterm and may vary considerably over even a generation or may be continually contested (e.g. Fowles 2002, 26), so that what we see in the archaeological record is either an aggregate of that contest, with actual practice and desired practice indistinguishable, or one side of the contest manifest more overtly in material ways. Moreover, a political ethos is established by multiple factors: it is “the political, social

9. From kin to class – and back again! Changing paradigms of the early polity and religious ideologies and practices that produce and/or express and/or perpetuate a group’s self-conception, world views, place within that world, internal organization, and operation” (Porter 2009, 213), whereas political practice comprises just the last two components of that list. It is what people actually do, whatever the structural and ideological content of a society suggest to us it is that they should be doing, or even that they think they are doing. However, the same sorts of terms might well be employed to describe actual political practice as are used for ethos. More often than not, I suspect, it is ethos that we recover from the material record (and even texts), but we mistake it for practice because of our assumptions about correlations between material categories, such as architectural form and socio-political organisation (see, for example, Porter 2007a, 84; and cf. Smith 2003, 230–231). What is more, ancient representations of ethos must be carefully separated from our own. Large-scale architecture determined to be secular equals a palace; a palace equals at minimum an elite segregated from society and holding a monopoly of power. But heterarchical political practice might be hidden because other authority structures such as ‘elders’ do not necessarily have a dedicated space in which they function, while extensive monumental architecture may convey a more powerful public authority than necessarily exists. Although arguments might be made for the continued utility of some form of classificatory system (Fowles 2002, 13–14), my aim here is not to identify a particular kind of political unit or categorise systems of power distribution, but to recognise the multiple sets of social and political networks in which people operate that come together to constitute the nature of any given polity – a close reading of the data in order to produce an “inter-emic” (Campbell 2007, 6) view of existence. I have deliberately refrained from adducing archaeological correlates to any of these descriptive terms because not only do I wish to avoid any implication of reification, but also because recent discussions of materiality (e.g. Miller 2005) indicate that categories of material culture and even individual objects may contain multiple and conflicting situations within themselves, especially “when one chooses to express one’s dissent with a situation through the very institutions [or artefacts (my insertion)] that ultimately recreate it” (Dobres and Robb 2000, 9). One should neither choose which aspect is most relevant nor explain one away in favour of the other; nor, in fact, should one even attempt to resolve the contradiction. Both are equally present and equally valid, and the question concerns only what happens when such contradictions are engaged. A high degree of monumentalism, for example, is seen variously as the product of chiefly behaviour and an indicator of the state, and it is not that one interpretation is necessarily wrong but that both situations are possible and can occur at the same time. Similarly, a certain form of basket may embody both oppression and resistance (Meskell 2005), so that the point

75

here is to think about the wide range of possibilities inherent in any given body of material rather than assume that one model should fit (meaning that another does not). The four perspectives and accompanying descriptive terms proposed here are only some among many that might be brought to bear as a way of conveying the nature of any particular polity under consideration. Space precludes detailed demonstration of this approach and its outcomes, but a brief summary of some of its implications for one site is possible. At Ebla, religious, administrative and productive functions and activities located in the city lie at the core of a network of related functions and activities located around and outside the city itself, maintained in part by the distribution of members of the ruling family, officials and their residences throughout the countryside in smaller towns (Archi 1990, 53; 1992, 25), and in part by royal rituals of pilgrimage (Fronzaroli 1992; 1993; Archi 2005, 90; Porter 2007a). Rather than reading Ebla as the northern archetype of the highly urbanised and centralised state in the 3rd millennium, the archaeological and textual evidence actually presents a decentralised polity in both morphology and practice. Morphologically the polity of Ebla is composed of a series of contiguous and non-contiguous elements, ranging from the as yet unlocated URU.BAR (translated as ‘suburbs’) in which many of the workers of the palace are thought to live (Archi 1982, 212; Arcari 1988, 128) to more remote dependent villages in its territory and the separate subordinate towns, such as Carchemish, it controlled. Political practice seems equally complex and may be familial in that power is shared across an extended family structure; heterarchical in that there is some indication that parallel systems of authority exist (but cf. Stone 1999 for a different application of heterarchy and hierarchy); and decentralised if the distributed components of the polity are effectively integrated through an equivalent spatial distribution of power as represented, for example, by the residences of princes and other officials. Because the archaeological sources come from one component of that spatial system (the mound of Tell Mardikh, which clearly does not comprise the entire city, let alone the polity of Ebla) and the written sources come from an even more restricted place, the so-called ‘palace’, we certainly have a disproportionate view of the power of the palace’s occupant, the ‘king’ and the range of his control (cf. Biga 1995). Yet the texts also give evidence of other bodies that play a part in governance, even if the lack of equivalent representation of their actions and responsibilities obscures their precise role. Several factors evident in the documentation indicate the complexity of political practice and ethos at Ebla, as well as the moments of disjuncture between them. For example, the repeated use of EN-EN, plural ‘kings’ (Archi 2001) and the near equivalence of operable power between minister and king raise questions about the assumption of autocracy at Ebla, and might be interpreted as power sharing

76

Anne Porter

between minister as belonging to a subsidiary branch of the house/lineage/family, and the head of that body, however it is constituted, the king. Another possibility is the idea of a corporate dynasty where rule is in some way passed across a plurality of royals (Michalowski 1988, 271) or collateral lines (Biga and Pomponio 1987, 61). The crown is common property within this extended family even though it was held by one member at a time. But the evidence for ancestor practices at Ebla (Archi 1986; 1988a; 1988b; 2002; Porter 2000) indicates that descent is part and parcel of establishing royal lineages rather than intra-generational ties and so perhaps operates to restrict the people who could lay claim to kingship; the distribution of the holdings of junior members of the line outside the city might also serve to limit competition for kingship. At the same time, I take EN.EN to refer to previous kings and to indicate their importance as complicit, active even, in the continued practice of governance. So while there may be an ethos of extended family rule, an intent, actual practices end up far more restrictive in outcome. Meanwhile, the presence of elders in the background of royal authority raises other questions of political practice, and although Marchesi (2006, 14) claims that the term ABxÁŠ-sù indicates not elders but a class of officials, nevertheless the ABxÁŠ-sù are often present at moments of decision making although their views are not represented in the palace sources. This might suggest they have their own authority, if in a different sphere. It may be a situation of heterarchy where two parallel structures of power and/or authority exist side by side, in congruence, or perhaps more likely, in competition. It may be a function of dispersed morphology where ABxÁŠ-sù constituted part of the administration in locations removed from Ebla itself, and who were present in the city when issues concerning the polity as a whole were at stake. Rather than conclude that because at Ebla there are ‘elders’ (who are in any case a group not necessarily based on kinship at all), this was essentially, or originally, a tribal society (as does Stein 2004, 74, following Klengel 1992, 33), it is necessary to think more complexly about what the cooccurrence of king and elders means for ethos, configuration, practice and morphology. Equally, rather than assume that because there is a large and elaborate structure located on an acropolis that there is an authoritarian power structure present engendered by socio-economic hierarchies (class), it is necessary to think more complexly about the relationship between material culture and ethos, configuration, practice and morphology (e.g. Porter 2007b, 102–105). Conversely, considering the evidence in terms of ethos, configuration, practice and morphology allows us to think more complexly about the early polity. Polity cannot be understood through categorisation of a few components considered indicative of this way of being or that, only through the various ways a multiplicity of components interconnect.

References Adams, R. McC. 1966. The Evolution of Urban Society. Chicago, University of Chicago Press. Algaze, G. 1993. The Uruk World System: The Dynamics and Expansion of Early Mesopotamian Civilization. Chicago, University of Chicago Press. Arcari, E. 1988. The administrative organization of the city of Ebla. In H. Hauptmann and H. Waetzold (eds) Wirtschaft und Gesellschaft von Ebla: Akten der Internationalen Tagung, Heidelberg 4–7 November 1986, 125–130. Heidelberg, Heidelberger Orientverlag. Archi, A. 1982. About the organization of the Eblaite state. Studi Eblaiti 5, 201–220. Archi, A. 1986. Die ersten zehn Könige von Ebla. Zeitschrift für Assyriologie 76, 213–217. Archi, A. 1988a. Cult of the ancestors and tutelary god at Ebla. In Y. L. Arbeitmen (ed.) Fucus: A Semitic/Afrasian Gathering in Remembrance of Albert Ehrman, 109–110. CILT 58. Amsterdam and Philadelphia, John Benjamins. Archi, A. 1988b. Testi amministrativi: Registrazioni di metalli e tessuti (archivio L.2769). ARET 7. Rome, Missione archeologica italiana in Siria. Archi, A. 1990. Agricultural production in the Ebla region. Les Annales Archéologiques Arabes Syriennes 60, 50–55. Archi, A. 1992. The city of Ebla and the organization of the rural territory. Altorientalische Forschungen 19, 24–28. Archi, A. 2002. Jewels for the ladies of Ebla. Zeitschrift für Assyriologie 92(2), 161–199. Archi, A. 2005. The Head of Kura – the Head of ’Adabal. Journal of Near Eastern Studies 64(2), 81–100. Archi, A. and M.-G. Biga 2003. A victory over Mari and the fall of Ebla. Journal of Cuneiform Studies 55, 1–44. Astour, M. 1992. An outline of the history of Ebla (part 1). In C. Gordon (ed.) Eblaitica: Essays on the Ebla Archives and Eblaite Language 3, 3–82. Winona Lake (IN), Eisenbrauns. Biga, M.-G. 1995. Review of A. Archi, Five Tablets from the Southern Wing of Palace G. Journal of the American Oriental Society 115, 297–298. Biga, M.-G. and F. Pomponio 1987. Iš’ar-Damu, roi d’Ebla. Nouvelles assyriologiques brèves et utilitaires. Nouvelles assyriologiques brèves et utilitaires 4, 60–61. Blanton, R. 1998. Beyond centralization: Steps toward a theory of egalitarian behavior. In G. Feinman and J. Marcus (eds) Archaic States, 135–172. Santa Fe, Schools of American Research Press. Bonechi, M. 1997. Lexique et ideologie royale à l’époque protosyrienne. MARI 8, 477–535. Cooper, L. 2006. Early Urbanism on the Syrian Euphrates. New York, Routledge. Cooper, L. 2007. Early Bronze Age burial types and socio-cultural identity within the northern Euphrates valley. In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 55–70. Levant Supplementary Series 5. Oxford, Oxbow. Campbell, R. 2007. Blood, Flesh and Bones: Kinship and Violence in the Social Economy of the Late Shang. Unpublished Ph.D. thesis, Harvard University.

9. From kin to class – and back again! Changing paradigms of the early polity Dohmann-Pfälzner, H. and P. Pfälzner 1996. Untersuchungen zur Urbanisierung Nordmesopotamiens im 3. Jt. v. Chr.: Wohnquartierplanung und städtische Zentrumsgestaltung in Tall Chuera. Damaszener Mitteilungen 9, 1–13. Dobres, M.-A. and J. Robb 2000. Agency in archaeology: Paradigm or platitude? In M.-A. Dobres and J. Robb (eds) Agency in Archaeology, 3–17. London, Routledge. Durand, J.-M. 1992. Unité et diversités au Proche-Orient à l’époque Amorrite. In D. Charpin and F. Joannès (eds) La circulation des biens, des personnes et des idées dans le Proche-Orient ancien, 97–128. CRRAI 38. Paris, Éditions Recherche sur les Civilisations. Earle, T. 1997. How Chiefs Come to Power: The Political Economy in Prehistory. Stanford, Stanford University Press. Evans-Pritchard, E. E. 1940 [reprinted 1969]. The Nuer: A Description of the Modes of Livelihood and Political Institutions of a Nilotic People. New York and Oxford, Oxford University Press. Feinman, G. and J. Marcus (eds) 1998. Archaic States. Santa Fe (NM), Schools of American Research Press. Flannery, K. 1999. Process and agency in early state formation. Cambridge Archaeological Journal 9(1), 3–21. Fowles, S. 2002. From social type to social process: Placing ‘tribe’ in a historical framework. In W. Parkinson (ed.) The Archaeology of Tribal Societies, 13–33. Ann Arbor, International Monographs in Prehistory. Fronzaroli, P. 1992. The ritual texts of Ebla. In P. Fronzaroli (ed.) Literature and Literary Language at Ebla, 163–185. Quaderni di Semitistica 18. Florence, Università di Firenze. Fronzaroli, P. 1993. Testi rituali della regalità (Archivio L.2769). ARET 9. Rome, Missione Archaeologica Italiana in Siria. Fronzaroli, P. 2003. Testi di Cancelleria. I Rapporti can le Città. Archivi reali di Ebla : Testi XIII. Rome, Missione Archaeologica Italiana in Siria. Giddens, A. 1981. A Contemporary Critique of Historical Materialism. Vol. 1: Power, Property and the State. Berkeley: University of California Press. Hempelmann, R. n.d. a. Domestic cult of the ancestors in Early Bronze Age Syria. Manuscript in preparation. Hempelmann, R. n.d. b. Die Ausgrabungen im Bereich K. In J.-W. Meyer (ed.) Ausgrabungen in Tell Chu ra in Nordostsyrien II: Vorbericht der Grabungskampagnen 1998–2004. Manuscript in preparation. Holy, L. 1996. Anthropological Perspectives on Kinship. London, Pluto. Johnson, G. 1978. Information sources and the development of decision-making organizations. In C. L. Redman, M. J. Berman, E. V. Curtin, W. T. Langhorne, Jr., N. M. Versaggi and J. C. Wanser (eds) Social Archaeology: Beyond Subsistence and Dating, 87–112. New York, Academic Press. Klengel, H. 1992. Syria, 3000–300 B.C.: A Handbook of Political History. Berlin, Akademie. Marchesi. G. 2006. LUMMA in the Onomasticon and Literature of Ancient Mesopotamia. History of the Ancient Near East 10. Padova, Sargon. McCorriston, J. 1997. The fiber revolution: Textile intensification, alienation and social stratification in ancient Mesopotamia. Current Anthropology 38(4), 517–549.

77

Meskell, L. 2005. Introduction: Object orientations. In L. Meskell (ed.) Archaeologies of Materiality, 1–17. Oxford, Blackwell. Michalowski, P. 1988. Thoughts about Ibrium. In H. Hauptmann and H. Waetzoldt (eds) Wirtschaft und Gesellschaft von Ebla: Akten der Internationalen Tagung, Heidelberg 4–7 November 1986, 267–277. Heidelberg, Heidelberger Orientverlag. Miller, D. 2005. Materiality: An introduction. In D. Miller (ed.) Materiality, 1–50. Durham (NC): Duke University Press. Morgan, L. 1877 [reprinted 1985]. Ancient Society: or, Researches in the Lines of Human Progress from Savagery through Barbarism to Civilization. Classics of Anthropology. Tuscon, University of Arizona Press. Peletz, M. 1995. Kinship studies in late twentieth-century anthropology. Annual Review of Anthropology 24, 343–372. Peltenburg, E. 1999. The living and the ancestors: Early Bronze Age mortuary practices at Jerablus Tahtani. In G. del Olmo Lete and J.-L. Montero Fenollós (eds) Archaeology of the Upper Syrian Euphrates: The Tishrin Dam Area, 427–442. Aula Orientalis Supplementa 15. Barcelona, Editorial Ausa. Peltenburg, E. 2007a. New perspectives on the Carchemish sector of the Middle Euphrates River Valley in the 3rd millennium BC. In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 1–24. Levant Supplementary Series 5. Oxford, Oxbow. Peltenburg, E. 2007b. Enclosing the ancestors and the growth of socio-political complexity in Early Bronze Age Syria. In G. Bartoloni and M. G. Benedettini (eds) Atti del Convegno Internazionale ‘ Sepolti tra i vivi: Evidenza ed interpretazione di contesti funerari in abitato’, Roma, 26–29 April, 2006, 91–123. Scienze dell’Antichità 14(1). Rome, Università degli studi di Roma ‘La Sapienza’. Pollock, S. 1991. Of priestesses, princes and poor relations: The dead in the Royal Cemetery of Ur. Cambridge Archaeological Journal 1(2), 171–189. Porter, A. 2000. Mortality, Monuments and Mobility: Ancestor Traditions and the Transcendence of Space. Unpublished Ph.D. thesis, University of Chicago. Porter, A. 2002a. The dynamics of death: Ancestors, pastoralism and the origins of a third millennium city in Syria. Bulletin of American Schools of Oriental Research 325, 1–36. Porter 2002b. Communities in conflict: Death and the contest for social order in the Euphrates River valley. Near Eastern Archaeology 65, 156–173. Porter, A. 2007a. Evocative topography: Experience, time and politics in a landscape of death. In G. Bartoloni and M. G. Benedettini (eds) Atti del Convegno Internazionale ‘ Sepolti tra i vivi: Evidenza ed interpretazione di contesti funerari in abitato’, Roma, 26–29 April, 2006, 71–90. Scienze dell’Antichità 14(1). Rome, Università degli studi di Roma ‘La Sapienza’. Porter, A. 2007b. You say potato, I say …typology, chronology and the origin of the Amorites. In C. Marro and C. Kuzucuoglu (eds) Sociétés humaines et changement climatique à la fin du Troisième Millénaire: une crise a-t-elle eu lieu en HauteMésopotamie? 69–115. Varia Anatolica XVIII. Paris, de Boccard. Porter, A. 2009. Beyond dimorphism: Ideologies and materialities of kinship as time-space distanciation. In J. Szuchman (ed.) Nomads, Tribes, and the State in the Ancient Near East: Cross-

78

Anne Porter

Disciplinary Perspectives, 199–223. Oriental Institute Seminars 5. Chicago, Oriental Institute of the University of Chicago. Sahlins, M. D. 1968. Tribesmen. Englewood Cliffs (NJ), PrenticeHall. Sasson, J. 1998. The king and I: A Mari king in changing perceptions. Journal of the American Oriental Society 118(4), 453–470. Schwartz, G. 2007. Status, ideology and memory in thirdmillennium Syria: “Royal” tombs at Umm al-Marra. In N. Laneri (ed.) Performing Death: Social Analyses of Funerary Traditions in the Ancient Near East and Mediterranean, 39–68. Oriental Institute Seminars 3. Chicago, Oriental Institute of the University of Chicago. Service, E. 1962. Primitive Social Organization. New York, Random House. Smith, A. T. 2003. The Political Landscape: Constellations of Authority in Early Complex Polities. Berkeley, University of California Press. Stein, G. 1994. Economy, ritual and power in ‘Ubaid Mesopotamia. In G. Stein and M. Rothman (eds) Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity, 35–46. Monographs in World Archaeology 18. Madison, Prehistory Press. Stein, G. 1998. Heterogeneity, power, and political economy: Some current research issues in the archaeology of Old World complex societies. Journal of Archaeological Research 6(1), 1–44. Stein, G. 2004. Structural parameters and sociocultural factors in the economic organization of northern Mesopotamian urbanism in the third millennium B.C. In A. G. Feinman and L. Nichols (eds) Archaeological Perspectives on Political Economies, 61–79. Salt Lake City, University of Utah Press. Steinkeller, P. 1999. Land tenure conditions in third millennium Babylonia: The problem of regional variation. In M. Hudson and B. Levine (eds) Privatization in the Ancient Near East and Classical World, vol. 1, 289–329. Peabody Museum of Archaeology and Ethnology Bulletin 5. Cambridge, Harvard University Press. Steinkeller, P. 2004. A history of mashkan-shapir and its role in the kingdom of Larsa. In E. Stone and P. Zimansky (eds) The Anatomy of a Mesopotamian City, 26–42. Winona Lake (IN), Eisenbrauns.

Stephen, F. and E. Peltenburg 2002. Scientific analyses of Uruk ceramics from Jerbalus Tahtani and other Middle-Upper Euphrates sites. In J. N. Postgate (ed.) Artefacts of Complexity: Tracking the Uruk in the Near East, 173–190. Cambridge, British School of Archaeology in Iraq. Stone, E. 1999. The constraints on state and urban form in ancient Mesopotamia. In M. Hudson and B. Levine (eds) Urbanization and Land Ownership in the Ancient Near East, 203–227. Peabody Museum of Archaeology and Ethnology Bulletin 7. Cambridge, Harvard University Press. Surenhagen, D. 1986. The dry farming belt: The Uruk period and subsequent developments. In H. Weiss (ed.) The Origins of Cities in Dry-Farming Syria and Mesopotamia in the Third Millennium B.C., 7–43. Guilford, Four Quarters. Trigger, B. 2003. Understanding Early Civilizations. Cambridge, Cambridge University Press. Tsukimoto, A. 1985. Untersuchungen zur Totenpflege (kispum) im alten Mesopotamien. Alter Orient und Altes Testament 216. Neukirchen-Vluyn, Neukirchener Verlag. Van der Mieroop, M. 2004. A History of the Ancient Near East ca. 3000–323 BC. Oxford, Blackwell. Wattenmaker, P. 1994. Political fluctuations and local exchange systems in the ancient Near East: Evidence from the Early Bronze Age settlements at Kurban Höyük. In G. Stein and M. Rothman (eds) Chiefdoms and Early States in the Near East: The Organizational Dynamics of Complexity, 193–208. Monographs in World Archaeology 18. Madison, Prehistory Press. Whitley, J. 2002. Too many ancestors. Antiquity 76, 119–26. Yoffee, N. 1993. Too many chiefs? (or, safe texts for the ’90s). In N. Yoffee and A. Sherratt (eds) Archaeological Theory: Who Sets the Agenda?, 60–78. Cambridge, Cambridge University Press. Yoffee, N. 1997. The obvious and chimerical: City-states in archaeological perspective. In D. Nichols and T. Charlton (eds) The Archaeology of City States: Cross-Cultural Approaches, 255–263. Washington DC, Smithsonian Institution. Yoffee, N. 2005. Myths of the Archaic State. Cambridge, Cambridge University Press. Zagarell, A. 1986. Trade, women, class and society in ancient western Asia. Current Anthropology 27, 415–430.

10 DIFFERENT MODELS OF POWER STRUCTURING AT THE RISE OF HIERARCHICAL SOCIETIES IN THE NEAR EAST: PRIMARY ECONOMY VERSUS LUXURY AND DEFENCE MANAGEMENT Marcella Frangipane

The Near East, which has been the theatre of the first emergence of politically centralised societies, clearly shows different evolutionary pathways in this process, leading to very different social, economic and political systems, depending on the original structural features of the societies in which this process occurred and the conditions of the environment. One of the crucial distinguishing factors was the presence or absence of an associated phenomenon of urban growth. There are pronounced differences between urban and nonurban systems of central political governance, particularly entailing a different degree and a different need for integration between the various social and economic components. Integration and interdependence, which grow stronger as territories become more urbanised, engender a need for central authority to play a mediating role between the parties in the economy conduction. In non-urbanised societies, conversely, the rise of a central political power is usually less intrinsically necessary for the functioning of the socio-economic system. The discriminating factors for the formation of highly urbanised and centralised systems appear to be mainly of two types: 1) environmental: the growth of large concentrations of people in urban settlements is only possible if there are conditions for a highly productive agriculture, generating large and continuously increasing surpluses; 2) social: a system of social and kinship relations emphasising the separate and competitive role of individual household units, fostering the legitimation of differences within which higher status members of the society can emerge. It is this intrinsic status of social privilege, accepted by all, which can deeply legitimise the leaders appointed to wield political power and, above all, administer the wealth of the community, even perhaps partly appropriating some

of it for their own maintenance. One structure which clearly exemplifies this type of society is the conical clan system (Kirchhof 1959). The various combinations and dialectical relations between these two categories of factors (social and environmental) produced different economic and political systems in the prehistoric Near East, which resulted in very different types and stability of central power structures. In this paper I shall consider three well known examples which appear to be very meaningful in this respect.

Southern Mesopotamia: The Elite Control of Staple Economy In southern Mesopotamia, from the first occupation of the alluvial plain around the end of the 7th to the beginning of the 6th millennium cal BC, both of the conditions mentioned above – great agriculture potential and a social/kinship system containing the germ of hierarchies – seem to have been met. The territory is extremely varied from the point of view of subsistence resources (large areas of land for extensive cereal farming, areas for horticulture, range-lands, coasts and rivers for fishing), but above all it has extremely high agricultural potential. However, there were also serious risk factors due to the inclement climate (hot and arid), which may have caused some drought years and soil salinisation (Adams 1966, 48–59; 1981, 1–26; Pollock 1999, 28–44). On the whole, the climate was fairly homogeneous, but on the other hand the specific conditions in each zone differed because of the variability of the sources and courses of water, as Adams has clearly shown (1981, 3–11, 14–22). All this must have helped the formation of societies

80

Marcella Frangipane

which were both very cohesive and coordinated in order to address common problems, but at the same time with strong internal competition to hoard the best resources. Both factors may have led to the need for central institutions or chiefs to mediate the circulation of subsistence products between areas of economic specialisation, and to deal effectively with the hazards (Algaze 2008, 40–68). I would suggest that the Ubaid society in Lower Mesopotamia already showed this structure, perhaps dating back to its earliest phases (Stein 1994). The size and layout of the houses at Tell el Oueili (Huot 1989; 1991) from the earliest periods of occupation reveal the organisation of communities into large, quite separate and distinct households, perhaps large families, which were subsequently to characterise later Mesopotamian society. The central and southern Mesopotamian settlements of the 6th millennium show little sharing of domestic activities in the open spaces between the houses or in communal facilities, but in the south there is the appearance of important buildings for public ceremonial events (Bernbeck 1995, 14–17; Frangipane 2007a). While we know very little about the specific function of the ‘temples’ at Eridu, their architecture and content suggest that they were in any event public places where activities, probably related to a kind of ritualised redistribution of food, were performed (Safar et al. 1981). It is possible that the people performing public activities with social and economic functions were preeminent individuals or households who were recognised as such by their communities by virtue of having been created by a conical or pyramidal social/kinship structure. Possible support for this hypothesis can be seen at the small village of Tell Abada in the Hamrin region (Jasim 1989), which despite being outside the southern alluvial plain is almost the only Ubaid site extensively excavated so far. There were no temples here, and the houses appear to have followed the same general architectural model and performed the same kind of activities, though different in size. But one of these houses shows peculiar traits in terms of architecture and associated evidence (Jasim 1989; Forest 1996, 59) and might be interpreted as the residence of a preeminent family. This house, moreover, was probably seen as symbolically representing the whole community, as is suggested by the concentration of numerous children’s burials specifically under its floor. There may have been another house by the side of this one, which was also distinct from the majority of the village houses in terms of size and architectural features, suggesting a possible stratification of social status typical of kinship and hierarchical genealogical systems (Frangipane 1996, 124). It is this model of society that became fully established in the Uruk period. The specific features of 4th millennium Lower Mesopotamia seem to have been a great development of the basic features just described for the Ubaid society. But by expanding the privileges in the economic sphere,

the growth itself of hierarchical relations transformed society into something new. The final outcome was very strong economic centralisation with regular flows of goods and labour towards the places where public activities were performed. The centralisation of resources may have taken place by obtaining tributes and/or concentrating means of production, such as land and livestock (ideologically perhaps the land and livestock of the gods administered by their representatives), which could only be productive by making people work for the central authorities. This system may have produced accumulation of wealth in terms of staple goods, basically food. But the basis for the functioning of this type of centralised economy was a substantial outlay of resources – especially foodstuffs – to the population to compensate their work. ‘Redistribution’, which has been traditionally considered the parallel aspect of ‘centralisation’, was not always associated with a real accumulation of goods in the hands of central authorities (cf the Neolithic redistribution in communal storehouses; Akkermans and Duistermaat 1996), but in the early state systems it indeed became an effective tool for the investment in labour by the elites, acting as a sort of entrepreneur (Frangipane 2000a). The increasingly more sophisticated procedures of administrative control, which were further developed to manage this intense movement of goods and activities by producing the emergence of an expanding class of officials, made it possible to broaden considerably the sphere of power exercised by the elites, also in territorial terms. For a long time, centralised goods continued to be mainly food and means of subsistence production (Yoffee 1995; Frangipane 1996; Liverani 1998; Pollock 1999, 79–116). This can also be inferred from the contents of the numerous Uruk pictographic tablets (Nissen 1986; Nissen et al.1993), and the main basis for the expanding wealth of elites continued to be the control of labour. The distribution of rations became the main feature of Uruk society, and the generalised distribution of meals in public (and, to a lesser degree, also private) places, as suggested by the exponential increase in mass-produced bowls, points to the provision of labour services by workers, and hence to increasing inequalities. This flow of goods and labour certainly attracted more people to the places where they were concentrated, boosting urbanisation. This was made possible by the potential for expanding agriculture in Lower Mesopotamia, which was able to maintain a large urban population. But the crucial factor for its development was, in my opinion, the structure of Mesopotamian society itself, which fostered the unequal organisation of production and consumption and the rise of a solid, centrally directed system. The assumed stratified structure in terms of large households of different social status would have helped to maintain and strengthen the hierarchical relationships between families and communities, giving a solid basis to social elites, who could embody

10. Different models of power structuring at the rise of hierarchical societies in the Near East authority and, thanks to competition for resources, gradually acquire economic privileges. The ideologically ‘natural’ character of these inequalities must have found its best expression in the ceremonial aspects of public activities, which were largely concentrated in monumental, probably sacred areas. The close relationship of the social elites with the gods, typical of the hierarchical kinship systems, gave them the full right to exercise their authority. The undisputed recognition of a common ‘authority’ would also have made it possible to establish a stable network of relations between villages and the emerging urban centres throughout the territory, thereby guaranteeing the flow of food and labour, which alone would have made it possible for the full urban model to become established and to gradually and constantly develop.

From Egalitarian to Hierarchical in Northern Greater Mesopotamia: A Deep Structural Change The organisational structure of the communities in northern Mesopotamia and eastern/south-eastern Anatolia from the earliest occupation of Jezira in the 7th millennium cal BC until the establishment of the Halaf culture in the 6th millennium appeared to be radically different (Frangipane 1996, 53–87; 2007a, 154–164). Smaller villages with a mixed agricultural, livestock, and hunting economy, in which the role of individual households was hardly recognisable in the architectural structures of the settlements and were therefore not primary in social and perhaps also economic terms, often had large buildings for common storage, evidencing collective ways of storing and redistributing food supplies and surpluses (Akkermans 1996; Akkermans and Duistermaat 1996). This appears as a society with small units scattered across a wide territory with almost no evidence of hierarchies, and engaged in different subsistence activities varying from one zone to another, sometimes also with specialisation and the resultant need for mutual cooperation and collective ways of managing food (Verhoeven 1999, 203–220; Frangipane 2007a, 154–161). These must therefore have been very cohesive, economically integrated tribal and basically egalitarian communities, probably composed of various and distinct socio-economic components. They had no temples or preeminent buildings, and they showed no evidence of social stratification. This situation changed radically with the spread in the north of the Ubaid cultural and social model during the course of the 5th millennium. This is not the appropriate place to examine why and how this southern model spread northwards (Breniquet 1996). But the outcome that can be seen throughout the area of Upper Mesopotamia and the Middle and Upper Euphrates as far as the eastern Anatolian mountains was the emergence of societies with

81

elites who wielded their power in expanding centres with administrative control over food and labour, above all in monumental public buildings. In several cases these were sacred or ceremonial buildings. Tepe Gawra (Tobler 1950; Rothman 2002) is the site showing the greatest evidence of this newly arising situation around the beginning of the 4th millennium (Late Chalcolithic 1–2). The system of competition between households also seems to have emerged in the northern regions, as evidenced by the development of the mass production of bowls and the adoption of the administrative control system based on sealing goods in the private sphere, as well as the public one. The increasingly widespread use of sealings and seals in the level XII and level XI houses at Tepe Gawra and in the houses at the Late Ubaid site of Degirmentepe in Anatolia (Esin 1989; 1994), together with the widespread adoption of mass-produced bowls, indicates transactions between households and not only the management of goods in the public places in which a central authority was becoming institutionalised. The Gawra XII White Room House, which was larger than the others, had more sealings than the others, and was also characterised by the presence of numerous child burials under the floors, very closely resembles the ‘house of the chieftain’ in the southernmost Ubaid village of Tell Abada. This suggests the presence of preeminent families there with community leadership functions. This Mesopotamian-type system became fully entrenched by around the middle of the 4th millennium (Late Chalcolithic 3) when the archaeological documentation reveals sites that were probably dominant over their region, such as Tell Brak and Arslantepe, which also had imposing monumental public architecture. Tell Brak appears to have been a real and large centre before the mid-4th millennium (Oates and Oates 1997; Emberling and McDonald 2003; Oates et al. 2007) and must have had high status figures exercising a political control over the area; Arslantepe had a preeminent zone on the highest part of the ancient mound with buildings of the elites, probably residential buildings, with fully local architecture and a large temple or ceremonial building with a tripartite Mesopotamian-type layout used mainly for distributing meals under administrative control (Frangipane 2000b; 2003). Both these sites show material cultures which were basically independent from any direct southern Mesopotamian influence. In other words, there seems to have been a process of assimilating and appropriating the southern socio-economic model, and adapting it to northern traditions and conditions. The new and different expansion of southern groups into the north during the Middle (Late Chalcolithic 4) and Late (Late Chalcolithic 5) Uruk phases took place on this cultural and socio-economic substrate (once again, I shall not go into the reasons and nature of this phenomenon and the related debate). What occurred in the latter half of the 4th millennium throughout the whole region was

82

Marcella Frangipane

a considerable development of the centralised economic and redistribution system essentially based, according to the archaeological evidence, on the control of the primary commodities and labour. Here again, this management system had powerful ideological legitimation though at the end of the period, together with an evident increase in the power of the central authorities, a greater economic, administrative and perhaps also political independence from the religious and ceremonial sphere seems to appear in certain Upper Euphrates contexts. Evidence of this is the development of the so-called ‘palatial’ system at the end of the 4th millennium at Arslantepe where public storerooms with redistribution activities and a sophisticated system of administrative control over the centralised circulation of goods were architecturally and functionally separated from the religious ‘temple’ sphere, demonstrating that they were distinctly autonomous in terms of their management (Frangipane 1997; Frangipane et al. 2007). These were therefore societies dominated by political and economic leaders, in some case very powerful, who were able to wield authority over the staple economy of the population of the surrounding villages and concentrate the means of production and labour. Their wealth seems to consist mainly of staple products though it certainly also enabled them, as it did in Mesopotamia, to finance other activities. But the main difference between what were similar centralised systems in the north and the south of Greater Mesopotamia was their different capacity to sustain a fully fledged urbanisation process (Algaze 2008). I believe that urban growth was accomplished only in the Khabour region (Oates and Oates 1993; Wilkinson and Tucker 1995; Gibson et al. 2002; Oates 2002; Ur 2002; Oates et al. 2007) where the environmental conditions made it possible for a profitable and expanding agriculture that could support a large urban population. Neither the Middle to Upper Euphrates or the Upper Tigris underwent any real process of urbanisation, even during the 3rd millennium (Cooper 2006b). The absence of urbanisation led to a less stable power system, which remained more vertically dominated and probably less radically linked to the stratified structure of a society in need of mediation and integration. The northern elites probably remained rather detached from the people who, in the majority of regions except for the Khabour area, seem to have continued to be basically rural and pastoralist. Once again, Arslantepe reflects this situation very well. It became a powerful religious, political, administrative and economic centre with a huge capacity for centralising resources and labour, but it still retained the dimensions of the previous period and indeed may have further reduced them. There was no urbanisation, and we must therefore presume that the vast numbers of people engaged in the ‘palace’ activities were not resident there, apart from perhaps a few of them. Rather they came from the villages and the

different population groups living in or visiting the plain, including transhumant pastoralists who were certainly present in the Malatya area at the end of the 4th millennium (Palumbi 2003; 2008; Frangipane and Palumbi 2007). Further support to this interpretation comes from the iconographies of the more than two hundred seals that were used on the thousands of sealings found in the palace at Arslantepe. These iconographies are extremely varied and show that widely differing traditions, styles and motifs were all used together, despite belonging in general to the northern glyptic tradition. This probably suggests that different sections of the population, and perhaps also different ethnic groups with their own economic and administrative identities, converged on the palace to perform the activities managed there (Frangipane et al. 2007, 475–477). Due to the lack of urbanisation, the production practices of the people in the Malatya region did not substantially change with the establishment of a kind of early state power management. Even though the villages would certainly have had to pay tribute and provide labour to the central authorities, the way of life in the villages was probably not substantially affected. This, together with the likely pressure exerted by pastoralists, who were organising themselves throughout the territory, and the parallel expansion of the Transcaucasian groups, eventually brought the whole system down. This collapse affected the whole of the Middle and Upper Euphrates Valley, coinciding with the well documented collapse of the Uruk system of relations. A similar collapse did not, in my opinion, occur in the Khabour and eastern Jezira where the urban structures seem to have undergone a re-organisational ‘crisis’ and in fact expanded further during the succeeding Ninevite 5 period (Schwartz 1994; Matthews 2003); nor did the political life and managerial system of central authority collapse there. The Middle and Upper Euphrates groups, who had lived in areas where farmland was in less plentiful supply and where the economy was more mixed, were perhaps less deeply affected by the introduction of southern models of society and therefore basically kept their traditional clan/ tribe-based structure, even when the system of economic centralisation and redistribution became established in the 4th millennium (Cooper 2006a; 2006b; Frangipane 2007b; Peltenburg 2007, 11–18). Whereas the urbanised Ninevite society continued to develop, in the Euphrates regions the economic centralisation system utterly disappeared, and power took other forms of a more political-defensive type, more symbolically linked to a ‘warrior’ ideology. The wealth of the dominant classes was now displayed in funerary rituals and in an abundance of metal items and weaponry (Peltenburg 1999; Sertok and Ergeç 1999; Frangipane et al. 2001; Porter 2002; Schwartz et al. 2006), symbolical and ideological aspects that were almost wholly absent from the cultures of Uruk inspiration.

10. Different models of power structuring at the rise of hierarchical societies in the Near East

Western Anatolia: The Emergence of Paramount Political Elites A very different type of society and power structuring is recognisable in the early ‘urban’ societies of the Early Bronze Age in western/central Anatolia. The first fully sedentary communities based on an agricultural and livestock economy occupying the southern region of central/ western Anatolia in the Neolithic seem to have been cultural units with a limited geographical extension, characterised by more or less agglutinated villages made up of fairly small dwellings with distinctive ground plans in each sub-region. Domestic equipment was situated almost entirely within the houses or in internal courtyards (central Anatolia) or directly linked to the dwellings, adjacent to the external walls (Lakes Region) as they were external projections of the individual domestic units (Özdoğan and Başgelen 2007). In the case of a village like Çatal Höyük, the open spaces were exclusively internal yards belonging to groups of houses arranged around them in a compact formation (Mellaart 1967; Hodder 2006). Storage was domestic. Buildings were not found to have been used for collective activities, and no public places have been found where authority was exercised or rituals carried out on the part of the whole community. The symbolic and ritual aspects were scattered and entrusted to individual domestic units, with various expressions ranging from very common figurines to extraordinary mural paintings and plastic reliefs found in many buildings at the site (Matthews 2002; Özdoğan 2002a, 257; Verhoeven 2002). These villages appear to have been based on a solid egalitarianism, with each household unit exhibiting much greater economic independence than was the case in the Neolithic societies of Jezira. However, the sense of the community as a whole seems to have been stronger than in southern Mesopotamia, as is suggested by the dense packing of the buildings in the settlements and the cohesive function of the strict standards set in the household rituals and behaviours. Unfortunately, the data on the Chalcolithic period in these regions of Anatolia are still scarce though there is evidence of a possible continuous transition to the Early Chalcolithic society in central Anatolia (Gérard 2002, 108–109). A developmental link with this model of society can be recognised in the early hierarchical societies of 3rd millennium western Anatolia. Early Bronze settlements in this region continued to remain quite small in terms of Mesopotamian urban standards, and the basic economic autonomy of the domestic units also seems to be evidenced from the lack of any central storage of foodstuffs and any evidence of regular administrative practices connected with centralised control of the customary circulation of goods (Özdoğan 2002b). Storerooms or concentrations of handicraft activities in the places of power may have existed in the Central Complex of Karatas (Mellink 1966; 1972;

83

1973; Warner 1994, 177–179) or in the citadel at Troy I and II (Blegen et al. 1950; 1951; Korfmann 2006), but no redistribution activities or economic transactions involving mobilisation of staple products to finance these activities were linked to them, at least judging from the available archaeological evidence. The storage places in central building complexes seem to have simply been deposits of goods for the elites in the form of accumulation of wealth, as in Troy II, or merely food supply in their residences, as it seems was the case at Küllüoba (Efe 2003). The rare cases of possible concentration of handicraft activities in the places of power, as may be inferred by the numerous spindle-whorls found in the area of the Central Complex at Karatas, probably indicate some sort of privileged relationships of the craftsmen with the central elites rather than implying a centralised system of production. The development of handicraft must indeed have involved the communities as a whole, as evidenced from the numerous workshops found in various settlements, such as Thermi, Poliochni and Küllüoba. Though there is therefore evidence to indicate forms of wealth accumulation by the elites, this wealth was not mobilised and regularly put into circulation as part of an economic interaction with the community. The lack of mass produced bowls is also interesting in this connection. There was also a lack of any very prominent central shrines with which the whole community could identify and refer to while evidence of cultic practices or forms of rituals, where they are documented at all, refer only to domestic, or slightly larger, spheres, as was possibly the case at Beycesultan Levels XVI–XIV (Lloyd and Mellaart 1962, 36–55). The funerary customs exhibited a wide range of rituals (cist graves, pithos graves, inhumation, cremation, etc.), which were only partly regionally diversified and were not matched by significant differences in terms of the wealth of the furnishings or the emphasising of social status. This diversity may have been linked to a variety of ethnic and cultural components, which were probably related to the similarly variegated picture in the Neolithic. These communities seem to have been based on a clan structure that was not very stratified in terms of complex social differentiation. Socio-political elites and preeminent social figures, however, certainly existed in the Early Bronze Age, representing themselves and their right to exercise authority, living in places that were kept separate from the rest of the community, and manifesting themselves in the architectural form of fortified ‘citadels’ or upper towns built in the centre of the inhabited areas. This development can be observed during the 3rd millennium at several sites, such as Troy, Karatas, and Limantepe on the coast or Küllüoba inland (Erkanal 1996; Efe 2003). In the most prominent of these places, the citadel of Troy II, there is evidence of concentrations of luxury objects, particularly those made of

84

Marcella Frangipane

metal. The big boost given to metallurgy in these societies (Efe 2002) probably came from the incentives given to these activities by the elites, who were the main purchasers of these goods, and possibly also from the protection they were able to guarantee to trade routes. The fairly widespread use of fortifications and the fortified compact arrangement of some villages (e.g. Küllüoba early phase and Demircihüyük; see Korfmann 1983) suggest an endemic conflict in these societies, which appears to confirm the hypothesis that they were small units traditionally competing with each other, now perhaps mostly along the trade routes. These were therefore rather vertically structured societies with groups exercising political power over the small communities and over quite limited territories but which, perhaps through alliances, managed to protect and favour trade by activating a kind of ‘wealth finance’ system whose features have been clearly described by D’Altroy and Earle (1985). The spread of the so-called ‘Troy culture’ over a vast territory, which extended to the south-east as far as Cilicia (Mellink 1989; Efe 2007) and contained clearly distinct yet closely correlated regional units, fits very well with this model.

Concluding Remarks In this attempt to emphasise the salient and distinguishing features of the early hierarchical societies in the Near East, two main, contrasting models have emerged: The first were the Mesopotamian-type societies, with elites who centralised primary goods and labour, accumulated wealth in the form of staple products, and established a system of regular reinvestment in various activities by redistributing these staple products, or part of them, to increasingly broader sections of the population in the form of remuneration for their labour. The centralised organisation in this case became almost an ‘entrepreneurial’ system, creating a very close interdependency between all of its social components. The consequences of this system were (conditions permitting) urbanisation, the emergence of sophisticated administrative procedures, and bureaucracy. The origin of the elites and their privileges were usually deeply entrenched in the socioeconomic system. It was this model that led to the birth of the state. The second type was that of western Anatolian societies with political/military type leaders who seem to have managed small territories and who were probably viewed ideologically in terms of their role as the defenders and representatives of the community. Interference by these elites in the basic production system of the general population appears to have been virtually non-existent, whereas elites may have played a very important role in protecting the trading routes for raw materials and in supporting craftsmen.

Their financial system was therefore based essentially on handicraft items and luxury goods, above all metals, and their political role was probably linked to their ability to guarantee access to the supply of raw materials in a weakly hierarchical society made up of small territorial units in a basically mountainous environment. This type of society did not evolve in an urban direction and did not exhibit features of the early state. The historical outcomes of these two trajectories, regardless of the many regional differences within each of them, were also very different, and it was only the expansionism of the later large empires that eliminated or narrowed the differences.

References Adams, R. McC. 1966. The Evolution of Urban Society. Chicago, Aldine. Adams, R. McC. 1981. Heartland of Cities: Surveys of Ancient Settlement and Land Use on the Central Floodplain of the Euphrates. Chicago, University of Chicago Press. Algaze, G. 2008. Ancient Mesopotamia at the Dawn of Civilization: The Evolution of an Urban Landscape. Chicago, University of Chicago Press. Akkermans, P. M. M. G. (ed.) 1996. Tell Sabi Abyad: The Late Neolithic Settlement. 2 vols. Leiden, Nederlands HistorischArcheologisch Instituut te Istanbul. Akkermans, P. M. M. G. and K. Duistermaat 1996. Of storage and nomads: The sealings from the Late Neolithic Sabi Abyad. Paléorient 22(2), 17–32. Bernbeck, R. 1995. Lasting alliances and emerging competition: Economic developments in early Mesopotamia. Journal of Anthropological Archaeology 14, 1–25. Blegen, C. W., J. L. Caskey and M. Rawson 1950. Troy I: The First and Second Settlement. Princeton, Princeton University Press. Blegen, C. W., J. L. Caskey and M. Rawson 1951. Troy II: The Third, Fourth and Fifth Settlement. Princeton, Princeton University Press. Breniquet, C. 1996. La disparition de la culture de Halaf. Paris, Éditions Recherche sur les Civilisations. Cooper, L. 2006a. The demise and regeneration of Bronze Age urban centers in the Euphrates Valley of Syria. In G. M. Schwartz and J. J. Nichols (eds) After Collapse: The Regeneration of Complex Societies, 18–37. Tucson, University of Arizona Press. Cooper, L. 2006b. Early Urbanism on the Syrian Euphrates. New York and London, Routledge. D’Altroy, T. N. and T. K. Earle 1985. Staple finance, wealth finance and storage in the Inka political economy. Current Anthropology 26(2), 187–206. Efe, T. 2002. The interaction between cultural/political entities and metalworking in western Anatolia during the Chalcolithic and Early Bronze Ages. In Ünsal Yalçin (ed.) Anatolian Metal II. Der Anschnitt 15, 49–65. Bochum, Deutsches BergbauMuseum.

10. Different models of power structuring at the rise of hierarchical societies in the Near East Efe, T. 2003. Küllüoba and the initial stages of urbanism in western Anatolia. In M. Özdoğan, H. Hauptmann and N. Basgelen (eds) From Villages to Cities: Studies presented to Ufuk Esin, 265–282. Istanbul, Arkeoloji ve Sanat Yayinlari. Efe, T. 2007. The theories of the ‘Great Caravan Route’ between Cilicia and Troy: The Early Bronze Age III period in inland western Anatolia. Anatolian Studies 57, 47–64. Emberling, G. and H. MacDonald 2003. Excavations at Tell Brak, 2001–2002. Iraq 65, 1–75. Erkanal, H. 1996. Early Bronze Age urbanization in the coastal region of western Anatolia. In Y. Sey (ed.) Housing and Settlement in Anatolia: A Historical Perspective, 70–82. Istanbul, Ege Yayinlari. Esin, U. 1989. An early trading center in eastern Anatolia. In K. Emre, B. Hrouda, M. Mellink and N. Özgüç (eds) Anatolia and the Ancient Near East, 135–141. Ankara, Türk Tarih Kurumu Basımevi. Esin, U. 1994. The functional evidence of seals and sealings of Degirmentepe. In P. Ferioli, E. Fiandra, G. G. Fissore and M. Frangipane (eds) Archives before Writing: Proceedings of the International Colloquium, Oriolo Romano, October 23–25, 1991, 59–81. CIRAAS 1. Rome, Ministero per i beni culturali e ambientali, Ufficio centrale per i beni archivistici. Forest, J.-D. 1996. Mesopotamie: L’apparition de l’etat. Paris, Éditions Paris-Méditerranée. Frangipane, M. 1996. La nascita dello Stato nel Vicino Oriente. Roma-Bari, Laterza. Frangipane, M. 1997. A 4th millennium temple/palace complex at Arslantepe-Malatya: North-south relations and the formation of early state societies in the northern regions of Greater Mesopotamia. Paléorient, 23(1), 45–73. Frangipane, M. 2000a. The development of administration from collective to centralized economies in the Mesopotamian world: The transformation of an institution from ‘system-serving’ to ‘self-serving’. In G. Feinman and L. Manzanilla (eds) Cultural Evolution: Contemporary Viewpoints, 215–232. New York, Kluwer Academic. Frangipane, M. 2000b. Origini ed evoluzione del sistema centralizzato ad Arslantepe: dal ‘Tempio’ al ‘Palazzo’ nel IV millennio a. C. ISIMU 3, 53–78. Frangipane, M. 2003. Developments in fourth millennium public architecture in the Malatya Plain: From simple tripartite to complex and bipartite pattern. In M. Özdoğan, H. Hauptmann and N. Başgelen (eds) From Villages to Cities: Studies Presented to Ufuk Esin, 147–169. Istanbul, Arkeoloji ve Sanat Yayinlari. Frangipane, M. 2007a. Different types of egalitarian societies and the development of inequality in early Mesopotamia. World Archaeology 39(2), 151–176. Frangipane, M. 2007b. The establishment of a Middle/Upper Euphrates EB I culture from the fragmentation of the Uruk world: New data from Zeytinli Bahçe Höyük (Urfa, Turkey). In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 122–141. Levant Supplementary Series 5. Oxford, Oxbow. Frangipane, M., G. M. Di Nocera, A. Hauptmann, P. Morbidelli, A. M. Palmieri, L. Sadori, M. Schultz and T. Schmidt-Schultz 2001. New symbols of a new power in a ‘royal’ tomb from

85

3000 BC Arslantepe, Malatya (Turkey). Paléorient 27(2), 105–139. Frangipane, M. and G. Palumbi 2007. Red-black ware, pastoralism, trade, and Anatolian-Transcaucasian interactions in the 4th–3rd millennium BC. In B. Lyonnet (ed.) Les Cultures du Caucase (VIe–IIIe millénaires avant notre ère): Leurs relations avec le Proche-Orient, 233–255. Paris, CNRS Éditions. Frangipane, M., P. Ferioli, E. Fiandra, R. Laurito, and H. Pittman 2007. Arslantepe Cretulae: An Early Centralised Administrative System Before Writing. Arslantepe vol. V. Rome, Università di Roma ‘La Sapienza’. Gérard, F. 2002. Transformations and societies in the Neolithic of central Anatolia. In F. Gérard and L. Thissen (eds) The Neolithic of Central Anatolia, 105–109. Istanbul, Ege Yayinlari. Gibson, McG., M. Maktash, J. A. Franke, A. Al-Azm, J. C. Sanders, T. Wilkinson, C. Reichel, J.. Ur, P. Sanders, Al Salameh, C. Hritz, B. Watkins and M. Kattab 2002. First season of SyrianAmerican investigations at Hamoukar, Hasekeh province. Iraq 64, 45–68. Huot, J.-L. 1989. Ubaidian villages of Lower Mesopotamia: Permanence and evolution from Ubaid 0 to Ubaid 4 as seen from Tell el Oueili. In E. F. Henrickson and I. Thuesen (eds) Upon This Foundation: The Ubaid Reconsidered, 19–42. Carsten Niebuhr Publications 10. Copenhagen, Museum Tusculanum Press. Huot, J.-L. (ed.) 1991. ‘Oueili, travaux de 1985. Paris, Éditions Recherche sur les Civilisations. Hodder, I. 2006. Çatalhöyük: The Leopard’s Tale. Revealing the Mysteries of Turkey’s Ancient ‘Town’. London, Thames & Hudson. Jasim, S. 1989. Structure and function in an Ubaid Village. In E. F. Henrickson and I. Thuesen (eds) Upon this Foundation: The Ubaid Reconsidered, 79–90. Carsten Niebuhr Publications 10. Copenhagen, Museum Tusculanum Press. Kirchhoff, P. 1959. The principles of clanship in human society. In M. H. Fried (ed.) Readings in Anthropology, 260–270. New York, Crowell. Korfmann, M. 1983. Demircihüyük: Die Agenise der Ausgrabungen 1975–78. Architecture, Stratigraphie und Befunde 1. Mainz on Rhein, Philipp von Zabern. Korfmann, M. (ed.) 2006. Troia: Archaeologie eines Siedlungshügels und seiner Landshaft. Mainz on Rhein, Philipp von Zabern. Liverani, M. 1998. Uruk, la prima città. Rome-Bari, Laterza. Lloyd, S. and J. Mellaart 1962. Beycesultan I: The Chalcolithic and Early Bronze Age Levels. London, British Institute of Archaeology at Ankara. Matthews, R. 2002. Homogeneity versus diversity: Dynamics of the central Anatolian Neolithic. In F. Gérard and L. Thissen (eds) The Neolithic of Central Anatolia, 91–97. Istanbul, Ege Yayinlari. Matthews, R. (ed.) 2003. Excavations at Tell Brak. Vol. 4: Exploring a Regional Centre in Upper Mesopotamia, 1994– 1996. Cambridge, British School of Archaeology in Iraq and McDonald Institute for Archaeological Research. Mellaart, J. 1967. Çatal Höyük: A Neolithic Town in Anatolia. London, Thames & Hudson. Mellink, M. 1966. Excavations at Karatas-Semayük in Lycia, 1965. American Journal of Archaeology 70, 246–255.

86

Marcella Frangipane

Mellink, M. 1972. Excavations at Karatas-Semayük and Elmali, Lycia, 1971. American Journal of Archaeology 76, 257–269. Mellink, M. 1973. Excavations at Karatas-Semayük and Elmali, Lycia, 1972. American Journal of Archaeology 77, 293–297. Mellink, M. 1989. Anatolian and foreign relations of Tarsus in the Early Bronze Age. In K. Emre, M. Mellink, B. Hrouda and N. Özgüç (eds) Anatolia and the Ancient Near East. Studies in Honour of Tahsin Özgüç, 319–331. Ankara, Türk Tarih Kurumu. Nissen, H. J. 1986. The archaic texts from Uruk. World Archaeology 17(3), 317–333. Nissen. H. J., P. Damerow and R. K. Englund 1993. Archaic Bookkeeping: Early Writing and Techniques of Economic Administration in the Ancient Near East. Chicago, University of Chicago Press. Oates, D. and J. Oates 1993. Excavations at Tell Brak, 1992–93. Iraq 55, 155–199. Oates, D. and J. Oates 1997. An open gate: Cities of the fourth millennium BC. Cambridge Archaeological Journal 7, 287– 297. Oates, J. 2002. Tell Brak: The 4th millennium sequence and its implications. In J. N. Postgate (ed.) Artefacts of Complexity: Tracking the Uruk in the Near East, 111–122. London, British School of Archaeology in Iraq. Oates, J., A. McMahon, P. Karsgaard, S. Al Quntar and J. Ur 2007. Early Mesopotamian urbanism: A new view from the north. Antiquity 81, 585–600. Özdoğan, M. 2002a. Defining the Neolithic of central Anatolia. In F. Gérard and L. Thissen (eds) The Neolithic of Central Anatolia, 253–258. Istanbul, Ege Yayinlari. Özdoğan, M. 2002b. The Bronze Age in Thrace in relation to the emergence of complex societies in Anatolia and in the Aegean. Anatolian Metal II. Der Anschnitt 15, 67–76. Özdoğan, M. and N. Basgelen (eds) 2007. Türkiye’ de Neolitik Dönem. Istanbul, Arkeoloji ve Sanat Yayinlari. Palumbi, G. 2003. Red-black Pottery: Eastern Anatolian and Transcaucasian Relationships around the mid-fourth millennium BC. Ancient Near Eastern Studies 40, 80–134. Palumbi, G. 2008. The Red and Black: Social and Cultural Interactions of the Upper Euphrates Communities in the 4th and 3rd Millennium BC. Studi di Preistoria Orientale 2. Roma, Università di Roma ‘La Sapienza’. Peltenburg E. 1999. The living and the ancestors: Early Bronze Age mortuary practices at Jerablus Tahtani. In G. del Olmo Lete and J.-L. Montero Fenollós (eds) Archaeology of the Upper Syrian Euphrates: The Tishrin Dam Area, 427–442. Aula Orientalis Supplementa 15. Barcelona, Editorial Ausa. Peltenburg, E. 2007. New perspectives on the Carchemish sector of the Middle Euphrates river valley in the 3rd millennium BC.

In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 1–24. Levant Supplementary Series 5. Oxford, Oxbow. Pollock, S. 1999. Ancient Mesopotamia. Cambridge, Cambridge University Press. Porter, A. 2002. The dynamics of death: Pastoralism and the origins of a third-millennium city in Syria. Bulletin of the American Schools of Oriental Research 325, 1–36. Rothman, M. S. 2002. Tepe Gawra: The Evolution of a Small Prehistoric Center in Northern Iraq. University Museum Monographs 112. Philadelphia, University Museum, University of Pennsylvania. Safar, F., M. A. Mustafa and S. Lloyd 1981. Eridu. Baghdad, Republic of Iraq, Ministry of Culture and Information, State Organization of Antiquities and Heritage. Schwartz, G. M. 1994. Before Ebla: Models of pre-state political organization in Syria and northern Mesopotamia. In G. Stein and M. S. Rothman (eds) Chiefdoms and Early States in the Near East, 153–174. Madison, Prehistory Press. Schwartz, G. M., H. H. Curvers, S. S. Dunham, B. Stuart and J. A. Weber 2006. A third millennium BC elite mortuary complex at Umm el-Marra, Syria: 2002 and 2004 excavations. American Journal of Archaeology 110, 630–641. Sertok, K. and R. Ergeç, 1999. A new Early Bronze Age cemetery: Excavations near the Birecik dam, preliminary report (1997– 98). Anatolica 25, 87–107. Stein, G. 1994. Economy, ritual and power in Ubaid Mesopotamia. In G. Stein and M. S. Rothman (eds) Chiefdoms and Early States in Near East, 35–46. Madison, Prehistory Press. Tobler, A. J. 1950. Excavations at Tepe Gawra. Philadelphia, University of Pennsylvania Press. Ur, J. A. 2002. Surface collection and offsite studies at Tell Hamoukar, 1999. Iraq 64, 15–43. Verhoeven, M. 1999. An Archaeological Ethnography of a Neolithic Community. Leiden, Nederlands Historisch-Archeologisch Instituut te Istanbul. Verhoeven, M. 2002. The transformation of society: The changing role of ritual and symbolism in the PPNB and the PN in the Levant, Syria and south-east Anatolia. Paléorient 28(1), 5–13. Warner, J. L. 1994. Elmali-Karatas II. The Early Bronze Age Village of Karatas. Bryn Mawr College Archaeological Monographs. Bryn Mawr (PA), Bryn Mawr College. Wilkinson, T. J. and D. J. Tucker 1995. Settlement Development in the North Jazira, Iraq: A Study of the Archaeological Landscape. London, British School of Archaeology in Iraq. Yoffee, N. 1995. Political economy in early Mesopotamian states. Annual Review of Anthropology 24, 281–311.

11 STATES OF HEGEMONY: EARLY FORMS OF POLITICAL CONTROL IN SYRIA DURING THE THIRD MILLENNIUM BC Lisa Cooper

Introduction The second half of the 3rd millennium BC saw the full adoption of urban settlements in Syria. Syrian cities grew to cover large areas, and they were accompanied by all the characteristics of urban society. Particularly noteworthy for this period of urbanisation is the site of Ebla, located in western Syria. During the 3rd millennium BC, Ebla grew to around 56 ha, one of the largest cities in the region. It also acquired extensive territories around the city, which included agricultural fields and grazing land for its thousands of sheep and goats. Excavations at Ebla revealed third millennium occupation which featured a lower town and a central high ‘acropolis’ upon which stood a magnificent and sprawling mud brick palace labelled ‘Palace G’. This was the residence of the king of Ebla. By far the most important discovery made in Palace G was that of archive rooms filled with thousands of cuneiform tablets which document the business and political affairs of the Ebla kings and their palace officials over a period of about 50 years dating to the 24th century BC, just before the palace was violently destroyed by fire (Archi and Biga 2003, 1). Scholars of ancient Ebla have been struck by the number of other cities, towns and territories which are mentioned in the Ebla tablets, and the Ebla kings’ efforts to enter into relationships with the people of those places, either through trade partnerships, political alliances or outright conquests. From these tablets, it is clear that during its fluorescence, Ebla’s power was felt over an extremely large part of the Near East. As wealthy and renowned as Ebla was during this period, however, it has been difficult to understand precisely the mechanisms by which it exerted its influence and supremacy over a wide region, and the extent to which it actually controlled regions of the Near East, especially various

parts of Syria, which are most frequently mentioned in the tablets. While it is clear that Ebla received the allegiance of many cities and their territories, especially during periods of warfare with other regional powers, or entered into partnerships with other polities, it remains to be clarified just what form these allegiances or alliances took. Did these other polities become dependants of Ebla, relinquishing their own economic and political autonomy to the Ebla king and his high officials? Or did such polities simply recognise the supremacy of Ebla, and while performing certain obligations for it, remain essentially independent? Scholars who have used different and sometimes conflicting terms to describe the nature of Ebla’s power have frustrated our understanding of Ebla’s power. Some refer to Ebla as a city-state while others prefer to call it a regional state or territorial state, thus laying emphasis on the wider control it is perceived to have exerted over a wide area (Michalowski 1985, 301; Archi 1992, 24; Milano 1995, 1221; Thuesen 2000, 59). At its most extreme, the term ‘empire’ has been used. In this regard, Ebla is seen as an ancient example of a Near Eastern superpower with supremacy over a vast area, employing various mechanisms to subjugate and control subject territories (Matthiae 1977; Astour 1989, 140). It is not our intention at this time to enter into a detailed discussion of the uses of terms such as city-state, territorial state and empire, although such a discussion may be fruitful, and would allow Ebla to be brought as a comparative case within the realm of current anthropological studies which investigate the nature of different types of complex societies throughout the Near East and the rest of the world (e.g. Trigger 2003; Yoffee 2005). For our purposes, we are comfortable calling Ebla a state, or a nascent state, without specifying further what kind of state it was. But the focus

88

Lisa Cooper

here is not what Ebla should be called, but what it did and how it operated. We are particularly interested in examining the nature of Ebla’s hegemony over other polities in Syria. A special focus shall be placed on the area of the Euphrates Valley of Syria, in the eastern part of Ebla’s sphere of influence. The principal reason for exploring this region is the available evidence, both in the form of the textual record provided by the Ebla tablets, as well as recent archaeological investigations, which illuminate Ebla’s political and economic activities in this region. The following therefore represents such an attempt to draw out information from these sources, and to provide some provisional conclusions about the state of Ebla’s hegemony.

Textual Sources from Ebla The inscribed cuneiform tablets from Ebla provide a rich source of information about the economy, religion, kingship and political activities of Ebla over a short time during the 24th century when it was ruled by three kings (Archi and Biga 2003 provide the most up-to-date chronology of Ebla kings). Of particular interest here are tablets which provide information about the territories and kingdoms to the east with which Ebla had some type of relationship, be it an alliance with an independent regional power or a polity that it endeavoured to subordinate or control in some way. The tablets indicate that this eastern region included all of the territories on either side of the Euphrates River along its entire course as it runs through Syria. It also included some territories to the north of the Syro-Turkish border. From the tablets, we shall summarise what we know about the nature of Ebla’s political relationship with these territories. Did Ebla actually exercise some form of hegemony over these regions, and if it did, what form of hegemony did this take? According to the tablets, Ebla dealt with at least three categories of polities. These are outlined as follows:

1. Polities ruled by kings (ens) that were independent of Ebla These kingdoms often rivalled the economic and political strength of Ebla. Some entered into wars with her, or became formidable neighbours which Ebla strove to keep in check. The kingdom of Mari, located further down the Euphrates River, near the present Syro-Iraqi border, was one of Ebla’s greatest rivals. Like Ebla, Mari pushed into the area of the Middle Euphrates Valley because of the economic opportunities it afforded, the result being that these two regional powers vied for supremacy in the region for several decades during the 24th century BC (Archi and Biga 2003, 2). A series of victories by Mari resulted for a time in the kingdom of Ebla having to pay Mari a heavy

tribute of silver and gold (Milano 1995, 1226). Despite these payments and the abandonment of some of Ebla’s interests in the Euphrates Valley, however, there is no indication that Mari interfered with the organisation and structure of the Ebla kingdom. Eventually, Mari’s strength weakened, and after open hostilities with that state, Ebla was able to win back its control or the allegiance of many of the territories of the Middle Euphrates (Archi and Biga 2003, 13–26). Although Ebla boasts of its defeat of Mari in battle, this victory never included a political or economic takeover of territory in the heart of Mari’s kingdom. The kingdom of Abarsal (for its suggested location, see Archi 1989, 16–17; Astour 1992, 32–33; Archi and Biga 2003, 10; Bunnens 2007, 48–50) remained independent of direct Eblaite control although it found itself in frequent contact with that state and had to compromise some of its authority on the occasion of Ebla’s political supremacy. The terms of Abarsal’s relationship to Ebla can be best seen in the so-called ‘Treaty of Abarsal’ drawn up at a time when Ebla was enjoying political ascendancy and considerable territorial expansion to the east (Archi and Biga 2003, 10). Although this is a treaty which formalises the relationship between two independent regional powers, it is clear from the tone and content of the treaty that Abarsal’s position at this time was subordinate to Ebla (Astour 1989, 147). Abarsal promised the smooth passage of merchants and messengers through its territory (Astour 1989, 147; 1992, 33; Bunnens 2007, 49). Moreover, while Ebla could send its travelling merchants to Abarsal, Abarsal had no right to do the same with regard to Ebla (Astour 1989, 148). Despite these unequal terms, however, no text lists Abarsal as being “in the hand” of the king of Ebla; that is, Ebla never directly controlled it. Abarsal’s independent relationship is also indicated in other parts of the treaty in which the territories belonging to Ebla are carefully listed alongside those which were separately governed by Abarsal (Archi 1989, 16; Milano 1995, 1227–28). Moreover, when we consider the actual quantities of tribute made by Abarsal to Ebla in the period of time after the date of treaty and after its initial payment of silver, we find that they were rather insignificant (Astour 1989, 148). In fact, they are far less numerous than the deliveries of gifts sent by Ebla to the city of Abarsal (Archi 1989, 17). We suspect that Ebla was desirous to maintain a friendly relationship with Abarsal, particularly given Ebla’s uneasy peace with its rival Mari, which also had territorial ambitions in this part of northern Syria/southern Anatolia. The Ebla tablets frequently mention the kingdom of Emar, which found itself in a close relationship with Ebla. Emar, whose capital city is identified with Tell Meskene on the western bank of the ‘great bend’ of the Euphrates River less than 100 km downstream from Carchemish, would have been situated very close to the frontier between Ebla and Mari’s sphere of influence (Archi and Biga 2003, 10).

11. States of hegemony: Early forms of political control in Syria during the third millennium BC

89

While Ebla’s influence over the city and territory of Emar was considerable, Emar was never under the direct control of Ebla and remained an independent polity with its own rulers (Archi 1990, 24). We know especially about the queen whose name was Tisha-Lim (Archi 1990, 24; Astour 1992, 46). This queen was known to own real estate, mainly tilled fields in the area of several villages (Milano and Rova 2000, 724 n. 22). Two royal decrees also report that Tisha-Lim purchased land in the regions around the towns of IrPES and Gurrabal, known to be in the possession of the Ebla king (Archi 1990, 25–26; Milano and Rova 2000, 724, and 724 n. 21). In these decrees it is made clear that Ebla recognised Tisha-Lim’s sovereignty over these territories and that it relinquished all rights of control over its citizens (Archi 1990, 26). This documentation testifies to a strong relationship between Ebla and Emar, and yet Emar does appear to have remained politically and economically independent of Ebla.

2. Polities that fell under the indirect control of Ebla but were ruled by their own king (en)

Fig. 11.1 Map of Syria during the period of the Ebla Archives. Excavated sites are accompanied by a dot. Postulated locations of polities mentioned in the Ebla tablets are indicated in italic capitals (e.g. HADDU).

These are polities which are listed in some way as being under the control of Ebla. They are often described, as in the so-called ‘Treaty between Ebla and Abarsal’ as being “in the hand” of the king of Ebla, indicating Ebla’s control over them (Edzard 1992). These would have been independent regional kingdoms before their subjugation by Ebla, with their own capital cities, towns, villages and territories. They were all ruled by kings, or ens as they are defined in the Ebla texts. The following list provides some of the names of such kingdoms, which appear to have existed in some location to the east and north-east of Ebla, on or near the Euphrates River in Syria or Anatolia, and possibly as far east as the Balikh River Valley. Their subject status to Ebla may have arisen as a result of their location on trade routes of great importance to Ebla or their proximity to other important regional powers such as Mari and Nagar, against which they would have buffered Ebla. The kingdoms include Burman, Gasur, Gudadanum, Haddu, Hazuwan, Kablul, Kakmium, NIrar and Ra’aq (for the posited locations of these polities, see Fig. 11.1; Archi 1989, 16; Astour 1992, 27, 32–35; Milano 1995, 1227; Bonechi 1998, 226, 234; Milano and Rova 2000, 722–730, 742; Archi and Biga 2003, 14, n. 44). In most cases, after Ebla’s conquest of kingdoms ruled by an en, the local kings were left to rule rather than be replaced by an Ebla administrative official. In the absence of any documentation to the contrary, they continued to govern their domains in much the same way as they had before they fell under Ebla’s control. In a few instances, however, we learn that the power and authority of certain kingdoms were greatly reduced, and their former rulers were removed. Hazuwan, which formerly had a king and was listed alongside other prominent client kingdoms, seems

to have lost its independence at a certain point since the occurrences of a king (en) of Hazuwan stop. The same also seems to hold true with Kablul, another former kingdom with an en (Milano and Rova 2000, 730, including n. 50). In both cases, it is conjectured that these polities were brought under a more direct rule of Ebla. All of the kings of Ebla’s subject polities had to demonstrate their continuing loyalty by providing some form of tribute to Ebla. Deliveries of tribute frequently appear to have been made at the time of special events, such as when Ebla needed support for military expeditions against its neighbouring foe, Mari (Archi and Biga 2003, 15). Tribute took several forms, which included quantities of oil (Astour 1992, 31), various amounts of silver, gold, and bronze; textiles, especially clothing items, livestock, wine, as well as obligations such as providing soldiers for Ebla’s military contingents or sending corvée workers to Ebla (Astour 1989, 149; 1992, 33, 44–46; Milano and Rova 2000, 732–733). Although not overly large or burdensome, these tributes would have signified the kingdoms’ continuing fidelity to Ebla and their acknowledgement of its political supremacy. Ebla reciprocated with its own delivery of ‘gifts’ to its subject kingdoms as rewards for their continuing fidelity. The Ebla archives possess numerous tablets which record such deliveries of gifts to its allies, and we are left with the impression that maintaining friendly relations with other kingdoms was of extreme importance to Ebla, especially near the border zone with Mari, with whom it was frequently

90

Lisa Cooper

at war. It is perhaps significant that the subject kingdoms named above are often listed in the same deliveries which included kingdoms such as Emar, which are considered fully independent of Ebla (Archi and Biga 2003, 16, 19, 22). Given this situation, we wonder if there really is as strong a distinction between the two categories of Ebla’s hegemony described above as we have made them out to be in this study.

3. Regions or former kingdoms that were under the direct control of Ebla There are cities which are never reported as having been ruled by their own king. Such is the case with Carchemish on the Euphrates River, located at the present Syro-Turkish border. Carchemish is mentioned in several Ebla documents, and archaeological remains testify to its occupation during the second half of the 3rd millennium BC although the size of its settlement is still the source of heavy conjecture (Cooper 2006, 55–56; Peltenburg 2007, 16; Falsone and Sconzo 2007). In addition to deliveries of textiles to Ebla (Lacambre and Tunca 1999, 591) we also learn from the Abarsal treaty that Carchemish was “in the hand” of the king of Ebla (Lacambre and Tunca 1999, 591; Bunnens 2007, 45). It seems that for much of the period of the Ebla archives Carchemish was politically dependent on Ebla and thus was not allowed its own king (Bunnens 2007, 45). IrPES and Gurrabal, two towns and accompanying territories in the vicinity of Emar (possibly to the west of Emar, between the Giabuul Plain and the Euphrates; Milano and Rova 2000, 724 n. 20), appear also to have fallen under the direct control of the Ebla state since some of their lands were sold to the queen of Emar (Milano and Rova 2000, 724, including n. 21). It is also interesting that members of the Ebla aristocracy had estates and houses in these regions (Astour 1992, 45; Milano and Rova 2000, 724 n. 21).

Summary of Textual Evidence The Ebla tablets demonstrate that Ebla was engaged in a variety of relationships with the territories of the Euphrates Valley of Syria that included: a) alliances with independent kingdoms; b) semi-control over kingdoms that recognised Ebla’s supremacy and had to pay tribute; and c) direct ownership and control over certain kingdoms and their territories. The nature of these relationships does not appear to be dependent upon geographical distance from Ebla. Indeed, some of Ebla’s direct possessions (e.g. irPES and Gurrabal) appear in the vicinity of Emar, probably the furthest territory away from it. We suspect that these types of relationships, which occur in a kind of patchwork manner, had much to do with the strength of the other various polities and their resistance or openness to entering into diplomacy or exchanges with Ebla (Peltenburg 2007, 11).

Archaeological Evidence for Ebla’s Presence and Hegemony in the Euphrates Valley of Syria We are well informed about the antiquity of the area of the Syrian Euphrates Valley between Emar in the south and the Syro-Turkish border in the north due to the number of excavations carried out on ancient sites in the area, particularly over the past 40 years. Excavations have revealed that the river valley was quite heavily populated during the 3rd millennium BC, particularly during the period when Ebla was a flourishing kingdom. The archaeological record for the Middle Euphrates Valley is very rich, and it is not our intention to deal here with all of its details. Rather, we have selected only aspects of material culture which can possibly be linked to Ebla’s presence, and we will try to examine the nature of that presence, particularly as it pertains to some form of control or attempts at subordination by that polity. Only material evidence in the form of city defences, landscape data pertaining to agricultural practices, public and private buildings, funerary remains, and small objects that can be dated to approximately 2450–2300 BC, will be considered (Cooper 2006, 15–20). The following paragraphs list and discuss aspects of human activity and their material manifestations which might be linked to the presence of Ebla in the Middle Euphrates Valley, and which may point to that polity’s attempts to control or subordinate the populations of the region.

1. Military activities, manifested in garrisons for soldiers, outposts, and well fortified settlements No structure has yet been positively identified as a military garrison for troops, nor has any evidence brought to light an installation that can be reconstructed as a military outpost or watchtower. There is ample evidence, however, for fortified settlements whose occupations can be dated approximately to the period of Ebla’s ascendancy in the 24th century BC. These settlements’ primary function was to house the local civilian populations of the Euphrates region. The settlements of Jerablus Tahtani, Tell ‘Abd, Tell Banat (with the Jebel Bazi citadel), Tell es-Sweyhat, Munbaqa, Tell Halawa A, Tell Habuba Kabira and Selenkahiye were all fortified during this time (Heusch 1980, 174–75; Orthmann 1981, 9; 1989, 18, 37; Eichler et al. 1984, 73; Machule et al. 1986, 81–3; Finkbeiner 1994, 116; 1997, 100; Peltenburg et al. 1996, 8; 2000, 71; Armstrong and Zettler 1997, 18; Peltenburg 1999a, 101; Van Loon 2001, 3.51 and 3.86–89; Otto 2006; Danti and Zettler 2007, 177–78; ). Their defences often consisted of ramparts and glacis constructed on the exterior faces of the settlements, along with thick walls and towers. The Euphrates fortifications may have been related in some way to the current political situation of the time

11. States of hegemony: Early forms of political control in Syria during the third millennium BC in which large regional powers, notably Ebla, Mari and possibly Abarsal, were vying for control and fought several battles over this region of Syria. Settlements which were controlled by or recognised the supremacy of one of these regional powers would have been particularly vulnerable to attacks. Further confirmation of the region’s vulnerability within the context of the Ebla/Mari hostilities might be found in the evidence from Jebel Bazi at the Tell Banat complex, where the gate building appears to have suffered some form of violent attack precisely around the time that Ebla experienced its sudden demise (Otto 2006, 11). The destruction of the Bazi citadel may have been carried out by Mari (Otto 2006, 23). Although this is intriguing evidence, there is little demonstrable proof that the construction of any of the Euphrates fortifications can be linked to the initiatives or sponsorship of one specific regional power such as Ebla or Mari. Moreover, while some of the fortifications’ construction may be approximately dated to the time of Ebla’s period of supremacy, still others were simply refurbishments or the strengthening of existing constructions which go back much earlier in time. Thus, attacks, raids and/or warfare on Euphrates sites appear to have been an endemic problem in this region and were not restricted to a single time period. In light of this fact, it seems more likely that many of these defensive projects were simply part of an initiative on the part of the local inhabitants of the settlements, who were perpetually concerned about the defence of their own people and possessions.

2. Tribute payments In the case of the Euphrates Valley of Syria, tribute probably consisted of agricultural products, namely grain, which was grown in abundance on the river terraces immediately above the Euphrates River, and sheep and goat products such as wool, the result of pastoralist activities practised in the upland steppe above the river valley (Cooper 2006, 36–41). Unfortunately, these products do not preserve well over time, and archaeological evidence is slim. Nonetheless, indirect evidence, in the form of landscape studies, may be a way in which to document the nature and intensity with which a given territory utilised its land for various agricultural or pastoral activities. For the Euphrates Valley, geomorphological investigations combined with site survey and the mapping of off-site sherd scatters have been undertaken in the area around the site of Tell es-Sweyaht (Danti 1997, 2000; Wilkinson 2004). They have revealed that the upland steppe east of Sweyhat received a slightly higher amount of rainfall per year, showing a potential for agriculture (Danti 1997, 85). Interestingly, a survey of the area also led to the discovery of an ancient site (Tell Jedi) which had been occupied during the late Early Bronze Age. This evidence thus testifies to the utilisation of these

91

agriculturally productive regions of increased rainfall and water runoff even as early as the 3rd millennium BC (Danti 2000, 276). We wonder if this may be linked to the increased need for agricultural products on the part of Tell es-Sweyhat during the period of the supremacy of regional powers such as Ebla when the settlement was faced with the additional burden of tribute, and had to extend its cultivation zone far out into the upland plateau to generate the surplus in agricultural products needed to fulfil this demand.

3. Administrative activities Eblaite administrative personnel could have been posted to this frontier area of the Euphrates Valley in order to assist with the collection of tribute from the local inhabitants and to ensure peace and stability in the region. Archaeologically, such a presence might take the form of buildings where official administrative activities could be carried out as well as the residences of foreign governors. Such buildings were probably larger and more elaborate than domestic structures and may have featured non-local architectural styles which set them apart from the other typical Euphrates-style buildings. These buildings may have contained exotic objects that linked them to a foreign presence such as Ebla. To date, only two large-scale secular buildings, which may have served as administrative centres that existed during the period of the Ebla archives, have been exposed through archaeological excavations: Building 6 at Tell Banat (Porter 2002a, 27) and the Southern Mansion at Selenkahiye (Van Loon 2001, 3.37, fig. 3.7). Unfortunately, neither of the buildings, in terms of their architectural style or their contents, points to Ebla in any significant way. The Southern Mansion at Selenkahiye did yield a large number of seal impressions, suggesting administrative or economic activities in the house, but these seals have parallels to those from all over the Near East, not simply Ebla (Mazzoni 1992, 13–77; Van Loon 2001, 12.495). Such evidence points to widespread connections, possibly through trade, but it does not demonstrate any exclusive administrative or economic links to Ebla.

4. Elite emulation Apart from the lack of demonstrable links between the large-scale secular buildings at Tell Banat and Selenkahiye and elite contexts at Ebla, no compelling connection can be made between elite funerary remains from Euphrates sites and those from Ebla. Such elite graves principally take the form of elite shaft and chamber tombs, distinguished by their large size, their visibility and the quality of the materials used to construct them (Thureau-Dangin and Dunand 1936, 96; Peltenburg et al. 1995, 7–13; 1996, 13–14; 2000, 69–71; McClellan and Porter 1999, 109–110; Porter 2002a,

92

Lisa Cooper

18–19). Parallels to these funerary monuments are hard to find beyond the Euphrates Valley. These remains testify to regionally specific funerary traditions which included very particular beliefs about the afterlife, the status and significance of the individuals who were buried there, and the importance of the living communities who continued to maintain and venerate these monuments (Peltenburg 1999b; Porter 2002b). The luxury grave goods which accompanied the deceased individuals of the monumental shaft and chamber tombs, especially at Tell Banat and Jerablus Tahtani, provide a somewhat different picture. These valuable objects consisted of items such as decorated ostrich eggs, a limestone statue wig, and a wide variety of beads, pendants, pommels, and plaques made out of ivory, shell, rock crystal, lapis lazuli, silver and gold (Peltenburg et al. 1995, 11–12; McClellan and Porter 1999, 110). Most tomb objects share similarities to other goods found at a variety of sites throughout the Near East, including distant Troy in Anatolia (Peltenburg et al. 1995, 12; McClellan and Porter 1999, 110). Such farflung contacts surely reflect the network of long-distance exchanges that the people of the Euphrates sites were engaged in at this time. Moreover, these grave goods, which served to enhance the prestige of the persons with whom they were associated, would seem to indicate an ideological framework that was shared by elites from all over the Near East, not simply those from one city such as Ebla (McClellan and Porter 1999, 110). In conclusion, it is difficult to detect any instances of emulation on the part of the elites of the Euphrates Valley of Syria that can be related specifically to Ebla. While maintaining very distinct customs related to funerary beliefs and traditions, the ways in which elites sought to showcase their status through various objects of adornment borrowed from a repertoire of elite behavioural customs and associated material trappings that were known throughout the Near East during this period of long-distance communications and exchanges. Such evidence demonstrates that while various kingdoms may have been striving to bring other regions under their political control, some social customs and their expressions in material culture transcended political boundaries.

Conclusions Having presented the available textual and archaeological evidence for the area of the Middle Euphrates, some conclusions can be made regarding the nature and intensity of Ebla’s hegemony over this region during the period of its strength in the 24th century BC. Ebla’s hold over the Euphrates was loose and transitory. While many of the tablets make the claim that Ebla ruled supreme over this region, particularly after its

supposed re-conquest of this region after the wars with Mari, this supremacy did not translate to many visible transformations in the region’s material culture. There is little evidence to indicate that Ebla had any concentrated or prolonged presence in the region, as evidenced by the paucity of remains suggesting the implantation of military or administrative personnel. The majority of Euphrates settlements and their local populations appear to have remained essentially undisturbed by the chain of outside events, although we do see attempts to increase the settlements’ defensive capabilities, which could relate to the wars fought in this region between the superpowers of Ebla and Mari. It is also possible that settlements had to intensify their agricultural activities in order to produce surplus goods destined as tribute for their overlords. But such activities do not appear to have economically crippled the local communities. Indeed, the second half of the third millennium BC, both during and after the period of Ebla’s fluorescence, appears as a prosperous time for the populations of the Syrian Euphrates, when settlements grew to their largest extent and were characterised by the most elaborate architectural embellishments and the acquisition of valuable goods from around the Near East. If the Euphrates Valley’s relationships with Ebla and other regional powers did anything at this time, it was to broaden its links to the wider Near East and all of the economic opportunities such an exposure afforded. These relationships also brought the Euphrates region into the wider cultural network of the Near East where a number of social customs and their material cultural manifestations were shared and copied, particularly among elites. If Ebla did have some kind of political hegemony over the Euphrates Valley, as many of the tablets appear to testify, this could be described as ‘hegemony without sovereignty’. This type of political strategy is best summarised by Kolata, who along with other scholars has recognised this phenomenon elsewhere in the world, particularly in the context of early states and empires (Berdan et al. 1996, 150; Trigger 2003, 113–119; Kolata 2006, 210–12). ‘Hegemony with sovereignty’ entails the domination of populations without actually administering them directly. Rather, power and influence are exercised by the mere demonstration of military strength and cultural superiority. These demonstrations are thought to ensure the political and economic subordination of the subject populations without having to pay exorbitant costs of maintaining control through a constant military presence, administrative officials, or the colonisation of the subject territory (Kolata 2006, 210) We suggest that this model may have been applicable to the situation of Ebla’s relationship to the Middle Euphrates Valley as outlined above. We hesitate, however, to use the term ‘strategy’ to describe this type of political phenomenon as we wonder if Ebla engaged in any conscious or deliberate programme to enforce its supremacy over the Euphrates

11. States of hegemony: Early forms of political control in Syria during the third millennium BC region. Rather, it is suspected that Ebla’s acceptance of tributes and other gifts, as well as its entry into diplomatic ties with local polities and its acquisition of land in various territories, came about in a somewhat ad hoc fashion when favourable opportunities availed themselves to assert a superior position over amenable or militarily vulnerable Euphrates settlements. Moreover, this type of hegemony disappeared almost as quickly as it had arrived. Within only a few decades, Ebla and all of the territories over which it had claimed supremacy had succumbed to the ambitions of other Near Eastern regional kingdoms, first in the form of its long standing rival Mari, and later by the rulers of the powerful Akkadian dynasty.

References Archi, A. 1989. La ville d’Abarsal. In M. Lebeau and P. Talon (eds) Reflets des deux fleuves.Volume de mélanges offerts à André Finet, 15–19. Leuven, Peeters. Archi, A. 1990. Imâr au IIIème millénaire d’après les archives d’Ebla. MARI 6, 21–38. Archi, A. 1992. The city of Ebla and the organization of its rural territory. Altorientalische Forschungen 19, 24–28. Archi, A. and M. G. Biga 2003. A victory over Mari and the fall of Ebla. Journal of Cuneiform Studies 55, 1–44. Armstrong J. A. and R. L. Zettler 1997. Excavations on the high mound (Inner Town). In R. L. Zettler (ed.) Subsistence and Settlement in a Marginal Environment: Tell es-Sweyhat, 1989– 1995 Preliminary Report, 11–32. MASCA Research Papers in Science and Archaeology 14. Philadelphia, Museum Applied Science Center for Archaeology, University of Pennsylvania Museum of Archaeology and Anthropology. Astour, M. C. 1989. The geographical and political structure of the Ebla empire. In H. Waetzoldt and H. Hauptmann (eds) Wirtschaft und Gesellschaft von Ebla, 139–158. Heidelberg, Heidelberger Orientverlag. Astour, M. C. 1992. An outline of the history of Ebla (Part 1). In C. H. Gordon and G. A. Rendsburg (eds) Eblaitica: Essays on the Ebla Archives and Eblaite Language, vol. 3, 3–82. Winona Lake (IN), Eisenbrauns. Berdan, F. F., R. Blanton, E. Boone, M. Hodge, M. Smith and E. Umberger 1996. Aztec Imperial Strategies. Washington, Dumbarton Oaks Research Library & Collection. Bonechi, M. 1998. Remarks on the IIIrd millennium geographical names of Syrian Upper Mesopotamia. In M. Lebeau (ed.) About Subartu. Studies devoted to Upper Mesopotamia. Vol. 1 – Landscape, Archaeology, Settlement, 219–241. Subartu 4. Turnout, Brepols. Bunnens, G. 2007. Site hierarchy in the Tishrin Dam area and third millennium geopolitics in Northern Syria. In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 43–54. Oxford, Oxbow. Cooper, L. 2006. Early Urbanism on the Syrian Euphrates. London, Routledge. Danti, M. D. 1997. Regional survey and excavations. In R. Zettler (ed.) Subsistence and Settlement in a Marginal Environment.

93

Tell es-Sweyhat, 1989–1995 Preliminary Report, 85–94. MASCA Research Papers in Science and Archaeology 14. Philadelphia, Museum Applied Science Center for Archaeology and University of Pennsylvania Museum of Archaeology and Anthropology. Danti, M. D. 2000. Early Bronze Age Settlement and Land Use in the Tell es-Sweyhat Region, Syria. Unpublished Ph.D. thesis, University of Pennsylvania. Danti, M. D. and R. Zettler, R. 2007. The Early Bronze Age in the Syrian north-west Jezireh: The Tell es-Sweyhat region. In E. Peltenburg (ed.) Euphrates River Valley Settlement: The Carchemish Sector in the Third Millennium BC, 164–183. Oxford, Oxbow. Edzard, D. O. 1992. Der Vertrag von Ebla mit A-bar-QA. In P. Fronazaroli (ed.) Literature and Literary Language at Ebla, 187–217. Florence, Universita di Firenze. Eichler, S., D. R. Frank, D. Machule, G. Mozer and W. Pape 1984. Ausgrabungen in Tall Munbaqa 1983. Mitteilungen der Deutschen Orient-Gesellschaft 116, 65–93. Falsone, G. and P. Sconzo, P. 2007. The ‘champagne cup’ period at Carchemish. A review of the Early Bronze Age levels on the Acropolis Mound and the problem of the Inner Town. In E. Peltenburg (ed.) Euphrates River Valley Settlement. The Carchemish Sector in the Third Millennium BC, 73–93. Oxford, Oxbow. Finkbeiner, U. 1994. Abd. In H. Weiss (ed.) Archaeology in Syria. American Journal of Archaeology 98, 116–117. Finkbeiner, U. 1997. Abd. In H. Weiss (ed.) Archaeology in Syria. American Journal of Archaeology 101, 97–100. Heusch, J. -C. 1980. Tall Habuba Kabira im 3. und 2. Jahrtausend: Die Entwicklung der Baustruktur. In J.-C. Margueron (ed.) Le Moyen Euphrate, 159–178. Leiden, E. J. Brill. Kolata, A. L. 2006. Before and after collapse: Reflections on the regeneration of social complexity. In G. M. Schwartz and J. J. Nichols (eds) After Collapse: The Regeneration of Complex Societies, 208–221. Tuscon, University of Arizona Press. Lacambre, D. and Ö. Tunca 1999. Histoire de la vallée de l’Euphrate entre le barrage de Tišrin et Karkemiš aux IIIe et IIe millénaires av. J.-C. In G. Del Olmo Lete and J.-L. Montero Fenollós (eds) Archaeology of the Upper Syrian Euphrates. The Tishrin Dam Area. Proceedings of the International Symposium Held at Barcelona, Jan. 28th–30th, 1998, 587–603. Barcelona, Editorial Ausa. Machule, D., K. Karstens, H.-H. Klapproth, G. Mozer, W. Pape, P. Werner, R. Mayer, R. Mayer-Opificius and M. Mackensen 1986. Ausgrabungen in Tall Munbaqa 1984. Mitteilungen der Deutschen Orient-Gesellschaft 118, 65–93. Matthiae, P. 1977. Ebla: An Empire Rediscovered. London, Hodder and Stoughton. Mazzoni, S. 1992. Le impronte su giara eblaite e siriane nel Bronzo Antico. Roma, Università degli studi di Roma ‘La Sapienza’. McClellan, T. L. and A. Porter 1999. Survey of excavations at Tell Banat: Funerary practices. In G. Del Olmo Lete and J.-L. Montero Fenollós (eds) Archaeology of the Upper Syrian Euphrates. The Tishrin Dam Area. Proceedings of the International Symposium Held at Barcelona, Jan. 28th–30th, 1998, 107–116. Barcelona, Editorial Ausa. Michalowski, P. 1985. Third millennium contacts: Observations

94

Lisa Cooper

on the relationships between Mari and Ebla. Journal of the American Oriental Society 105, 293–302. Milano, L. 1995. Ebla: A third-millennium city-state in ancient Syria. In J. M. Sasson (ed.) Civilizations of the Ancient Near East, 1219–1230. New York, Scribner. Milano, L. and E. Rova 2000. Ceramic provinces and political borders in Upper Mesopotamia in the late Early Dynastic period. In S. Graziani (ed.) Studi sul Vicino Oriente Antico dedicati alla memoria di Luigi Cagni, 709–749. Naples, Istituto Universitario Orientale. Orthmann, W. 1981. Halawa 1977–1979. Bonn, Habelt. Orthmann, W. 1989. Halawa 1980–1986. Bonn, Habelt. Otto, A. 2006. Archaeological perspectives on the localization of Naram-Sin’s Armanum. Journal of Cuneiform Studies 2006, 1–26. Peltenburg, E. 1999a. Tell Jerablus Tahtani 1992–1996: A summary. In G. Del Olmo Lete and J.-L. Montero Fenollós (eds) Archaeology of the Upper Syrian Euphrates. The Tishrin Dam Area. Proceedings of the International Symposium Held at Barcelona, Jan. 28th–30th, 1998, 97–105. Aula Orientalis Supplementa 15. Barcelona, Editorial Ausa. Peltenburg, E. 1999b. The living and the ancestors: Early Bronze Age mortuary practices at Jerablus Tahtani. In G. Del Olmo Lete and J.-L. Montero Fenollós (eds) Archaeology of the Upper Syrian Euphrates. The Tishrin Dam Area. Proceedings of the International Symposium Held at Barcelona, Jan. 28th–30th, 1998, 427–442. Barcelona, Editorial Ausa. Peltenburg, E. 2007. New perspectives on the Carchemish sector of the Middle Euphrates River Valley in the 3rd millennium BC. In E. Peltenburg (ed.), Euphrates River Valley Settlement. The Carchemish Sector in the Third Millennium BC, 1–24. Oxford, Oxbow. Peltenburg, E., S. Campbell, P. Croft, D. Lunt, M. A. Murray and

M. E. Watt 1995. Jerablus-Tahtani, Syria, 1992–4: Preliminary report. Levant 27, 1–28. Peltenburg, E., D. Bolger, S. Campbell, M. A. Murray and R. Tipping 1996. Jerablus-Tahtani, Syria, 1995: Preliminary report. Levant 28, 1–25. Peltenburg, E., E. Eastaugh, M. Hewson, A. Jackson, A. McCarthy and T. Rymer 2000. Jerablus Tahtani, Syria, 1998–9: Preliminary report. Levant 32, 53–75. Porter, A. 2002a. The dynamics of death: Ancestors, pastoralism, and the origins of a third-millennium city in Syria. Bulletin of the American Schools of Oriental Research 325, 1–36. Porter, A. 2002b. Communities in conflict: Death and the contest for social order in the Euphrates River Valley. Near Eastern Archaeology 65, 156–173. Thuesen, I. 2000. The city-state in ancient western Syria. In M. H. Hansen (ed.) A Comparative Study of Thirty City-State Cultures, 55–65. Copenhagen, Royal Danish Academy of Sciences and Letters. Thureau-Dangin, F. and M. Dunand 1936. Til-Barsib. Paris, Geuthner. Trigger, B. G. 2003. Understanding Early Civilizations. Cambridge, Cambridge University Press. Van Loon, M. N. 2001. Selenkahiye. Final Report on the University of Chicago and University of Amsterdam Excavations in the Tabqa Reservoir, Northern Syria, 1967–1975. Istanbul, Nederlands Historisch-Archaeologisch Instituut. Wilkinson, T. J. 2004. On the Margin of the Euphrates: Settlement and Land Use at Tell es-Sweyhat and in the Upper Lake Assad Area, Syria. Chicago, Oriental Institute of the University of Chicago. Yoffee, N. 2005. Myths of the Archaic State. Cambridge, Cambridge University Press.

PART 3

TECHNOLOGY, ECONOMY AND SOCIETY

12 A HOUSEHOLD AFFAIR? POTTERY PRODUCTION IN THE BURNT VILLAGE AT LATE NEOLITHIC TELL SABI ABYAD Olivier Nieuwenhuyse

Professor Edgar Peltenburg’s outstanding record as an archaeologist shows two recurring elements: his scholarly interest in the social dynamics of technology and style in prehistoric societies; and his propensity for researching a sunny Mediterranean island. However, he has also explored a prehistoric site at Jerablus Tahtani in the heart of Upper Mesopotamia. In this paper I shall investigate the organisation of prehistoric pottery production in Upper Mesopotamia. I shall limit the discussion to one major key site for the Late Neolithic, Tell Sabi Abyad, situated in the valley of the Balikh, a tributary of the Euphrates in northern Syria. The history of the mound of Tell Sabi Abyad is highly complex. I shall review one of the more extensively excavated occupation levels at this site, Level 6, which is located on the south-eastern slopes of the site’s main mound in what has been termed Operation 1 (for an overview, see Akkermans et al. 2006). Also known as the “Burnt Village” because of the conflagration that reduced the Level 6 village to ashes around 6000 BC (all dates in this paper are calibrated dates BC), this settlement was inhabited at a time when communities across northern Syria, southeastern Anatolia and northern Iraq had initiated far-reaching changes in the way pottery was made and used. Not long before the Burnt Village, at about 6200 BC, a long era of rigorously plain, coarsely finished ceramics had come to an end with the introduction, initially in limited quantities, of painted pottery vessels. Soon afterwards, between c. 6100 and 5900 BC, the ceramic assemblage changed rapidly and fundamentally, from being dominated by mostly plain, plant-tempered ceramics to mainly consisting of painted Fine Ware (Akkermans 1993; Nieuwenhuyse 2007). This discovery has sparked a lively debate, not just with regard to the details of the ceramic sequence or the

precise chronological synchronisations between key sites, but also with regard to the assumptions underlying our interpretations. In Upper Mesopotamia a major obstacle to understanding the organisation of prehistoric societies is the tendency to classify the evidence according to rigid culture-historical frameworks as a basis for comparative studies and reconstructions of long-term social evolution. Textbooks on Near Eastern prehistory are often organised in separate sections dealing with such entities as the ProtoHassuna, the Hassuna, the Samarra, the Halaf and the Ubaid cultures. Recent fieldwork on the Late Neolithic shows the inadequacies of the existing framework. Most of the elements traditionally ascribed to the ‘Halaf package’ can now be traced back to the centuries preceding the Halaf period. In terms of the ceramics, it has been shown that what is seen as ‘Halaf pottery’ emerged gradually from a preceding Transitional (or Proto-Halaf) stage in which Hassuna and Samarra stylistic traits dominated (Campbell 1992, 1998; Akkermans 1993, 1997; LeMière and Nieuwenhuyse 1996; Cruells and Nieuwenhuyse 2005; Nieuwenhuyse 2007). The Level 6 Burnt Village was inhabited at the start of the Transitional period between Pre-Halaf and Early Halaf. Rather than trying to pigeonhole this particular community into one culture-historical framework or the other, this paper explores the technological and social aspects of pottery production. What were the raw materials, tools and the technological knowledge needed, and how did people procure them? Were ceramic vessels made locally or were they imported from elsewhere? Is there evidence of pottery production within the village that archaeologists may identify as such?

98

Olivier Nieuwenhuyse

Models of Late Neolithic Ceramic Production Scholars have presented divergent views on how pottery production was organised in the Upper Mesopotamian Late Neolithic. This heterogeneity reflects the different research interests of the scholars involved and their theoretical backgrounds as well as the diverse cultural mosaic during the Late Neolithic. To a fair measure it also reflects the slow, unequal development of research. Presently far more work has been done on the later stages of the Pottery Neolithic, which are characterised by an array of attractively decorated pottery styles such as the Hassuna, Samarra and Halaf. In contrast, much less is known about the earlier stages of the Pottery Neolithic, c. 6900–6200 BC, which have only most recently emerged as major research foci in their own right. When it comes to the Halaf period, 5900–5300 BC, one influential perspective was outlined by Watson and LeBlanc, who suggested that the Halaf painted ‘luxury’ ceramics were symbols of prestige exchange between established elites, for which they used the term ‘chiefdom’ (Watson and LeBlanc n.d.; Watson 1983; Redman 1978). Watson and LeBlanc used statistical measurements of similarity between painted motif frequencies from a range of Halaf sites to infer patterns of social interaction. Strong empirical support for the view of Halaf ceramics as part of organised networks of exchange came from work by Davidson and McKerrel (1976, 1980; Davidson 1977). Using neutron activation analyses (NAA) of Halaf sherds and local clays from a variety of sites and wadis in north-eastern Syria, they argued that a number of Halafian sites had been ceramic production centres. These centres supplied ‘satellite’ villages with ceramic vessels, and at the same time they were engaged in the mutual exchange of high-quality painted pottery. This view of the Halaf was criticised from the 1980s onwards with the rapid accumulation of new data from the field. New fieldwork made it clear that in fact there was precious little evidence to suggest that Halaf societies had been socially complex in the traditional sense of the word (Akkermans 1993). Apart from the presumed ‘prestige’ ceramics there were few unequivocally identifiable luxury goods, the spatial layout of villages did not immediately point to social differentiation, and the burial record suggested a heterogeneous yet non-hierarchical approach to death (Campbell 1992; Akkermans and Schwartz 2003). Akkermans (1993, 319) concluded that the painted pottery was not involved in the establishment of any sort of social hierarchy. Furthermore, Davidson and McKerrel’s interpretations were subjected to devastating criticism by Galbraith and Roaf (2001). Re-evaluating the earlier data with updated statistical methods, Galbraith and Roaf concluded that although ceramics were certainly exchanged in the Halaf period, the earlier NAA data do not support the specific model proposed by Davidson and McKerrel. As to the period immediately preceding the Halaf, most

scholars would accept that the Hassuna and Samarra styles reflect different subsistence adaptations and attendant social organisations. In this view, the Samarra pottery style reflects socially more advanced organisations based on irrigation agriculture along the two major river systems in Central Mesopotamia, the Euphrates and the Tigris. In contrast, groups in the north with the Hassuna pottery style continued to practise dry farming agriculture and did not develop hierarchical social systems (Oates 1972; 1973; Huot 1994; Breniquet 1996; Aurenche and Kozlowski 1999). Whereas pottery production came into the hands of specialists in the Samarra territories, Hassuna pottery production continued to be practiced at the level of the individual household. While the Samarra potters developed a standardised set of complex design structures, design structures in the north were kept much simpler and did not become standardised (Bernbeck 1994). The Hassuna groups imported the occasional high-quality vessel from their Samarra neighbours and made cheapish imitations (Lloyd and Safar 1945). Mortensen (1970) argued that the stylistic differences between Hassuna and Samarra ceramics reflected different modes of production. Whereas the bulk of the Hassuna pottery was felt to be made locally, Mortensen suggested that the most complexly decorated items were made by travelling Samarran specialists (1970, 118–121). This discussion has been strongly reinvigorated recently by the discovery in Syria and south-eastern Turkey of a gradual sequence of ceramic change leading to the Halaf pottery tradition. This transition is characterised by a modest introduction and subsequent rapid increase of painted Fine Ware made in what may be termed the Hassuna-Samarra style if broadly applied (Campbell 1992, 1998; Akkermans 1993). During the Transitional period innovations in ceramic technology, vessel shape, and decorative style led to what archaeologists have termed Early Halaf pottery (Cruells and Nieuwenhuyse 2005). In other words, what archaeologists have treated as separate culture-historical entities can also be seen, in Upper Mesopotamia at least, as part of a continuum of ceramic innovation (Nieuwenhuyse 2007). Presently, a lively debate focusses on what precisely should fall within the definitions of Samarra and Hassuna pottery (Bernbeck 2009), the identification of the technological chaînes opératoires (operational sequences) underlying the various pottery groups (Van As et al. 1998; Nieuwenhuyse et al. 2001; Robert et al. 2009), and the issue of ceramic exchange (LeMière and Picon 1987; 1999; LeMière and Picon 2009) among other points.

Pottery Production in the Level 6 Burnt Village The Level 6 Burnt Village has been excavated over an area of almost 1400 m2. It was constructed on the south-eastern slopes of a much older mound that dates back to the early

12. A household affair? Pottery production in the Burnt Village at Late Neolithic Tell Sabi Abyad

99

Fig. 12.1 Plan of the Level 6 Burnt Village at Tell Sabi Abyad (drawing by M. Kriek in Akkermans and Nieuwenhuyse forthcoming).

stages of the Pottery Neolithic (Akkermans et al. 2006). The Level 6 remains have been radiocarbon dated to c. 6000 BC. At this time the north-eastern part of Tell Sabi Abyad, too, was inhabited, but the western part of the mound appears to have been empty. A dense concentration of large multiroomed buildings was characteristic (Fig. 12.1). It has been persuasively argued that these were used largely for storage by a semi-pastoralist population (Akkermans and Duistermaat 1997; Verhoeven 1999). Between 400 and 670 people may have relied on the village, with perhaps some 120 people residing there permanently (Verhoeven 1999, 211–213).

It is important to note that although there is much similarity from one multi-roomed building to the next, the buildings were certainly not all equal in architectural composition or size. Intriguingly, some building complexes were conspicuously larger than others. These may have been the domiciles of households that in this particular episode of the village history were economically more successful than others. Interestingly, although kilns of various shapes and sizes have been found dispersed through the Level 6 village, they are concentrated in two courtyards on either side of the largest complex in the village (Building II). Before

100

Olivier Nieuwenhuyse

concluding that the inhabitants of this building controlled essential pottery production facilities, we should note that so far the distribution of small finds does not suggest a clear differentiation (Verhoeven 1999). By and large, each complex appears to have had a broadly similar inventory of material goods prior to the conflagration. Tools and facilities that may have been used for the production of pottery are dispersed widely throughout the Level 6 village and cannot be associated with any building complex in particular. Nor does the composition of the ceramic assemblage, although certainly highly variable from one building complex to another, suggest any patterns that we may interpret as reflecting unequal access to specific pottery groups (Nieuwenhuyse 2007). It may be difficult, perhaps impossible, at this stage to identify with certainty the tools that people used for pottery production. Since the local production was probably carried out largely at a household level and at a low intensity, the necessary tools may have been unspecialised to begin with (Peacock 1982; Van der Leeuw et al. 1987; Costin 1991). Production facilities, such as pits for clay storage or a covered space for drying vessels, may be virtually indistinguishable from pits and storage rooms used for other activities. No cache of unfired vessels waiting to be fired has been recovered from the Burnt Village. So far it has not been possible to locate any spatial clusters of artefacts that enable the identification of production facilities (Verhoeven 1999; Nieuwenhuyse 2007). How, then, do we know for certain that ceramics were produced at Tell Sabi Abyad itself? Firstly, although such finds are very scarce, misfired and warped ceramic fragments are occasionally found. Secondly, the abundance of ceramic vessels throughout the settlement, their bulkiness, and their fragility in transport all argue for local production. Finally, the microscopic and chemical analysis of ceramics and local clays strongly suggests that a large part of it was made locally (LeMière 1989; Van As et al. 1998; LeMière and Picon 2009). Most raw materials were locally available. Ethnographic comparisons suggest that the potters from Tell Sabi Abyad collected their clays within walking distance of the village (Arnold 1985). We shall never be able to locate the exact locations, however, as several metres of sediment have accumulated over the Late Neolithic field level (Wilkinson 1996). Most of the samples of clay collected near the current course of the Balikh River, close to the village, have reasonable to excellent properties for making pottery. The main pigment for slipping and painting the vessels was ochre. Small pieces of ochre are regularly found in the Burnt Village, and some of the basalt tools show traces of the grinding of this pigment. Ochre, or iron-oxide haematite, was probably available on the limestone terraces of the Balikh valley, which occur less than 10 km from the site (Akkermans 1993, 274). Ceramic production was in all probability seasonally organised and integrated within the broader annual cycle of

activities, festivities and the movements of people. In the highly seasonal climate of Upper Mesopotamia the cold, wet winter months would have been unfavourable for drying pots and fuel. In the winter and spring, moreover, various agricultural activities would have laid claims upon the available labour time. Within the semi-pastoralist economy these seasons would have been socially intense as the larger part of the community probably stayed within or close to the village. The summer would have been the likely time of the year for ceramic production. A large part of the population would have left the village with the herds. Those who stayed behind would have engaged in ceramic production. Straw left over from the harvest would have made convenient tempering material (Matson 1974; Akkermans 1993, 275– 279; Bernbeck 1994, 263; Eiland 2003, 337). Dramatic changes in vessel morphology occurred during the Transitional period. Interestingly, these do not seem to be associated with the invention of new shaping techniques. Vessels of all shapes and sizes seem to have been made with coiling, a technique known already since the first introduction of pottery around 6900 BC. Moulds were most probably used for shaping convex bases as well as the sharply carinated profiles that became increasingly popular during the Transitional period. First, the base was pressed into the mould, after which the carinated body was built up with coils. The mould may simply have been the re-used base fragments of a discarded vessel (Van As and Jacobs 1989; Van As et al. 1998). Abundant traces of sculpting, scraping, smoothing and burnishing point to a variety of surface-finishing techniques. Vessels were sometimes decorated with techniques such as stabbing, impressing or incising. The tools needed for this may have been fairly simple and need not have been exclusively limited to pottery production. So far no usewear analyses have been carried out on the bone awls and needles, the flint or obsidian tools, or the ceramic scrapers and ‘loamers’ that occur in great numbers at the site. It seems likely that in addition to a whole range of other possible functions some of these items were also employed to modify ceramic vessels (Fig. 12.2). Non-industrial potters in ethnographic settings have sometimes been shown to be emotionally attached to the tools they use. These may have been transmitted over generations, and so they can acquire great social value (De Boer and Lathrap 1979; Dillingham 1992). Pebbles with a glossy working surface are indeed attested at Tell Sabi Abyad, but what they were used for remains to be established. A remarkable technological breakthrough at this time was the development of firing strategies that increased the potters’ control over temperatures and oxygen fluctuations in the kiln. This allowed them to produce dark-on-light painted Fine Ware (Steinberg and Kamili 1984; Noll 1991; Robert et al. 2009). Unfortunately, we are hard pressed to identify the pottery kilns used by the Burnt Village potters.

12. A household affair? Pottery production in the Burnt Village at Late Neolithic Tell Sabi Abyad

101

Fig. 12.2 Left: Flint and obsidian tools from the Level 6 Burnt Village possibly used for pottery production (b, d). Scraping a fine clay (b, c) yields traces similar to those observed on Burnt Village Fine Ware sherds (a) (courtesy Loe Jacobs, Faculty of Archaeology, Leiden University). Right: A large, circular, domed kiln from the Level 6 Burnt Village with an opening at ground level. The kiln was found completely preserved, but the top was removed during the excavation (Marc Verhoeven as a scale, 1.81 m).

Characteristic for the later stages of the Transitional period, for instance, is the keyhole-shaped kiln with a narrow, arched, combustion chamber and a domed heating chamber with stones on the floor, partly sunk into the ground to save fuel. They are usually found with only the lower part still intact, however, and their function remains elusive. Verhoeven and Kranendonk (1996, 82) suggest that the smaller examples were used for roasting meat, but the larger examples may well have been used for firing pottery. A similar potters’ kiln was reported from Tell Ziyada on the Khabur River (Buccellati et al. 1991). Another type is larger, circular or sometimes oval with an opening in the wall at floor level (Fig. 12.2). These beehiveshaped kilns could be up to 3 m in diameter. The interior was usually hard and red-burnt, suggesting that they reached high temperatures. Elsewhere in Upper Mesopotamia, the first examples of circular kilns with a two storey construction appear at this stage (Merpert and Munchaev 1993) in which a perforated grid separated the lower firing chamber from the upper space holding the ceramic vessels. So far, no examples of this type have been attested in the Burnt Village.

Coming from Far Away Certainly not all pottery vessels in the Burnt Village were locally produced, nor were all raw materials local in origin. For instance, it was characteristic during the Transitional period to decorate coarse plant-tempered vessels by scratching them with bitumen. Geochemical analysis of the pigment showed that it came from two distinct sources in northern Iraq, c. 500 km away from the site (Connan et

al. 2004; Nieuwenhuyse et al. 2004). It is most likely that these coarse vessels were locally made and decorated with non-local pigments (Fig. 12. 3, nos 5–6). There is sound evidence for non-local pottery within the village even if the exact locations of origin remain to be established. Ongoing work by LeMière shows that part of what is termed Standard Fine Ware (SFW) came from elsewhere. Two other Fine Ware groups, termed Orange Fine Ware (Fig. 12.3, nos 10–12) and Fine-Painted Ware, may also be of non-local origin (LeMière and Nieuwenhuyse 1996; LeMière 2000; 2001). These two categories comprise only a very small minority within the assemblage, but their presence is significant. LeMière and Picon (2008) also suggest that the proportion of imported SFW may have been highest in the early stages of the Transitional period. A regular exchange of painted Fine Wares, then, was implicated in the ceramic developments observed during the Transitional period. SFW bears close similarities to Hassuna and Samarra pottery from northern and central Iraq (Fig. 12.3, nos 13–20), but it would be too far-fetched at this stage to conclude that ‘Samarran impulses’ (Akkermans 1993) initiated the Transitional period. Although it is certainly possible that the occasional vessel travelled to the Burnt Village from as far away as central Mesopotamia, it is more likely that most of the exchange was conducted among the Transitional period villages that are now being detected in survey work across upper Mesopotamia (Nieuwenhuyse 2000; Nieuwenhuyse and Wilkinson 2007; Erdalkıran 2008). Perhaps the clearest case of non-local ceramics within the Burnt Village is the Dark Face Burnished Ware (DFBW), which includes about 4% of the assemblage (Fig. 12.3, nos 7– 9). This heavily mineral-tempered ware is petrographically,

102

Olivier Nieuwenhuyse

Fig. 12.3 Selected examples of pottery from the level 6 Burnt Village: (1–4) decorated Standard Ware; (5–6) bitumen-painted Standard Ware; (7–9) Dark-Faced Burnished Ware; (10–12) Orange Fine Ware; (13–20) Standard Fine Ware.

12. A household affair? Pottery production in the Burnt Village at Late Neolithic Tell Sabi Abyad chemically and stylistically distinct from everything else at the site (LeMière and Picon 1987; 1999; Bader et al. 1994; Nieuwenhuyse 2007; Diebold forthcoming). From its origin, probably within the south-eastern Turkish Zagros, it was widely distributed across the upper Mesopotamian plains. Both functional and socio-symbolic factors may have stimulated demand. Functionally, this was excellent ‘cooking ware’ in which vessels were repeatedly re-shaped from jars into hole-mouth pots (Fig. 12.3, no. 9). Much of the DFBW consisted of serving vessels; in the Burnt Village this pottery may have held the role of exotic serving-and-display ware before the emergence of the painted Fine Wares.

Concluding Remarks The Burnt Village provides an intriguing case study of ceramic production and consumption in upper Mesopotamia at the close of the 7th millennium. Ceramic production and consumption was far from a self-sufficient, localised affair. Certainly, most of the vessels that circulated in the village were made locally by individual families that organised their work at the household level. Direct evidence for local production is scarce at best, but this in itself fits well within the expectations from generalised models of household production. Pottery production facilities and tools that can be identified are dispersed through the various buildings and open areas. There is little evidence to suggest ‘elite’ sponsorship of, or control over, production. At the same time, certain vital raw materials could only be gained through exchange with other groups, and DFBW and some of the Fine Ware came to the village from elsewhere. Alongside production systems aiming at the local community, some ceramic production will have been organised in a manner more akin to the household industry mode, involving several producing groups aiming at a larger range of regional or even supra-regional consumers. A dynamic, very active orientation to inter-regional networks and relationships lay at the heart of Late Neolithic societies after c. 6200 BC. Communities such as the Burnt Village participated in supra-regional networks of exchange in which various goods and ideas circulated over huge distances. Approaches to painted pottery styles in the upper Mesopotamian Late Neolithic have sometimes tended to promote a view of material culture as essentially reflecting given life-ways or regional identities. At Sabi Abyad, however, much ceramic innovation observed during the Transitional period centred upon elaborately painted Fine Ware: vessels excellently suitable for social display within the context of serving food and drink. Pottery styles in the Burnt Village were actively involved in creating and maintaining social identities.

103

References Akkermans, P. M. M. G. 1993. Villages in the Steppe: Later Neolithic Settlement and Subsistence in the Balikh Valley, Northern Syria. Ann Arbor, International Monographs in Prehistory. Akkermans, P. M. M. G. 1997. Old and new perspectives on the origins of the Halaf culture. In O. Rouault and M. Wäfler (eds) La Djéziré et l’Euphrate syriens de la Protohistoire à la fin du second millénaire av. J.-C., 55–68. Paris, Éditions Recherche sur les Civilisations. Akkermans, P. M. M. G. and K. Duistermaat 1997. Of storage and nomads: The sealings of Late Neolithic Tell Sabi Abyad, Syria. Paléorient 22(2), 17–44. Akkermans, P. M. M. G. and O. P. Nieuwenhuyse (eds) forthcoming. Excavations at Late Neolithic Tell Sabi Abyad: Interim Report on the Archaeological Research (1994–1999) in the Balikh Valley, Syria. Turnhout, Brepols. Akkermans, P. M. M. G., R. Cappers, C. Cavallo, O. P. Nieuwenhuyse, B. Nilhamn and I. Otte 2006. Investigating the early Pottery Neolithic of northern Syria: New evidence from Tell Sabi Abyad. American Journal of Archaeology 110(1), 123–156. Akkermans, P. M. M. G. and G. M. Schwartz 2003. The Archaeology of Syria: From Complex Hunter-Gatherers to Early Urban Societies (ca. 16,000–300 BC). Cambridge, Cambridge University Press. Arnold, D. E. 1985. Ceramic Theory and Cultural Process. Cambridge, Cambridge University Press. Aurenche, O. and S. K. Kozlowski 1999. La Naissance du Néolithique au Proche-Orient. Paris, Éditions Errance. Bader, N. O., V. A. Bashilov, M. LeMière and M. Picon 1994. Productions locales et importations de céramique dans le Djebel Sinjar au VIe millénaire. Paléorient 20(1), 61–68. Bernbeck, R. 1994. Die Auflosung der häuslichen Produktionsweise. Berlin, Dietrich Reimer. Bernbeck, R. 2008. Taming time and timing the tamed. In J. M. Córdoba, M. Molist, M. C. Perez, I. Rubio and S. Martinez (eds) Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East, Madrid 3–8 April, 2006, 709–728. Isimu 10. Madrid, Universidad Autónoma de Madrid. Breniquet, C. 1996. La disparition de la culture de Halaf; les origines de la culture d’Obeid dans le Nord de la Mésopotamie. Paris, Éditions Recherche sur les Civilisations. Buccellati, G., D. Buia and S. Reimer 1991. Tell Ziyada: The first three seasons of excavation (1988–1990). Bulletin of the Canadian Society for Mesopotamian Studies 21, 31–61. Campbell, S. 1992. Culture, Chronology and Change in the Later Neolithic of North Mesopotamia. Unpublished Ph.D. thesis, University of Edinburgh. Campbell, S. 1998. Problems of definition: The origins of the Halaf in north Iraq. In M. Lebeau (ed.) About Subartu: Studies Devoted to Upper Mesopotamia, 39–52. Subartu 4. Turnhout, Brepols. Connan, J., O. P. Nieuwenhuyse, A. van As and L. Jacobs 2004. Bitumen in early ceramic art: Bitumen-painted ceramics from Late Neolithic Tell Sabi Abyad (Syria). Archaeometry 46(1), 115–124. Costin, C. L. 1991. Craft specialization: Issues in defining, documenting, and explaining the organization of production. In

104

Olivier Nieuwenhuyse

M. B. Schiffer (ed.) Advances in Archaeological Method and Theory 3, 1–56. Tuscon, University of Arizona Press. Cruells, W. and O. P. Nieuwenhuyse 2005. The Proto-Halaf period in Syria: New sites, new data. Paléorient 30(1), 47–68. Davidson, T. E. 1977. Regional Variation within the Halaf Culture. Unpublished Ph.D. thesis, University of Edinburgh. Davidson, T. E. and H. McKerrel 1976. Pottery analysis and Halaf period trade in the Khabur headwaters region. Iraq 38, 45–56. Davidson, T. E. and H. McKerrel 1980. The neutron activation analysis of Halaf and ‘Ubaid pottery from Tell Arpachiyah and Tepe Gawra. Iraq 42, 45–56. De Boer, W. R. and D. W. Lathrap 1979. The making and breaking of Shipibo ceramics. In C. Kramer (ed.) Ethnoarchaeology: Implications of Ethnography for Archaeology, 102–138. New York, Columbia University Press. Diebold, B. forthcoming. Exchange and Interaction in the Late Neolithic of Northwestern Mesopotamia. Ph.D. thesis, Yale University. Dillingham, R. 1992. Acoma and Laguna Pottery. Santa Fe, School of American Research Press. Eiland, M. 2003. Ceramics and society. In R. Matthews (ed.) Excavations at Tell Brak Vol. 4: Exploring an Upper Mesopotamian Regional Centre, 1994–1996, 321–362. London, British School of Archaeology in Iraq. Erdalkıran, M. 2008. The Halaf ceramics in Şirnak area, Turkey. In J. M. Córdoba, M. Molist, M. C. Perez, I. Rubio and S. Martinez (eds) Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East, Madrid 3–8 April, 2006, 755–766. Isimu 10. Madrid, Universidad Autónoma de Madrid. Galbraith, J. and M. Roaf 2001. Does the Neutron Activation Analysis of Halaf Pottery Provide Evidence of Trade? A Reexamination of Davidson and McKerrell’s Investigation of Halaf Pottery from the Khabur Basin in Eastern Syria. LSE Department of Statistics Research Report. London, London School of Economics. Huot, J. L. 1994. Les premiers villageois de Mésopotamie: du village a la ville. Paris, Armand Collin. LeMière, M. 1989. Clay analyses of the prehistoric pottery. In P. M. M. G. Akkermans (ed.) Excavations at Tell Sabi Abyad. Prehistoric Investigations in the Balikh Valley, Northern Syria, 233–235. British Archaeological Reports International Series 468. Oxford, British Archaeological Reports. LeMière, M. 2000. L’occupation proto-Hassuna du Haut-Khabur occidental d’après la céramique. In B. Lyonnet (ed.) Prospection Archéologique Haut Khabur Occidental (Syrie du N.E.) 1, 127–149. Beirut, Institut français d’archéologie du ProcheOrient [IFAPO]. LeMière, M. 2001. The Neolithic pottery from Tell Kosak Shamali. In Y. Nishiaki and T. Matsutani (eds) Tell Kosak Shamali Vol. I. The Archaeological Investigations on the Upper Euphrates, Syria. Chalcolithic Architecture and the Earlier Prehistoric Remains, 179–211. Tokyo, University Museum, University of Tokyo. LeMière, M. and O. P. Nieuwenhuyse 1996. The prehistoric pottery. In P. M. M. G. Akkermans (ed.) Tell Sabi Abyad: The Late Neolithic Settlement, 119–284. Istanbul, Nederlands Historisch Archeologisch Instituut. LeMière, M. and M. Picon 1987. Productions locales et circulation

des céramiques au VIème millénaire au Proche Orient. Paléorient 13(2), 137–151. LeMière, M. and M. Picon 1999. Les débuts de la céramique au Proche Orient. Paléorient 24(2), 5–26. LeMière, M. and Picon, M. 2008. A contribution to the discussion on the origins of the Halaf culture from chemical analyses of pottery. In J. M. Córdoba, M. Molist, M. C. Perez, I. Rubio and S. Martinez (eds) Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East, Madrid 3–8 April, 2006, 729–734. Isimu 10. Madrid, Universidad Autónoma de Madrid. Lloyd, S. and F. Safar 1945. Tell Hassuna: Excavations by the Iraq Government General of Antiquities in 1943 and 1944. Journal of Near Eastern Studies 4, 255–289. Matson, F. R. 1974. The archaeological present: Near Eastern village potters at work. American Journal of Archaeology 78, 345–347. Merpert, N. Y. and R. M. Munchaev 1993. Yarim Tepe I. In N. Yoffee and J. J. Clark (eds) Early Stages in the Evolution of Mesopotamian Civilization: Soviet Excavations in Northern Iraq, 73–114. Tuscon, University of Arizona Press. Mortensen, P. 1970. Tell Shimshara: The Hassuna Period. Copenhagen, Munksgaard. Nieuwenhuyse, O. P. 2000. Halaf settlement in the Khabur headwaters. In B. Lyonnet (ed.) Prospection Archéologique Haut Khabur Occidental (Syrie du N.E.) 1, 151–260. Beirut, Institut français d’archéologie du Proche-Orient [IFAPO]. Nieuwenhuyse, O. P. 2007. Plain and Painted Pottery. The Rise of Regional Ceramic Styles on the Syrian and Northern Mesopotamian Plains. Papers on Archaeology of the Leiden Museum of Antiquities 3. Turnhout, Brepols. Nieuwenhuyse, O. P., J. Connan, A. van As and L. Jacobs 2004. Painting pots with bitumen at Late Neolithic Tell Sabi Abyad (Syria). Neo-Lithics 2(3), 22–25. Nieuwenhuyse, O. P., L. Jacobs, A. van As, T. Broekmans and M. Adriaens 2001. Making Samarran Fine Ware: Technological observations on the ceramics from Tell Baghouz (Syria). Paléorient 27(1), 147–165. Nieuwenhuyse, O. P. and T. J. Wilkinson 2007. Late Neolithic settlement in the area of Tell Beydar, (NE Syria). In M. Lebeau and A. Suleiman (eds) Beydar Studies I, 268–303. Subartu 21. Turnhout, Brepols. Noll, W. 1991. Alte Keramiken und ihre Pigmente. Studien zu Material und Technologie. Stuttgart, Schweizerbart. Oates, J. 1972. Prehistoric settlement patterns in Mesopotamia. In P. J. Ucko, R. Tringham and G.W. Dimbleby (eds) Man, Settlement and Urbanism, 299–310. London, Duckworth. Oates, J. 1973. The background and development of early farming communities in Mesopotamia and the Zagros. Proceedings of the Prehistoric Society 39, 147–181. Peacock, D. P. S. 1982. Pottery in the Roman World: An Ethnoarchaeological Approach. London, Longman. Redman, C. L. 1978. The Rise of Civilization: From Early Farmers to Urban Society in the Ancient Near East. San Francisco, Freeman. Robert, B., A. Lassalle and R. Chapoulie 2008 New insights into the ceramic technology of the Proto-Halaf (‘Transitional’) period by using physico-chemical methods. In J. M. Córdoba, M.

12. A household affair? Pottery production in the Burnt Village at Late Neolithic Tell Sabi Abyad Molist, M. C. Perez, I. Rubio and S. Martinez (eds) Proceedings of the 5th International Congress on the Archaeology of the Ancient Near East, Madrid 3–8 April, 2006, 736–752. Isimu 10. Madrid, Universidad Autónoma de Madrid. Steinberg, A. and D. C. Kamili 1984. Paint and paste studies of selected Halaf sherds from Mesopotamia. In P. M. Rice (ed.) Pots and Potters: Current Approaches in Ceramic Archaeology, 187–208. Institute of Archaeology Monograph 24. Los Angeles, Institute of Archaeology, University of California. Van As, A. and L. Jacobs 1989. Technological aspects of the prehistoric pottery. In P. M. M. G. Akkermans (ed.) Excavations at Tell Sabi Abyad. Prehistoric Investigations in the Balikh valley of northern Syria, 215–221. British Archaeological Reports International Series 468. Oxford, British Archaeological Reports. Van As, A., L. Jacobs and O. P. Nieuwenhuyse 1998. The Transitional Fine Ware pottery of Tell Sabi Abyad, Syria: A pilot study. Newsletter of the Department of Pottery Technology 14/15, 25–47. Leiden, Leiden University Press. Van der Leeuw, S. E., A. J. Spruijt and V. A. Shelton-Bunn 1987. Ceramic production. In R. W. Brandt, W. Groenman-van Waateringe and S. E. van der Leeuw (eds) Assendelver Polder Papers 1, 225–264. CINGULA 10. Amsterdam, Universiteit

105

van Amsterdam, Albert Egges van Giffen Instituut voor Praeen Proto-historie. Verhoeven, M. 1999. An Ethnographical Ethnography of a Late Neolithic Community: Space, Place and Social Relations at Tell Sabi Abyad, Syria. Istanbul, Nederlands Historisch Archeologisch Instituut. Verhoeven, M. and P. Kranendonk 1996. The excavations: Stratigraphy and architecture. In P. M. M. G. Akkermans (ed.) Tell Sabi Abyad: The Late Neolithic Settlement, 25–118. Istanbul, Nederlands Historisch Archeologisch Instituut. Watson, P. J. 1983. The Halafian culture: A review and synthesis. In T. C. Young, P. E. L. Smith and P. Mortensen (eds) The Hilly Flanks and Beyond, 231–250. Chicago, University of Chicago Press. Watson, P. J. and S. A. LeBlanc n.d. Excavation and analysis of Halafian materials from southeastern Turkey: The Halaf period reexamined. Paper presented at the 72nd Annual Meeting of the American Anthropological Association, New Orleans, Nov. 28–Dec. 2, 1973. Wilkinson, T. J. 1996. Sabi Abyad: The geoarchaeology of a complex landscape. In P. M. M. G. Akkermans (ed.) Tell Sabi Abyad: The Late Neolithic Settlement, 1–24. Istanbul, Nederlands Historisch Archeologisch Instituut.

13 LATE CYPRIOT CERAMIC PRODUCTION: HETERARCHY OR HIERARCHY? Louise Steel

There have been various technological studies of Cypriot pottery (most recently Åström 2001; Karageorghis 2001; Crewe 2004; 2007; Knappett and Kilikoglou 2007); however, the organisation of LC pottery production has received less attention (but see Keswani 1991). This article examines changing pottery production during the LBA. It is proposed that prior to the 13th century BC there was considerable variation in pottery production, and only in the 13th and 12th centuries did ceramic production cut across regional boundaries, a corollary of increasing urbanisation and centralised control (all dates in this paper are cal BC).

Pottery Production and Social Complexity Various archaeological studies have sought to explain changes in pottery production within an evolutionary paradigm (Peacock 1977; Rice 1981; Underhill 1991; Loney 2000). Changes are viewed as evidence for technological improvement and the rise of complex societies. Consequently, there is a tendency to relate increasing specialisation of pottery production to increasing social complexity (Rice 1981). In an ethnoarchaeological study Peacock (1982) proposed various modes of production related to different stages of social complexity: small-scale, village-based societies were typified by domestic household production, whereas highly standardised mass production within specialist workshops was characteristic of social complexity and urbanised societies (part-time specialisation within chiefdoms and full-time specialisation in states; see Underhill 1991). The development of a wheelmade technology in particular is associated with economic specialisation and mass production within an urban context (Arnold 1985, 206–208; Crewe 2007, 209). Other developments include

better firing techniques and more effective control of reducing/oxidising conditions (Arnold 1985, 123). Recent approaches, however, have questioned the automatic association between the development of complex sociopolitical structures and technological advance (Loney 2000) and relate specialised production to heterarchical social organisation rather than centralised economic control within a vertical hierarchy (Schoep and Knappett 2004, 26–27). The adoption of new technologies is voluntary, and people may choose to do so for a variety of reasons which need not be predictable or ‘rational’ (Loney 2000). Pottery production is essentially conservative. Potters are reluctant to change their technological base, and once they have identified suitable clays, tempers and fuels, and suitable modes of production, they continue to exploit these with little internal impetus for change (Rice 1984, 243–244; 1987, 461–465). Arnold (1985, 205–206) notes that the manufacture of pottery involves mastering a considerable number of motor habits; this is a long process and the skills involved are more easily acquired during childhood. Indeed, the motor habits of a particular mode of production (hand-built, mould-made, wheelmade) are so deeply ingrained (Arnold 1985, 206) that they may act as significant barriers to change (1985, 221). Arnold also observes that cultural attitudes and beliefs might impede transmission of new styles, forms and decoration (1985, 223). Hence, changes in pottery production indicate significant cultural transformations that need not necessarily be related to increasing complexity in social organisation.

Late Cypriot Social Organisation During the 2nd millennium BC, Cyprus underwent a number of socio-economic and demographic transformations associated with an increase in population and greater

13. Late Cypriot ceramic production: Heterarchy or hierarchy?

107

Fig. 13.1 Map of LBA sites.

participation within East Mediterranean trade networks (see Fig. 13.1 for map of principal sites). The transitional MC III–LC I period was a time of social upheaval (Merrillees 1971) typified by the abandonment of settlements and a destruction horizon. Coupled with these changes is evidence for the emergence of new social elites, who identified themselves through elaborate funerary ritual (Keswani 2004, 80). Present understanding of LC settlement is based on a tripartite model developed by Catling (1962, 142–143; see also Keswani 1993; 1996; Knapp 1997; 2008), comprising large coastal towns and small rural and mining settlements in the hinterland. By the 14th century the towns were the dominant economic and political force within the surrounding landscape. Despite the rich archaeological record, our understanding of the political organisation of LBA Cyprus is limited. We might posit some form of centralised administration located within the urban centres; nonetheless, it is problematic trying to interpret socio-political organisation within conventional, anthropologically derived models. Peltenburg (1996) has argued for the emergence of a state on Cyprus prior to LC II, possibly centred on Enkomi, and associated with increasing exploitation and

external distribution of copper. Certainly, a number of possible signifiers of social complexity can be identified at Enkomi in LC IB: a monumental building serving a number of economic functions (the fortress); Cypro-Minoan tablets; local seal production; and wheelmade pottery. The island’s topography, dominated by the vast Troodos mountain range that was largely impassable and unexploited in antiquity, impeded inter-regional communication (Peltenburg 1991, 107) and presumably the establishment of a unified LC state. Instead, a series of regional polities might be postulated similar to the Iron Age city kingdoms (Iacovou and Michaelides 1999). In fact, there is very little evidence for a hierarchically organised state on Cyprus during the LBA (Smith 1994; Keswani 1996, 234). Instead, socio-political organisation can best be characterised as a series of complex chiefdoms, demonstrating no more than two levels of control (Smith 1994, 33–36). Moreover, the notion of the state itself is problematic (Knappett 1999, 616, 618–619; Adams 2006, 1–2), especially the inherent socio-evolutionary perspective and the overriding focus on economic and political organisation within the urban context at the expense of regional level analysis. Certainly, the LC material does not fit the model of a territorial state,

108

Louise Steel

based on a hierarchical network of administrative units over a wide region, and under the control of a single ruler. Instead, the smaller polity of a city state, based on an urban core and hinterland, appears more attractive (Trigger 2003, 92–119). Another possible model is the ‘village-state’, which is based on kinship linearity, a redistributive economy and ceremonial enactment (Maisels 1990). Whatever the socio-political organisation of LBA Cyprus, the process of urbanisation implies increasing social complexity and as a corollary has important implications for the organisation of craft production, including pottery (Griffin 1965, 104, 113; Sahlins 1972, 101; Rice 1981; Feinman et al. 1984, 297–302, 322).

Late Cypriot Ceramic Production CHANGES DURING THE TRANSITIONAL MC III–LC I PERIOD The advent of the LC period is marked by the appearance of new wares, White Slip (WS), Base Ring (BR), and Monochrome (Merrillees 1971), replacing the traditional Red Polished pottery used for some 500 years. The adoption of these new wares occurred at a varying pace in different parts of the island (Merrillees 1971; Manning 2001, 81; Crewe 2007, 214), reflecting the persistence of strong regional identities against a backdrop of dramatic social change. Two distinct ceramic traditions emerged from the east-west cultural split in the island: the canonical LC wares developed in the northwest of the island while the traditional MC tradition survived in the east. Although the eastern pottery tradition appears archaising, new styles developed here (Bichrome, Red on Black and Red on Red); however, these did not have a lasting impact. The initial use of the LC wares was closely associated with ceremonial practices; the new wares were predominantly used in tombs in the Morphou region but were only gradually adopted in the surrounding settlements (Merrillees 1971, 57). A similar situation is apparent in the east; at Enkomi, for example, the new styles were primarily used as grave goods throughout LC I and were only adopted within household contexts in LC II (Crewe 2007, 214). Alongside the new wares, traditional vessels were also transformed. Most striking is the preference for flat bases and ring bases for jugs and jars although the traditional round-based jug persisted into LC I. This undoubtedly reflects changing household practices relating to the use and storage of these vessels (Crewe 2007, 227–228). Shapes such as the krater reflect Levantine influence (Amiran 1970, 99–101) and illustrate the adoption of exotic new social customs related to feasting (Crewe 2007, 228). The Plain White handmade storage jars attested in eastern Cyprus (Pilides 1996, 108) demonstrate the adoption of new storage strategies at a time when the organisation of the household was undergoing significant change (Bolger 2003, 32–36). Regional variation might indicate that pottery production

continued to be village-based, perhaps as a small-scale household industry (Crewe 2007, 216, 227). However, there is significant evidence for pottery production within specialist workshops, suggesting the co-existence of at least two modes of production in LC I (Underhill 1991, 13; Loney 2000, 651). The clearest evidence for a pottery workshop is at Morphou-Toumba tou Skourou (Vermeule and Wolsky 1990); this undoubtedly relates to specialised production discussed by Peacock (1982, 9, 25–26; Underhill 1991, 14, table 2). The excavators suggest that the local elite controlled production (Vermeule and Wolsky 1990, 101); however, rather than imposing a hierarchical model of control, we might explore horizontal differentiation in economic specialisation similar to that proposed for Protopalatial Crete (Schoep and Knappett 2004, 26–27). The wholesale transformation of the pottery tradition reflects choices made by the LC potters and throws light upon their technological advances. The potters chose to exploit new clay beds in the copper-rich Troodos region (Courtois 1977), the wheel/turntable was used for certain elements (Crewe 2004; 2007), and increased firing temperatures suggest a kiln-based technology. An experimental phase with new clay beds and firing technologies is evident in the variable finish of Proto WS (Cadogan et al. 2001, 80–81); similar experimentation is documented at Episkopi-Phaneromeni with Drab Polished, Proto BR and BR I wares (Herscher 2001, 18). Specialist potters became increasingly proficient with these new technologies throughout LC I, resulting in further changes to their product. Greater control over the kiln allowed the WS potters to achieve temperatures of 1100˚ centigrade; ironically, however, their improved skill ultimately resulted in an apparent decline in quality. The creamy-white smectitic slip used for WS I had a tendency to blister at this higher temperature and was replaced by a micaceous chloritic clay that gave a duller finish to the ware (Aloupi et al. 2001, 23). The higher temperatures also resulted in reducing conditions, and the characteristic WS I bichrome decoration (achieved by the use of an oxidised iron-based red and a manganese-based brown/black pigment) fell out of favour (Aloupi et al. 2001, 18). Thus the changes observed in the surface treatment of WS I and II are intrinsically linked to the advancement of technology and not necessarily to the tastes or choices of consumers. Certain developments reflect foreign influence at a time when the island was increasingly involved in overseas trade. Chief amongst these is the adoption of the wheel in eastern Cyprus. Masson explicitly links the introduction of the wheel to the settlement of Levantine artisans (1976, 152–153). The development of the Bichrome Wheelmade and Red Lustrous Wheelmade wares has been attributed to influence from Anatolia and Syria (Åström 2001, 134–135; Karageorghis 2001, 145–148, 153) while certain decorative aspects of Bichrome Wheelmade may have been adopted

13. Late Cypriot ceramic production: Heterarchy or hierarchy? from the Cyclades (Artzy 2001, 163–166). The consensus is that wheelmade pottery developed in Cyprus within a climate of foreign contact and exchange of ideas; however, the mode of transmission is unclear (Crewe 2007, 225–226). Ethnographic studies indicate that new potters are more likely to adopt the practices of the communities into which they moved in response to different resources and local demand, rather than impose new techniques on the resident population (Rice 1984, 245). The transfer of technological skills, in fact, is very difficult to achieve (Loney 2000, 649), and the very different motor habits used to produce handmade and wheelmade pottery represent a major obstacle to the transmission of expertise (Arnold 1985, 221), perhaps explaining why these wares had limited impact on Cypriot production. The changes apparent in the LC I ceramic tradition are socially and culturally significant. These effectively constitute the disappearance of an ancient tradition that had persisted on the island for some 500 years. At one level the development of new wares illustrates the adoption of new technologies, new modes of manufacture, and the exploitation of new resources, reflecting shifting relations with the physical and technological environment (Binford 1962, 219; Rice 1984, 252). The exploitation of new materials implies interruption to the procurement of the preferred MC resources, perhaps a result of demographic shift. Potters needed to seek out new resources, the properties of which they were not familiar with, and thus had to develop new skills (Courtois 1977; Rice 1984, 244; Vaughan 1991, 119; 1994, 91). Increasing population likewise had a significant impact on pottery production. Production by household specialists was probably no longer sufficient to meet the demand of a larger, urbanbased population (Rice 1984, 244–245, 257). Accordingly, Cypriot potters developed new strategies to cope with greater demand. Thus population stress and urbanisation were the impetus for the move away from domestic production or household industries to full-scale ceramic production at specialised centres (Feinman et al. 1984, 302). Nonetheless, this was not simply an economic transformation; the underlying cultural changes are equally significant (Crewe 2007, 225). Material culture expresses the articulation of social groups (Binford 1962, 219; Wiessner 1983; 1985; Rice 1984, 252), and the adoption of new forms and decorative styles reveals changing social practices. The choice, preparation and consumption of food and drink are deeply embedded in cultural traditions; hence the utensils needed for these practices might be particularly resistant to change (Rice 1984, 245; 1987, 463). Consequently, the introduction of new pottery forms in LC I may mirror new social practices linked to changing diet. Subtle transformations continue throughout the LC period, such as the disappearance of the spout after LC IIB (Fig. 13.2, no. 2; Lagarce and Lagarce 1985, fig. 17, no. 68; South et al. 1989, fig. 52) and the introduction of the strainer spout in

109

LC IIIA (Hadjisavvas 1991, fig. 17, nos 1–2, 4–5), perhaps reflecting further changes in food processing techniques and diet. The new wares were initially appropriated in funerary (Merrillees 1971, 57; Crewe 2007, 214) and religious contexts (al-Radi 1983). This would suggest that the new styles were specifically related to innovations in ceremonial practices used to renegotiate and reaffirm social relations at a time of social upheaval. Ethnographic studies support the notion that transformations in ceremonial behaviour, especially when associated with changing fortunes of elite groups, will impact upon pottery production (Rice 1984, 246, 252); moreover, the process of urbanisation, while poorly understood, is probably the key to understanding many changes in pottery manufacture (1984, 248–249) and seems particularly pertinent in the MC–LC period. LC II: CERAMIC REGIONALISM By LC II a common ceramic tradition was established throughout Cyprus (Fig. 13.3) “although subtle distinctions in style can still be detected” (Herscher 1984, 27). Despite the apparent standardisation there is some evidence for continued regional variation, which has important implications for understanding the organisation of pottery production and inter-regional exchange. One issue is whether pottery production was centralised and controlled by the urban elite or if small-scale production within village workshops persisted (Knapp and Cherry 1994, 8, 159–161). No potters’ quarters have been identified in the main urban centres, but regional workshops have been identified at Athienou (Dothan and Ben-Tor 1983) and Sanidha (Todd and Hadjicosti 1991; Todd and Pilides 1993; 2001). Following a detailed study of Plain White pottery, Keswani posits a workshop centred at Enkomi, which not only supplied the town but also other communities in eastern Cyprus, and possibly even the residents of Kalavasos-Ayios Dhimitrios (Keswani 1999). However, the mechanisms by which the products of these workshops were disseminated throughout the island are unclear. While there is an “absence of any apparent palatial system of centralised production and distribution of goods” (Vaughan 1994, 91), large monumental buildings, which had an economic, redistributive role, are attested at several urban centres. These may well have regulated an exchange network for the products of regional workshops, including pottery (Keswani 1993; Knapp and Cherry 1994, 159–161). At first glance the WS II style (Fig. 13.2) appears to be uniform throughout the island. The ubiquitous form was the hemispherical bowl with wishbone handle, decorated in the Normal Style; nonetheless, “subtle distinctions in style … can still be detected” (Herscher 1984, 27; Table 13.1). The diversity of WS forms attested in the Kalavasos/Maroni region is unparalleled on the island (Fig. 13.2, no. 5; Russell 1989; South and Steel 2001, fig. 3) and perhaps reflects proximity to workshops located in the southern foothills of the Troodos. Particularly noteworthy is the preference in LC

110

Louise Steel

Fig. 13.2 White Slip II pottery (not to scale): (1) WS II Normal Style hemispherical bowl, Kalavasos-Ayios Dhimitrios no. 167; after South et al. 1989, fig. 49; (2) WS II Transitional spouted bowl, Kalavasos Ayios Dhimitrios no. 137; after South et al. 1989, fig. 52; (3) WS II Normal Style hemispherical bowl, Kalavasos-Ayios Dhimitrios no. 369; after South et al. 1989, fig. 57; (4) WS II Late hemispherical bowl, Kalavasos-Ayios Dhimitrios no. 930; after South et al. 1989, fig. 6; (5) WS II jug, Kalavasos-Ayios Dhimitrios no. 160; after South et al. 1989, fig. 49.

IIA for drinking sets typically decorated in the Transitional WS II style (Steel 1998, 290). Other regional variants include WS IIA (Popham 1972, 446–447, fig. 51), found primarily in south-west and occasionally in south-central Cyprus. Closely related is the distinctive Parallel Line Style (Popham 1972, fig. 56, no. 4; Russell 1989, 2–3, figs 3–4). Only small quantities have been identified in north-west and eastern Cyprus (Catling 1957, fig. 19, no. 182; Dikaios 1969, pl. 63, nos 10, 12, 17, 25; pl. 76, no. 31); however, it comprises around one-third of the WS recovered at Kalavasos-Ayios Dhimitrios, perhaps suggesting local production (Russell 1989, 3). The distribution of these styles clearly points to multiple production centres (Knapp and Cherry 1994, 57);

this is further reiterated by chemical analyses (Artzy et al. 1981; Knapp and Cherry 1994, 57–59). The proximity of the probable production centres (Kouklia, Kourion, and Kalavasos) to the Troodos massif where basaltic clays compatible with the WS fabric are found (Courtois 1977; Courtois and Velde 1989, 74; Knapp and Cherry 1994, 59) suggests that the main WS workshops were located in the hinterland of these urban sites, perhaps controlled by the urban elites. There was widespread demand for WS both in Cyprus and the Levant, indicative of an efficient distribution system possibly instituted by the polities of the south and west of the island. “The technical ceramic standards manifested by a

13. Late Cypriot ceramic production: Heterarchy or hierarchy?

White Slip II Styles Transitional White Slip II (Fig. 13.2, no. 2; Russell 1986, 42)

Area South central (Kalavasos region)

White Slip Normal (Fig. 13.2, nos 1, 3, 5; Russell 1989, 2) White Slip IIA (Popham 1972, 446–447, fig. 51)

Island wide

Parallel Line (Popham 1972, fig. 56(4); Russell 1986, 44–53; Russell 1989, 2–3, figs 3– 4) White Slip II Late (Fig. 13.2, no. 4; Russell 1989, 2)

SW and southcentral Cyprus (Palaepaphos, Kourion, Kalavasos) Kalavasos region

Island wide

Stylistic components Carefully executed ladder pattern with lines of equal length Frontal orientation of decoration Wavy band on lip and handle Subsidiary decoration: chains of carefully executed lozenges, cross-hatched lozenges, double crosshatched lozenges Free field motifs: dotted rosettes, horned discs, lozenges, group of four dots Ladder pattern with thicker outer bands (might have subsidiary decoration, eg. hooked chain) Vertical element: ‘palm motif’ (hastily executed vertical zigzag) horizontal element below rim: wavy line & chain of double cross-hatched lozenges Horizontal element below rim: chain of doublecrossed hatched lozenges framed by double parallel bands; also dotted zigzag motif and dotted scallop pattern Vertical element: groups of 4 parallel bands Horizontal element below rim: groups of four hastily executed, short, parallel lines overlapping; Vertical element: groups of four hastily executed, parallel lines Table 13.1 White Slip II styles.

111

Shapes Deep bowls, deep spouted bowls, tankards, kraters, hemispherical bowls & rare footed bowls

Hemispherical bowls

Hemispherical bowls

Various forms: e.g. jars, tankards, spindle bottles and hemispherical bowl

Hemispherical bowls

Fig. 13.3 Base Ring pottery (not to scale): (1) BR II carinated cup; after Mee and Steel 1998, n. 66, pl.11; (2) BR II jug with painted decoration, after Mee and Steel 1998, no. 203, pl. 38. (3) BR I juglet (bilbil); after Mee and Steel 1998, no. 52, pl. 9; (4) BR II juglet with painted decoration; after Mee and Steel 1998, no. 202, pl. 38; (5) Bucchero jug; after Mee and Steel 1998 no. 67, pl. 11.

112

Louise Steel

significant quantity of Base Ring Ware represented a remarkable achievement by potters” (Vaughan 1994, 86), displaying confident use of a very plastic clay to hand build thin-walled vessels fired at high temperatures (Vaughan 1991; 1994). There is some evidence for double firing (Vaughan 1991, 122); this entailed greater investment in time and effort and increased risk in the firing process. Paradoxically, the progressive technical achievements of the BR potters resulted in an apparent decline in the ware’s finish: the lustrous BR I vessels formed from fine-grained pastes were supplanted by vessels made from a coarse-grained paste and with a matte surface finish. Alongside this, the range of forms contracted to the small juglet and the Y-shaped carinated cup (Fig. 13.3). These changes, especially the homogeneous profile of the ware, are suggestive of increasing standardisation and centralisation of production, and as a corollary an extensive and efficient distribution network (Vaughan 1994, 86). Petrographic analysis reveals three distinct regional groups, representing overlapping areas of manufacture (Vaughan 1987; 1991; 1994). It is not clear, however, how the BR potters achieved such homogeneity. Possibly, itinerant potters worked between these regions, carefully selecting suitable local clay sources. BR evidently was deemed highly desirable and was prescribed for certain cultural activities, most notably in the religious sanctuaries where the carinated cups were used for liquid libations (Vaughan 1991, 124). These cups were so intrinsically linked to LC religious practices that even after BR production had ceased in the 12th century the form was still being made in plain ware (Courtois 1971, fig. 94; Jones and Catling 1986, 595). Current evidence therefore supports highly skilled potters working within regional workshops during LC II rather than production at the level of a household industry. There is no indication that urban elites controlled production. Instead a heterarchical system of production might be posited, effectively comprising small groups of independent workshops. Nonetheless, the distribution of the wares implies efficient supply networks throughout the island; it is possible that the urban elites controlled the circulation of these wares between the production centres and their economic hinterland. LATE CYPRIOT IIC: MASS PRODUCTION AND STANDARDISATION Scale and competition have been identified as two critical factors which influence pottery production (Feinman et al. 1984, 300). A smaller number of large workshops, massproducing standardised pottery, flourish within centralised, politically controlled economies; alongside greater standardisation less effort is expended in decoration (Rice 1981; Feinman et al. 1984, 299–302). Although the LC “data are not sufficiently fine-grained to determine whether centralised production increased through time” (Knapp and Cherry 1994, 161), the technical and morphological evidence support increasing uniformity of pottery production by the

13th century, contemporary with the floruit of the urban centres. LC IIC pottery corresponds well with a model of mass production. In contrast to the varied forms current in LC I–IIB, by LC IIC both WS and BR were characterised by a limited and standardised morphology (Steel 1998, 292) and display considerable uniformity. WS production was restricted to hemispherical bowls and Base Ring to carinated cups. Alongside this there was increasing standardisation in the decoration of the WS bowls: the subsidiary cross-hatched lozenges, hooked lozenges, and rows of dots characteristic of the ware in LC IIA–B disappeared (Fig. 13.2, no. 1); instead the bowls were more simply decorated with the ladder pattern (Fig. 13.2, no. 3), suggesting articles decorated rapidly on a turntable. The end result is exemplified by the seemingly hasty banded decoration of WS II Late bowls (Fig. 13.2, no. 4; Table 13.1; Popham 1972, 456, fig. 57; Russell 1989, 3, figs 3, 6; South and Steel 2001, fig. 5). The apparent decline in surface treatment is paralleled by the later BR products; while the earlier BR I forms were burnished or polished, the later products of the workshops were either simply smoothed, wiped and left unslipped or were dipped in a matte slip (Vaughan 1991, 124, table 12.2). Ultimately, the traditional BR workshops could not withstand the forces of mass production. Despite continuing demand it could not be met by hand-made production. Instead, potters began to experiment with a less plastic clay with large inclusions, which was more suitable to being thrown on the wheel, the result being mass-produced wheel-made imitations of the carinated cup attested at Enkomi and Kouklia (Courtois 1971, fig. 94; Jones and Catling 1986, 595). By c. 1100 BC Bucchero (a derivative of BR) was being made from a pale firing, less plastic paste, which was covered by a dark slip to achieve the appearance of BR, the final “abandonment of the specialised skills of the traditional Cypriot potters” after some 400 years (Vaughan 1994, 92). The Aegeanising WPWM III (Fig 13.4) pottery further illustrates centralised, mass production of pottery within the urban environment (Sherratt 1991, 191). Chemical analyses suggest several production centres at Enkomi, Kition and Kouklia (Jones and Catling 1986, 606). This ware mimicked the Mycenaean imports popular in Cyprus throughout the LC II period although only a limited range of shapes were produced. These largely comprise kraters and bowls (Sherratt 1991, 192), which were easily formed on the wheel and decorated on turntables, the ultimate rapidly produced ceramic product. Likewise, Cypriot potters were unable to reproduce the fine lustre of the surface slip and painted decoration that was typical of the Mycenaean imports, and the decoration was highly standardised and formulaic. By the time of the Bronze-Iron Age transition (LC IIIB–CG I) wheelmade decorated pottery was being mass produced on a large scale and a homogeneous, island-wide ceramic style was established (Steel 1994).

13. Late Cypriot ceramic production: Heterarchy or hierarchy?

113

Fig. 13.4 White Painted Wheelmade II shallow bowls (not to scale); after South et al. 1989, fig. 12. (1) Kalavasos-Ayios Dhimitrios no. 1036; (2) Kalavasos-Ayios Dhimitrios no. 1034; (3) KAD no. 1037; (4) Kalavasos-Ayios Dhimitrios no. 1035.

Conclusions The changing dynamics of LC pottery production are associated with demographic and sociopolitical changes. The technical advances apparent in LC I indicate increasing specialisation of production and the establishment of regional workshops, producing goods for internal consumption and external trade. By LC II there was an effective distribution system circulating pottery throughout the island. Possibly distribution was centrally controlled, but throughout most of the LC period the evidence points to a heterarchical production and distribution system. By the 13th century, however, there is considerable evidence for increasing standardisation and specialisation typical of mass production. This process peaks in LC IIIA, marking the final stage in Cypriot Bronze Age ceramic production. It is no coincidence that mass production of wheelmade pottery coincided with the height of urbanisation and foreshadowed the homogeneous Iron Age ceramic industry.

Acknowledgements I would like to thank the editors for inviting me to contribute to this volume. I worked at Kissonerga-Mosphilia for Eddie in 1988 through which I became interested in Cypriot ceramics. Between 1995 and 1998 I held a British Academy postdoctoral fellowship at the University of Edinburgh

where I was lucky to experience Eddie’s guidance in the fields of Cypriot and Near Eastern archaeology. An early version of this paper was prepared during this period and accepted for publication in a special volume of BASOR in 1998. I am grateful to Eddie for his comments on the early drafts. Unfortunately, this publication was not to see fruition. A revised version of the paper was accepted for publication in 2000 in Embedded Technologies, but again the publication process was interrupted. It is a pleasure now to be able to revise the paper in the light of recent developments in Late Cypriot ceramics, and to offer it in appreciation of Eddie’s significant contribution to Cypriot archaeology. Thanks are due to the anonymous reviewer for comments and to Steve Thomas (University of Wales Lampeter), who read a final draft of the paper.

List of Abbreviations BR CG LBA LC MBA MC WPWM III WS

Base Ring Cypro-Geometric Late Bronze Age Late Cypriot Middle Bronze Age Middle Cypriot White Painted Wheelmade III White Slip

114

Louise Steel

References Adams, E. 2006. Social strategies and spatial dynamics in neopalatial Crete: An analysis of the north-central area. American Journal of Archaeology 110, 1–36. Aloupi, E., V. Perdikatsis and A. Lekka 2001. Assessment of the White Slip classification scheme based on physico-chemical aspects of the technique. In V. Karageorghis (ed.) The White Slip Ware of Late Bronze Age Cyprus. Proceedings of an International Conference Organized by the Anastasios G. Leventis Foundation, Nicosia, in Honour of Malcolm Wiener. Nicosia 29th–30th October 1998, 15–26. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 2. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Al-Radi, S. M. S. 1983. Phlamoudhi Vounari: A Sanctuary Site in Cyprus. Studies in Mediterranean Archaeology 65. Göteborg, Paul Åströms Förlag. Arnold, D. E. 1985. Ceramic Theory and Cultural Process. Cambridge, Cambridge University Press. Amiran, R. 1970. Ancient Pottery of the Holy Land, from its Beginnings in the Neolithic Period to the End of the Iron Age. New Brunswick (NJ), Rutgers University Press. Artzy, M. 2001. A study of the Cypriote Bichrome Ware: Past, present and future. In P. Åström (ed.) The Chronology of BaseRing Ware and Bichrome Wheel-made Ware. Proceedings of a Colloquium Held in the Royal Academy of Letters, History and Antiquities, Stockholm, May 18–19, 2000, 157–174. Stockholm, Royal Academy of Letters, History and Antiquities. Artzy, M., I. Perlman and F. Asaro 1981. Cypriote pottery imports at Ras Shamra. Israel Exploration Journal 31, 37–47. Åström, P. 2001. Bichrome Hand-made Ware and Bichrome Wheelmade Ware on Cyprus. In P. Åström (ed.) The Chronology of Base-Ring Ware and Bichrome Wheel-made Ware. Proceedings of a Colloquium Held in the Royal Academy of Letters, History and Antiquities, Stockholm, May 18–19, 2000, 131–142. Stockholm, Royal Academy of Letters, History and Antiquities. Åström, P. (ed.) 2001. The Chronology of Base-Ring Ware and Bichrome Wheel-made Ware. Proceedings of a Colloquium Held in the Royal Academy of Letters, History and Antiquities, Stockholm, May 18–19, 2000. Stockholm, Royal Academy of Letters, History and Antiquities. Binford, L. R. 1962. Archaeology as anthropology. American Antiquity 28, 217–225. Bolger, D. 2003. Gender in Ancient Cyprus: Narratives of Social Change on a Mediterranean Island. Walnut Creek (CA), Altamira. Cadogan, G., E. Herscher, P. Russell and S. Manning 2001. MaroniVournes: A long White Slip sequence and its chronology. In V. Karageorghis (ed.) The White Slip Ware of Late Bronze Age Cyprus. Proceedings of an International Conference Organized by the Anastasios G. Leventis Foundation, Nicosia, in Honour of Malcolm Wiener. Nicosia 29th–30th October 1998, 75–88. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 2. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Catling, H. W. 1957. The Late Bronze Age pottery. In J. du Plat Taylor, Myrtou Pigadhes: A Late Bronze Age Sanctuary in Cyprus, 26–59. Oxford, Ashmolean Museum.

Catling, H.W. 1962. Patterns of settlement in Bronze Age Cyprus. Opuscula Atheniensia 4, 129–169. Courtois, J.-C. 1971. Le sanctuaire du Dieu au L’ingot d’EnkomiAlasia (Chypre). In C. F. A. Schaeffer (ed.) Alasia I, 151–362. Paris, Klincksieck. Courtois, L. 1977. Céramique et métallurgie anciennes: Le bol à lait de Chypre. Bulletin Académie et Société des Sciences 16, 9–17. Courtois, L. and B. Velde 1989. Petrographic studies of pottery (preliminary report). In A. K. South (ed.) Kalavasos Ayios Dhimitrios II. Ceramics, Objects, Tombs, Specialist Studies. Vasilikos Valley Project 3, 73–77. SIMA 71(3). Göteborg, Paul Åströms Förlag. Crewe, L. 2004. Social Complexity and Ceramic Technology on Late Bronze Age Cyprus: The New Evidence from Enkomi. Unpublished Ph.D. thesis, University of Edinburgh. Crewe, L. 2007. Sophistication in simplicity: The first production of wheelmade pottery on Late Bronze Age Cyprus. Journal of Mediterranean Archaeology 20, 209–238. Dikaios, P. 1969. Enkomi. Excavations 1948–1958, vol. 1. Mainz on Rhein, Philipp von Zabern. Dothan, T. and A. Ben-Tor 1983. Excavations at Athienou, Cyprus. 1971–1972. QEDEM 16. Jerusalem, The Hebrew University. Feinman, G., S. A. Kowalewski and R. E. Blanton 1984. Modelling ceramic production and organisational change in the preHispanic valley of Oaxaca, Mexico. In S. E. van der Leeuw and A. C. Pritchard (eds) The Many Dimensions of Pottery: Ceramics in Archaeology and Anthropology, 295–333. Amsterdam, Universiteit van Amsterdam. Griffin, J. B. 1965. Ceramic complexity and cultural development: The eastern United States as a case study. In F. R. Matson (ed.) Ceramics and Man, 104–113. Chicago, Aldine. Hadjisavvas, S. 1991. LC IIC to LC IIIA without intruders: The case of Alassa-Pano Mandilaris. In J. A. Barlow, D. L. Bolger and B. Kling (eds) Cypriot Ceramics: Reading the Prehistoric Record, 173–180. University Museum Monographs 74. Philadelphia, University Museum, University of Pennsylvania. Herscher, E. 1984. The pottery of Maroni and regionalism in Late Bronze Age Cyprus. In V. Karageorghis and J. D. Muhly (eds) Cyprus at the Close of the Late Bronze Age, 23–28. Nicosia, A. G. Leventis Foundation. Herscher, E. 2001. Early Base Ring Ware from Phaneromeni and Maroni. In P. Åström (ed.) The Chronology of Base-Ring Ware and Bichrome Wheel-made Ware. Proceedings of a Colloquium Held in the Royal Academy of Letters, History and Antiquities, Stockholm, May 18–19, 2000, 11–21. Stockholm, Royal Academy of Letters, History and Antiquities. Iacovou, M. and D. Michaelides (eds) 1999. Cyprus: The Historicity of the Geometric Horizon. Nicosia, Archaeological Research Unit, University of Cyprus. Jones, R. E. and H. W. Catling 1986. Cyprus 2500–500 BC; the Aegean and the Near East 1500–1050 BC. In R. E. Jones (ed.), Greek and Cypriot Pottery. A Review of Scientific Studies, 523–625. Fitch Laboratory Occasional Papers 1. Athens, British School of Archaeology. Karageorghis, V. 2001. Bichrome Wheelmade Ware: Still a problem? In P. Åström (ed.) The Chronology of Base-Ring Ware and Bichrome Wheel-made Ware, 143–155. Proceedings of a Colloquium Held in the Royal Academy of Letters, History and

13. Late Cypriot ceramic production: Heterarchy or hierarchy? Antiquities, Stockholm, May 18–19, 2000. Stockholm, Royal Academy of Letters, History and Antiquities. Karageorghis, V. (ed.) 2001. The White Slip Ware of Late Bronze Age Cyprus. Proceedings of an International Conference Organized by the Anastasios G. Leventis Foundation, Nicosia, in Honour of Malcolm Wiener. Nicosia 29th–30th October 1998. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 2. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Keswani, P. S. 1991. A preliminary investigation of systems of ceramic production and distribution in Cyprus during the Late Bronze Age. In J. A. Barlow, D. L. Bolger and B. Kling (eds) Cypriot Ceramics: Reading the Prehistoric Record, 97–118. University Museum Monographs 74. Philadelphia, University Museum, University of Pennsylvania. Keswani, P. S. 1993. Models of local exchange in Late Bronze Age Cyprus. Bulletin of the American Schools of Oriental Research 292, 73–83. Keswani, P. S. 1996. Hierarchies, heterarchies, and urbanisation process: The view from Bronze Age Cyprus. Journal of Mediterranean Archaeology 9, 211–250. Keswani, P. S. 2004. Mortuary Ritual and Society in Bronze Age Cyprus. Monographs in Mediterranean Archaeology. London, Equinox. Knapp, A. B. 1997. The Archaeology of Late Bronze Age Cypriot Society: The Study of Settlement, Survey and Landscape. Glasgow, University of Glasgow Press. Knapp, A. B. 2008. Prehistoric and Protohistoric Cyprus: Identity, Insularity, Connectivity. Oxford, Oxford University Press. Knapp, A. B. and J. F. Cherry 1994. Provenience Studies and Bronze Age Cyprus: Production, Exchange and PoliticoEconomic Change. Monographs in World Archaeology 21. Madison, Prehistory Press. Knappett, C. 1999. Assessing a polity in protopalatial Crete: The Malia-Lasithi state. American Journal of Archaeology 103, 615–639. Knappett, C. and V. Kilikoglou 2007. Provenancing Red Lustrous Wheelmade ware: Scales of analysis and floating fabrics. In I. Hein (ed.) The Lustrous Wares of Late Bronze Age Cyprus and the Eastern Mediterranean: Papers of a Conference, Vienna 5th–6th of November 2004, 115–140. Denkschriften der Gesamtakademie 41. Contributions to the Chronology of the Eastern Mediterranean 13. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Lagarce, J. and E. Lagarce 1985. Alasia IV. Deux Tombes du Chypriote Récente d’Enkomi (Chypre). Tombes 1851 et 1907. Paris, Éditions Recherche sur les Civilisations. Loney, H. 2000. Society and technological control: A critical review of models of technological change in ceramic studies. American Antiquity 65, 646–668. Maisels, C. K. 1990. The Emergence of Civilization: From Hunting and Gathering to Agriculture. Cities and the State in the Near East. London and New York, Routledge. Manning, S. W. 2001. The chronology and foreign connections of the Late Cypriot I period: Times they are a-changin’. In P. Åström (ed.) The Chronology of Base-Ring Ware and Bichrome Wheel-Made Ware, 69–94. Proceedings of a Colloquium Held

115

in the Royal Academy of Letters, History and Antiquities, Stockholm, May 18–19, 2000. Stockholm, Royal Academy of Letters, History and Antiquities. Masson, E. 1976. A la recherche des vestiges Proche-Orientaux à Chypre : Fin du Bronze Moyen et début du Bronze Récent. Archäologischer Anzeiger 1976(2), 139–165. Mee, C. and L. Steel 1998. Corpus of Cypriote Antiquities 17: The Cypriote Collections in the University of Liverpool and the Williamson Art Gallery and Museum. Studies in Mediterranean Archaeology 20(17). Jonsered, Paul Åströms Förlag. Merrillees, R. S. 1971. The early history of Late Cypriote I. Levant 3, 56–79. Peacock, D. P. S. 1977. Ceramics in Roman and Medieval archaeology. In D. P. S. Peacock (ed.) Pottery and Early Commerce: Characterization and Trade in Roman and Later Ceramics. London, Academic Press. Peacock, D. P. S. 1982. Pottery in the Roman World: An Ethnoarchaeological Approach. New York, Longman. Peltenburg, E. J. 1991. Local exchange in prehistoric Cyprus: An initial assessment of picrolite. Bulletin of the American Schools of Oriental Research 282/283, 107–126. Peltenburg, E. J. 1996. From isolation to state formation in Cyprus, ca. 3500–1500 B.C. In V. Karageorghis and D. Michaelides (eds) The Development of the Cypriot Economy from the Prehistoric Period to the Present Day, 17–44. Nicosia, University of Cyprus and Bank of Cyprus. Pilides, D. 1996. Storage jars as evidence of the economy of Cyprus in the Late Bronze Age. In V. Karageorghis and D. Michaelides (eds) The Development of the Cypriot Economy from the Prehistoric Period to the Present Day, 107–124. Nicosia, University of Cyprus and Bank of Cyprus. Popham, M. R. 1972. White Slip ware. In The Swedish Cyprus Expedition IV(1C): The Late Cypriote Bronze Age, Architecture and Pottery, 431–471. Lund, The Swedish Cyprus Expedition. Rice, P. M. 1981. Evolution of specialized pottery production: A trial model. Current Anthropology 22, 219–240. Rice, P. M. 1984. Change and conservatism in pottery producing systems. In S. E. van der Leeuw and A. C. Pritchard (eds) The Many Dimensions of Pottery. Ceramics in Archaeology and Anthropology, 233–288. Amsterdam, Universiteit van Amsterdam. Rice, P. M. 1987. Pottery Analysis: A Sourcebook. Chicago, University of Chicago Press. Russell, P. J. 1989. The fine ware ceramics: The settlement deposits in the west, central, east and south-east areas. In A. K. South (ed.) Vasilikos Valley Project 3: Kalavasos-Ayios Dhimitrios II. Ceramics, Objects, Tombs, Specialist Studies, 1–11. Studies in Mediterranean Archaeology 71(3). Göteborg, Paul Åströms Förlag. Sahlins, M. 1972. Stone Age Economics. New York, de Gruyter. Schoep, I. and C. Knappett 2004. Dual emergence: Evolving heterarchy, exploding hierarchy. In J. C. Barrett and P. Halstead (eds) The Emergence of Civilisation Revisited, 21–37. Sheffield Studies in Aegean Archaeology 6. Oxford, Oxbow. Sherratt, E. S. 1991. Cypriot pottery of Aegean type in LC II–III: Problems of classification, chronology and interpretation. In J. A. Barlow, D. L. Bolger and B. Kling (eds) Cypriot Ceramics:

116

Louise Steel

Reading the Prehistoric Record, 185–198. University Museum Monographs 74. Philadelphia, University Museum, University of Pennsylvania. Smith, J. S. 1994. Seals for Sealing in the Late Bronze Age. Unpublished Ph.D. thesis, Bryn Mawr College. South, A. K., P. Russell and P. Keswani 1989. Vasilikos Valley Project 3. Kalavasos Ayios Dhimitrios II: Ceramics, Objects, Tombs, Specialist Studies. SIMA 71(3). Göteborg, Paul Åströms Förlag. South, A. and L. Steel 2001. The White Slip sequence at Kalavasos. In V. Karageorghis (ed.) The White Slip Ware of Late Bronze Age Cyprus. Proceedings of an International Conference Organized by the Anastasios G. Leventis Foundation, Nicosia, in Honour of Malcolm Wiener. Nicosia 29th–30th October 1998, 65–74. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 2. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Steel, L. 1994. Pottery production in Cyprus in the eleventh century B.C. In V. Karageorghis (ed.) Cyprus in the Eleventh Century B.C. Nicosia, Leventis Foundation. Steel, L. 1998. The social impact of Mycenaean pottery in Cyprus. Annual of the British School at Athens 93, 285–296. Todd, I. A. and M. Hadjicosti 1991. Excavations at Sanidha 1990. Report of the Department of Antiquities, Cyprus, 37–74. Todd, I. A. and D. Pilides 1993. Excavations at Sanidha 1992. Report of the Department of Antiquities, Cyprus, 97–146. Todd, I. A. and D. Pilides 2001. The archaeology of White Slip production. In V. Karageorghis (ed.) The White Slip Ware of Late Bronze Age Cyprus. Proceedings of an International Conference Organized by the Anastasios G. Leventis Foundation, Nicosia,

in Honour of Malcolm Wiener. Nicosia 29th–30th October 1998, 15–26. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 2. Vienna, Verlag der Österreichischen Akademie der Wissenschaften. Trigger, B. G. 2003. Understanding Early Civilizations: A Comparative Study. Cambridge, Cambridge University Press. Underhill, A. P. 1991. Pottery production in chiefdoms: The Longshan period in northern China. World Archaeology 23, 12–27. Vaughan, S. J. 1987. A Fabric Analysis of Late Cypriot Base Ring Ware: Studies in Ceramic Technology, Petrology, Geochemistry and Mineralogy. Unpublished Ph.D. thesis, University College London. Vaughan, S. J. 1991. Material and technical classification of Base Ring Ware: A new fabric typology. In J. A. Barlow, D. L. Bolger and B. Kling (eds) Cypriot Ceramics: Reading the Prehistoric Record, 119–130. University Museum Monographs 74. Philadelphia, University Museum, University of Pennsylvania. Vaughan, S. J. 1994. Base Ring Ware: A regional study in Cyprus. In A. B. Knapp and J. F. Cherry (eds) Provenience Studies and Bronze Age Cyprus. Production, Exchange and Politico-Economic Change, 86–92. Monographs in World Archaeology 21. Madison, Prehistory Press. Vermeule, E. and F. Z. Wolsky 1990. Toumba tou Skourou: A Bronze Age Potters’ Quarter on Morphou Bay in Cyprus. Cambridge (MA), Harvard University Press. Wiessner, P. 1983. Style and social information in Kalahari San projectile points. American Antiquity 48(2), 253–276. Wiessner, P. 1985. Style or isochrestic variation? A reply to Sackett. American Antiquity 50(1), 160–166.

14 THE DOMESTICATION OF STONE: EARLY LIME PLASTER TECHNOLOGY IN THE LEVANT Gordon Thomas

When considering the place of technology in prehistory, archaeological research tends to focus on practical aspects of how it is conducted, what its material correlates are or the manner in which it is transmitted between groups and across time. The origins of any particular technology are only vaguely considered and are usually assumed to be the adaptive product of trial and error over many generations. I have long been uneasy with this rather simplistic approach that makes many assumptions about the nature of technology and also about the way both human society and the human mind work. I hope within this essay to stretch our understanding a little of the origins of one particular technology, that of lime plaster making in the Levant, and although I can provide no definitive answer to the question, I hope nevertheless to present a possible way forward in our thinking by opening up the way we look at this technology and by questioning some of the assumptions we make about it. I will argue that lime plaster making is a complex, counter-intuitive technology which developed in the Near East during the Natufian period prior to the inception of the full Neolithic. I also argue against the fortuitous or casual development of such a technology, seeing it rather as resulting from specific relationships forged between people and various elements of their environment which emerged from new systems of belief and practice. In this I turn to the work of David Lewis-Williams and David Pearce (2005), as well as Jacques Cauvin (2000), as sources of inspiration.

this discussion I will refer to lime plaster only where it has been reliably identified by scientific analysis; all vague and generalised references to ‘plasters’ encountered in excavation reports are ignored as these frequently make no distinction between innovative mud technology and actual lime or gypsum plasters. In particular, I will rely on the work of W. David Kingery, Pamela B. Vandiver and Martha Prickett (1988) for their analysis of many of the most important early samples from the Near East. Lime plaster making is a complex and, at times, counterintuitive technology which requires a specific cycle of manufacture (Fig. 14.1). Quarried limestone chunks (CaCO3), pure calcium carbonate, are first burnt in a kiln (Fig. 14.2) with temperatures of 800–900˚ centigrade, effectively driving off all carbon dioxide and leaving a powdery substance

Lime Plaster Making The identification of lime plaster is a skill which is achieved only by experience but is more reliably achieved through specific chemical and microscopic analyses. Throughout

Fig. 14.1 The lime cycle showing the chemical changes to the material during the process.

118

Gordon Thomas

Fig. 14.2 A simple lime burning kiln. The kiln is about one metre in diameter and will produce enough material to cover the interior walls of one of the five metre diameter houses at the Lemba Experimental Village in Cyprus (Thomas 2004).

called quicklime (CaO) (Fig.14.3). When slaked in water the quicklime reacts violently and gives off a tremendous heat and smell, creating over a length of time lime putty or calcium hydroxide (Ca(OH)2) (Fig.14.4). Under modern, airtight conditions this material can be stored for considerable periods of time and indeed improves in condition and quality. However, as soon as the material is used and is therefore exposed to air, another reaction takes place in which the water is driven from the paste and carbon dioxide is reabsorbed from the air back into the material, thus re-establishing its original chemical composition, CaCO3, and creating a strong, durable plaster with chemical characteristics and properties similar to the original limestone. More advanced use of the material with the addition of aggregates or animal hair can further improve the performance of the lime plaster. Analyses of many samples of lime plaster from early sites in the Near East indicate that calcination at the burning stage was often incomplete, resulting in the inclusion of limestone fragments and pieces of charcoal within the final lime plaster (Kingery et al. 1988, 224–225, 228–231).

The Lime Plaster Industry in the Levant

Fig. 14.3 Powdered, burnt limestone direct from the kiln showing the disintegration of the limestone as it is turned into quicklime. Each chunk of calcined limestone weighs between one and two kilogrammes.

Fig. 14.4 Lime putty after slaking the quicklime in water for several weeks.

The Natufian period in the Levant (also known as the late Epi-Palaeolithic) is a period of remarkable change in which there was a shift from simple to complex subsistence systems involving the intensive use of certain plant resources and the development of increasingly complex relationships with specific animal food sources, most notably gazelle. New and quite variable tool technologies emerged, including, for the first time, ground stone tool technology, and there was also a florescence of figurative and decorative ‘art’. Significantly for this discussion, the early Natufian core area in north and central Israel and Jordan saw the emergence of the first large semi-permanent settlements. Semi-subterranean structures, some with a concentric ring of postholes, have been recorded at a number of sites across the region with the best preserved being found at ‘Ain Mallaha (Perrot 1960; 1966; 1975). This period is seen as the time in which the foundations of many of the key aspects of the ‘Neolithic revolution’, which characterised the succeeding periods, were established (Simmons 2007; Rollefson 2008; Olszewski 2008). It was also during this time that the knowledge of making lime plaster first emerged in the Near East. The lime plaster industry is thought to have emerged as a complex pyrotechnic process during the Natufian phase in the Levant c. 12,800–10,500 BP (i.e. before the archaeological present, c. 1950). One of the best preserved sites, as far as the architectural evidence is concerned, is the site of Hayonim Cave, dated to c. 12,400–12,000 BP, which produced evidence of a possible lime plaster ‘kiln’ (Kingery et al. 1988, 223). The excavator (Bar Yosef 1983) describes six rounded, semi-subterranean, stone-built structures

14. The domestication of stone: Early lime plaster technology in the Levant constructed in a cluster at the mouth of the cave. Most of these structures contained a hearth, but one in particular was 2.5 m in diameter and contained a hearth with a 0.20 m thick layer of a white, porous material (Bar Yosef 1983). This was identified through scanning electron microscopy and energy dispersive x-ray analysis as being almost pure lime plaster (CaCO3) with some residual limestone fragments (Kingery et al. 1988, 223). These buildings, as well as the slightly later structures built on the terrace in front of the cave (Valla 1998), produced several hearths set with limestone blocks and pits lined with limestone slabs. Limestone was also used quite extensively during the Natufian for ground stone mortars at Hayonim Cave and as slabs for tables or sometimes with cup-marks and set beside hearths as at Ramat Harif (Valla 1998). At the early Natufian site of ‘Ain Mallaha a remarkable set of buildings is preserved set into the hill slope such that the buildings are semi-subterranean with one side open downslope. The buildings are circular or semi-circular and are constructed in carefully set dry-stone walls with internal space subdivided by partitions represented by postholes and low dividing walls. Several hearths, each of slightly different form, are preserved in one of the best surviving houses often in association with limestone kerbs or slabs. Against one of the walls was a ‘bench’ covered with a degraded material, identified by Kingery and his colleagues as an incompletely calcined lime plaster which had been metamorphosed by prolonged exposure (Kingery et al. 1988, 228). However, Valla also describes a wall which at one point had been “… replaced by a covering of crushed limestone coated with red paint,” which he regards as “… one of the earliest examples of an artificial construction material (1998, 172). It is unclear whether or not these two things are the same feature. There is also some evidence that this technology was known by at least the Epi-Palaeolithic (c. 14,000 BP) and was used as “… an adhesive to assemble a tool from flint microliths…” from Lagama North VIII (Kingery et al. 1988, 219). However, this latter example is unusual in that it is so early, pre-dating Natufian developments, and it comes from a site in the Sinai which lies outside the core Natufian area. What is clear from this very brief summary of early lime plasters in the Levant is that during the Natufian period where it first emerged, limestone was being used extensively in the construction of houses and in fixtures and fittings within these houses. We also have evidence which places this limestone in conjunction with hearths and storage pits, frequently lining them or as a foundation stone within them. This close relationship between limestone, fire and possibly water in a domestic setting makes the fortuitous ‘discovery’ of the technology of lime plaster making an obvious and justifiable assumption. However, there are two main problems with this line of reasoning, as I will argue below. The first problem relates to the methods used for modern plaster identifications, and the second relates to modern concepts about the discovery and development of technology.

119

The Identification of Lime Plasters The first problem arises with the methods used to analyse early plasters. Kingery et al. (1988) base their identification on chemical analyses and on the microscopic determination of the micro-structure of the material. They were concerned primarily with the distinction between lime and gypsum plasters on the basis of the scanning electron microscope identification of either the needle-like morphology of gypsum plaster or the colloidal spherules of lime plaster. Microscopic analysis of lime plasters from sites in the Levant by Goren and Goldberg (1991) has revealed that previous analysis based only on chemical identifications can be misleading since it fails to distinguish between limestone and lime plaster, both of which appear as CaCO3. It is the appearance of the crystals and the presence of microfossils under the microscope which can determine the composition of the plaster and the extent of the use of true burnt lime. Kingery et al. make no mention of micro-fossils although they do describe incompletely calcined material, which may suggest some of the samples contained unburnt limestone. In many cases Goren and Goldberg were able to determine that burnt lime formed a much smaller percentage of the total bulk of the plasters than was thought to be the case, with 30% being a more realistic figure for most samples. They also suggest that firing temperatures for the production of lime were never as high or as prolonged as had previously been claimed. These findings are also borne out by my own research using experimental archaeology to reconstruct the châine opératoire of the lime plaster cycle (Thomas 2004). This considerably alters our perception of the early stages of the development of this technology and suggests that primitive lime plaster could potentially have been made under domestic conditions. However, this does not answer the question of how it was discovered; it merely indicates to us that the wherewithal to create lime plasters already existed by the early Natufian period. The second problem regarding concepts of the early discovery of lime plaster making concerns the assumptions we make about the role of ancient technology. When considering questions about the origins and development of complex technologies in antiquity we must also consider the ancient mind. Lewis-Williams and Pearce (2005) have led us into a very plausible interpretation of the workings of the ancient mind. In the absence of any analytical form of reasoning and of a world view informed by the scientific method we cannot assume that ‘discovery’ and invention in prehistory came about through trial and error or by experimental practice. We cannot assume that the path to domestication or to the development of a particular technology was goal oriented. That would presume that the ancient mind was able to theorise about a concept, for example lime plaster or domestic animals, for which it had no previous knowledge or experience. If we were to make

120

Gordon Thomas

such an argument we would run the risk of teleological reasoning whereby the end result would be considered to be the cause of a particular technology. Technology develops when human behaviour and the natural behaviour of plants and animals or the properties of materials coincide. The domestication of plants and animals and the exploitation of natural resources occur because people choose to put themselves into a close relationship with them. Lewis-Williams and Pearce suggest that the domestication of plants and animals itself had little to do with food procurement and the security of the food supply (2005, 139). They develop a very convincing argument characterising the emergence of domestication as a human-plant/animal relationship which evolved through shamanistic practices and beliefs. The relationship works and persists when the behaviour of both come together or coincide with one another. I suggest that the same sort of cognitive processes were in operation with the domestication of plants or animals and the exploitation of other aspects of the environment. However, in order to examine this statement it is first necessary to discuss some of the commonly held views or assumptions about the emergence of lime plaster technology.

Two Models of the Early ‘Discovery’ of Lime Plaster Technology What sort of cultural behaviour makes the acquisition and development of a particular technology more likely? This question addresses the issue of the link between cognitive functions, cultural practices, and the physical properties of materials that leads to the development of particular technologies. In the context of this essay we must ask what possible cognitive processes could have led people to burn lumps of limestone at very high temperatures, store the resulting powder in water for a length of time and then use it to create new types of objects and architectural features. I am wary of the notion that the emergence of complex technologies was a fortuitous or experimental process except in that it may have been the side result of other specific activities and beliefs. Kingery et al., however, appear to follow such a line of reasoning (1988, 220): A limestone hearth under a strong fire or a few limestone pebbles in a bright fire will have a surface layer transformed into quicklime (CaO) which, when mixed with water, gives off heat, reacts with hide or hair, and can be smeared on a surface to set as a hard, rock-like product….Thus, in lime rich areas, the ‘discovery’ of lime and gypsum plasters must have occurred many times.

I believe that there are problems with this sort of reasoning relating both to the logic used and to the practicalities of this technology. Firstly, there is the question of intent, which

I have outlined above, whereby we cannot assume the ability to theorise a final result. I suggest that in Kingery’s ‘campfire hypothesis’ there is no obvious sequence of events which could have led to the fortuitous discovery of lime plaster. Secondly, as anyone who has experimented with this technology will know, there are the practical aspects of lime burning whereby very strong, unpleasant and toxic fumes are released during the calcination process that would not have been tolerated around a domestic hearth or during cooking. So, it is likely that either the normal domestic hearth would not have been able to achieve temperatures high enough to allow calcination to take place or, if it was possible, that it did not occur within a domestic setting. In either case it makes the ‘campfire hypothesis’ for the discovery of lime plaster making improbable. For these reasons it is also unlikely that the lime plaster recorded on the hearth at Hayonim Cave was created in situ. Experimental work has shown that lime burning kilns do not necessarily leave a skim of lime plaster around the base of the kiln (Thomas 2004), which is the assumption implicit in describing the Hayonim hearth as a lime-burning kiln. Additionally, the high temperatures needed to achieve calcination would have ignited the structure within which the hearth was located, and the slaking process itself could not have occurred on the hearth itself. A more likely scenario would have the lime plaster being created elsewhere and then applied to the hearth as a final act. A second possible hypothesis, which I have not seen specifically articulated within a Near Eastern context, is the so-called ‘pot boiler hypothesis’. Across much of the British Isles during the Bronze and Iron Ages huge mounds of stones that have been used as pot boilers have accumulated over the centuries. These are stones which when heated in a fire and then plunged into water can be used to boil a container of water either for cooking or also possibly in saunas. Although this phenomenon has not specifically been identified in the Near East, there is some evidence that stones may have been used in a similar manner and, as a potential route for the discovery of lime plaster making, must be explored. On the face of it, this is a practice which has much to commend it as a potential source for the discovery of lime plaster making as it involves first heating stones in a fire before plunging them into water, both of which are necessary steps in the operational sequence for making lime plaster. It has the advantage over the ‘campfire hypothesis’ in that there is this obvious sequence of events which can be seen as the precursor of, or at least the basis for, the development of this technology. However, some of the arguments against the ‘campfire hypothesis’ pertain to the ‘pot-boiler hypothesis’ as well. Firstly, there is the question of intent and the risk of teleological reasoning, which would see the end result as driving the evolution of practice from the heating of pot-boiler stones for cooking to lime plaster making. Secondly, and probably more importantly from a

14. The domestication of stone: Early lime plaster technology in the Levant practical perspective, there are doubts that normal domestic fires could ever have heated the limestone to the required temperatures for calcination to take place, and there is also the problem that heated limestone, when plunged into water, reacts violently and dissolves, effectively contaminating the water or food. Ancient people would presumably have learned very quickly that all sedimentary stones are unsuitable as pot-boilers. So, the essential premise for a technology to develop – that repeated practice would lead to the accumulation of experience and knowledge necessary to develop a new technology – would not have occurred.

Experiment: The Way Forward? Clearly, if assumptions such as the campfire and the potboiler hypotheses are to be addressed as possible models of technological innovation, they need to be tested. With this in mind it is possible to devise a series of experiments and research projects designed to test specific aspects of ideas about this early industry in the Levant. The questions raised can be posed along the following lines: 1) Can we distinguish between crushed limestone paste and lime plaster in the archaeological record? 2) Can lime plaster be created using either the campfire or the pot-boiler hypotheses? 3) Can we identify the operational sequence of lime plaster making on Natufian sites? The first of these questions must address the potential issue about the identification of lime plasters mentioned above by carrying out a comprehensive set of analyses which build on the earlier work of Kingery, Vandiver and Prickett (Kingery et al. 1988). By replicating in a series of repeated experiments the practices associated with domestic hearths and cooking, such experimental texts could establish the validity of the assumptions about these two hypotheses. Finally, a new look at the archaeological material itself from the perspective of trying to identify the operational sequence of a specific technology would be a necessary and informative adjunct to any project of research. However, if through experimental testing we are to reject both of these assumptions or hypotheses about the early discovery of lime plaster making, is it possible to put in place an alternative way of looking at the technology or, at least, of theorising about its inception?

The Revolutionary Neolithic It is within the context of the emergence of proto-Neolithic aspects during the Natufian period and the critical role that technology played in its emergence that we must try to understand the place of the manufacture and use of lime plaster. Studies of technologies too frequently focus on the practical aspects of how they are conducted and how the

121

end result is used. Each technology, however, is situated within a social, cultural and ideational framework of belief and practice. Technologies do not emerge for practical reasons alone. The work of Pierre Lemonnier (1992; 2002) alerts us to the way in which technological practice can be deeply embedded within the social fabric and within belief systems and demonstrates how technological choice is not always informed by practical considerations of efficiency or usefulness. The development of a particular technology is usually along pathways forged by belief systems and concepts of cosmology, and it is through this that systems of cultural selection and choice come into play. David Lewis-Williams (2004) and with David Pearce (2005) have produced a very persuasive argument based on shamanistic beliefs and practice, beginning with the Upper Palaeolithic in Europe and reaching into the EpiPalaeolithic of the Near East, which link specific neurological functions within the human brain with belief and practice. They demonstrate possible mechanisms by which human experience (in this case, altered states of consciousness) can lead to the emergence of different belief systems and a change in the perception of the world and of the human relationship with it. In particular, they demonstrate how belief systems can bring people, plants, animals and materials into a closer, more dynamic relationship that brings about a fundamental alteration in the nature of these things: in other words, domestication. Cauvin (2000) also argues convincingly that the Neolithic in the Near East was not so much a revolution in technology and practice as a fundamental revolution in symbols and beliefs. To emphasise this, and in deference to Gordon Childe, he coined the term the ‘revolutionary Neolithic’, which entails wide ranging and deep seated changes in cultural beliefs and practices in which a very different way of seeing the world emerged. Like Lewis-Williams and Pearce he sees the early Neolithic as being characterised by a conflict of world views between mobile hunter-gatherers and sedentary farmers which was resolved within society through the elaboration of ritual practice. When turning to the development of technology we see that this highlights the serendipitous nature of discovery and invention in prehistory by which the coming together of diverse beliefs and practices can produce an outcome that is neither foreseen nor expected. In other words, the experience of shamanistic practice during the very late Palaeolithic led to a very different perception of the human relationship with the world, which in turn led to a fundamentally different form of relationship between people and their environment. This is at the core of the Neolithic experience. People began to perceive of the world differently and to treat things like plants, animals and other resources differently. This would apply as much to inanimate parts of the natural world as it would to living things, and in the context of this essay would apply equally to the use of stone and, in particular, limestone.

122

Gordon Thomas

Answering the question of why or how lime plaster was developed may also, therefore, partly answer the question about what the Neolithic is. Did the alteration in perception and behaviour which occurred with relation to the plant and animal kingdoms and led to their domestication also affect other things like limestone? Was stone ‘domesticated’ during the Natufian period in the Levant? In other words, through various social and possibly religious practices did people specifically start using limestone in a way that affected its composition and led to the discovery of lime plaster technology? There are no easy or obvious answers to these questions. Throughout this paper I have raised doubts concerning some of the assumptions made about the discovery and development of the lime plaster industry in the prehistoric Levant, and I have proposed ways of further investigating this phenomenon. I have suggested that the development of such a technology may have had much in common with changes in belief and behaviour that led to domestication firstly of plants and later of animals within the same time span. By investigating the realities of lime plaster production through experimental archaeology, and with the help of the analyses of prehistoric plasters by Gourdin and Kingery (1975), Kingery et al. (1988) and Goren and Goldberg (1991), the social role of the early lime plaster industry can be explored within the framework of changes that were taking place in the Levant during the Natufian period. Although experimental archaeology cannot give us a clear answer to the questions posed above, it does have the advantage of allowing us to develop our ideas about this fascinating period and this innovative technology from a better informed perspective.

References Bar Yosef, O. 1983. The Natufian in the Southern Levant. In T. Cuyler Young, P. E. L. Smith and P. Mortensen (eds) The Hilly Flanks and Beyond, 11–42. Studies in Ancient Oriental Civilisations 36. Chicago, Oriental Institute, University of Chicago.

Cauvin, J. 2000. The Birth of the Gods and the Origins of Agriculture. Translated from the French by T. Watkins. New Studies in Archaeology. Cambridge, Cambridge University Press. Goren, Y. and P. Goldberg 1991. Petrographic thin sections and the development of Neolithic plaster production in northern Israel. Journal of Field Archaeology 18, 131–138. Gourdin, W. H. and W. D. Kingery 1975. The beginnings of pyrotechnology: Neolithic and Egyptian lime plaster. Journal of Field Archaeology 2, 133–150. Kingery, W. D., P. B. Vandiver and M. Prickett 1988. The beginnings of pyrotechnology, part II: Production and use of lime and gypsum plaster in the Pre-Pottery Neolithic Near East. Journal of Field Archaeology 15, 219–240. Lemonnier P. 1992. Elements for an Anthropology of Technology. Anthropological Papers 88. Ann Arbor, Museum of Anthropology, University of Michigan. Lemonnier P. (ed.) 2002. Technological Choices: Transformations in Material Cultures since the Neolithic. London, Routledge. Lewis-Williams, D. 2004. The Mind in the Cave: Consciousness and the Origins of Art. London, Thames & Hudson. Lewis-Williams, D. and D. Pearce 2005. Inside the Neolithic Mind: Consciousness, Cosmos and the Realm of the Gods. London, Thames & Hudson. Olszewski, D. 2008. The Palaeolithic period, including the Epipalaeolithic. In R. Adams (ed.) Jordan: An Archaeological Reader, 35–69. London, Equinox. Perrot, J. 1960. Excavations at Eynan (‘Ein Mallaha): Preliminary report on the 1959 season. Israel Exploration Journal 10(1), 14–22. Perrot, J. 1966. Le gisement Natoufien de Mallaha (Eynan), Israel. L’Anthroplogie 70, 437–484. Perrot, J. 1975. Mallaha, Eynan. Paléorient 2, 485–486. Rollefson, G. 2008. The Neolithic period. In R. Adams (ed.) Jordan: An Archaeological Reader, 71–108. London, Equinox. Simmons, A. H. 2007. The Neolithic Revolution in the Near East: Transforming the Human Landscape. Tucson, University of Arizona Press. Thomas, G. D. 2004. Early lime plaster technology in the Near East: Experimental work at the Lemba Experimental Village, Cyprus. Experimentelle Archäologie in Europa 3, 91–100. Valla, F. 1998. The first settled societies – Natufian (12,500 – 10,200 BP). In T. Levy (ed.) The Archaeology of Society in the Holy Land, 169–187. London, Leicester University Press.

15 DOMESTICATION OF PLANTS AND ANIMALS, DOMESTICATION OF SYMBOLS? Danielle Stordeur

Introduction The most notable contribution of Jacques Cauvin to the understanding of the origins of the Neolithic is to have considered that a profound mental transformation, perceptible through a ‘revolution of symbols’ beginning in the Khiamian, was the principal driving force behind the advent of farming and herding (Cauvin 1977). Since his death, our knowledge has expanded, especially in the northern Levant, concerning occupations from the Prepottery Neolithic A (henceforth PPNA) to the Pottery Neolithic. Thus it is important to re-examine the evidence to learn whether other transformations may be inferred. What are the characteristics of the symbolism used by the populations which, from the PPNA Horizon to the early PPNB, were just beginning to master their environment? How do these symbols change, when in the middle PPNB agriculture became part of a genuine economic system? To attempt to answer these questions I will examine the cultures which succeeded each other from 9500 to 7000 BC (all dates in this paper are cal BC; phasing follows Aurenche et al. 1981): the PPNA (9500–8700 BC), the early PPNB (8700–8200 BC), the middle PPNB (8200–7500 BC), and the late PPNB (7500–7000 BC). I propose to discuss these phases chronologically in groups of two.

Between 9500 and 8200 BC: The Beginnings of Agriculture and Herding There are several arguments for linking the PPNA and the early PPNB, at least in the northern Levant where a transitional phase shows the progressive passage from one to the other (Stordeur and Abbès 2002). In this zone the principal architectural and social changes appeared in the PPNA. Agriculture remained pre-domestic in the two

phases, but animals began to be domesticated in the second (Helmer et al. 2005; Tanno and Willcox 2006). However, the two periods hardly differ in their symbolism. As recent evidence has mainly appeared in the northern Levant, we will start in this region and examine the southern Levant only for purposes of comparison.

Between 8000 and 7000 BC: Fully Agricultural Societies The middle and late PPNB may be considered jointly. The first is well defined and is set apart from the past by the fact that farming and herding had truly become part of an economic system. But there is nothing conclusive that sets the late PPNB apart in regard to its symbolism. There are clear differences between the northern and southern Levant. We will start this time in the south as it is there that new practices indicative of social and ideological changes are revealed.

Symbolism at the Time of the First Experiments in Farming For the sites of the northern Levant dated from 9500 to 8200 BC, there is no representation explicitly referring to the domesticated plants and animals which humans were beginning to master (Stordeur 2003; Helmer et al. 2004). What then were the themes which preoccupied them? Animals are represented in figurative representations or by their bones. For the aurochs, the largest animal of the period, deposits of bucrania are very frequent and found in collective contexts, almost always concealed, as at Mureybet (Cauvin 1997), Tell ‘Abr 3 (Yartah 2005a,b), Qaramel

124

Danielle Stordeur

(Mazurowski and Jammous 2000), Hallan Çemi (Rosenberg 1999) and Dja’de (Coqueugniot 2000). However, at Jerf el Ahmar (Stordeur 2003) bucrania of aurochs were hung on the walls of a small round house (Fig. 15.1, no. 1). The bucrania are sometimes enhanced. At Jerf el Ahmar, one of them was decorated with a necklace, while at Tell ‘Abr the horns were coated with earth before being concealed in a collective building. Representations of bucrania are engraved on small worked stones (Jerf el Ahmar: Fig. 15.1, no. 2) or large slabs (Tell ‘Abr 3). Representations of the entire animal are rarer. They emphasise the powerful appearance of the male aurochs, as at Göbekli (Fig. 15.1, no. 3; Schmidt 2002). The head of the animal is always turned to show the two horns. What does this animal evoke? The evidence appears to suggest a masculine symbol, as is the case in many cultures if the horns are considered (Valla 2003) rather than the specific animal. Moreover, gazelles (Tell ‘Abr 3) or rams (Göbekli) and also horned animals are sometimes represented. Among felines, panthers are represented at Tell ‘Abr 3 (Fig. 15.1, no. 4) where they are engraved on slabs embellishing the bench of a communal building. They are seen from above in a static position as at Jerf el Ahmar (Fig. 15.1, no. 11) and Qaramel (Mazurowski 2004), while in Anatolia they are sometimes shown in dynamic and aggressive attitudes. Thus on a pillar at Göbekli, a male panther, standing erect, shows its teeth (Fig. 15.1, no. 5). How should this large nocturnal feline be considered? The representation of the threatening animal contrasts strongly with that of the animal at rest. The fox, the only canid identified, figures on the pillars of Göbekli (Fig. 15.1, no. 3) and a grooved stone from Jerf el Ahmar (Fig. 15.1, no. 6). What does it evoke? Not considered to be frightening, the fox is attributed a wily character and frequently plays a double role. Its relation with death is sometimes referred to (myth of Orpheus). Among the birds, cranes (Fig. 15.1, no. 3) and bustards are represented, but birds of prey are the most frequent. At Jerf el Ahmar vultures are represented on small stones. Even when the bird is frontally depicted, the head is in profile so that the beak is clearly seen (Fig. 15.1, no. 6), like raptors in modern heraldry. While the vulture is a scavenger with a negative connotation in Europe, in certain oriental traditions its relation to death is enhanced. For example, in Zoroastrian funerary rites the bodies of the dead, deposited on the tops of towers, were taken by vultures. Such practices continue in the Himalayas where the dead are abandoned to vultures, symbolically charged with carrying them to the hereafter. Certain representations of the 10th millennium suggest the existence of a relation of this type. A bird of prey is perched on the heads of two human figures at Nevali Çori (Hauptmann 1999). A slab engraved with two headless human representations is framed by two steles sculpted in the form of raptor heads at Jerf el Ahmar (Fig.

15.1, nos 7–8). But the later site of Çatal Hüyük provides the clearest representations, depicting birds of prey in flight which appear to be carrying away headless human bodies (Fig. 15.1, no. 10; Mellaart 1967). It is notable in the archaeological examples that the connection between the vulture and the human specifically relates to the head. The bird is depicted frontally when it is perched on the heads and reversed when it is associated with headless humans. The very frequent representation of the snake is sometimes realistic as at Nevali Çori (Hauptmann 1999) where it decorates the back of a human head (Fig. 15.1, no. 11). The representation is often reduced to an undulating line ending in a triangle as at Jerf el Ahmar (Fig. 15.1, nos 6, 12), Dja’de, Tell ‘Abr 3 and Qaramel. The snake is omnipresent in the Snake Building at Göbekli (Schmidt 2002). Its symbolism is contrasting – dangerous animal, chthonic, healer – and it appears in all cases to be linked to death, directly or in reversed fashion. Finally, the scorpion is almost always considered primarily as dangerous although it is also regarded as a maternal symbol. It is found on a grooved stone from Jerf el Ahmar in an accurate and realistic style (Fig. 15.1, no. 13) and on a megalithic pillar at Göbekli (Schmidt 2007). It is noteworthy that its antennae are also evocative of horns. Representations of the human figure are rare in the northern Levant. At Mureybet female figurines are well attested but are not found on other PPNA sites. They become more numerous, however, in the early PPNB (Dja’de el Mughara, Nevali Çori) while animal figures become less frequent (Schmidt 1999; Helmer et al. 2004). At Göbekli an expression of femininity is present in an ‘immodest’ engraving of a woman with spread legs. The head and the body are always dissociated at Jerf el Ahmar, whether in representations or human remains. Several deposits of skulls are hidden in communal buildings while a headless skeleton lay upon the floor in one of them (Fig. 15.1, no. 9). Two small heads could be parts of statuettes with a groove at the back to allow them to be fixed to the bodies. Conversely, we have already mentioned the headless human representations from a communal building (Fig. 15.1, no. 7). In the southern Levant none of the animal themes of the north are found. Female representations are dominant. According to the evidence of material culture, it certainly appears that the north and the south are radically different from one another. The northern Levant is characterised by a unique system of symbols in which the animal themes form a coherent, powerful whole. In some cases the wild animals could be confused with domestic animals (wild boar/pig, aurochs/ox), but wild animals are dominant. The human figure, discreet at first, emerges more strongly with time. It is sometimes combined with an animal figure, and the couple thus formed is significant. Besides the ‘woman and bull’ couple identified by Cauvin, we have brought to light the link which unites

15. Domestication of plants and animals, domestication of symbols?

125

Fig. 15.1 The symbolic world of the Pre-pottery Neolithic A and the early Pre-Pottery Neolithic B. (1) house with bucrania at Jerf el Ahmar (photo by the author); (2) plaques engraved with bucrania from Jerf el Ahmar; (3) sculpted pillar at Göbekli Tepe: bull, fox and crane; (4) panthers engraved on a slab at Tell ‘Abr 3; (5) threatening panther at Göbekli Tepe; (6) grooved stone with fox, snakes and vulture from Jerf el Ahmar; (7–8) headless human and stele with sculpted bird of prey from a communal building at Jerf el Ahmar; (9) headless skeleton at Jerf el Ahmar; (10) painting with vultures and headless bodies at Çatal Hüyük (Mellaart 1967). (11) sculpted snake on the back of a human head from Nevali Çori (Hauptman 1999); (12) plaque engraved with snakes and a panther from Jerf el Ahmar; (13) grooved stone with scorpion from Jerf el Ahmar.

126

Danielle Stordeur

vultures with humans, and with death. In order to firmly establish that these animal and human ‘personages’ form a true ‘system of symbols’, a great deal of analysis remains to be done, but from our first attempts we can observe a recurring pattern. It resides in the ambiguous character of the dominant themes, which evoke both life and death, masculine and feminine. But is there a logical relation between this possible ‘system’ and the preoccupations of the populations which were beginning to master their environment? We have not seen a direct reference to this mastery. As the link during this first stage is far from being clear, it is necessary to examine what happens next in the story. Do we see real changes in symbolism among the populations which became true farming societies?

Symbolism in Fully Agricultural Societies Between the early PPNB and the middle PPNB, there is a rupture detectible everywhere, especially in the southern Levant, a region where influences from the north had begun to mix with earlier traditions. In examining the entire middle/late PPNB, we often begin with the evidence from the south, as it is there that new practices appear which are revelatory of a change in mentalities. Animal symbolism persists with representations and deposits of bones. Animal representations (Fig. 15.2, nos 1–2) throughout the Levant are similar. Everywhere, clayfigurines represent goats, sheep, oxen, pigs and more rarely foxes. These are almost always domestic animals. Thus a primary difference with the preceding period becomes apparent. Representations of powerful or dangerous animals have disappeared. The representations are small and stereotyped. From the southern to the northern Levant representations of oxen are similar at Tell Aswad (Fig. 15.2, no. 1), ‘Ain Ghazal (Rollefson 1983), Gritille (Voigt 1985), Mezraa Teleilat (Özdoğan 2007) and Halula (Fig. 15.2, no. 2; Molist-Montana 1996). Çatal Hüyük is an exception. Although late, it retains a link not only with the earlier animal themes of the northern Levant, but with an emphatic manner of representation. The ‘vulture/death’ association is very explicit (Fig. 15.2, no. 10). The other ‘dominant’ animals, the aurochs, the raptors and the felines, are still represented on a large scale. But even though the ‘symbolic actors’ are the same as before, the roles which they play indicate another relation between animals and humans (Mellaart 1967; Cauvin 1997; Forest 1993). Thus the panthers are represented not as threatening but as dominated, as seen, for example, on a statuette showing a woman seated on a chair composed of two panthers, her hands resting on their heads (Fig. 15.2, no. 3). Deposits of horns are found throughout the Levant. At Tell Aswad male goat horns were deposited in two pits at the entrance of a house. Young animals and young humans

received the same funerary treatment, a young sheep/goat and many human babies being buried in the walls of houses (R. Khawam, personal communication). The goat therefore appears to replace the ox in explicitly symbolic deposits. Other evidence seems to indicate a kind of complementarity between humans and horned animals as at Kfar HaHoresh (Goring-Morris 2005). It could be a bovid buried under a human grave, a headless skeleton of a gazelle deposited near a modelled (‘plastered’) human skull, or human bones disposed in the form of an animal silhouette. Throughout the Levant the modification of themes is accompanied by a transformation of the stylistic language. The animals figured are almost all domestic, and the manner of representing them also appears ‘domesticated’. The ox does not escape this transformation, but its importance is indicated by the fact that it is over-represented (D. Helmer, personal communication). The horned animals are evidence of both continuity and change. Although the ox appears to retain a certain prestige, its representation has become familiar. Moreover, male goat horns in the deposits are often substituted for those of an ox. An apparent continuity could also conceal profound changes. Thus at Çatal Hüyük, in spite of the obvious links with past traditions, it appears that the human-animal relation is reversed: it is man (or rather woman?) that appears to dominate the great cats. What about the human figure? Most of the figurines are female. Two groups stand out. The first consists of the statuettes made in the continuity of the old traditions. They are quite realistic with ample but well-proportioned forms and are found throughout the Levant. In the south they are very stereotyped (e.g. Tell Aswad and ‘Ain Ghazal: Fig. 15.2, nos 4–5). The second corresponds to human representations of simple geometric volume, emphasizing a selection of specifically female characteristics. But representations on a large scale exist in Anatolia and in the southern Levant. At ‘Ain Ghazal (Rollefson 1983) flat statues modelled in lime plaster on a reed core required a high technical investment. The face is polychrome. Bitumen on the top of the head suggests the attachment of hair. At Jericho (Garstang et al. 1935) the same type of statue shows a face with sculpted features with paint suggesting a beard or body painting. There are masks from many sites. An example from Nahar Hemar is painted with lines running from the centre of the face towards the temples. Lateral holes carry traces of bitumen, which could have served to attach hair (Bar Yosef and Alon 1988). The human representations of this phase are thus varied and may be classed in three groups: the first consists of female statuettes with norms evoking the past; the second consists of simple figures; and the third consists of statues and masks that indicate features, painting on the body, and hair. The funerary practices in the northern and southern Levant are clearly different from each other. In the north the burials

15. Domestication of plants and animals, domestication of symbols?

127

Fig. 15.2 The symbolic world of the middle and late Pre-pottery Neolithic B. (1) figurine of a bovid from Tell Aswad; (2) figurine of a bovid from Halula; (3) women with panthers from Çatal Hüyük (Mellaart 1967); (4) female statuette from Tell Aswad; (5) female statuette from ‘Ain Ghazal; (6–12) modelled skulls at Tell Aswad.

128

Danielle Stordeur

occur in the private sphere as at Halula (Molist-Montana 2007) while in the south collective funerary precincts appear. In the north funerary objects are omnipresent while in the south they are rare. Finally, it is only in the south that a particular practice characteristic of this period occurs: the modelling of skulls. Modelled skulls have been found on seven sites: Ramad, Aswad, ‘Ain Ghazal, Kfar HaHoresh, Beisamoun, Jericho and Yiftahel (H. Khalaili, personal communication). This treatment, which consists of modelling a face on the skull of the deceased, is reserved for rare individuals who appear to have belonged to a kind of social ‘elite’. The meaning of these practices remains elusive, but it can be approached by bringing together a multitude of observations on the contexts of the deposits, their structure, their associations with other remains or objects, and their accessibility. It is necessary to determine the sex and age of the individuals and the exact treatment of each of the skulls. What observations have we been able to bring together and what suggestions does ethnography offer us? At Tell Aswad the modelled skulls were found in collective contexts in two successive funerary areas. For each episode, a single burial contains at its base modelled skulls (Fig. 15.2, nos 6–12) buried under other human remains (Stordeur and Khawam 2007). The coating covers the entire face including the jaw. The eyes are represented closed, and the junction of the eyelids is indicated by a slit filled with black. The representation of the nose is careful (Fig. 15.2, nos 9–12). Technical and stylistic differences exist, but there is a certain degree of homogeneity within each group. For example, the ears are fine and realistic in the older group, but simpler in the later group. In the latter, the use of yellow is added to the red, white and black used earlier, and all the skulls are mounted on low bases. The modelling on their foreheads and temples was cut to form an indentation that suggests the implantation of hair (Fig. 15.2, nos 9, 12). This observation, coupled with the fact that the coating is turned back at the ears, suggests that hair was fixed to the modelled skulls of the later area. Sex and age have not yet been determined, but when they are determined account must be taken of the fact that the notion of ‘ancestor’ is not always related to age (as, for example, in New Guinea). The individuals whose skulls are modelled could have been chosen for the symbolic status of ancestor. Social standing, deeds of war and of power, as well as other factors could have guided the choice of a particular individual. This is the context which at the moment provides the most solid clues. All the modelled skulls at Tell Aswad, as they were found, were definitely out of reach, buried as they were at the bottom of graves and covered by other human remains. Elsewhere, as at Jericho, some were found placed on the floors of buildings, accessible and thus subject to manipulation. The ethnographic data agree on one thing – that the practice of modelling skulls consists of two stages. During the first, the modelled skull is kept in a communal

building. It is ‘brought out’ during ceremonies and perched on a mannequin. After two or three generations the modelled skull is buried, always in a collective context. Although at Jericho the destruction of the building corresponds perhaps to the first stage, at Aswad we are definitely in the second, but we know that the first was respected thanks to certain details: the existence of a broken nose that was repaired, and the absence of fragments which are missing from their coating and which would have fallen outside the tomb. The modelling of skulls required great technical skill and probably a high level of initiation. Those who carried out the modelling must have had a particular status, like those whose skulls were modelled. The society that conceived of such practices would have been structured with a differentiation in roles. An organisation by lineage may have existed, or even a hierarchy, although we do not know what the nature and the extent of the power held by its beneficiaries might have been. What is important here is to demonstrate the profound change in beliefs which this practice indicates, and to keep this in mind while integrating the representations and the masks into our research. Ethnology is helpful. We see that in the middle Sepik (New Guinea) four ‘subjects’ are decorated with standardised motifs and colours which refer to social classifications: bodies are painted for ceremonial dances; bodies of the deceased are painted; and modelled skulls and masks are decorated (C. Kocher, personal communication). These types of decoration could have been practised in the PPNB. We have in fact found common points in the use of colours on the statues, masks and modelled skulls, and have noted the usual preoccupation with inserting hair to complete the representation.

Conclusions What seems to become apparent with the new archaeological evidence from the Levant dating to between 9500 and 7000 BC is that a real change in symbolism and in imagination occurred around 8200 BC. A coherent system which expressed itself through representation of wild animals changed to one in which the human being was at the centre. This new system of thought appears to correspond to the moment when fundamental socio-economic changes took place. The populations of the PPNA and early PPNB experimented with agriculture and herding, and their material culture reflects a feverish spirit of invention. Their symbolic world, however, remained dominated by animal figures. The new importance accorded to the human figure accompanied the beginning of the middle PPNB and not the ferment which preceded it. It is in the middle PPNB that wild nature is domesticated in the fields and herds, and possibly also in human mentalities. But the break is neither total nor definitive. A continuous thread links the two periods which we have examined, and this possibly corresponds to myths which in certain cultures

15. Domestication of plants and animals, domestication of symbols? are still alive. But let us keep in mind that when we try to access the thoughts of prehistoric populations, we end up, despite our efforts, banging our heads against the wall. Eddie Peltenburg knows this well, but he also knows that we must not deprive ourselves of the pleasure of reflecting on such a subject. This is why I am very happy, after a long friendship, to offer him my humble interpretations.

Acknowledgements The CNRS, the Ministry of Foreign Affairs and the Directorate of Antiquities of Syria have supported the excavations, administratively and financially, the results of which have served as the basis for this article. I warmly thank my friends and colleagues: F. Abbès, D. Helmer, M. Molist, F. Valla and G. Willcox whose patient critical readings of the text have been essential to the completion of this essay. I would also like to thank the following colleagues for kindly permitting me to publish their photographs: K. Schmidt (Fig. 15.1, nos 3, 5: Göbekli Tepe), H. Hauptman (Fig. 15.1, no. 11: Nevali Çori), M. Molist (Fig. 15.2, no. 2: Halula), J. Mellaart (Fig. 15.1, no. 10: Çatal Hüyük) and G. Rollefson (Fig. 15.2, no. 5: ‘Ain Ghazal, photo by Curt Blair). Thanks are also due to M. Roux for the images of Jerf el Ahmar, to L. Dugué for the images of Tell Aswad, and to Elizabeth Willcox, who translated my original text into English.

References Aurenche, O., J. Cauvin, M.-C. Cauvin, L. Copeland, F. Hours and P. Sanlaville 1981. Chronologie et organisation de l’espace dans le Proche-Orient de 12000 à 5600 avant J.-C. In J. Cauvin and P. Sanlaville (eds) Préhistoire du Levant, 571–601. Paris, CNRS Éditions. Bar Yosef, O. and D. Alon 1988. Nahal Hemar cave: The excavations. ‘Atiqot 18, 1–30. Cauvin, J. 1997. Naissance des divinités, Naissance de l’agriculture. La révolution des symboles au Néolithique. Paris, CNRS Éditions. Coqueugniot, E. 2000. Dja’de (Syrie), un village à la veille de la domestication (seconde moitié du 9è millénaire av. J.-C.). In J. Guilaine (ed.) Les Premiers Paysans du Monde, 55–71. Paris, Errance. Forest, J.-D. 1993. Çatal-Hüyük et son décor: pour le déchiffrement d’un code symbolique. Anatolica Antiqua 2, 1–42. Garstang, J., J. P. Droop and J. Crowfoot 1935. Jericho: City and necropolis. 5th report. Liverpool Annals of Archaeology and Anthropology 22, 143–184. Goring-Morris, A. N. 2005. Life, death and the emergence of differential status in the Near Eastern Neolithic: Evidence from Kfar HaHoresh, Lower Galilee, Israel. In J. Clarke (ed.) Archaeological Perspectives on the Transmission and Transformation of Culture in the Eastern Mediterranean, 89–105. Levant Supplementary Series 2. Oxford, Oxbow.

129

Hauptmann, H. 1999. The Urfa region. In M. Özdoğan and N. Başgelen N. (eds) Neolithic in Turkey: The Cradle of Civilization: New Discoveries, vol. 1, 65–86. Istanbul, Arkeolojive Sanat Yayinlari. Helmer, D., L. Gourichon and D. Stordeur 2004. A l’aube de la domestication animale. Imaginaire et symbolisme animal dans les premières sociétés néolithiques du nord du Proche-Orient. Anthropozoologica 39(1) (Colloque HASRI: Domestications animales, Dimensions sociales et symboliques. Hommage à J. Cauvin), 143–163. Helmer, D., L. Gourichon, H. Monchot, J. Peters and M. Saña Segui 2005. Identifying early domestic cattle from the PrePottery Neolithic sites on the Middle Euphrates using sexual dimorphism. In J. D. Vigne, J. Peters and D. Helmer (eds) New Methods and the First Steps of Mammal Domestication. Proceedings of the 9th International Council of Archaeozoology. (Durham, 23–28 August, 2002), 86–94. Oxford, Oxbow. Mazurowski, R. F. and B. Jammous 2000. Tell Qaramel: Excavations 2000. In M. Gawlikowski and W. A. Daszewski (eds) Polish Archaeology in the Mediterranean: Reports 2000, 327– 341. Warsaw, Centrum Archeologii Srodziemnomorskiej. Mazurowski, R. F. 2004. Tell Qaramel: Excavations 2003. In M. Gawlikowski and W. A. Daszewski (eds) Polish Archaeology in the Mediterranean: Reports 2003, 355–370. Warsaw, Centrum Archeologii Srodziemnomorskiej. Mellaart, J. 1967. Çatal Hüyük: A Neolithic Town in Anatolia. London, Thames & Hudson. Molist-Montana, M. 1996. Tell Halula (Siria). Un yacimento neolitico del valle medio del Eufrates. Campanas de 1991 y 1992. Madrid, Ministerio de Educacion y cultura. Molist-Montana, M. 2007. Practicas funerarias y primeras sociedades agrícolas del Próximo Oriente: Caracterización y discusión como variable arqueológica de análisis. In J. Justel et al. (eds) Las aguas primigenias. El Próximo Oriente como fuente de civilización, 365–382. Zaragoza, IEIOP. Özdoğan, M. 2007. Mezraa-Teleilat. In M. Özdoğan and N. Başgelen (eds) Türkiye’ de Neolitik Dönem. Yeni kazilar, yeni bulgular, vol. 1, 189–201 and vol. 2, 159–194. Istanbul, Arkeoloji ve Sanat Yayinlari. Rosenberg, M. 1999. Hallan Çemi. In M. Özdoğan and N. Başgelen (eds) Neolithic in Turkey: The Cradle of Civilization: New Discoveries, vol. 1, 25–34. Istanbul, Arkeolojive Sanat Yayinlari. Rollefson, G. O. 1983. Ritual and ceremony at Neolithic Ain Ghazal (Jordan). Paléorient 9(2), 29–38. Schmidt, K. 1999. Boars, ducks and foxes: The Urfa project 1999. Neo-Lithics 3(99), 12–14. Schmidt, K. 2002. The 2002 excavations at Göbekli Tepe (southeastern Turkey): Impressions from an enigmatic site. Neo-Lithics 2(02), 8–13. Schmidt, K. 2007. Göbekli Tepe. In M. Özdoğan and N. Başgelen (eds) Neolithic in Turkey: The Cradle of Civilization: New Discoveries, vol. 1, 115–129 and vol. 2, 105–116. Istanbul, Arkeolojive Sanat Yayinlari. Stordeur, D. and F. Abbès 2002. Du PPNA au PPNB: Mise en lumière d’une phase de transition à Jerf el Ahmar (Syrie). Bulletin de la Société préhistorique française 99(3), 563– 595. Stordeur, D. 2003. Symboles et imaginaire des premières cultures

130

Danielle Stordeur

néolithiques du Proche-Orient (haute et moyenne vallée de l’Euphrate) In J. Guilaine (ed.) Arts et symboles du Néolithique à la Protohistoire: Hommage à J. Cauvin, 15–37. Paris, Errance. Stordeur, D. and R. Khawam 2007. Les crânes surmodelés de Tell Aswad (PPNB, Syrie): Premier regard sur l’ensemble, premières réflexions. Syria 84, 5–32. Tanno, K. and G. Willcox 2006. How fast was wild wheat domesticated? Science 31(311), 1886. Valla, F. 2003. Une urgence: donner du sens. Des sacrifices dans

le Natoufien et le PPNA proche-orientaux? In Volume d’ Hommage à Claude Masset et Jean Leclerc, 205–218. Société Archéologique de Picardie 21. Voigt, M. 1985. Village on the Euphrates. Expedition, 27(1), 10–24. Yartah, T. 2005a. Tell ‘Abr 3, un village du Néolithique Précéramique (PPNA) sur le Moyen Euphrate. Première approche. Paléorient 30(2), 141–158. Yartah, T. 2005b. Les bâtiments communautaires de Tell ‘Abr 3 (PPNA, Syrie). Neo-Lithics 1(05), 3–9.

16 HERDS LOST IN TIME: ANIMAL REMAINS FROM THE 1969–1970 EXCAVATION SEASONS AT THE CERAMIC NEOLITHIC SETTLEMENT OF PHILIA-DRAKOS SITE A, CYPRUS Paul Croft

Introduction The remarkable and continuous succession of field projects that Eddie Peltenburg has undertaken in Cyprus over four decades has contributed massively to our understanding of the early prehistory of the island. Participation in these projects has had an enormous influence on the way that successive generations of younger scholars think about and practice archaeology. Whilst these projects have focussed mainly on western Cyprus they began, prior to the Turkish invasion of 1974, with excavations on the north coast at Ceramic (Late) Neolithic Ayios Epiktitos-Vrysi (henceforth Vrysi) from 1969–1973. The results of investigations at Vrysi remain central to our understanding of the period and its chronology. Beginning several years before the Vrysi project another Ceramic Neolithic settlement, Philia-Drakos Site A (henceforth Philia-Drakos), was excavated by Trevor Watkins (1965–1970). Indeed, the young Peltenburg assisted here in his pre-Vrysi days (1967). Philia-Drakos is located inland, some 35 km to the southwest of Vrysi, and in view of the broad synchroneity and geographic proximity of the two sites comparison of the material recovered from them would seem particularly apt. However, the extent to which comparison has been possible and, indeed, a broader assessment of how the Philia-Drakos settlement fits into the Cypriot Ceramic Neolithic as a whole, has been hampered by the incomplete publication of the PhiliaDrakos excavations. With regard to the substantial quantity of animal remains excavated at Philia-Drakos, they have received virtually no attention for the forty years since their excavation, and little information has previously been available on them.

Only very recently (late 2008) was the task begun of systematically examining and documenting the assemblage. Excavations on several Cypriot Ceramic Neolithic sites, as well as a book focussing on the period (Clarke 2007), add up to a significant renewal of interest in recent years, impelling the writer to commence this long-intended study. It may be observed, as an aside, that for the 1969 animal bones from Philia-Drakos the timing of the study was fortunate indeed, for although some contextual information has been lost as a result of the ongoing deterioration of disintegrating, insectravaged brown paper bags and fading labels, the information was still salvageable for the greater part of the material. During the course of the five excavation seasons at PhiliaDrakos the excavator had remarked upon the abundance of faunal remains; the quantity of antlers in particular suggested that deer were important here (Watkins 1969, 35; 1970a, 241). Subsequently, the present writer was able to examine some of the bone recovered during the 1968 season, and figures of 71% deer, 17% pig and 11% caprines were generated from a sample of 252 postcranial fragments identified at this time (Croft 1991, 69). This frustratingly limited amount of information, with percentages based on a potentially inadequate small sample, is all that has been available to researchers (e.g. Wasse 2007, 61) up until now. Accordingly, one purpose of the present contribution is to upgrade the quality of information available, pending completion of the analysis of the Philia-Drakos bones and their eventual publication in detail. Additionally, the paper offers a comparison, at a general level, with the Vrysi faunal record and also that from Paralimni-Nissia, one of the recently excavated sites. Philia-Drakos was a village settlement located on

132

Paul Croft

the southern slope of the shallow Ovgos river valley, consisting of small sub-rectangular buildings with stonebuilt foundations supporting walls of mud or other materials. Beneath the buildings a remarkable series of rock-cut shafts, tunnels and chambers are of obscure, possibly ritual purpose. The settlement was at least partly surrounded by a substantial wall and ditch arrangement. It was occupied, probably discontinuously, from an initial phase of the Ceramic Neolithic period, characterised by Dark-Faced Burnished Ware and an absence of the Red on White painted pottery that characterises the three subsequent phases on the site (Watkins 1969; 1970b). Although not initially accepted by the excavator (Watkins 1973, 52), a single radiocarbon determination (Birm-72 5270±100 non cal BP or late 4th– early 5th millennium cal BC) on probable hearth material of phase 3 is credible (Peltenburg 1978, fig.5; Stanley Price 1979, 19–21, 31), but the duration of the occupation at Philia-Drakos is not known. The Ceramic Neolithic period in Cyprus probably began around 4700 cal BC, lasting some 700–800 years until the transition to the Early Chalcolithic around 4000/3900 cal BC (Clarke 2007, 22). The analysis of the Philia-Drakos animal remains is still in progress. Material from the 1969 and 1970 excavation seasons has all been identified and recorded, but nothing from the 1965 and 1967 seasons has yet been examined. The relatively small amount of 1968 material that the writer was previously able to examine, mentioned above, seems too little to be the entire collection from that season and, in any case, has been excluded from consideration here. Although it is derived from only a portion of the entire assemblage, it is hoped that the presentation here of some general information on the PhiliaDrakos animal bones, and preliminary discussion of it, will be viewed as pertinent at a time of escalating interest in the Cypriot Ceramic Neolithic.

The Philia-Drakos Animal Bones (1969–1970 Seasons) Because analysis of the Philia-Drakos animal bones is still in progress, and as contextual and phasing information are not presently available to the writer, the 1969–1970 animal bones are considered here as a single, undifferentiated assemblage. The assemblage of over 3000 identifiable fragments (Table 16.1) is heavily dominated by the remains of fallow deer (Dama mesopotamica) (66.5%). Caprines (22.6%, estimated to consist of roughly three quarters sheep and a quarter goats) and pigs (9.1%) are also abundant. In addition to these larger mammals, dog (1.2%), fox (0.5%) and cat (0.1%) are also present. Additionally, nonmammalian remains include three bird bones, a carapace fragment of a freshwater turtle, and nine claws of freshwater crab (Potamon sp.). The disparity between the frequencies of the main animals presented here (Table 16.1) and those

fallow deer caprines pig dog fox cat TOTAL

NISP 2224 756 304 39 18 2 3343

(66.5%) (22.6%) (9.1%) (1.2%) (0.5%) (0.1%)

Table 16.1 Identified mammalian bone fragments from PhiliaDrakos A.

that have previously been quoted elsewhere (e.g. Croft 1991, 69) probably reflects some combination of the inherent unreliability of the original small faunal sample, the exclusion of cranial fragments from it, and the writer’s greater experience of identifying fragmentary prehistoric Cypriot animal bones after three decades of practice.

Larger Mammals In view of the abundance of antler at Philia-Drakos, and its highly fragmented condition, it would clearly be inappropriate to count each identifiable scrap as an individual fragment for analytical purposes. Accordingly, antler has been incorporated by counting each shed antler base as an individual fragment, and if an excavation context yielded antler pieces that did not include a shed base (or seemed unlikely to be broken off from an already counted shed base), a count of one single antler fragment was recorded for this material also. Unshed antler bases, attached to pedicles/frontal bones, were counted as head fragments. This approach resulted in 86 countable pieces of antler being included in the NISP (Number of Identified Specimens) for deer and, by counting them in this way, it is hoped that exaggeration of the dominance of deer in the assemblage has been avoided. Twenty-four shed antler bases were recorded to which may be added 3 pedicles from which the antlers had clearly been shed to give a total of 27 shed antlers. More than twice as abundant, though, were unshed antlers (57 specimens plus a hacked pedicle implying the removal of an additional unshed antler). Identified caprine remains include 110 fragments (76%) that could be identified as sheep and 35 (24%) as goat, suggesting that sheep outnumbered goats at Philia-Drakos by about 3 to 1. Limited evidence for horncore shapes indicates that goat horncores were of the untwisted variety that is standard in Cyprus up until the Bronze Age whilst horncores of sheep resembled those of recent mouflon (Croft 2006, 270–71). The deer would have been hunted although doubtless subjected to some measure of game management (Croft

16. Herds lost in time: Animal remains from the Philia-Drakos Site A, Cyprus 1991; 2002). By contrast, the sheep, goats and pig were probably at least mainly domestic stock although their remains may include hunted, free-living individuals also.

Small Carnivores A substantial number of dog bones in context 209.2 represent most of a single skeleton, and were accordingly counted as a single fragment although a separately counted pelvic fragment from this context must derive from a second individual. Concentrations of 14 dog bones in S2.2 and 9 bones in 211.3 may well originally have derived from single individuals, and their inclusion as individually counted bones in the NISP has probably resulted in an overrepresentation of dog, which would otherwise seem to occur with a similar frequency to fox. Thirteen bones of larger mammals, primarily phalanges, displayed clear indications of having been gnawed by dogs, and many other damaged and abraded specimens could well have been gnawed. Two longbone shaft fragments of cat included a small burnt splintered fragment of the distal part of a humerus shaft, suggesting the possibility that the carcase had somehow been processed. Fox remains encompassed 18 bones scattered through a dozen different contexts.

Non-mammals Bird remains came principally from context 40.3, which yielded a substantial piece of ulna shaft that is attributable to one of the several geese (Anser sp.) of the region, and an additional longbone shaft fragment that might well derive from the same individual. A third bird bone, an immature proximal ulna of an unidentified medium-sized (pigeon sized) bird, came from 208.3. More distinctly aquatic and clearly reflecting the proximity of the Ovgos river, the freshwater turtle Mauremys rivulata (otherwise variously known as the stripe-necked turtle, Balkan pond turtle or Balkan terrapin) is presumably the creature represented by a gracile carapace fragment. Amongst the animal bones were also remains of freshwater crab (Potamon sp.), the persistence of which in the locality is attested by a fairly complete exoskeleton (from 35.4) that was clearly recent in origin. In view of the excavator’s comment on the abundance of crab remains (Watkins 1969, 35; 1970a, 241) it seems very possible that additional ancient crab remains were recovered and have been stored elsewhere, perhaps with the shells that were also mentioned as being numerous (Watkins 1970a, 241). Three out of the nine crab claws seen by the writer were burnt.

The Ayios Epiktitos-Vrysi Animal Bones The Ceramic Neolithic village settlement of Vrysi consisted

fallow deer caprines pig dog fox cat TOTAL

133

NISP 244 (37.2%) 330 (50.4%) 69 (10.5%) 6 (0.9%) 2 (0.3%) 4 (0.6%) 655

Table 16.2 Identified mammalian bone fragments from Ayios Epiktitos-Vrysi.

of superimposed small stone buildings clustered in artificial hollows located on a small headland on the north coast of Cyprus. A substantial ditch defined the landward edge of the settlement in its early phases (Peltenburg 1982). A series of 16 radiocarbon dates encompasses three phases of occupation, running from c. 4400 to 3900 cal BC (Peltenburg and Spanou 1999, table 1). A rather small amount of faunal material was reported by A. J. Legge from three seasons of excavation at Vrysi, and in view of his comment that the bones were “those so far identified” (1982, 76) it seems that this report is based on an unspecified proportion of what was actually recovered. Only 667 identifiable bone fragments of larger mammals were reported from Vrysi (Legge 1982, table 3), the material deriving overwhelmingly (over 95%) from the middle phase of occupation here (Legge 1982, 76). Legge suggested that the paucity of bone and the sparseness of middens on the site may have been due to bones and other rubbish having been thrown directly into the sea (1982, 78). The Vrysi faunal data, presented here as Table 16.2, have been abstracted from various of the tables presented by Legge and also from his text, and are presented here in such a way as to enable comparison to be made as directly as possible with the Philia-Drakos data (Table 16.1). Because some of the figures quoted in Table 16.2 do not precisely match figures presented by Legge, some explanation may be offered to prevent confusion. Most importantly, it should be noted that Table 16.2 includes all identified fragments (NISP) of each taxon combining, for the main animals, data for “bones” and for “teeth” (and jaws) that were presented separately by Legge (1982, tables 6–7). From consideration of Legge’s (1982, 401–2) description of the dog remains it is unclear how he settled on a count of five bones (1982, table 5), and this writer (following his own method) has inferred a total of six countable fragments of dog. For cat, the count of four fragments presented in Table 16.2 includes the three teeth described by Legge (1982, 85) in addition to the single non-dental item that he also tabulated (1982, table 5). The three items reported from Vrysi as “possibly roe deer” (Legge 1982, 76 and table 5) have been excluded from consideration here because numerous subsequent studies

Paul Croft

134

Fig. 16.1 Relative frequency of identified mammalian bone fragments at three Ceramic Neolithic sites in Cyprus.

of faunal remains from prehistoric Cyprus have failed to confirm the presence of this small deer, so Legge’s tentative attribution was almost certainly incorrect.

Discussion Even at this interim stage in the examination of the faunal remains from Philia-Drakos, the amount of identified material is very much greater than was reported from Vrysi (Tables 16.1–16.2). Indeed, it amounts to more bone by far than has previously been reported from the entire Ceramic Neolithic period in Cyprus. Whilst it is clear, then, that the Vrysi sample must be viewed as less reliable, it is not negligible in size, and it seems unlikely that the considerable degree to which its composition differs from that of the Philia-Drakos sample is completely due to the Vrysi sample being inadequately small. The contrast between the two samples lies in the relative abundance of deer and caprine remains. Lumped together, these ruminants account for nearly identical proportions of identified faunal remains (89% at Philia-Drakos, 88% at Vrysi), but considered individually, their proportions diverge widely: deer remains are much more abundant and caprine remains less abundant at Philia-Drakos than at Vrysi (Fig. 16.1). Working with the “unusually complete” information

provided by the excavator, who had classified contexts at Vrysi into three main types, Legge examined the taxonomic composition of the faunal samples by context type, revealing considerable variability (1982, 78–79). Amalgamating data presented separately for dental and other identified pieces by Legge (1982, tables 6–7), it may be calculated that 258 fragments identified from floor contexts included 23.6% deer, 69.8% caprines and 6.6% pig. By contrast, non-occupation building fills included deer and pig remains with twice the frequency (47.7% and 16.7% of 174 fragments) and caprine remains with half the frequency (35.6%). Middens, or “collections of rubbish”, the third main type of context, provide an even greater degree of contrast, including 73.6% deer remains and only 15.3% caprine, but the sample (72 fragments) was rather small to be viewed as reliable. Even so, it seems fair to suggest that faunal remains in midden deposits were probably deer-dominated to at least the same degree as the bones from building fills. The frequency data quoted here are illustrated in Fig. 16.2. Legge’s conclusion (1982, 77) regarding taxonomic composition that “the archaeological context of the bone sample appears to be all-important” must reflect a significant level of selectivity in the human behaviour that lies behind the observed variability. Perhaps it was simply the case that the larger bones of deer, more noticeable or more inconvenient underfoot, were more likely

16. Herds lost in time: Animal remains from the Philia-Drakos Site A, Cyprus

135

Fig. 16.2 Relative frequency of identified bone fragments in the main types of context at Ayios Epiktitos-Vrysi.

to be ‘tidied up’ from floors, ending up in fills or middens, or being dumped into the sea. (If discard directly into the sea involved a disproportionate amount of material that would otherwise have ended up in fill or midden contexts, the entire on-site assemblage will have become skewed in favour of the caprine remains, which are so prevalent in the floor contexts). More complex scenarios than mere sloppy housekeeping may be envisaged, however, entailing taxon-specific spatial patterning in meat preparation and consumption: was deer meat more regularly cooked and/or consumed outdoors whilst caprine meat tended more often to be prepared and eaten indoors? Naturally, consideration of such possibilities invites further speculation regarding culinary practices, and the social and symbolic contexts in which different kinds of meat may have been consumed by the Vrysi villagers. Whatever its cause, the prevalence of caprine remains and the paucity of deer bones in the Vrysi assemblage taken as a whole, by comparison with Philia-Drakos, is an observed fact. It is clear, however, that even if the taxonomic composition of the caprine-dominated Vrysi assemblage (as presented in Table 16.2) is accepted at face value, the considerably larger body size of deer means that they would have been by far the main providers of meat. If it is assumed, for the sake of making an estimate, that the average deer yielded 3.4 times the amount of meat as a caprine, and a pig

three times as much, then 61% of the meat at Vrysi may be estimated to have come from deer, compared with as little as 24% from caprines and 15% from pig. In the same way, meat supply figures of deer 82%, caprines 8% and pig 10% may be estimated for Philia-Drakos. Thus, although the contrast between a caprine-dominated faunal assemblage at Vrysi and a deer dominated one at Philia-Drakos is pronounced, consideration of relative meat yields arguably provides a better assessment of economic importance and suggests a more limited degree of disparity, reducing it to a matter merely of the extent which deer dominated the animal economies of the two sites. The only other significant bone assemblage from Cyprus that dates to Ceramic Neolithic times comes from ParalimniNissia (henceforth Nissia), a walled coastal settlement in the southeast of the island (Flourentzos 2008). The assemblage of 1034 identified bones of large mammals from Nissia comprised 77.1% deer, 16.7% caprines, 4.1% pig, 1.5% dog, 0.4% fox and 0.1% cat (Croft 2008, table 1). This is the same range of animals as attested at both Vrysi and Philia-Drakos, and their levels of representation within the Nissia sample are broadly similar to Philia-Drakos (Fig. 16.1). Relative contributions to meat supply at Nissia may be estimated as deer 90%, caprines 6% and pig 4%. At Vrysi fish remains were recovered in surprisingly small quantities in view of the coastal location of the site (Legge 1982, 86

136

Paul Croft

and table 3), but it is hard to believe that fishing really was of negligible importance here. It may well be that fish are seriously underrepresented in the Vrysi faunal assemblage, as has been inferred for Nissia (Croft 2008, 105). As at Vrysi, marine turtles seem also to have been exploited to some degree at Nissia (Legge 1982, 85 and table 3; Croft 2008, 105). The limited amount of additional faunal data available for Ceramic Neolithic Cyprus is insufficient to permit even a basic comparison of the sort presented above between Philia-Drakos, Vrysi and Nissia to be extended to include other sites. Faunal material from Dikaios’ old excavation at Sotira-Teppes seems not to have been saved systematically: both the original faunal list (Zeuner and Grosvenor Ellis 1961) and, later, Ducos (1965, 4) indicate an implausibly small amount of bone from this extensive excavation. Even though the figure of 76% deer at Sotira-Teppes quoted by Ducos is liable to be unreliable since it is clearly based on a total assemblage of only 25 identified fragments, it is broadly comparable with the frequencies of deer remains at Philia-Drakos and Nissia, emphasising the contrast with Vrysi, where deer remains did not predominate. Although Kalavasos-Pamboules, another of Dikaios’ old excavations, does possess a Ceramic Neolithic component, the faunal sample that was saved from this site is both incomplete (clearly only selected specimens were saved) and chronologically mixed, including material dating to the various phases of the Chalcolithic, so it is of limited value (Croft 2007, 70–72). Dikaios (1962, 106–112) did not mention having recovered bones from the neighbouring Ceramic Neolithic site of Kalavasos-Kokkinoyia, and recent excavations at this enigmatic site seem to confirm that it lacks animal bones (Croft 2007, 72); whilst the existence of an apparently anosseous site is most intriguing, it is unhelpful for present purposes. Finally, faunal remains from the 1990s excavations at Kantou-Kouphovounos remain to be published in detail, but preliminary comments indicate merely that deer, sheep, goat and pig were all present (Karali 2002, 467).

cal BC, Cypriot animal economies were characteristically dominated by deer, and the extent of this domination seems to have attained a maximum around the mid-5th millennium cal BC, i.e. during Ceramic Neolithic times (Croft 1991, 69, 75). Very high levels of deer dependency were maintained through into Chalcolithic times, apparently persisting for at least a millennium or so, and longer in some areas (Webb et al. in press). The decreasing capacity of deer hunting to support an expanding human population probably explains the gradual decline in the economic significance of deer, with human communities pressured into increasing reliance on domestic stock. Even though the study of faunal remains from PhiliaDrakos is not yet complete, and the data have not yet been considered in detail, it is already clear that the composition of the sample falls into the expected pattern of very heavily deer dominated animal economies. The identification of the Ceramic Neolithic as the period when dependence on fallow deer reached an exceptionally high level was originally based mainly on the rather sparse evidence from Philia-Drakos that was previously available, bolstered by the very thin evidence from Sotira-Teppes (Croft 1991, 69), but the upgraded information presented here for PhiliaDrakos along with recent results from Nissia place this conclusion on a much firmer footing. Since it seems likely that site-specific disposal practices have had an impact on the composition of the rather small published faunal sample from Vrysi, doubts exist regarding the degree to which it truly represents the animal economy. But even if the caprine-dominated composition of the Vrysi sample is accepted at face value, it is still clear that it reflects an animal economy in which deer, with their larger body size, were the most productive species. Thus, the difference between Vrysi and the apparently standard economies of Philia-Drakos and Nissia was less pronounced than might be concluded from the taxonomic composition of the samples of identified bone fragments. Indeed, it seems likely that the Vrysi animal economy, rather than being as eccentric as first appearances might suggest, was also of standard type for the Ceramic Neolithic.

Conclusions Following the introduction of fallow deer to Cyprus, probably by colonists who imported a Neolithic (PPNB) agro-pastoralist way of life around the mid-9th millennium cal BC (Croft 2003, 56–58; Peltenburg et al. 2001, 46), deer rapidly rose to a position of considerable economic significance. In the particular circumstances of Cyprus, the heavy dependence on deer that developed there came to define an insular pattern of economic behaviour not replicated on the contemporary mainland (Croft 1991, 63; Wasse 2007, 61). From an early stage of the Cypro-PPNB through to the inception of the Bronze Age in the mid-3rd millennium

References Clarke, J. 2007. On the Margins of Southwest Asia: Cyprus during the 6th to 4th Millennia BC. Oxford, Oxbow. Croft, P. W. 1991. Man and beast in Chalcolithic Cyprus. Bulletin of the American Schools of Oriental Research 282/283, 63–79. Croft, P. W. 2002. Game management in early prehistoric Cyprus. In Proceedings of the 25th International Congress of the International Union of Game Biologists, Limassol, Cyprus 3–7 September 2001. Zeitschrift für Jagdwissenschaft 48 (Supplement), 172–179. Croft, P. W. 2003. The animal bones. In E. Peltenburg (ed.) Lemba Archaeological Project III(1): The Colonisation and

16. Herds lost in time: Animal remains from the Philia-Drakos Site A, Cyprus Settlement of Cyprus. Investigations at Kissonerga-Mylouthkia, 1977–1996, 49–58. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Croft, P. W. 2006. Animal bones. In D. Frankel and J. M. Webb, Marki Alonia. An Early and Middle Bronze Age Settlement in Cyprus. Excavations 1996–2000, 268–281. Studies in Mediterranean Archaeology 123(2). Sävedalen, Paul Åströms Förlag. Croft, P. W. 2007. Fauna. In J. Clarke, P. Croft and C. McCartney, The 1940s excavations at Kalavasos-Kokkinogia and KalavasosPampoules, 70–72. Report of the Department of Antiquities, Cyprus, 45–86. Croft, P. W. 2008. Animal remains from Paralimni-Nissia. In P. Flourentzos, The Neolithic settlement of Paralimni, 101–118 (appendix A). Nicosia, Department of Antiquities, Cyprus. Dikaios, P. 1962. The Stone Age. In The Swedish Cyprus Expedition IV(1A), 1–204. Lund, Swedish Cyprus Expedition. Ducos, P. 1965. Le daim à Chypre aux époques préhistoriques. Report of the Department of Antiquities, Cyprus, 1–8. Flourentzos, P. 2008. The Neolithic Settlement of Paralimni. Nicosia, Department of Antiquities. Karali, L. 2002. The animals of prehistoric Cyprus. In W. H. Waldren and J. A. Ensenyat (eds) World Islands in Prehistory: International Insular Investigations, 465–471. V Deia International Conference of Prehistory. British Archaeological Reports International Series 1095. Oxford, British Archaeological Reports. Legge, A. J. 1982. The vertebrate fauna. In E. J. Peltenburg, Vrysi. A Subterranean Settlement in Cyprus. Excavations at Prehistoric Ayios Epiktitos-Vrysi 1969–73, 76–87, 401–414. Warminster, Aris & Phillips. Peltenburg, E. J. 1978. The Sotira culture: Regional diversity and cultural unity in Late Neolithic Cyprus. Levant 10, 55–74. Peltenburg, E. J. 1982. Vrysi. A Subterranean Settlement in Cyprus.

137

Excavations at Prehistoric Ayios Epiktitos-Vrysi 1969–73. Warminster, Aris & Phillips. Peltenburg, E., S. Colledge, P. Croft, A. Jackson, C. McCartney and M. A. Murray 2001. Neolithic dispersals from the Levantine corridor: A Mediterranean perspective. Levant 33, 35–64. Peltenburg, E. and S. Spanou 1999. Neolithic painted pottery from Ayios Epiktitos-Vrysi. Report of the Department of Antiquities, Cyprus, 13–34. Stanley Price, N. P. 1979. Early Prehistoric Settlement in Cyprus, 6500–3000 B.C. British Archaeological Reports International Series 65. Oxford, British Archaeological Reports. Wasse, A. 2007. Climate, economy and change: Cyprus and the Levant during the late Pleistocene to mid-Holocene. In J. Clarke, On the Margins of Southwest Asia: Cyprus during the 6th to 4th Millennia BC, 43–63. Oxford, Oxbow Books. Watkins, T. 1969. The first village settlements. Archaeologia Viva 1(3), 29–38. Watkins, T. 1970a. Fouilles de Philia-Drakos, Site A. In V. Karageorghis (ed.) Chronique des fouilles et découvertes archéologiques à Chypre en 1969, 234–243. Bulletin de Correspondance Hellénique 94(1), 191–300. Watkins, T. 1970b. Philia-Drakos Site A: Pottery, stratigraphy, chronology. Report of the Department of Antiquities, Cyprus, 1–9. Watkins, T. 1973. Some problems of the Neolithic and Chalcolithic period in Cyprus. Report of the Department of Antiquities, Cyprus, 34–61. Webb, J. M., D. Frankel, P. Croft and C. McCartney 2009. Excavations at Politiko-Kokkinorotsos. A Chalcolithic hunting station in Cyprus. Proceedings of the Prehistoric Society 75, 189–238. Zeuner, F. E. and A. Grosvenor Ellis 1961. Animal bones. In P. Dikaios, Sotira, 235–236 (appendix 3). Philadelphia, University Museum, University of Pennsylvania.

PART 4

AGENCY, IDENTITY AND GENDER

17 AGENCY IN THE PRE-POTTERY NEOLITHIC A

Bill Finlayson

Writing in 2000, Dobres and Robb observed that agency had become a buzzword with little debate about what it means or how it might be useful, and authors simply making “ad hoc appeals to the concept” (2000, 3), and this still appears to sum up references to agency in the archaeology of the Neolithic transformation of Southwest Asia. This paper attempts to move this discussion forward as agency theory has great potential for developing the sort of local history that is fundamental to our developing conceptions of the Neolithic transformation. There is a tension between those who see agency as the fruits of empowered, intentional actions, and those who see the significance of agency lying far more in the unintended consequences of actions. The many different ways of viewing agency lead directly to differing opinions of its utility and purpose, such as the differences between Hodder’s perception of agency as pertaining to the individual (1984; 1999) and Barrett’s development of agency theory in light of Gidden’s theory of structuration (Giddens 1979; Barrett 1988; 2000). Barrett argues for knowledgeable agents who construct their social world while their actions are conditioned and constrained by that social world. Agents should be conceived as ‘decentred subjects’ not individuals, and “agency refers to the actions of individual social actors embedded within a broader socio-cultural and ecological setting” (Joyce 2000, 71). Fowler (2000) argues that Hodder’s 1999 approach to the individual is modern and universalist. Emphasising the individual rather than the dialectic with structure tends to lead to over-active actors and the recreation of modern western individuals. Gero (2000) has criticised the application of the agent as empowered individual, the idea that only some people have agency and some have more power than others. Morals, language and history emerge through society, not as individual

attributes, even though they may be expressed differently by different people. Agency is not faceless or generalised; it is mediated by cultural context, history, gender, age, lineage, and class. It is embodied, but it is also embedded, historically situated. Agency is a cultural process which is involved in the creation, negotiation and transformation of both a sense of personhood and collective identities. The agent does not construct, but is ‘constructed through’ material culture. Hodder (2000, 22) has noted that there is an opposition between descriptions of human ‘behaviour’ and the idea of ‘agency’. Human behaviour is rational and therefore predictable where, for example, people respond to climate change in a logical, rule-like manner. Although rational, it emphasises the animal-like quality that is behind optimal foraging theory and environmental determinism. In contrast, agency lies behind human choice, the power to act, an informed choice made within a specific context. It is important that we do not view structure and agent as a dichotomy but a duality (Barrett and Fewster 2000). In a sense this is key to our use of agency theory in the Neolithic transformation where we are often left debating which comes first, whether it be sedentism or cultivation, social change or climate. This chicken and egg style debate resolves within agency as the dialectic between processes.

The Use of Agency Theory in the Pre-Pottery Neolithic A I believe that one of our objectives in studying the Neolithic transformation has to be to develop local histories, detailed accounts of what is occurring above the level of individual site, but far below the wide ranging universal models that are often put forward (e.g. Cauvin 2000). The big narratives

142

Bill Finlayson

operate at a high level of abstraction – climate change, economic processes, the rise of ideologies – and part of the purpose of studying agency is to escape from these bland abstractions. If we accept that agency is historically specific and not a universal process like behaviour, historically situated agents practicing the routines of daily life represent a useful way of examining the Pre-Pottery Neolithic A (henceforth PPNA). We can now provide detailed accounts of the processes that took place during the Neolithic transformation, but our ability to explain these remains uncertain. The overarching models tend to describe the symptoms or secondary phenomena (Finlayson 2007) and are divorced from a human scale and the detailed archaeological evidence that is becoming available. People are either assumed to react to external stimulate as automata or become subsumed within a process, and the potential of human agency in making choice is denied. Hodder (1990), Renfrew (2003) and Watkins (2002) all argue that not very much happened before the Neolithic, which makes the inception of the process hard to comprehend. Watkins (2004) suggests that if biological evolution (‘nature’) was the primary driving force in early human development, to be replaced by culture in the modern world, there must have been a time in the past when the balance shifted. Robb sees this as a flawed argument, observing that even in the Palaeolithic people behaved in cultural ways that made the physical world “inhabitable and negotiable” (2004, 138). Watkins has argued that a cognitive revolution in the late Epipalaeolithic allowed people for the “first time to formulate and articulate a ‘world-view’ in which people could situate themselves in relation to each other, to their place in the world” (2005, 84). This capacity is clearly vital for agency. Although he regards cognitive and cultural developments seen in the Neolithic as taking place over the last 20,000 years, Watkins believes that until about 10,000 years ago human minds still worked differently from the way they work now (2005). Such a late revolution appears unlikely; aspects of continuity from early Epipalaeolithic glimpses, such as Ohalo II, to the heights of the early Natufian suggest otherwise, and the continuity from late Natufian to early PPNA and onto PPNB do not suggest any moment of such radical change but a gradual process that clearly has its naissance in the Upper Palaeolithic. Watkins appears to be describing a symptom of change, stating that the difference at the end of the Epipalaeolithic was “one critical environmental factor…the social world of the permanent, sedentary community” (2005, 87). We can understand this either as until those communities developed, people had no agency, and were indeed entirely subject to adaptation; or that people chose to live in communities, and that they already had the ability to do this. This is central to understanding that much of human cognitive ability emerged much earlier, in the human revolution, and that where communities had previously become larger, as in

the Early Natufian or the European Upper Palaeolithic, they adopted richly symbolic lifestyles. However, as noted by Rowley-Conwy, semi-sedentary complex-hunter gatherers arise and vanish from the ethnographic record and are not an evolutionary stage towards agriculture (2001). We can accept that people had agency before the Neolithic, and look to agency to try to gain insights into the beginnings of the Neolithic transformation. Ethnography furnishes us with a generalised context, one which does not provide much scope for change other than externally forced. If we use modern analogues, context is deeply flawed (Finlayson 2009), and we should not assume that the Kebaran was a modern hunter-gatherer society resembling the San; similarly, we should not assume that the Natufian resembled North West Coast Indians or that the early Neolithic corresponded to an orientalist vision of traditional farmers. These snap-shot static visions of the present make unlikely models for the past, and being assembled as a series of opposed social systems, offer little understanding to the process of transformation. There remains an underlying assumption of some evolutionary imperative that leads from one state to another (the band, tribe, chiefdom, state of Service 1962) generally as an adaptive shift programmed by an external force such as climate change. These generalised, non-specific and timeless societies lack the differentiation and variability which appear so central to most hunter-gatherer societies. Through practice, agency has the potential to initiate change. Agent centred approaches allow for internally created social change rather than adaptation to external forces. This differs from many models of the Neolithic transformation which refer to the ‘pulling’ and ‘pushing’ of external forces, such as climate, population and resources, and where people are reduced to automata directly responding to these external stimulae without the mediation of their social structures or choices. Even where people come into the equation, responding by making ‘rational’ or behavioural choices, they remain largely invisible and passive in the process. Evolutionary explanations tend to assume that, given the right circumstances, everything will fall into place; moreover, there is an underlying assumption of development – deduced from hindsight. This leaves no room for human choice.

Agency in the Pre-Pottery Neolithic A Farming (and cultivation of wild plants) makes demands in terms of social practice (Bender 1989) to do with sedentism and organised storage, and as agriculture develops, these changing practices become more onerous with land clearance, soil maintenance, ownership, and water provision. Many of these conditions are only nascent in the PPNA, but Bender’s underlying issue concerns whether

17. Agency in the Pre-Pottery Neolithic A the technology of farming causes social complexity or is a product of complexity. Many of the features associated with traditional small-scale farming are seen equally in complex hunter-gatherer societies. Indeed, it has often been argued that farming can only have arisen from such complex huntergatherers (although see Finlayson 2009). Domestication and farming appear in many ways to be symptoms of the social change, technologies necessary for allowing people to act the way they wished. Boyd (2005) has suggested that at the heart of the Neolithic transformation is a change in the way people eat; the acquisition of food is a social activity, and that as a consequence the transformation must be examined in terms of social practice and not generalised strategies. His description of the historical and practical knowledge required (organisation, distribution and consumption) makes agency vital to the transformation. Material culture is central to expressing and constructing agency. In the PPN there is a burgeoning material vocabulary, and the developing use of architecture that we see during the early Neolithic clearly added to the context in which embodied actors reproduced and changed their societies. Watkins refers to the developing settlements as “theatres of memory” (2004), an important concept that he interprets in terms of external symbolic storage. He does not continue the use of his metaphor of theatre by linking it to the actors and says he wants to “duck the question” of why people were choosing to live in larger, more sedentary settlements (2004, 98). The difficulty with this approach is that it risks seeing the ‘external symbolic storage’ created through architecture as just that, external; a material resource to be controlled and not part of the dialectic of agency. The idea of “literally constructing new worlds of the imagination that they could inhabit” (Watkins 2004, 105) appears to represent a case beyond hyper-agency and intentionality, and despite his recognition of the long Epipalaeolithic ‘prelude’, Watkins appears to deny the historical context in which people were acting. It implies a greater degree of complete knowledge about the world than an agent is likely to have. By ducking the question, Watkins’ powerful ideas regarding architecture lose their potential for explanation. The culturally negotiated concept of individuality appears likely to have been in considerable flux over the period when communities grow more settled and larger. Much of the debate regarding agency in the Neolithic has assumed the presence of hyper-agents or aggrandisers pushing social change, such as Hayden’s model (2004). If one wants to avoid the trap of evolutionary imperatives, it may be more appropriate to understand the notion of the unintended consequences of action. This is not to deny intentionality to actors, but people in the Natufian and early Neolithic did not consciously set out to develop the fully developed agrarian lifestyle of the late Neolithic and onwards. The significance of ‘aggrandisers’ depends as much on those “motivated to follow them” (Clark 2000, 99) as on the aggrandiser, and it

143

is not only elites (or would be elites) who have agency. The assumption that increasingly large and settled communities have an inevitable evolutionary logic, and that these then form the context for change as new mechanisms for coping with such permanent human aggregations become necessary, is problematic. A more fundamental question is why people moved into these communities, containing the potential for increased conflict, more difficulty in conflict resolution, competition over diminishing resources, and so on. We have no evidence in either the PPNA or PPNB that these communities were being pulled together by an increasingly hierarchical society. Much of our evidence in the PPNA actually suggests an increasing notion of community, collective and corporate, and perhaps we can see here a period of subsumption of the individual that is very different from the fairly individualistic society of some hunter-gatherers, and that such a reduction in individuality may have been a necessary part of increased co-residency. Shanks and Tilley (1987) discuss the importance of symbolic contexts in terms of individuality, and this too may provide a link with what is happening in the increasing material symbolism present in the Neolithic. Their observation that “meaning is precarious; its reproduction may result in its reconstitution” (1987, 74) is important. Despite the increased visibility of material culture in the Neolithic, we should not assume a single set of meanings across time and space. There are highly visible differences in material culture within the early Neolithic world of Southwest Asia, between major geographical provinces (e.g. the use of naturalistic images in the Upper Euphrates region compared to the use of more abstract images and greater interest in skulls in the southern Levant), as well as considerable inter-site variation within these large regions. We need to be wary of assuming a uniform pattern of behaviour because we know what will happen next in terms of the gradual sharing of an agricultural package through the later Neolithic. We should avoid interpreting actions through hindsight, as in descriptions of ‘proto-agriculture’ or seeing the PPNB as the intended consequence of the architectural beginnings made in the PPNA and Natufian. Agency will have performed in each locale within the local historically specific context, producing a highly variable mosaic of interacting communities. We easily fall into the trap of evolutionary imperatives, and our literature is full of assumptions about logical patterns of development. Evolution and development are built into the way we approach our study (Shanks and Tilley 1987), but such simplistic models tend to collapse in the face of the evidence, which shows that the Neolithic does not arise from an increasingly Neolithic Early Natufian but from an increasingly un-Neolithic Late Natufian, and that a village farming society arises from the collapse of PPNB mega-sites that do not herald proto-urban organisation. Burial evidence appears to represent a direct way to

144

Bill Finlayson

approach people. Wright (1978) used Binford’s ethnographic approach (1971) to identify a generalised situation from the ethnographic present, arguing that burial data from el-Wad provided evidence for social ranking developed to maintain order with increasing population and sedentism. His approach has been criticised by both Belfer-Cohen (1995) and Boyd (2001), and much is based on the idea that a skeleton is an individual, and that the burial of this individual reflects his or her living self. As Boyd observes, not only was Wright’s approach empirically flawed, but mortuary practices may not reflect social categories, instead being evidence for “social practices in which the living and the dead come together” (Boyd 2001, 197). Boyd argues that this requires us to abandon the quest for generalised models, and to move to studying practice and agency, a path suggested by Kuijt in his consideration of PPNA mortuary practice when he proposed that it is a “form of public action, a social drama designed and constructed by the living, often to elicit community participation” (Kuijt 1996, 315). Both Verhoeven (2002) and Kuijt have argued that secondary burials in the PPN were planned in advance and were extremely powerful in reinforcing communal, collective identity, with Verhoeven suggesting that primary inhumations in architecture in the Pottery Neolithic (henceforth PN) may have been intended to emphasise emerging individual households. They believe that burial rituals are important for reproducing social structure and providing a social regulatory function. Kuijt in particular (1996) has argued that burial practices in the late Natufian and PPNA indicate deliberate attempts to prevent or limit the emergence of hierarchical power, a resistance that continued into the Middle PPNB. In the PPNA there are both single primary and collective secondary burials. The partial burial of bodies and the manipulation of bones by adding and subtracting different skeletal parts may all suggest that the burials we see are not about individuals. This has suggested to some that the underlying ideas may have been about lineage and possibly the beginning of ideas of ‘home’ (e.g. Watkins 1990). Watkins (2004) argues that the treatment of bodies, especially skulls, was part of the way people symbolised the relationships between the living and the dead. However, as a development of Kuijt’s thoughts, burial may not be about death and ancestors but about living agency within the present, i.e. the people who are manipulating the remains. The more important relationships that were being symbolised were to do with the living through the practices of burial as an ongoing performance. That said, mediation with the spirit world can be an act of authority, and modifying ritual practices is potentially an important way of creating change. In some ways PPNA burials suggest a lack of strong individual personhood in society. In southern Jordan, at Wadi Faynan Site 16 (henceforth WF16), there is a clear association between sub-floor burials within structures and built-in ground stone tools (Finlayson and Mithen 2007)

while at Dhra’ burials seem to be placed around the outside of structures or placed though their walls (Finlayson et al. 2003). A focus on these practices within sites and between close neighbours is potentially more important than looking at gross differences in symbolic forms between the northern and southern Levant, or by ascribing the importance of these practices to their subsequent development in the PPNB with its rich repertoire of skull deformation and plastered skulls. The significance in the PPNA is what people were doing in the PPNA, and variation within broad themes suggests both that there was a widely held set of beliefs and knowledge, and that different communities were able to reproduce and create ideas of identity by acting within these structures. One of the important aspects of ritual is that the drama involved creates powerful social forces. Ritual can serve as a way of reproducing the corporate or collective identity (although there is the potential for such powerful situations to be appropriated by emerging elites). Variations in ritual can also serve to express identity within communities (potentially within ‘households’ and lineages) and between communities. The variations we see within the PPNA world are not simply indicative of the presence of different traditions but of the expression of identity. As noted above, Watkins has used burial to look at the idea of ‘home’ as part of his consideration of Wilson’s analysis of the significance of architecture (Wilson 1988; Watkins 1990). This has ramifications into ideas of household, which can be used as a means of discussing small-scale societies and considering the different types of people who make up a household while avoiding family or other kin-based terms that are hard to see archaeologically. However, within the PPNA we need to be careful using such terms and referring to structures as dwellings. At WF16 the interior of structures appears small, and the presence of visibly marked burials and ground stone tools makes these unlikely candidates for ‘homes’. At Dhra’ structures vary in purpose, with the main structures either being granaries or being dominated by builtin food processing tools (Kuijt and Finlayson 2009). While we may assume that some of these buildings functioned in part as sleeping areas at certain times during the course of the year, there is no evidence that this was a primary function. Furthermore, given the absence of internal hearths from most structures and the evidence for much activity outside the structures in areas conventionally described as middens (Mithen et al. in press), we can assume that much day-to-day living occurred outdoors and we need to avoid using the “house as a proxy … where archaeologists write of excavating the household where they really mean they have excavated buildings and spaces that form part of domestic life” (Hendon 2004, 275). There appears to be a huge amount of variation within PPNA architecture, with a constant and active rebuilding and modification to structures. The publicly visible granaries at Dhra’ may represent an ideal of an unconcealed resource available for sharing,

17. Agency in the Pre-Pottery Neolithic A enforcing the social status quo within a changing environment. The granaries indicate intentionality and planning. There is considerable work effort put into these structures, but even more there is a commitment to place, to the cultivation of wild plants and to the storage of a harvest (Kuijt and Finlayson 2009). This commitment to place may also be reflected in the built-in food processing tools and, at WF16, in the burial rites. The continuing architectural fluidity may indicate that there was a counterbalancing desire to reproduce at least a fictitiously mobile society, providing a socially needed ambiguity and flexibility. Practice as formed by agency need not coincide with ideal structures, and people are clearly working within and manipulating an existing context, unaware, of course, that such cultivation practices and storage may in the long term lead to domestication and a far greater degree of sedentism. PPNA middens are also created public spaces that do not seem to limit movement and visibility. Space is kept open and public in the PPNA. Space is created by people and reflects intentionality through the creation of socially meaningful spaces – the stage people move through. Hodder argues that the increased materialisation associated with the Neolithic and sedentism allowed for the materialisation of social structure, making it objectified and therefore more readily modified, and that people were more self-aware, making agency more apparent: “This shift towards a centring of human agency comes about as the inverse of the entanglement process. Material entanglement creates the possibility for greater human intervention that lies behind the processes of storage, sedentism and domestication” (2004, 51). Hodder tends to argue from a late PPNB/PN perspective, and there are problems with this argument as the processes of developing storage, sedentism and domestication began thousands of years earlier. From the apparently highly public granaries, external hearths and small interior floor spaces we can suggest that much of the standard interpretation of architecture as creating barriers and privacy has not occurred within the PPNA. There is no evidence for hidden storage. Huntergatherer practices, such as sharing of foods, remain possible. The fluidity of architectural forms suggests that even if settlements became more permanent, the architecture was used to maintain a connection to a more mobile way of life.

Conclusions Agency theory provides a way for us to free ourselves from generalised bland modeling and make full use of our diverse and fine-grained archaeological data from meticulous excavation to write local histories. These not only allow us to create the building blocks for larger scale accounts, but also enable us to consider the role of people

145

within the process. Agency enables variation by focusing on differences between cultural rules and actual practices. The study of agency allows us to explore the meaning of variation. Agency allows us to avoid having to rely on abstract processes as the driving forces of change. Climate may lead to change, but it is humans who select responses, using their existing world to make the new one. The PPNA arises from a Late Natufian context. There is no doubt that Holocene climate conditions provided new opportunities to which PPNA people reacted in a Late Natufian/PPNA manner, unintentionally creating the emerging Neolithic structures and contexts by transforming their existing world – theatres not of memory, but of practice.

References Barrett, J. 1988. Fields of discourse: Reconstituting a social archaeology. Current Anthropology 7(3), 5–16. Barrett, J. C. 2000. A thesis on agency. In M.-A. Dobres and J. E. Robb (eds) Agency in Archaeology, 61–68. London, Routledge. Barrett, J. C. and K. J. Fewster 2000. Intimacy and structural transformation: Giddens and archaeology. In C. Holtorf and H. Karlsson (eds) Philosophy and Archaeological Practice: Perspectives for the 21st century, 25–38. Göteborg: Bricoleur. Belfer-Cohen, A. 1995. Rethinking social stratification in the Natufian culture: The evidence from burials. In S. Campbell and A. Green (eds) The Archaeology of Death in the Ancient Near East, 9–16. Oxford, Oxbow. Bender, B. 1989. The roots of inequality. In D. Miller, M. Rowlands and C. Tilley (eds) Domination and Resistance, 83–95. London, Unwin. Binford, L. R. 1971. Mortuary practices: Their study and their potential. In J. A. Brown (ed.) Approaches to the Social Dimensions of Mortuary Practice, 6–29. Washington DC, Society for American Archaeology. Boyd, B. 2001. The Natufian burials from el-Wad, Mount Carmel: Beyond issues of social differentiation. Journal of the Israel Prehistoric Society 31, 185–200. Boyd, B. 2005 . Transforming food practices in the Epipalaeolithic and Pre-Pottery Neolithic. In J. Clarke (ed.) Archaeological Perspectives on the Transmission and Transformation of Culture in the Eastern Mediterranean, 106–112. Levant Supplementary Series 2. Oxford, Oxbow Books and Council for British Research in the Levant. Cauvin, J. 2000. The Birth of the Gods and the Origins of Agriculture. Translated from the French by T. Watkins. Cambridge, Cambridge University Press. Clark, J. E. 2000. Towards a better explanation of hereditary inequality: A critical assessment of natural and historic human agents. In M.-A. Dobres and J. E. Robb (eds) Agency in Archaeology, 92–112. London, Routledge. Dobres, M.-A. and J. E. Robb 2000. Agency in archaeology: Paradigm or platitude? In M.-A. Dobres and J. E. Robb (eds) Agency in Archaeology, 3–17. London, Routledge.

146

Bill Finlayson

Finlayson, B. 2007. Sedentism and the village: Social motivations for the beginning of farming. Studies in the History and Archaeology of Jordan 9, 467–474. Amman, Department of Antiquities, Jordan. Finlayson, B. 2009. The ‘complex hunter-gatherer’ and the transition to farming. In N. Finlay, S. McCartan, N. Milner and C. Wickham-Jones (eds) From Bann Flakes to Bushmills: Papers in Honour of Professor Peter Woodman, 172–185. Prehistoric Society Research Paper 1. Oxford, Oxbow. Finlayson, B., I. Kuijt, T. Arpin, M. Chesson, S. Dennis, N. Goodale, S. Kadowaki, L. Maher, S. Smith, M. Schurr and J. McKay 2003. Dhra’ Excavation Project 2002: Interim report. Levant 35, 1–38. Finlayson, B. and S. Mithen (eds) 2007. The Early Prehistory of Wadi Faynan, Southern Jordan, Archaeological Survey of Wadis Faynan, Ghuwayr and al-Bustan, and Evaluation of the Pre-Pottery Neolithic A site of WF16. Wadi Faynan 1, Levant Supplementary Series 4. Oxford, Oxbow Books and Council for British Research in the Levant. Fowler, C. 2000. The individual, the subject, and archaeological interpretation. In C. Holtorf and H. Karlsson (eds) Philosophy and Archaeological Practice: Perspectives for the 21st Century, 107–133. Göteborg, Bricoleur. Gero, J. M. 2000. Troubled travels in agency and feminism. In M.-A. Dobres and J. E. Robb (eds) Agency in Archaeology, 34–39. London, Routledge. Giddens, A. 1979. Central Problems in Social Theory. London, Macmillan. Hayden, B. 2004. Sociopolitical organisation in the Natufian: A view from the northwest. In C. Delage (ed.) The Last Hunter-gatherers in the Near East, 263–308. British Archaeological Reports International Series 1320. Oxford, British Archaeological Reports. Hendon, J. A. 2004. Living and working at home: The social archaeology of household production and social relations. In L. Meskell and R.W. Preucel (eds) A Companion to Social Archaeology, 272–286. Oxford, Blackwell. Hodder, I. 1984. Archaeology in 1984. Antiquity 58, 25–32. Hodder, I. 1990. The Domestication of Europe: Structure and Contingency in Neolithic Societies. Oxford, Blackwell. Hodder, I. 1999. The Archaeological Process: An Introduction. Oxford, Blackwell. Hodder, I. 2000. Agency and individuals in long-term processes. In M.-A. Dobres and J. E. Robb (eds) Agency in Archaeology, 21–33. London, Routledge. Hodder, I. 2004. Neo-thingness. In J. Cherry, C. Scarre, and S. Shennan (eds) Explaining Social Change: Studies in Honour of Colin Renfrew. McDonald Institute Monographs. Cambridge, McDonald Institute of Archaeological Research. Joyce, A. A. 2000. The founding of Monte Alban: Sacred propositions and social practices. In M.-A. Dobres and J. E. Robb (eds) Agency in Archaeology, 71–91. London, Routledge. Kuijt, I. 1996. Negotiating equality through ritual: A consideration of Late Natufian and Pre-Pottery Neolithic period mortuary practices. Journal of Anthropological Archaeology 15(4), 313–336.

Kuijt, I. and B. Finlayson 2009. Inventing storage: Evidence for the earliest pre-domestication granaries 11,000 years ago in the Jordan Valley. Proceedings of the National Academy of Sciences of the United States of America 106(27), 10,966–10,970. Washington DC, National Academy of Sciences. Mithen, S., B. Finlayson, M. Najjar, E. Jenkins, S. Smith, S. Hemsley, D. Maricevic, N. Pankhurst, L. Yeomans and H. AlAmarat in press. Excavations at the PPNA site WF16: A report on the 2008 season. Forthcoming in Annual of the Department of Antiquities of Jordan. Renfrew, C. 2003. Figuring It Out. London, Thames & Hudson. Robb, J. 2004. The extended artifact and the monumental economy: A methodology for material agency. In E. DeMarrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World, 131–139. McDonald Institute Monographs. Cambridge, McDonald Institute of Archaeological Research. Rowley-Conwy, P. 2001. Time, change and the archaeology of hunter-gatherers: How original is the ‘original affluent society’? In C. Panter-Brick, R. Layton and P. Rowley-Conwy (eds) Hunter-gatherers: An Interdisciplinary Perspective, 39–72. Cambridge, Cambridge University Press. Service, E. 1962. Primitive Social Organization: An Evolutionary Perspective. New York, Random House. Shanks, M. and C. Tilley 1987. Social Theory and Archaeology. Cambridge, Polity. Verhoeven, M. 2002. Transformation of society: The changing role of ritual and symbolism in the PPNB and the PN in the Levant, Syria and south-east Anatolia. Paléorient 28(1), 5–14. Watkins, T. 1990. The origins of house and home? World Archaeology 21(3), 336–347. Watkins, T. 2002. Re-thinking the Neolithic Revolution and its consequences. In C. Breniquet and C. Kepinski (eds) Études Mésopotamiennes: Recueil des Textes Offert à Jean-Louis Huot, 489–507. Paris, Éditions Recherche sur les Civilisations. Watkins, T. 2004. Architecture and ‘theatres of memory’ in the Neolithic of southwest Asia. In E. De Marrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World, 97–106. McDonald Institute Monographs. Cambridge, McDonald Institute for Archaeological Research. Watkins, T. 2005. The Neolithic revolution and the emergence of humanity: A cognitive approach to the first comprehensive world-view. In J. Clarke (ed.) Archaeological Perspectives on the Transmission and Transformation of Culture in the Eastern Mediterranean, 84–88. Levant Supplementary Series 2. Oxford, Oxbow Books and Council for British Research in the Levant. Wilson, P. 1988. The Domestication of the Human Species. New Haven, Yale University Press. Wright, G. A. 1978. Social differentiation in the early Natufian. In C. L. Redman, M. J. Berman, E. V. Curint, W. T. Langhorne, N. M. Vergassi and J. C. Wanser (eds) Social Archaeology: Beyond Subsistence and Dating, 201–233. New York, Academic Press.

18 UNDERSTANDING SYMBOLS: PUTTING MEANING INTO THE PAINTED POTTERY OF PREHISTORIC NORTHERN MESOPOTAMIA Stuart Campbell

Like several other contributors to this volume, some of the first classes I took as an undergraduate were taught by Eddie Peltenburg and, again in common with many others, I have worked on projects with him in both Cyprus and Syria over many years since then. It would be difficult to either quantify or overestimate the extent to which I have been influenced by him. It is both a pleasure and an honour to make a contribution to this volume. The painted ceramics of the late Neolithic in northern Mesopotamia are some of the most elaborate and attractive decorated pottery in prehistory. The overwhelming majority of the decoration is geometric, sometimes with what seems like an endless parade of motifs and subtle variations. Rare examples stand out as very different, with much more naturalistic decoration depicting people, animals, structures and artifacts in scenes whose power and significance seems to us to be much more immediately recognisable. This paper argues that much of this decoration, both abstract and figurative, carried meaning and that these meanings endowed the ceramics with a social agency of their own (cf Gell 1998). Understanding the ways in which the agency could be exercised can provide a key to understanding how society of late Neolithic northern Mesopotamia was constituted. Over a period from just before 6000 cal BC to a little after 5000 cal BC, the pottery of north Mesopotamia is characterised by extensive and sometimes elaborate painted decoration. Although it has traditionally been divided into different cultures or phases, the Samarran, Halaf and Ubaid, it may be more profitable to think of it as a broad ceramic phase characterised by that domination of painted decoration, reflecting both a stylistic expression that came into use c. 6200 cal BC and declined c. 4,750 cal BC, “l’ère de la céramique peinte” (Huot 1994, 63) and the social milieu within which it had meaning and significance.

In the past, the painted decoration of this general phase has primarily been analysed and interpreted typologically and chronologically. Decoration has been used to define cultural or chronological groupings, and it has been sub-divided into many individual motifs which have been examined for their symmetry. Similarity in motif assemblages has explicitly or implicitly been used to look at group identity and differentiation. Little attention has been paid, however, to what was actually meant by the symbolism of individual motifs or combinations of motifs and the degree of sophistication or convention in the messages that the decoration could convey, although this is key to understanding how decoration was both used and adopted. In other words, form has been prioritised at the expense of meaning. Although it will remain impossible to comprehend fully exact meanings from prehistoric material, it is perhaps possible to gain insights into the types of meaning that were present and something of how they functioned within a wider system of symbolic communication. The contrasts and links between the predominant abstract, geometric decoration and the much rarer, naturalistic decoration can act as a powerful tool to gain conceptual leverage on this wider system. Throughout the period, most of the decoration on the painted pottery is geometric and abstract (e.g. Fig. 18.1). Here I wish to explore one possible way of understanding the choices, combination and meanings of the geometrical and apparently abstract motifs as symbols that, at times at least, had explicit meanings, both individually and in groups. The much rarer examples of decoration with depictions of naturalistic scenes contrast strongly with this predominant geometric decoration. In archaeological publications, the two categories of decoration have generally been considered

148

Stuart Campbell

Fig. 18.1 Typical Halaf vessels from Arpachiyah decorated with geometric motifs (after Mallowan and Cruickshank Rose 1935, fig. 60, no. 5 and fig. 61, no. 2).

separately, with the more naturalistic depictions often separated out from the rest of the ceramics as prize finds. I wish instead to explore the way in which the two types of decoration may be understood as different aspects of the same system of communication with a complementary role in pre-urban social interaction and integration. Although it is certainly true that there may have been considerable variations in both time and space, for simplicity I will make little effort here to incorporate regional or chronological subtleties. Most of my examples come from the pottery manufactured and decorated in the Halaf style. This is largely due to convenience. There have, of course, been other approaches to this challenge. Mallowan famously outlined a sequence of development for the bucrania motif, running from naturalistic to highly abstract, and argued that a similar process of stylisation may have occurred with other motifs as well (Mallowan and Cruickshank Rose 1935, 154–165). Where the complete process of schematisation is not attested, this is difficult to demonstrate for many motifs and, in any case, need not correlate with significance or meaning. Even where motifs represent abstractions of what was once naturalistic, they need not have deeply symbolic meanings; for example, the suggestion that has been made many times that crosshatching may originate as an attempt to depict basketry (e.g. Mallowan and Cruickshank Rose 1935, 153; Wengrow 2001). If this suggestion has merit, the link might be deeply meaningful or it might be relatively trivial – or it might point to meanings that were shared between different media. There was probably not a simple way that meaning was

communicated. Decoration on pots doubtless conveyed information in different ways and at multiple levels. Different aspects of the decoration might possess very different significance. Thus, not only have various analytical approaches been taken; they may also help us to reconstruct different types of meaning. Hole, Bernbeck and Nieuwenhuyse have explored the significance of the structure of the decorative scheme (Hole 1984; Bernbeck 1994; 1999; Nieuwenhuyse 2007). Elements of composition, such as symmetry and repetition, may have been important (e.g. von Wickede 1986; Melville 2005). The analysis of individual motifs themselves has a particularly detailed history of study (e.g. LeBlanc and Watson 1973; Davidson 1977; Campbell 1992; Irving 2001). Each approach may be seen as complementary to the others by focussing on different aspects of the design. However, all of these studies have emphasised typologies and generalised structure. Although meaning has been considered, it has been treated as a rather general concept, often in a manner drawing implicitly or explicitly on similar approaches to the analysis of style (Conkey and Hastorf 1990). These approaches can certainly help us understand both aspects of identity and the ways in which a potter conceptualised and executed a design. They tell us less about what meanings these elements may have carried. Although they have shed light on important aspects of the decoration of pottery in northern Mesopotamia, they have not generally been part of an effort to construct a general theory of what decoration meant and how it functioned as a mechanism of social communication.

18. Understanding symbols: Putting meaning into painted pottery Schmandt-Besserat has recently suggested that compositions of pre-4th millennium painted pottery focussed on filling space according to rules of aesthetics, whether the compositions used geometric or naturalistic decoration (2007, 5–22). Although she acknowledges that prehistoric decoration carried meaning, she suggests that it was generalised rather than something which could be complex and dynamic. She draws on parallels with language, especially written language, to suggest that scenes are only explicitly narrative in the 4th and 3rd millennia. Thus “... preliterate pottery composition formed an all-over pattern meant to be apprehended as a whole, or globally, those of the literate period were to be viewed analytically” (2007, 24) and “Preliterate pottery paintings could only evoke an idea” (2007, 25). In contrast, I would argue that it is not that Neolithic decoration could not support a narrative but that the narrative needed to be deciphered and explained; that the process of extracting and recreating meaning would have been a process of social interaction. Nieuwenhuyse has recently proposed such a theory (2007, 206–212). His interpretation emphasises structured sets of oppositions between bounded-unbounded, naturalisticabstract, repetitive-discontinuous designs. Designs with bounded, continuous and geometrical attributes are suggested to have had an ‘outward’ social orientation while the unbounded, discontinuous and ‘figurative’ styles were directed ‘inward’ at local and domestic activities and meanings. While there is much to embrace in this proposal, it should also perhaps be noted that the interpretation of painted, naturalistic decoration on pottery depends heavily on the interpretation of representational depictions on other media, such as wall paintings, seals, figurines and applied decoration on pots (Nieuwenhuyse 2007, 210). Comparisons across media have been neglected in the past, so this inclusive approach is very welcome. It does not necessarily follow, however, that the same rules and audiences were observed in all cases; painted pottery may have had different considerations. However, the discussion presented here is not incompatible with Nieuwenhuyse’s proposals. In the rare cases where naturalistic or figurative decoration is present on the late Neolithic pottery of north Mesopotamia, it has often been interpreted in isolation. A range of interpretations have been put forward for different examples. In contrast to the geometric decoration, it has usually been assumed that figurative designs did carry important social meanings. Thus representational designs have been identified as carrying ritual meaning, including the depiction of deities and supernatural beings (IppolitoniStrika 1990; 1996; Breniquet 1992; Forest 1996; Cauvin 2000). In a stimulating analysis, Garfinkel interpreted a series of human figures as dancers (2003). Despite their immediate impact on the observer, it is probably a mistake to treat the naturalistic designs as completely separate from the more general geometric

149

motifs. Some types of decoration, such as bucrania, can be considered in both categories as it is used along a spectrum from naturalistic to stylised. Furthermore, there is little evidence that the prehistoric potters maintained a rigid division. Almost all pots with representational designs also have elements of geometric decoration, sometimes used to frame naturalistic scenes but perhaps often used to reinforce the fact that the pot remains a pot by retaining the most typical geometric elements, such as a band around the vessel rim. It may be profitable to explore a more integrated approach where naturalistic and geometric decorations are not seen as completely separate. “Visual representation refers both to the act of portraying, symbolizing or presenting the likeness of something, and to the use of the resultant image “to ‘re-present’, imagine, describe, define, understand, fix, construct, organise, regulate and even transform the world as we perceive it” (Skeates 2007, 199). Given an appropriate social context, both geometric and naturalistic motifs can function in this way. The difference between the abstract image and the naturalistic example can be one of degree – the representation in the former case may be more formalised, more embedded in convention and also potentially hidden. The key constituent of the abstract image may not be obvious, with less meaningful elaboration hiding the more significant core that delivers the real meaning. These meanings can be overt, but they can also be obscured and elaborated by the addition of further elements. This places a great deal of emphasis on the social context in which decoration was created and displayed. The range of meanings encoded in the decoration of a vessel was undoubtedly complex, and its comprehension was equally certainly dependent on the observer. More broadly, the meaning would have been created by the setting – the occasion of consumption of food and drink, the participants and their interaction. The meanings would have emerged from social discourse (cf Bernbeck 1999), both spoken and unspoken. Some elements of the meaning would certainly have operated on the level of familiarity and identification, simply on the level of ‘is my pottery like your pottery?’. Other meanings might well have been associated with function, both of the vessel and the way it was used, and were possibly reinforced by variables such as types of food and cooking methods. However, it is possible to argue that the combination of vessel shape, structure of the decoration and the particular motifs might carry more explicit meanings, perhaps associated with specific concepts and narratives. The clearest indication of this comes from the exceptional vessel/figurine from the Halaf levels of Yarim Tepe II (Fig. 18.2). This figurine was found in a pit, broken in pieces and associated with burning (Merpert and Munchaev 1987). It seems possible that it had actually been treated in a way that is analogous to human funerary treatment, which also some-

150

Stuart Campbell

Fig. 18.2 Vessel figurine from Yarim Tepe II (after Munchaev and Merpert 1981, fig. 98).

times has elements of burial, fragmentation and burning (Campbell 2008). The removable head was not found with the rest of the pot, perhaps because it was made of organic material or perhaps because it was deliberately separated from the body, a practice which could also parallel the occasional special treatment given to human skulls. It does not seem contentious to argue that it was a figurine with high symbolic value, which had a use in specific rituals in which presumably both the ability to fill the figurine/vessel with liquid and its removable head would have had a significant role. It is probable that it represented a specific mythical or supernatural being who would have figured in narratives of importance in systems of society and belief. Assuming the figurine/vessel did have an important status, it follows that the decoration on this vessel is not random but had been selected for very specific reasons that may have amplified the meanings attached to the person or being represented. Some of the decoration is broadly naturalistic, such as the hair and possible armlets, and may be associated with the woman depicted in the vessel or the role that she performed. The elements of the figurine/vessel that are particularly relevant here are the ones that aren’t obviously naturalistic although it is possible that they were associated with body paint or tattooing. These are motifs

which had been added with particular purpose and to add particular meanings to the figure. They were both relevant to the person represented and conveyed additional information that was probably quite explicit in intent and meaning. In particular, the significant motifs are the rosette or flower depicted in the navel and the dotting that fills the exaggerated pubic area. Furthermore, although the overall artefact is very naturalistic, there are no feet or legs. This is not simply a technical requirement as a roughly contemporary figurine/ vessel at Domuztepe has very well modelled legs and feet (Campbell 2004). On the Yarim Tepe II figurine, instead of feet, there is a flange with a row of upturned triangle motifs running around it. These non-naturalistic elements are particularly interesting because they also occur in the geometric decoration on pots that otherwise would not appear particularly unusual. Although all the elements do occur in isolation, they are used in the same combinations with surprising frequency. Rosettes probably occur most frequently on Halaf pottery in association with areas filled with dots, either in alternate panels or chequer board patterns. This repeats the association of the rosette or flower with dots in the pubic triangle of the figurine/vessel at Yarim Tepe II. Strikingly, one of the main vessel types that often has alternating panels of rosettes

18. Understanding symbols: Putting meaning into painted pottery

151

Fig. 18.3 Bowls with flanged bases, rosettes and dots from Umm Qseir (after Tsuneki and Miyake 1998, fig. 26, nos 1 and 9).

Fig. 18.4 Bukrania motifs on Halaf pottery from Arpachiyah (after Mallowan and Cruickshank Rose 1935, fig. 76, nos 2 and 4).

Fig. 18.5 The ‘dancing ladies’ motif on the interior rim of bowls from Khirbet Garsour (original illustration by the author).

and dots along the interior of the rim also has a flanged base which can be decorated with up-turned triangle motifs (Fig. 18.3). Examples can be cited from both Yarim Tepe II and Umm Qseir in north-east Syria (Tsuneki and Miyake 1998, fig. 26, nos 1, 9). I would suggest that the pots with the same combination of motifs that we see on the Yarim Tepe II figurine/vessel may either draw on precisely the same meanings or even represent the same woman, whether supernatural or mythological, in a much more abstract form. The decoration needs to be understood as partaking in the same mythologies or narratives as the being represented by the figurine. While this example is outstanding, there are other indications that some motifs may carry specific meanings. The most obvious is the well known bucrania (Mallowan and Cruickshank Rose 1935, 154–165). Although the bulls’ horns are often highly schematic, they still appear on a very wide range of Halaf pottery in a form recognisable to us, almost always embedded in otherwise geometric decoration (Fig. 18.4). Although they have received less attention, a similar

spectrum running from naturalistic to stylised can be observed in other motifs such as mouflon horns and birds which are also most commonly integrated with abstract motifs. Similarly, the distinctive motifs that appear round the interior rims of both Samarran and early Halaf pottery and are generally known as ‘dancing ladies’ are often seen in various stages of stylisation (Fig. 18.5). It is possible that the ultimate level of stylisation of this motif is the simple swags that are the most frequent decoration on the same part of the vessel on late Halaf pottery (e.g. Fig. 18.1, a). While meaning might have been replaced by convention during the long process of abstraction and schematisation, I would suggest that it is more likely that the meaning was retained but no longer required the full form to be depicted or perhaps even understood. The process of abstraction may well have taken other more naturalistic depictions and hidden them in geometric motifs whose symbolism cannot be accessed by archaeologists but may have been no less potent by being obscured. This pattern of encoded meanings can possibly be extended further. Some of the classic Halaf patterns are

152

Stuart Campbell

Fig. 18.6 Motifs showing animals on Halaf pottery at Arpachiyah (after Mallowan and Cruickshank Rose 1935, fig. 77: 1, 9 and 16).

Fig. 18.7 Naturalistic scenes on Halaf pottery from (a) Tell Halaf (after von Oppenheim 1943); and (b) Domuztepe (photo by the author).

made up of dots (see the rather different discussion in Nieuwenhuyse 2007, 207). Dots, however, tend to occur only in particular places, sometimes in combination with other geometric motifs but particularly in association with depictions of animals and humans (e.g. Figs 18.4, 18.6 and 18.7, a). The appearance of dots on the Yarim Tepe II figurine/vessel is again relevant. In all these cases, the dots appear as a secondary or background element. They may be adding meaning or value to the primary element, perhaps a concept of animation or drawing attention to it as a major actor in some otherwise hidden narrative. Abstract but meaningful decoration need not always derive from a naturalistic original. Although there are hints that suggest the significance of some motifs, most of the meanings must inevitably escape archaeologists. Nonetheless, based on the examples cited, it seems possible that much of the apparently abstract, geometric decoration may have had more or less complex meanings. On one level, pottery decoration may simply have been about the familiar (i.e. isochrestic meanings; see Sackett 1990). On another level, explicit meanings could be

decoded and used to convey social narratives – discourses that could link events and episodes in socially significant ways, and that encapsulated ways of understanding the world, society and the place of the individual or group within it. These narratives might embody folklore, dreams and the everyday experience of the world; frequently they might have mythological or supernatural elements. Because of the degree of abstraction in most of the decoration, meaning may often have been relatively fixed, imposing a high level of convention, so that it might have been best used to relate established themes. Elements might also be juxtaposed to challenge existing narratives and create new variants, but this understanding might only be possible when there was also a personal narrative to explain what might otherwise have been simply odd. Certainly the use of conventional elements would have constrained the introduction of novel subjects and limited the scope for new narratives to be introduced. Abstract, geometric motifs therefore may have functioned to reinforce or modify social conventions, not to initiate new understandings of the world.

18. Understanding symbols: Putting meaning into painted pottery

153

Fig. 18.8 Depiction of houses on a Halaf pot from Domuztepe (photo by the author; decoration is partially reconstructed based on repeating elements).

A possibly similar context of use can be seen in the highly decorated chichi beer bowls in Ecuadorian Amazon, the creation and decoration of which are critical aspects of a wife’s role (Bowser 2000). The abstract decoration on the vessels represents features of mythology, including spirits, animals, plants and stars, as well as family relationships and the connection between a woman and her dream world. “The key symbols of female identity in Achuar and Quichua belief systems – manioc, pottery clay, garden soil, and the garden spirit – are linked through language, myth, and song . . . On a daily basis, a woman’s act of serving chicha in a pottery bowl to her husband or brother makes reference to this cluster of key symbols” (Bowser 2000, 228). Within this framework, designs are deeply personal and individual. Innovations and interpretation of designs are an active topic of discussion by both men and women. In the prehistoric pottery of Mesopotamia, extensive naturalistic decoration is unusual. As already discussed, it is most often absorbed into the geometric patterns on vessels. The more striking examples of naturalistic decoration are very different. Not only is the design more obviously representational, but the structure is usually much more open (e.g. Fig. 18.7). Large areas of the vessel can be filled and different naturalistic elements are usually combined to create scenes, such as the combination of houses, birds and trees in Fig. 18.8. While some of this might simply be style relating to an individual potter, perhaps demonstrating technical ability, I propose that more often its function may have been to introduce new types of meanings and new narratives which could not be created using the more stylised geometric motifs. Because these narratives were new, they had to be made much more explicit. Naturalistic decoration therefore

may have functioned to introduce new social narratives, and to replace and extend existing social conventions. The depiction of naturalistic scenes, including people, animals and places, might have been associated with control. Representational images can be powerful and dangerous, and the vessels carrying these depictions may have been highly active social agents in themselves. In time, as the new narratives themselves became conventional, the naturalistic depictions had the potential to become more abstract and perhaps eventually be absorbed in the much larger and more common category of abstract, stylised or geometric designs. The power of innovation may have been significant. Not only may the depiction of naturalistic scenes have created powerful objects, but it could also have been a direct challenge to conventional social narratives. As a powerful mechanism through which convention could be challenged, it might have constituted a threat to established cosmologies and social order. Consequently, its use might only have been open to certain individuals acting in particular contexts. While naturalistic motifs are generally very rare throughout the period, there is one substantial context at Domuztepe where they are remarkably common. This is the ‘Ditch’, which is not in fact a single feature as the name suggests but a long series of linear cuts and re-cuts along an axis of c. 30 m. Although the activity may have continued for well over 100 years, most of the pottery in the refuse that made up the fill of the ‘Ditch’ seems to be Halaf Ia in date. What is remarkable is the quantity of naturalistic decoration, to the extent that it actually dominates the pottery assemblage. While examples occur with apparently headless bodies (Campbell 2004), dancing ladies (Campbell 2008, fig. 2, no. 4), animals and many other motifs, it is the depictions

154

Stuart Campbell

Geometric/Abstract Common Used in social contexts with active set of meanings that could relate to personal histories, storytelling and/or mythologies. Encoding of meanings is inflexible and with a framework of convention. Can be used to modify or challenge old narratives through innovative juxtaposition but within existing framework. Reinforces existing framework. May suggest common cosmologies and shared social narratives within area of use, which is sometimes very wide.

Naturalistic/representational Very rare and exceptional. Used in social contexts with active set of meanings that could relate to personal histories, storytelling and/or mythologies. Draws on pre-existing encoding of meanings by use of some geometric decoration but not constrained by it. Potential to create completely new narratives and meanings. Potential to challenge and transform existing framework. Understanding may be very contextual and local to particular regions in which new narratives appear.

Table 18.1 Roles of geometric and naturalistic decoration in conveying meaning.

that show houses with trees standing between them and usually birds perched on the roofs (Fig. 18.8) that are the most common, with perhaps 20 or more vessels carrying variants of this scene. We need to excavate more extensively to fully understand the contemporary pottery at Domuztepe. However, it seems probable that the pottery in the ‘Ditch’ represents a specific context of use, the refuse from which was disposed of in one location, perhaps because it was in some way ‘dangerous’ or ‘powerful’ and needed to be controlled after its use and breakage. This may suggest a particular domain within which new social narratives were being advanced or an authority which was using pottery decoration as an active agent of change. If the interpretation proposed above is correct, we can see the painted decoration on the pottery of the late Neolithic in north Mesopotamia from a new perspective, as part of a system of communication where vessels gained agency that was created and deciphered through social narratives. This gives the ceramics a significant and active role in the way in which society functioned and the ways that social conventions were conveyed and enforced. Although both abstract, geometric decoration and representational designs functioned in ways that were closely related, they may have represented opposite ends of the same system, with the ability to convey different types of meanings (Table 18.1). By being more standardised and representing accepted cosmologies, the stylised, geometric motifs may have been meaningful over much wider regions. While this correlates with the wide spread of certain motif combinations, such as the association of flowers and dots or the appearance of ‘dancing ladies’, it also poses the question of the extent to which stylistic similarities in pottery decoration reflect shared social narratives and mythologies. If the more naturalistic decoration was used to convey new narratives, it may have been much more local in impact, perhaps requiring

more verbal interpretation, and possibly reflecting the intent of individuals or small corporate groups to introduce new ways of understanding the world.

References Bernbeck, R. 1994. Die Auflösung der Häuslichen Produktionsweise: das Beispiel Mesopotamiens. Berlin, Dietrich Reimer. Bernbeck, R. 1999. Structure strikes back: Intuitive meanings of ceramics from Qale Rostam, Iran. In J. E. Robb (ed.) Material Symbols: Culture and Economy in Prehistory, 90–111. Occasional Paper 26. Carbondale, Center for Archaeological Investigations. Bowser, B. J. 2000. From pottery to politics: An ethnoarchaeological study of political factionalism, ethnicity, and domestic pottery style in the Ecuadorian Amazon. Journal of Archaeological Method and Theory 7(3), 219–248. Breniquet, C. 1992. A propos du vase halafien de la Tombe G2 de Tell Arpachiyah. Iraq 54, 69–78. Campbell, S. 1992. Culture, Chronology and Change in the Later Neolithic of North Mesopotamia. Unpublished Ph.D. thesis, University of Edinburgh. Campbell, S. 2004. Domuztepe 2004 excavation season. Anatolian Archaeology 10, 4–6. Campbell, S. 2008. Feasting and dancing: Gendered representation and pottery in later Mesopotamian prehistory. In D. Bolger (ed.) Gender through Time in the Ancient Near East, 53–76. Lanham (MD), AltaMira. Cauvin, J. 2000. The Birth of the Gods and the Origins of Agriculture. Translated from the French by T. Watkins. Cambridge, Cambridge University Press. Conkey, M. W. and C. A. Hastorf (eds) 1990. The Uses of Style in Archaeology. Cambridge, Cambridge University Press. Davidson, T. E. 1977. Regional Variation within the Halaf Ceramic Tradition. Unpublished Ph.D. thesis, University of Edinburgh. Forest, J.-D. 1996. Mesopotamie: l’apparition de l’Etat VIIe–IIIe millénaires. Paris, Méditerranée.

18. Understanding symbols: Putting meaning into painted pottery Garfinkel, Y. 2003. Dancing at the Dawn of Agriculture. Austin, University of Texas Press. Gell, A. 1998. Art and Agency: An Anthropological Theory. Oxford, Clarendon. Hole, F. 1984. Analysis of structure and design in prehistoric ceramics. World Archaeology 15(3), 326–347. Huot, J.-L. 1994. Les Premiers Villages de Mésopotamie: Du Village à la Ville. Paris, Armand Colin. Ippolitoni-Strika, F. 1990. A bowl from Tell Arpachiyah and the tradition of portable shrines. Mesopotamia 25, 147–174. Ippolitoni-Strika, F. 1996. Halafian art, religion, society: The funerary bowl from Arpachiyah. The fringed square as a “sacred rug”. Mesopotamia 31, 5–31. Irving, A. C. 2001. A Contextual Study of Ceramic Evidence for Social Relations and Change during the Halaf-Ubaid Transition. Unpublished Ph.D. thesis, University of Manchester. LeBlanc, S. and P. J. Watson 1973. A comparative statistical analysis of painted pottery from seven Halafian sites. Paléorient 1, 119–136. Mallowan, M. E. L. and J. Cruickshank Rose 1935. Excavations at Tall Arpachiyah, 1933. Iraq 2, 1–178. Melville, D. J. 2005. Aspects of symmetry in Arpachiyah pottery. In R. Sarhangi and R.V. Moody (eds) Renaissance BanffBridges: Mathematical Connections in Art, Music and Science (Proceedings 2005), 131–136. Pheonix (AZ), Bridges Visual Art Exhibit. Merpert, N. I. and R. M. Munchaev 1987. The earliest levels at Yarim Tepe I and Yarim Tepe II in northern Iraq. Iraq 49, 1–37.

155

Munchaev, R. M. and N. I. Merpert 1981. Earliest Agricultural Settlements of Northern Mesopotamia. Moscow, Nauka. Nieuwenhuyse, O. P. 2007. Plain and Painted Pottery. The Rise of Late Neolithic Ceramic Styles on the Syrian Plains. Brussels, Brepols. Sackett J. R. 1990. Style and ethnicity in archaeology: The case for isochrestism. In M. W. Conkey and C. Hastorf (eds) The Uses of Style in Archaeology, 32–43. Cambridge, Cambridge University Press. Schmandt-Besserat, D. 2007. When Writing Met Art: From Symbol to Story. Austin, University of Texas Press. Skeates, R. 2007. Abstract and figurative representation and the politics of display in Neolithic southeast Italy. In C. Renfrew and I. Morley (eds) Image and Imagination: A Global Prehistory of Figurative Representation, 199–210. Cambridge, McDonald Institute for Archaeological Research, University of Cambridge. Tsuneki, A. and Y. Miyake (eds) 1998. Excavations at Tell Umm-Qseir in Middle Khabur Valley, North Syria. Tsukuba, Department of Archaeology, Institute of History and Anthropology, University of Tsukuba. Von Oppenheim, M. F. 1943. Tell Halaf 1, Die Prähistorischen Funde. Berlin, De Gruyter. Von Wickede, A. 1986. Die Ornamentik der Tell Halaf-Keramik: Ein Beitrag zu ihrer Typologie. Acta Praehistoria et Archaeologica 18, 7–32. Wengrow, D. 2001. The evolution of simplicity: Aesthetic labour and social change in the Neolithic Near East. World Archaeology 33(2), 168–188.

19 GENDER AND SOCIAL COMPLEXITY IN PREHISTORIC AND PROTOHISTORIC CYPRUS Diane Bolger

All narratives of the past are gendered, whether consciously or not. Without explicitly considering gender, however, we are more likely to base our interpretations of past societies upon unmediated assumptions reflecting modern western beliefs and practices. Not only does this distort our understanding of men’s and women’s roles and relations in the remote past, but it legitimises their normative status in contemporary society. The present paper is dedicated to Eddie Peltenburg’s long-standing interest in the social, political and economic development of early societies in the ancient Near East, particularly in Cyprus, as well as his concern with interpreting archaeological evidence within particular historical and stratigraphic contexts. I shall begin by looking at some of the ways in which traditional narratives of the past have served to limit our understanding of gender, identity and the sexual division of labour during the Neolithic and Chalcolithic periods of the island. I shall then consider some of the interfaces between gender and social complexity during the Bronze and Early Iron Ages. By examining evidence for gender relations through time, and interpreting those changes within specific contextual frameworks, it is possible to move beyond universalist assumptions to a more nuanced understanding of social development among the island’s early prehistoric and protohistoric communities.

Transcending Binary Gender Categories in Early Cypriot Society Traditional narratives of the past have often assumed that men and women had sharply distinct roles even during the early phases of prehistory. Among archaeologists working in Cyprus, this tendency is exemplified by the

generic interpretation of early anthropomorphic figurines as symbols of female fertility and/or motherhood. Such a narrow view constrains our ability to appreciate the complex relationships between gender, the body and social identity in the Cypriot past by ignoring other gender categories (such as male, ambiguous and dual-sexed) and by limiting women’s economic and social roles, as well as their bodily experiences, to their biological capacity for reproduction. It also fails to acknowledge that gender in many societies is a much more sophisticated construct than a simple male/ female polarity. Images of male as well as female bodies are present in the island’s Neolithic and Chalcolithic repertoires, as are dual-sexed and ambiguously sexed examples (Fig. 19.1; for a range of types, see illustrations in Dikaios 1953 and 1961; and in Peltenburg et al. 1991; 1998; and 2003). Given the overt sexual symbolism of some of the figurines, they are likely to have had as much to do with sexuality as fertility or birth. There is also evidence to suggest that some figurines depicted various stages of male and female lifecycles, perhaps marking critical transformations in gender identities throughout an individual’s life (Bolger 2003, chap. 4). Gender archaeology over the past ten to fifteen years has been highly critical of binary categories of sex and gender, focussing instead upon the considerable degree of spatial and temporal variability in the archaeological record (e.g. Tringham and Conkey 1998; Hamilton 2000; Bolger 2003, chap. 4; Mina 2007; Croucher 2008; Daems 2008; Joyce 2008). It is no longer acceptable to lump all examples together under a single interpretative umbrella since it is widely recognised that gender, identity and the body are unstable categories that are subject to frequent change (Butler 1999). Characterisations of early figurative art in Cyprus as eternally feminine, static and unchanging,

19. Gender and social complexity in prehistoric and protohistoric Cyprus

Fig. 19.1 Dual-sexed figurine of the Chalcolithic period from Building 1 at Lemba-Lakkous (after Peltenburg et al. 1985, frontispiece).

symbolic of fertility or maternity, representing mother goddesses or precursors to the goddess Aphrodite, give rise to ahistorical narratives of the past that ignore contextual constraints and obscure the dynamic character of bodily representation (e.g. Orphanides 1990; Karageorghis and Karageorghis 2002). A second common misconception concerning gender in ancient societies is the notion that a sexual division of labour emerged at an early stage in prehistory and rapidly led to the differential involvement of men and women in technology, production and exchange (Bolger 2003, chap. 3; Bolger 2010). One particularly common example is the view that women were the principal pottery makers in early household based production since pottery manufacture is regarded as an ‘interruptible’ task that can easily be accommodated within the domestic schedules conventionally associated with women’s work (e.g. Arnold 1985, 100–101). Theories

157

concerning gender and pottery production in early agropastoral societies, made popular during the 1960s and 1970s by Deetz, Longacre, Whallon and others, were also based on a priori assumptions concerning descent and postmarital residence patterns (e.g. Deetz 1968; Whallon 1968; Longacre 1981). Ceramic designs were used as evidence for establishing those patterns, and since it was assumed that women made pots, it followed that exogamic practices involving the movement of women should foster greater levels of intra-site variation and lesser degrees of inter-site variation as new ideas and techniques were transmitted between villages. Joanne Clarke has challenged this hypothesis by comparing motifs on Red-on-White ceramics from the Cypriot Late Neolithic sites of Ayios Epiktitos-Vrysi, Klepini-Troulli and Philia-Drakos site A (2002). Her results show that the two closest sites (Vrysi and Troulli) exhibit weaker stylistic links than those more distant from one another (Vrysi and Philia), thus contradicting the Deetz/Longacre hypothesis and suggesting that exogamic marital practices are not likely to have been responsible for the diffusion of stylistic elements in ceramic production on the island during its early phases. As Clarke observes, the models adopted by Deetz, Longacre, Whallon and others “highlight the theoretical inconsistencies of archaeological interpretation with regard to both gender based divisions of labour in prehistoric subsistence strategies, and socio-cultural interactive patterning” (2002, 253). Collective and cooperative efforts rather than sexually segregated patterns in the organisation of labour are strongly suggested by the results of experimental work with Chalcolithic pottery at the Lemba Experimental Village where the replication of Early Chalcolithic pottery vessels has shown that the manufacture of hand-made pottery of the 4th and early 3rd millennia entailed a long and complex operational sequence that would have demanded the collaborative efforts of men, women and probably even children (Bolger 2003, chap. 3; Bolger and Shiels 2003). In addition, the absence of firing facilities such as pits or kilns, as well as the collection of clays and tempering materials from remote sources, indicates that some stages of pottery production took place at considerable distances from home. Stereotypical narratives of the past that envision men as the primary makers and users of stone tools and women as potters fail to recognise gender as a dynamic process that is historically and contextually situated and therefore subject to frequent spatial and temporal variation.

Gender and Social Complexity in the Cypriot Bronze Age While there is some evidence for low levels of social differentiation, status distinction and prestige hierarchy

158

Diane Bolger

Fig. 19.2 Red Polished plank figurines of the Early-Middle Cypriot periods (after Morris 1985, figs 178–185, 211–212, 214, 228–229).

during the Neolithic and Chalcolithic periods in Cyprus (Peltenburg 2002), there is little evidence to suggest separate roles for males and females. During the Cypriot Bronze Age, however, we might expect to see some evidence for a gendered division of labour, particularly in light of sociopolitical changes that occurred during the second half of the 3rd millennium as the island became part of a wider sphere of cultural interaction; and during the 2nd millennium as the result of the development of trade and metallurgy, which are widely regarded as catalysts for accelerated rates of socio-economic complexity, a more centralised political administration, and the growth of densely populated urban centres (Knapp 1986; 1993; 2008; Keswani 2004; Steel 2004). In fact, recent research on gender during the Cypriot

Bronze Age has not confirmed that expectation (Bolger 2004). A study of EC-MC plank figurines by Talalay and Cullen (2002) addresses this question for the late 3rd to early 2nd millennia and provides a good example of the ways in which figurative art can be given more specific cultural meanings through contextual analysis (Fig. 19.2). Although some of the plank figures have been found in settlements, they are more frequently associated with large chamber tombs located in discrete cemetery complexes and utilised repeatedly as collective burial facilities over several generations. The latter are very different from the simple mortuary facilities of earlier periods, which with few exceptions comprise individual pit graves that were

19. Gender and social complexity in prehistoric and protohistoric Cyprus

Fig. 19.3 Left: the ‘ingot god’ from Enkomi (after Tatton-Brown 1979, fig. 102). Right: the Bomford figurine (courtesy of the Ashmolean Museum, Oxford).

159

used only once and were situated within settlements. This new collective burial system signifies important changes in social organisation, and the placement of figurines into tombs may have been intended to underscore an individual’s membership in corporate kinship groups. Talalay and Cullen propose that the standardised, ‘genderless’ forms of the plank figures, as well as their repetitive decorative style, may also have been designed purposely to reinforce ideologies of corporate identity. As a result, aspects of individual identity, such as sex and gender, may have been deliberately obscured or suppressed (for an opposing view, see Knapp and Meskell 1997). While arguments for gender ambiguity within the contexts of EC and MC mortuary ritual are compelling, we cannot assume that this was the case in settlement contexts, or in later phases of the Bronze Age, which are characterised by the island’s increasing participation in international trade networks and by an accelerated demand by elite groups for exotic luxury goods. The latter were very likely imbued with symbolic as well as economic value, displayed and manipulated in order to enhance personal or group status,

Fig. 19.4 Plan of Tomb 11 at Kalavasos-Ayios Dhimitrios (courtesy of Alison South).

160

Diane Bolger

and used to legitimise the positions of emerging elites within an increasingly hierarchical social structure (e.g. Knapp 1986; 2008, chap. 4; Peltenburg 1996; Keswani 2004, chap. 5; Steel 2004, chap. 6). Very little has been said in these discussions, however, concerning the agency or identity of the groups who were involved in those crucial developments, with most scholars adopting a vague terminology (such as the use of the term ‘elite’) that foregrounds the role of active players (tacitly assumed to be males) in the growth of socioeconomic complexity (Gero 2000; Bolger 2003, chap. 1). It is significant in this context that the Ingot god figurine from Enkomi (Fig. 19.3, left) rather than the Bomford figurine (Fig. 19.3, right) is commonly used to illustrate these arguments, again revealing an unspoken belief that men rather than women (or both) were the principal agents of social change (for exceptions see Bolger 2003, chap. 4; and Knapp 2008, 173–186). Clearly the changes in gender relations that accompanied the emergence and development of complex society in Cyprus are issues worthy of further investigation. Given the lack of well stratified settlement evidence for large segments of the Bronze Age, changes in the interfaces between gender and socio-economic complexity are more readily assessed through the evidence of mortuary data. In mortuary rituals of the LBA, polarised gender constructs are suggested by the exclusive burial of males and/or females at a number of sites, such as Akhera, Toumba tou Skourou and Kalavasos-Ayios Dhimitrios (Bolger 2003, 171–175). Simple correlations between segregated burial customs and low female status are contradicted, however, by the luxurious nature and copious numbers of finds in female graves, such as the remarkable burial of two adult females together with three infants in T 11 at Kalavasos-Ayios Dhimitrios (Fig. 19.4; see also Goring 1989; South 1997). The lack of clear-cut patterns is in keeping with gendered approaches in anthropology and history, which emphasise diversity rather than uniformity of gender constructs and socio-political organisation in the transition to state level society (e.g. Gailey 1987; Silverblatt 1988; Ehrenreich et al. 1995; Joyce and Hendon 2000). Perhaps the most important aspect of gender suggested by burial evidence is the differential treatment of elite and non-elite women: female burials at a number of sites were furnished with considerable wealth, but women of lower status are archaeologically invisible. The same cannot be said of lower status males, who were accorded burial rights at a number of sites, such as Enkomi (French T 10) and Ayios Iakovos-Melia (T 8) where burials of adult males are associated with locally made pottery but lack metalwork and other prestige items (Fischer 1986, 32–35). Rather than fostering status distinctions between males and females, the growth of urbanism and more complex forms of socio-political organisation appears to have had a polarising effect on women, dividing them into a number

of disparate and competing sub-groups. This suggests that the social construction of gender may have been more closely linked to the wealth and rank of individuals and their affiliated groups than to biologically or culturally determined ‘masculine’ and ‘feminine’ traits (for further details, see Bolger 2003, chap. 6). The existence of a centralised political apparatus on the island during the LBA continues to be a topic of considerable debate (for recent discussions, see Knapp 2008, chaps. 4, 6; and Bolger 2009), but it seems likely that heterarchical rather than hierarchical forms of social organisation fostered variable patterns of gendered behaviour in different regions and localities (Bolger 2004, chaps. 6–7). This may help to explain the seemingly contradictory evidence in the mortuary record of the LBA in which high proportions of elite female burials occur at some sites but not at others, a pattern which has been observed by Keswani (2004, 141). It also suggests that while some women in LC society managed to attain positions of high status, it was probably by virtue of their class and kinship affiliation rather than their sex or gender that they managed to do so.

Gender and Status in the Early Iron Age: Continuity or Change? Major disruptions at the end of the LBA in Cyprus and the eastern Mediterranean, due in part to displaced populations from the Aegean, are marked by new settlement patterns, pottery types and burial facilities; moreover, the appearance of elite tombs containing warrior equipment (e.g. Skales T 76, Alaas T 12, and Kaloriziki T 40) suggests that military prowess played an important role in socio-political developments on the island from the end of the LBA into the EIA (Coldstream 1989; Steel 2002). In light of these dramatic changes, which also involved processes of acculturation through the assimilation and amalgamation of different ethnic groups, Steel has argued for a significant shift in gender relations during the EIA (2002, 105). In the final section of this paper I will attempt to assess this proposition by looking at evidence of EIA mortuary ritual and pottery iconography. Given the lack of settlement evidence for the LBA/IA transition and the succeeding CG I period, these classes of evidence furnish the most useful data. Bioarchaeological evidence for gender in EIA Cyprus is limited due to the poorly preserved and inadequately recorded state of the skeletal material. However, several sites, such as Gastri-Alaas, Kouklia-Skales and KourionKaloriziki, provide some insights into the treatment of elite males and females. The cemetery of Gastri-Alaas (LCIIIB) shows the appearance of a new tomb type, the chamber tomb with dromos (Karageorghis 1975). All burials were inhumations, and all but one were single interments, but only a small proportion of the skeletal material has been sexed. In

19. Gender and social complexity in prehistoric and protohistoric Cyprus

161

Fig. 19.5 Plan of Tomb 40 at Kourion-Kaloriziki (after McFadden 1954, fig. 8).

most cases only the presence or absence of skeletal material was noted, but sex was recorded for the principal burials in three of the tombs: T 14, an adult male associated with a Bucchero Wheelmade cup and a Proto-bichrome stirrup jar; T 15, an adult female buried with ten pottery vessels, including a Mycenaean three-handled jar and a lentoid flask of ‘foreign ware’, as well as a pair of gold earrings; and T 16, an adult female whose many associated finds include 18 pottery vessels, an ivory disc, two bronze fibulae, a bronze pin, and a pair of gold earrings. Despite the limited nature of this evidence, it is significant that the two females were very lavishly furnished while the male burial had relatively few objects. In accordance with current approaches in anthropology and archaeology, which stress the central role of the living community in the treatment of the dead (e.g. Parker Pearson 1999), this pattern suggests a high degree of continuity in communal values regarding the status of elite females throughout the LC period. The rich cemetery at Kouklia-Skales (Karageorghis 1983) should provide ample scope for understanding patterns of gender and mortuary ritual during the Iron Age, but problems of preservation, excavation, and recording pose serious limitations. In addition, the practice of multiple burial programmes makes it difficult to associate particular individuals with particular sets of grave goods.

The osteological report published in the Skales report by Schulte-Campbell (appendix XII) summarises the analytical results on 20 individuals (9 males, 4 females, 5 adults of unknown sex, and 1 child), but it is not complete: only a portion of the anthropological material was studied, and there are numerous inconsistencies between the numbers and sexes of individuals provided in the tomb descriptions of the main text and those in appendix XII. Despite these limitations, it is possible to gain some idea of the wealth and status of individuals buried in these tombs. In the five cases where sexed burials can be associated with particular sets of grave goods, four are adult females (Tombs 72, 78, 88, 93) and one is an adult male (T 79). This suggests that elite women formed a substantial proportion, if not a majority, of the burial population at the site. Headshaping, which was widely practiced by elite groups during the LBA in Cyprus (Bolger 2003; Lorentz 2008), has been recorded on two CG skulls: an adult male (T 49, b) and an adult female (T 93, 36a). Both display the deliberate ‘postbregmatic’ pattern common in LBA contexts, suggesting that the practice of cranial modification continued to be applied to both male and female infants during the CGI period. I turn now to what is arguably the richest and most important tomb of EIA Cyprus, T 40 at Kourion-Kaloriziki (Fig. 19.5). The tomb was looted in 1903, but nearly 50

162

Diane Bolger

Fig. 19.6 Cypro-Geometric I plate in Proto White Painted ware from Tomb 58 at Kouklia-Skales (after Iacovou 1988, fig. 78).

years later, with help from one of the looters, McFadden managed to re-locate it; his subsequent excavations led to the discovery of a rock-cut bench at the south end of the chamber which the looters had missed (1954). A number of exceptional in situ finds were found on top of the bench, including a bronze urn containing the cremated remains of an adult female and luxury items, such as gold and bronze ornaments, ceramic drinking equipment, and a bronze drinking set (Steel 2002, 112). A similar urn recovered from the earlier looting operations was assumed to have originally contained the cremated remains of her male partner, despite the lack of evidence for a burial within it; this ‘invisible’ male is still widely regarded as the bearer of the famous sceptre and therefore a king, making the female his consort. McFadden’s proposal of this hypothetical male has thus become a ‘fact’ cited in much subsequent archaeological literature to support notions of kingship and royalty (e.g. Coldstream 1989, 333). If we ignore the obvious gender biases inherent in this interpretation and consider the evidence itself, there are strong grounds to argue for the continued social recognition of elite women during the EIA. The emergence of Proto White Painted ware during the LC IIIB period (Iacovou 1988) issued in a new stylistic repertoire which Steel cites as evidence for newly emerging gender constructs (2002, 112). Several of the painted figures on PWP vessels can clearly be identified as males engaged in activities (i.e. hunting, fishing and other activities involving animals), which Steel characterises as a ‘macho’ style (Fig. 19.6). However, some of the male figures on these vessels appear to be engaged in less ‘macho’ endeavours: one is shown holding a flower (Iacovou 1988, cat no. 34) while

another plays a lyre (1988, cat. no. 29). None of the human figures on PWP pottery is unquestionably female, but in the corpus published by Iacovou (1988) only two of the four figures on LC IIIB examples and two of the five figures on CGI examples are unambiguously male, i.e. depicted with a penis and/or a beard (1988, cat. nos 19, 29, 33–34). It is not entirely clear how the gender of the other examples has been determined, but in some cases it is likely to have been inferred from their dress, their well-developed calf and thigh musculature, and the presence or absence of weaponry. The lack of clear cut sex indicators on the majority of the painted figures, however, begs the question of why some were depicted as active, aggressive, unambiguous males while the gender of others is far less certain; such diversity of expression should encourage us to adopt a more flexible and nuanced approach to our understanding of gender during the EIA. While the evidence for gender in Early Iron Age Cyprus is limited, and needs to be examined in greater detail than has been possible here, it is difficult to confirm that “a significant change in gender relations among elite groups” took place at this time (Steel 2004, 113). As we have seen, burial evidence from three key sites (Gastri-Alaas, KoukliaSkales, and Kourion-Kaloriziki), as well as iconographic evidence on Proto White Painted vessels, fails to support the notion of a dramatic shift in gender relations during the LC IIIB and CG I periods, at least among elite groups; unfortunately, we still know very little about the gender roles and identities among non-elites of either period.

Conclusions By considering changes in gender constructs within specific temporal and contextual frameworks it is possible to move beyond generic models based on unmediated assumptions and explore the dynamic inter-relationships between gender and other aspects of culture such as technology, ritual, material culture and socio-economic complexity (Joyce 2008). As we have seen, there is little evidence for polarised gender categories during the earlier phases of Cypriot prehistory. While there are some indications of increasing distinctions between male and female roles during the LBA, differences within those gender groups are as visible and significant as divisions between them. And despite the considerable social transformations that mark the transition to the EIA, gender constructs do not appear to have altered significantly, at least among the upper echelons of society. The overall picture that emerges is one in which abstract models linking increasing levels of socio-economic complexity to a decline in female status cannot easily be sustained. During all phases of Cypriot prehistory, local and regional patterns are likely to have prevailed over islandwide developments, and this lack of cultural uniformity

19. Gender and social complexity in prehistoric and protohistoric Cyprus undermines attempts to apply universal categories of gender and social identity to ancient Cypriot communities. By examining the evidence contextually, and by highlighting variability in social behaviour, it is possible to move beyond abstract categories such as ‘state’, ‘pre-state’, ‘male’ and ‘female’ to uncover patterns of gender based on particular temporal and spatial constraints. Here, as elsewhere, links between sex, gender and other aspects of society must be demonstrated rather than assumed. In contrast to unilinearevolutionary models in which socio-political trajectories are seen to emerge gradually from one century or millennium to the next, contextual approaches are non-linear and recursive, and as a result are more finely attuned to the richness and diversity of human behaviour that the archaeological record of Cyprus so strikingly conveys.

Abbreviations CG EC EIA LBA LC MC T

Cypro-Geometric Early Cypriot Early Iron Age Late Bronze Age Late Cypriot Middle Cypriot Tomb

References Arnold, D. E. 1985. Ceramic Theory and Cultural Process. Cambridge and New York, Cambridge University Press. Bolger, D. 2003. Gender in Ancient Cyprus: Narratives of Social Change on a Mediterranean Island. Gender and Archaeology Series. Walnut Creek (CA), AltaMira. Bolger, D. 2009. Review of Prehistoric and Protohistoric Cyprus: Identity, Insularity, and Connectivity by A. B. Knapp. Antiquity 83, 550–551. Bolger, D. 2010. The dynamics of gender in early agricultural societies of the Near East. Signs 35.2, 503–531. Bolger, D. and J. Shiels 2003. The pottery. In E. Peltenburg (ed.) The Colonisation and Settlement of Cyprus: Investigations at Kissonerga-Mylouthkia, 1976–1996, 133–168. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Butler, J. 1999. Gender Trouble: Feminism and the Subversion of Identity. London and New York, Routledge. Clarke, J. 2002. Gender, economy and ceramic production in Neolithic Cyprus. In D. Bolger and N. Serwint (eds) Engendering Aphrodite: Women and Society in Ancient Cyprus, 251–263. CAARI Monographs 3. Boston, American Schools of Oriental Research. Coldstream, J. N. 1989. Status symbols in Cyprus in the eleventh century BC. In E. Peltenburg (ed.) Early Society in Cyprus, 325–335. Edinburgh, Edinburgh University Press. Croucher, K. 2008. Ambiguous genders? Alternative interpret-

163

ations: a discussion of case studies from the Pre-pottery Neolithic and Halaf periods. In D. Bolger (ed.) Gender through Time in the Ancient Near East, 21–51. Gender and Archaeology Series. Lanham (MD), AltaMira. Daems, A. 2008. Evaluating patterns of gender through Mesopotamian and Iranian human figurines. In D. Bolger (ed.) Gender through Time in the Ancient Near East, 77–117. Gender and Archaeology Series. Lanham (MD), AltaMira. Deetz, J. 1968. The inference of residence and descent rules from archaeological data. In L. R. Binford and S. R. Binford (eds) New Perspectives in Archaeology, 41–48. Chicago, Aldine. Dikaios, P. 1953. Khirokitia. Oxford, Oxford University Press. Dikaios, P. 1961. Sotira. Philadelphia, University of Pennsylvania Press. Ehrenreich, R. M., C. L. Crumley, and J. E. Levy (eds) 1995. Heterarchy and the Analysis of Complex Societies. Archaeological Papers of the American Anthropological Association 6. Washington DC, American Anthropological Association. Fischer, P. 1986. Prehistoric Cypriot Skulls. Studies in Mediterranean Archaeology 75. Göteborg, Paul Åströms Förlag. Gailey, C. W. 1987. From Kinship to Kingship: Gender Hierarchy and State Formation in the Tongan Islands. Austin, University of Texas Press. Gero, J. 2000. Troubled travels in agency and feminism. In M.A. Dobres and J. Robb (eds) Agency in Archaeology, 40–50. London and New York, Routledge. Goring, E. 1989. Death in everyday life: Aspects of burial practice in the Late Bronze Age. In E. Peltenburg (ed.) Early Society in Cyprus, 95–105. Edinburgh, Edinburgh University Press. Hamilton, N. 2000. Ungendering archaeology: Concepts of sex and gender in figurine studies. In M. Donald and L. Hurcombe (eds) Representations of Gender from Prehistory to the Present, 17–30. New York, MacMillan. Iacovou, M. 1988. The Pictorial Pottery of Eleventh Century B.C. Cyprus. Studies in Mediterranean Archaeology 78. Göteborg, Paul Åströms Förlag. Joyce, R. 2008. Ancient Bodies, Ancient Lives: Sex, Gender and Archaeology. London, Thames & Hudson. Joyce, R. and J. A. Hendon 2000. Heterarchy, history and material reality: “Communities” in late Classic Honduras. In M. A. Canuto and J. Yaeger (eds) The Archaeology of Communities: A New World Perspective, 143–160. New York, Routledge. Karageorghis, V. 1975. Alaas: A Protogeometric Necropolis in Cyprus. Nicosia, Department of Antiquities, Cyprus. Karageoghis, V. 1983. Palaepaphos-Skales: An Iron Age Cemetery in Cyprus. Ausgrabungen in Alt-Paphos auf Cypern 3. Constance, Universitätsverlag Konstanz. Karageorghis, V. and J. Karageorghis 2002. The great goddess of Cyprus or the genesis of Aphrodite in Cyprus. In S. Parpola and R. M. Whiting (eds) Sex and Gender in the Ancient Near East, 263–282. Winona Lake (IN), Eisenbrauns. Keswani, P. 2004. Mortuary Ritual and Society in Bronze Age Cyprus. Monographs in Mediterranean Archaeology. London, Equinox. Knapp, A. B. 1986. Copper Production and Divine protection: Archaeology, Ideology and Social Complexity on Bronze Age Cyprus. Studies in Mediterranean Archaeology Pocket-book 42. Göteborg, Paul Åströms Förlag. Knapp, A. B. 1993. Social complexity: Incipience, emergence and

164

Diane Bolger

development on prehistoric Cyprus. Bulletin of the American Schools of Oriental Research 292, 85–106. Knapp, A. B. 2008. Prehistoric and Protohistoric Cyprus: Identity, Insularity and Connectivity. Oxford, Oxford University Press. Knapp, A. B. and L. Meskell 1997. Bodies of evidence on prehistoric Cyprus. Cambridge Archaeological Journal 7(2), 183–204. Longacre, W. A. 1981. Kalinga pottery: An ethnoarchaeological study. In I. Hodder (ed.) Pattern of the Past, 49–66. New York, Cambridge University Press. Lorentz, K. 2008. From life course to longue durée: Headshaping as gendered capital? In D. Bolger (ed.) Gender through Time in the Ancient Near East, 281–311. Gender and Archaeology Series. Lanham (MD), AltaMira. McFadden, G. 1954. A Late Cypriote III tomb from Kourion Kaloroiziki no. 40. American Journal of Archaeology 58(2), 131–142. Mina, M. 2007. Figurines without sex: People without gender? In S. Hamilton, R. Whitehouse and K. Wright (eds) Archaeology and Women: Ancient and Modern Issues, 263–282. Walnut Creek (CA), Left Coast. Morris, D. 1985. The Art of Ancient Cyprus. Oxford, Phaidon. Orphanides, A. 1990. The meaning and function of the Bronze Age terracotta anthropomorphic figurines from Cyprus. Report of the Department of Antiquities, Cyprus, 45–50. Parker Pearson, M. 1999. The Archaeology of Death and Burial. Gloustershire, Sutton. Peltenburg, E. 1996. From isolation to state formation in Cyprus, c. 3500–1500 B.C. In V. Karageorghis and D. Michaelides (eds) The Development of the Cypriot Economy: From the Prehistoric Period to the Present Day, 17–44. Nicosia, University of Cyprus and Bank of Cyprus. Peltenburg, E. 2002. Gender and social structure in prehistoric Cyprus: A case study from Kissonerga. In D. Bolger and N. Serwint (eds) Engendering Aphrodite: Women and Society in Ancient Cyprus, 53–66. CAARI Monograph Series 3. Boston, American Schools of Oriental Research. Peltenburg, E. J., D. Baird, A. Betts, S. Colledge, P. Croft, C. Elliott, T. Lawrence, D. A. Lunt, K. Niklasson, J. Renault-Miskovsky, J. S. Ridout-Sharpe, E. Slater, J. D. Stewart and C. Xenophontos 1985. Lemba Archaeological Project I: Excavations at LembaLakkous, 1976–1983. Studies in Mediterranean Archaeology 70(1). Göteborg, Paul Åströms Förlag.

Peltenburg, E., A. Betts, D. Bolger, P. Croft, C. Elliott, E. Goring, M. A. Murray, J. S. Ridout-Sharpe and G. Thomas 1991. Lemba Archaeological Project II(2): A Ceremonial Area at Kissonerga. Studies in Mediterranean Archaeology 70(3). Sävedalen, Paul Åströms Förlag. Peltenburg, E., D. Bolger, P. Croft, E. Goring, B. Irving, D. Lunt, S. W. Manning, M. A. Murray, C. McCartney, J. S. RidoutSharpe, G. Thomas, M. E. Watt and C. Elliott-Xenophontos 1998. Lemba Archaeological Project II(1A): Excavations at Kissonerga-Mosphilia, 1979–1992. Studies in Mediterranean Archaeology 70(2). Jonsered, Paul Åströms Förlag. Peltenburg, E., D. Bolger, S. Colledge, P. Croft, S. C. Fox, E. Goring, A. Jackson, D. Lunt, C. McCartney, M. A. Murray, J. Ridout-Sharpe, G. Thomas and M. E. Watt 2003. Lemba Archaeological Project III(1): The Colonisation and Settlement of Cyprus. Investigations at Kissonerga-Mylouthkia, 1976– 1996. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Silverblatt, I. 1988. Women in states. Annual Review of Anthropology 17, 427–460. South, A. 1997. Kalavasos-Ayios Dhimitrios 1992–1996. Report of the Department of Antiquities, Cyprus, 151–176. Steel, L. 2002. Wine, women and song: Drinking ritual in Cyprus in the Late Bronze and Early Iron Ages. In D. Bolger and N. Serwint (eds) Engendering Aphrodite: Women and Society in Ancient Cyprus, 105–119. CAARI Monograph Series 3. Boston, American Schools of Oriental Research. Steel, L. 2004. Cyprus before History: From the Earliest Settlers to the End of the Bronze Age. London, Duckworth. Talalay, L. and T. Cullen 2002. Sexual ambiguity in Early-Middle Cypriot plank figures. In D. Bolger and N. Serwint (eds) Engendering Aphrodite: Women and Society in Ancient Cyprus, 181–195. CAARI Monograph Series 3. Boston, American Schools of Oriental Research. Tatton-Brown, V. 1979. Cyprus BC: 7000 Years of History. London, British Museum. Tringham, R. E. and M. W. Conkey 1998. Rethinking figurines: A critical view from archaeology of Gimbutas, the “goddess” and popular culture. In L. Goodison and C. Morris (eds) Ancient Goddesses, 22–45. London, British Museum. Whallon, R. P. 1968. Investigations of late prehistoric social organization in New York state. In L. R. Binford and S. R. Binford (eds) New Perspectives in Archaeology, 223–244. Chicago, Aldine.

20 THE PAINTING PROCESS OF WHITE PAINTED AND WHITE SLIP WARES: COMMUNITIES OF PRACTICE Louise C. Maguire

There are many instances in archaeology of not seeing the wood for the trees, but committed research and passionate determination can obtain a clearer picture from evidence available. The success of this approach has been demonstrated by Professor Eddie Peltenburg, who poses challenging questions to any type of data from archaeological fieldwork to the minutiae of small finds, whether freshly excavated or hidden in the darker recesses of museum basements. He has communicated his enthusiasm for thorough investigations to his students, and as one of them I have been persistent and inquisitive in my research into Cypriot Middle and Late Bronze Age ceramics. This paper is an attempt to search beyond the pottery types primarily used to classify and date material. It presents a systematic reconstruction of the painting processes of White Painted ware. The analysis has been carried out by reconstructing the complete brushstroke sequence on each pot. It reaches the conclusion that the major differences between the sequences found on White Painted wares from the Karpas, the eastern Mesaoria and the rest of the island are a result of the positioning of the pot during the painting process. The differences in the positioning of the pots, either upright or upside down, indicate that there was a significant shift of practice between the Middle and Late Bronze Ages. This will be demonstrated and discussed.

Communities of Practice Hardin (1977; 1983) and Van Keuren (1999, 8) conclude that potters can borrow design elements with some ease. Design structures, however, or the sequences used to paint the pot, are indicative of subconscious learning behaviours which signal that potters were closely interacting within

the same community. Van Keuren discovered that nonlocal potters could learn from local potters in spite of any existing language barriers. While non-local potters may have produced almost identical pots, it can be observed that the design execution sequence is not copied (1999, 51–52). Rather, non-local potters retained their own identity through a design sequence pertaining to their own learning framework even though their communities had merged. Many of these behaviours are adopted through copying motor skills and are learned by children (Crown 2001) as embedded behaviour within a learning framework which may last for generations. Researchers studying the behaviour of potters who produced the Classic Stallings pottery assemblages of Georgia and South Carolina discovered that the orientation of punctured decoration was significant (Sassaman and Rudophi 2001). Handedness could be reconstructed from the orientation of the punctured decoration. The geographical percentages of handedness, they suggested, reflected matrilocal post-marital residence among Stallings riverine communities. Over the course of fifteen generations, certain communities displayed a greater proportion of left-oriented vessels, above the statistical norm for populations, ancient or modern (Sassaman and Rudophi 2001). These types of behaviours, which originate from a learning framework by copying motor skills, have also been defined by anthropologists as increasing participation within “communities of practice” (Lave and Wenger 1991, 49–50; Wenger 1998, 45) where communities are the learning unit. Research has extended beyond pottery groups to textiles where the direction of cordage could be reconstructed from impressions in pottery (Minar 2001). The results show that groups with very different patterns of final twist direction are not likely to have originated from the same cultural or technological traditions, which has implications for the

166

Louise C. Maguire

correlates of material culture and embedded behaviour (Minar 2001, 398–399). In the case of the Cypriot ceramic material we are slightly more restricted in our ability to measure generational intervals since the bulk of painted whole vessels required for analysis comes from tomb groups rather than settlement contexts. Nevertheless, we do have a large number of whole vessels which we can use to reconstruct the painting processes from the Middle and Late Bronze Ages across the island. We will presume that individual potters were living and practicing in the same areas immediate to the propensity of ware type (e.g. the bulk of Red on Black pottery in Cyprus is from the Karpas and suggests that it was made in this region). We can then attempt to trace the subconscious brushstroke behaviour of these individual potters, which can be studied as collective community practice across the island. An initial survey of WP, PWS and WS wares has revealed that there were significant differences in the design execution sequences both within ware groups and across the traditional regional groups (Maguire 2009a). It was observed that the directionality of brushstrokes encircling the vessels was either left to right or right to left, or bidirectional, but the reasons for these differing practices were not fully understood. In order to fully comprehend the painting process of each vessel, all the brushstrokes visible on a pot have been reconstructed by a potter using schematic illustration. The results are presented here. In the course of reconstructing the full design composition on each vessel, it became apparent that it was possible to suggest the position that the pot had been held in to complete both the banding (by hand and not rotative kinetic energy, henceforth RKE) and the individual elements or friezes within framed bands.

Positioning Practice There are two main levels of observation in determining the position in which a pot has been held while being painted. The first is the direction of the brushstrokes. Brushstroke direction can be determined by the pool of paint at the start of the brush and the tapering of the stroke as the paint is released from the brush or tool. As the paint is expended, the paint applied often becomes weaker. As the pot is observed in the upright position, the brushstrokes proceed left to right or anti-clockwise around the vessel. A starting point for the potter is often a feature such as a handle or a spout. If the pot has been turned upside down or on its side, the brushstrokes run right to left or clockwise. Some pots have been turned on their side, and this can be determined in conjunction with the positioning of elements of decoration along with brushstroke direction. We have presumed that the dominant hand held the brush or tool (such as sticks or bird feathers)

while the other hand held the pot; future analyses may be able to determine handedness. The second factor in determining the position of the pot at painting is the layout of the design friezes around the vessel. In most cases the potters worked without markers, estimating the number of elements required to finish a frieze encircling the vessel. This would lead to either an over- or underestimation of the number of elements required to complete a frieze, anomalies which can be recorded. The potter would often reduce the size of the element to squeeze in an element at the end of the frieze, or the potter would misjudge the number of elements in a chequer pattern and leave a blank square. In both of these examples it is possible to work out the start and finish of a frieze and therefore the direction of the frieze as the potter painted. If we combine these analyses with the direction of the brushstrokes, we can begin to estimate the position of the pot at painting. Two examples from two differing practices are presented below.

White Painted Positioning Karpas: Boghaz The painting process of a WP V amphora from Boghaz is illustrated in Fig. 20.1. This particular amphora, which displays a distinctive WP chequered pattern, does have a PWS fabric, which leads to the conclusion that the origins of PWS may lie with this variant (Webb 1997, 90, no. 408). The vessel belongs to the WP tradition of the Karpas and eastern Mesaoria. Several examples of WP IV and WP V, stylistically similar to the Boghaz piece, come from Galinoporni [e.g. Cyprus Museum T 2 (1956) no. 17]; T 1 (1956) no. 5, see Crewe forthcoming, fig. 5.5); Nitovikla (T 2, no. 47, see Gjerstad et al. 1934, pl. CVIII.11); and Ayios Iakovos (T 9, see Nys and Åström 2005, 30 and, 238, pl. 13). The design structure of the vessel is divided into four friezes encircling the vessel. The frieze immediately below the rim is filled with a row of lozenges. The remaining friezes are filled with a chequered lattice. With the pot in the upright position, the brushstrokes can be observed to run from right to left around the vessel. This is true for the encircling pairs of bands, which act as a frame for the lozenges and hatched chequers, as well as for the encircling wavy line, which alternates with the bands of decoration. It is not exactly clear where the starting point of the horizontal double framing line is on each frieze. The direction of the brushstroke, right to left, does indicate that the potter held the pot upside down with the predominant hand painting in an anti-clockwise direction. To stop the paint becoming smudged by holding the vessel in the hand, the potter probably inserted his/her fist or whole hand into the body of the vessel to turn it to the next painting angle. The encircling bands were done in nearly parallel strokes

20. The painting process of White Painted and White Slip wares: Communities of practice

Fig 20.1 Schematic reconstruction of the painting process of a WP V vessel from Boghaz (University of New England, 74/13/1).

167

168

Louise C. Maguire

with stop and start points often in the same place as the potter developed a rhythm. The encircling bands were painted first, while the pot was upside down. To complete the painting process, the pot was then held either upright or with the vessel on its side while the potter finished the decoration within the framed bands. In the angle of the vessel shown in Fig. 20.1, a, it is easy to see that the starting point of the first frieze was the row of lozenges below the rim. The lozenges were created by a grid of crosses. However, as the potter progressed around the vessel, it is evident that the spacing and width of the crosses for completing the pot was overestimated. Consequently, the potter has squeezed in two smaller crosses which float in the pattern and are not connected in to the previous rhythmic crosses (see Fig 20.1, d). The remaining three bands of decoration containing the lattice chequer squares have been laid out in a grid pattern with either vertical brushstrokes (pot upright) or right to left brushstrokes (pot on its side), most of which go straight through the horizontal dividing line. The first frieze of chequers below the lozenges starts to the right of the handle and progresses around the vessel. However, a compensating anomaly exists where the chequers have been extended at the end point to meet the start point (Fig. 20.1, b). This is evident on the second frieze of chequers, but in this instance the potter has started at the vertical line below the handle and worked right to left (pot upside down or on side); the compensating anomaly of an extended sixelement chequer and two adjacent blank squares is clearly obvious (Fig. 20.1, b, second frieze). Most of the wavy bands are created with squiggle brushstrokes of single curves, but double curves have been used on the band beneath the top frieze (Fig. 20.1, a).

North, Northwest and Centre: Kythrea The painting process of a WP V Fine Line Style jug from Kythrea Tomb 1 (Åström 1972, 70, fig. XVII.3) is illustrated in Fig. 20.2. This jug belongs to a group of vessels which have a very distinctive fine line decoration using a limited repertoire of motifs in a wide selection of vessel types. It has been thought that their execution is so distinctive that they could belong to individual artists or workshops (Maguire 1991). The distribution of Fine Line Style extends from Lapithos-Kylistra, Toumba tou Skourou and Akhera in the north to Nicosia-Ayia Paraskevi, Kythrea, and Politiko in the centre and Livadhia and Klavdhia in the southeast. It even extends to Tell el-Dabca in the Nile Delta where three fragments have been discovered (Maguire 2009b, 31–32). The distinctive motifs of the Fine Line Style ware have also been found on handmade Tell el-Yahudiyeh ware globular juglets (Maguire 2009b, 24–25). The design structure of the vessel illustrated in Fig. 20.2 is similar to the Boghaz vessel previously discussed in that the potter demarcates the vessel into zones. In this case,

the vessel is divided by framed encircling bands around the neck and widest part of the vessel. This large frieze is then subdivided into vertical panels filled in with alternating themes of cross hatching, either full panels or lozenges. The significant difference between the two vessels is that the banding (not using RKE) in the case of Kythrea T 1 was probably started to the right of the handle in a left to right direction while the pot was in an upright position. The pot was continuously turned and probably held by the undecorated base or neck. The panels were also completed while the vessel was in an upright position, starting to the right of the handle and progressing around the pot. The final panel to the left of the handle has been compromised and is smaller and narrower than the previous panels. The majority of WP V Fine Line Style vessels have been painted while the vessel was upright although some have been turned upside down to complete small elements of decoration.

Sample Over 200 whole vessels were analysed using brushstroke analysis. These vessels were sampled from two main WP groups: Group 1, WP II–VI from the north, centre and southeast of the island; and Group 2, WP III–V, WP III–IV Wavy Line Style of the Karpas/Mesaoria region. Group 1 is characterised by brushstrokes produced with a single brush and banding of the vessel to demarcate zones. The banding brushstrokes run from left to right (from an observational angle), indicative of predominantly upright positioning. This group comprises an extensive variety of shapes, including bowls, jugs, tankards, flasks, composite vases and animal shapes, with widespread distribution across the island. This distribution includes the WP Cross Line Style, WP Pendent Line Style and WP V Broad Band wares of the centre and south-east of the island. The WP III–V and WP III–IV Wavy Line Style of Group 2 are very similar in general stylistic appearance to Group 1 wares. Although the vessel functions are probably similar, the overall shapes are quite different. However, Group 2 is characterised by cross-hatched motifs within friezes marked by encircling bands (single or double) but displays consistent right to left directionality (pot upside down) in banding but sometimes upright positioning for finishing decorative bands. Parallel wavy lines, more common on RoB ware, are often carried out using a multiple tool. One particularly interesting example of this ware, which offers a critical link to the behaviour of PWS and RoB, is a large WP V jug from Ayios Iakovos-Melia (Gjerstad et al. 1934, 336, pl. CIX.1; Maguire 2009a, fig. 5.4). It comprises multiple brushstrokes in a RoB fashion on the neck and base of the vessel and in ladder patterns in chequered squares, but the directionality of the banding is mixed. It is a unique example of how a piece has been fashioned using techniques and behaviour of both WP and RoB on the same vessel.

20. The painting process of White Painted and White Slip wares: Communities of practice

Fig 20.2 Schematic reconstruction of the painting process of WP V FLS vessel from Kythrea Tomb 1 (Cyprus Museum A710).

169

170

Louise C. Maguire

Results The results of the analyses are presented in Fig. 20.3. For Groups 1–3 it is clear that the upright position is the predominant one for holding the vessel with the occasional occurrence of mixed positions, which may be peculiar to specific practices in WP V Fine Line Style. In Group 2, WP wares from the Karpas and eastern Mesaoria, there are very few vessels that have been fashioned with the vessel in the upright position for the banding lines. Of those that are upright, some appear to be imports to this area from the north (e.g. WP V FLS, Ayios Iakovos T 9 no. 2; see Åström 1972, fig. XVII.9). It is not surprising that the positioning practice of the WP wares of the Karpas peninsula is different from the rest of the island, given the distinctiveness of other ware groups from this area (e.g. RoB). The shapes of the WP wares of Group 2 stand out from the rest of the island but are at home in the stylistic repertoire of the RoB wares to which they seem closely linked. Chronologically for Group 1, the WP II wares of Lapithos-Vrysi tou Barba and Vounous, for example, are predominantly painted in the upright position, and the practice continues in WP III, IV and V at Deneia (Frankel and Webb 2007). This continuity demonstrates that despite population growth and migratory fluctuation, and amidst diversity and variation in technological changes, the painting process remained the same. For Group 2, while the available evidence suggests that the WP Wavy Line Style and some WP IV may have been present in MCII, the majority of WP vessels are WP V, perhaps indicating a date of MCIII or later. In comparison to Group 1, therefore, Group 2 has a relatively shorter lifespan.

Proto White Slip and White Slip Ware Positioning In the Late Bronze Age the painting practices of Proto White Slip and White Slip wares reflect those of Group 2 rather than those of Groups 1 and 3. PWS and WS wares also differ from Group 1 in that they are fashioned using multiple brushes, commonly used in RoB ware. A brief survey of the PWS and WS wares of the Late Bronze Age indicates that P/WS vessels were predominantly painted with the vessel held upside down and the brushstroke direction right to left, with occasional turns to the upright position (Maguire 2009a). Fig. 20.4 is a schematic illustration reconstructing the painting sequence on a PWS bowl (Morris 1985, pl. 29b). The brushstrokes run from right to left around the vessel. Both the wavy line below the rim and the framed bands of the rope lattice have been completed while the pot was help upside down. The incidental motifs were probably done while the vessel was upright. Multiple brushes or devices, probably made with sticks or chicken feathers, were used for the double framed encircling bands (2 brush). The hatching within the bands was done with a

Fig. 20.3 Proportions of upright, upside down and mixed painting practices in WP wares.

20. The painting process of White Painted and White Slip wares: Communities of practice

171

Fig. 20.4 Schematic reconstruction of the painting process of a PWS vessel (after Morris 1985, 38, pl. 29).

fine 2 brush for the horizontal lines and the oblique lines were made of 8 brushfuls. This painting process is typical of the bulk of P/WS vessels. A vessel from Morphou-Toumba tou Skourou is an example of PWS painted while the pot is help upside down (Vermeule and Wolsky 1990, 269, T III.8, pl. 159C). A WS I example from Palaeopaphos-Teratsoudhia shows that the banding lines have been carried out while the vessel was upside down (T 105 Chamber B(iii); see Karageorghis 1990, 48, pl. VIII). A WS I bowl from Ayia Irini-Paleokastro has been worked in both positions, upside down and upright (T 21.34; see Pecorella 1977, 148, fig. 363b) and a WS II vessel from Maroni-Vournes has also been held upside down (Cadogan et al. 2001, fig. 14). This community practice of holding the vessel upside down to be painted continued for centuries in the case of WS wares. This aspect of longevity has not been fully addressed but has been alluded to in many studies (e.g. Morris 1985, 36; Karageorghis 2001, 7). WS is a truly remarkable stylistic phenomenon and was manufactured and distributed extensively in Cyprus, the Levant, Egypt and the Aegean for a period of some four centuries or more (Eriksson 2007, 13, table 1B). Incredibly, in this 400 year period the repertoire of handmade shapes was very conservative – bowl, jar, jug, tankard, bottle and krater – and within these types there is a large amount of standardisation (Popham 1972, 441; Morris 1985, 37). Continuing to fashion handmade vessels when devices using rotational energy were widely adopted in the Levant and Mediterranean reiterates so-called conservatism, but essentially a formula of organisational intensity is more likely to have been economically viable. The decorative elements, while progressively sloppier towards the end of White Slip, are nevertheless instantly recognisable and the constituent impermeable thick white slip very durable. The

encircling bands, rope lattice, lozenges, horned motifs and multiple dots are varied, but the overall stylistic impression is essentially static and a White Slip piece is easily recognised. The sheer volume of pottery produced suggests mass production even before production locales such as Sanidha-Moutti tou Ayiou Serkou were excavated. In light of this stylistic longevity, it seems even more pertinent to emphasise this behavioural transformation of the painting process at the end of the Middle Bronze Age.

Discussion From this survey it is clear that at the end of the Middle Bronze Age the practice of painting a WP vessel holding it in the upright position for banding was the most common practice in Cyprus. Consistently holding the vessel upside down was the preferred practice only in the Karpas and eastern Mesaoria. The PWS and WS wares, which eventually replaced the WP wares, were also painted holding the vessel upside down but notably for a period of several centuries. Much of the WP production of the Middle Bronze Age has been seen against a background of inter-regional communities. Some households practiced pottery making, and continuity was maintained most likely through matrilocal residence. Some communities were in closer interaction than others, but all are likely to have been linked to some degree through the copper industry (Frankel 1974). However, the major shift from these WP diversified products to the more standardised P/WS and other uniform LC wares has often been interpreted as a change from small unit based household production to specialised workshop output

172

Louise C. Maguire

(Steel 2004, 163). A comparative shift from a diversified output in jugs and juglets across the Mediterranean and Egypt to a standardised pan-Cypriot distribution network of Base Ring juglets may also indicate increased organisation in production to suit consumption (Maguire 2009b, 64). The break in household pottery production has been linked to the unsettled period between MCIII and LCI during which a certain amount of hybridisation occurred (Crewe 2007, 216; Knapp 2008, 114; Frankel and Webb 2007, 106) along with the abandonment of existing settlements and the appearance of new coastal settlements. What cannot be underestimated is the widespread production, distribution and adoption of P/WS wares in areas where community practice was the opposite and had been for several centuries. Did potters equipped with the knowledge of high temperatures, thick slips, multiple brush tools, bichrome techniques and the practice of painting pots upside down also move into existing settlements and the new settlements? If practices are ultimately linked to the learning frameworks of pottery-making communities, might these communities have influenced the learning framework through the creation of specialised workshops or through their own residence in new areas? The impetus for these potters to initially move to new areas may be the result of increased economic necessity through the exploitation of copper or the harvesting and distribution of indigenous produce for export; the longevity of the communities of practice may be due to sustained familial learning frameworks, albeit on a large scale or highly specialised workshops of apprentices. Elsewhere I have suggested that the PWS and WS practice of painting the pot upside down originated in the WP of the Karpas and eastern Mesaoria and the multiple brushes of the RoB ware tradition (Maguire 2009a). The embedded behaviour of the potters of both ware groups is very similar, and these potters are very likely to have originated from the same community. This is the behaviour of a highly organised community of potters already specialising in the preparation and processing of clays to achieve homogeneous fine fabrics, producing a limited range of shapes (predominantly bowls), and presumably achieving speedier completion rates with the use of multiple tools. Even though PWS and RoB wares were produced in different parts of the island, the fact that they were exported in large numbers to sites in the Levant (e.g. Tell el-cAjjul) at the earliest inception of the production of PWS further corroborates a link between these two ware groups. RoB and RoR wares are also found at sites in the west where the earliest PWS has been excavated, such as Akhera (see Karageorghis 1965, 99, fig. 26.97) and Pendayia (Karageorghis 1965, 29, fig. 9.38). We can surmise that the RoB and WP potters of the Karpas were already working within specialised workshops, producing a limited number of shapes, using consistent firing temperatures, and executing a limited design structure.

Conclusions This paper has presented a reconstruction of the painting processes of White Painted Ware based on analysing the brushstroke sequences on 200 WP vessels. The results show that there are fundamental differences in the communities of practice between the WP of the north, centre and south of the island (where the pots were painted while held upright) and the WP of the Karpas and eastern Mesaoria (where the pots were painted held upside down). These differences are significant since PWS and WS practices originate from the practices and learning framework of the Karpas and were also painted while held upside down. The implications are that a complete transformation of communities of practice occurred at the end of the Middle Bronze Age, a shift which should be added to elements of social transformation which seemed to be happening at this formative period.

Acknowledgements I am extremely grateful to Prof. Manfred Bietak, first speaker of the special research program SCIEM 2000 (The Synchronization of Civilizations in the Eastern Mediterranean in the 2nd Millennium BC) and Asst. Prof. Irmgard Hein, who have enabled this research. I would also like to thank Dr. Pavlos Flourentzos for permission to research and publish material from Kythrea, and Dr. Watters of the University of New England for providing the photography in Fig 20.1. I am indebted to Elizabeth Hird, who studied the painting process and drew the schematic illustrations in Figs 20.1, 20.2 and 20.4; and to Lubica Zelenkova for compiling the layouts in those same figures.

Abbreviations FLS LC MC PWS RoB RoR T WP WS

Fine Line Style Late Cypriot Middle Cypriot Proto White Slip Red on Black Red on Red Tomb White Painted White Slip

References Åström P. 1972. The Middle Cypriote Bronze Age. The Swedish Cyprus Expedition IV(1B). Lund, Swedish Cyprus Expedition. Cadogan, G., E. Herscher, P. Russell and S. Manning 2001. Maroni-

20. The painting process of White Painted and White Slip wares: Communities of practice Vournes: A long White Slip sequence and its chronology. In V. Karageorghis (ed.) The White Slip Ware of Late Bronze Age Cyprus. Proceedings of an International Conference organized by the Anastasios G. Leventis Foundation, Nicosia, in Honour of Malcolm Wiener. Nicosia 29th–30th October 1998, 75–88. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 2. Vienna. Crewe, L. 2007. Sophistication in simplicity: The first production of wheelmade pottery on Late Bronze Age Cyprus. Journal of Mediterranean Archaeology 20(2), 209–238. Crewe, L. forthcoming. Tomb 1 (1956) at Galinoporni and the Middle–Late Cypriot transition in the Karpas Peninsula. Forthcoming in Report of the Department of Antiquities, Cyprus. Crown, P. 2001. Learning to make pottery in the prehispanic American southwest. Journal of Anthropological Research 57, 451–469. Eriksson, K. 2007. The Creative Independence of Late Bronze Age Cyprus: An Account of the Archaeological Importance of White Slip Ware in Assessing the Relative Chronology of Late Bronze Age Cyprus and the Island’s Historical Links with the Societies of the Eastern Mediterranean during this Period. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 10. Vienna. Frankel, D. 1974. Middle Cypriot White Painted Pottery: An Analytical Study of the Decoration. Studies in Mediterranean Archaeology 42. Göteborg, Paul Åströms Förlag. Frankel, D. and J. Webb 2007. The Bronze Age Cemeteries at Deneia in Cyprus. Studies in Mediterranean Archaeology 135. Sävedalen, Paul Åströms Förlag. Gjerstad, E., J. Lindros, E. Sjöqvist, and A. Westholm 1934. The Swedish Cyprus Expedition: Finds and Results of the Excavations in Cyprus 1927–1931, Volume I. Stockholm, Swedish Cyprus Expedition. Hardin, M. 1977. Individual style in San José pottery painting: The role of deliberate choice. In J. N. Hill and J. Gunn (eds) The Individual in Prehistory: Studies of Variability in Prehistoric Technology, 109–136. New York, Academic Press. Hardin, M. 1983. The structure of Tarascan pottery painting. In D. K. Washburn (ed.) Structure and Cognition in Art, 8–24. Cambridge, Cambridge University Press. Karageorghis, V. 1965. Nouveaux documents pour l’étude du bronze récent à Chypre. Études Chypriotes III. Paris, de Boccard. Karageorghis, V. 1990. Tombs at Palaepaphos. Nicosia, A. G. Leventis Foundation. Karageorghis, V. (ed.) 2001. The White Slip Ware of Late Bronze Age Cyprus. Proceedings of an International Conference organized by the Anastasios G. Leventis Foundation, Nicosia, in Honour of Malcolm Wiener. Nicosia 29th–30th October 1998. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 2. Vienna. Knapp, A. B. 2008. Prehistoric and Protohistoric Cyprus: Identity,

173

Insularity and Connectivity. Oxford, Oxford University Press. Lave, J. and E. Wenger 1991. Situated Learning: Legitimate Peripheral Participation. Cambridge, Cambridge University Press. Maguire, L. C. 1991. The classification of Middle Bronze Age pottery: Wares, styles … workshops? In J. A. Barlow, D. Bolger and B. Kling (eds) Cypriot Ceramics: Reading the Prehistoric Record, 59–66. University Museum Monographs 74. Philadelphia, University Museum, University of Pennsylvania. Maguire, L. C. 2009a. Design execution sequences in White Painted, Proto White Slip and White Slip pottery. In I. Hein (ed.) The Formation of Cyprus in the 2nd Millennium B.C., 39–48. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 20. Vienna. Maguire, L. C. 2009b. Tell el-Dabca XXI: The Cypriot Pottery and its Circulation in the Levant. Ausgrabungen in Tell el-Dabca, Manfred Bietak (Hrsg.). Untersuchungen der Zweigstelle Kairo des Österreichischen Archäologischen Instititutes Kairo. Vienna. Minar, C. 2001. Motor skills and the learning process: The conservation of cordage final twist direction in communities of practice. Journal of Anthropological Research 57(4), 381–405. Morris, D. 1985. The Art of Ancient Cyprus. Oxford, Phaidon. Nys, K. and P. Åström 2005. Cypriote Antiquities in Public Collections in Sweden: Malmö, Lund and Göteborg. Corpus of Cypriote Antiquities 28. Studies in Mediterranean Archaeology 20(28). Sävedalen, Paul Åströms Förlag. Pecorella, P. E. 1997. Le tombe dell’età del bronzo tardo della necropoli a mare di Ayia Irini “Paleokastro.” Biblioteca di Antichità Cipriote 4. Rome, Consiglio nazionale delle ricerche, Istituto per gli studi micenei ed egeo-anatolici. Popham M. R. 1972. White Slip ware. In P. Åström, The Late Cypriote Bronze Age, 431–471. Swedish Cyprus Expedition IV(1C), 431–71. Lund, Swedish Cyprus Expedition. Sassaman, K. E. and W. Rudophi 2001. Communities of practice in the early pottery traditions of the American southeast. Journal of Anthropological Research 57(4), 407–425. Steel, L. 2004. Cyprus Before History: From the Earliest Settlers to the End of the Bronze Age. London, Duckworth. Van Keuren, S. 1999. Ceramic Design Structure and the Organization of Cibola White Ware Production in the Grasshopper Region, Arizona. Arizona State Museum Archaeological Series 191. Tucson, Arizona State Museum. Vermeule, E. D. T. and F. Z. Wolsky 1990. Toumba tou Skourou. A Bronze Age Potters’ Quarter on Morphou Bay in Cyprus. Cambridge (MA), Harvard University Press. Webb, J. 1997. Cypriote Antiquities in Australian Collections I. Corpus of Cypriote Antiquities 18. Studies in Mediterranean Archaeology 20(18). Jonsered, Paul Åströms Förlag. Wenger, E. 1998. Communities of Practice: Learning, Meaning, and Identity. Cambridge, Cambridge University Press.

21 THE CERAMIC INDUSTRY OF DENEIA: CRAFTING COMMUNITY AND PLACE IN MIDDLE BRONZE AGE CYPRUS Jennifer M. Webb

Introduction Ceramic decorative styles have been extensively used to explore prehistoric social relations, particularly in North American and Mesoamerican studies (e.g. Plog 1980; 1990; 2003; MacDonald 1990; Hegmon 1995; Graves 1998; Bartlett and McAnany 2000). This approach has rarely, however, been applied to pottery wares of the Cypriot Bronze Age where concern with seriation, typology and spatio-temporal distribution largely continues to hold sway. Decorative motifs have an idiosyncratic, artistic dimension, reproduce shared traditions, and implicitly or explicitly convey membership in social groups (Wobst 1977; Wiessner 1983; 1985; 1989; Plog 1990, 2003; Graves 1998). As such they have the potential to reveal the ‘relative identities’ (Wiessner 1990, 107) of those who produced, used and viewed ceramic vessels at varying scales of spatial and social distance and to define individual, sub-group and community membership within small scale societies. Recent studies have moved beyond the concept of the physical site to view communities as ‘networks of interaction’ involving social reproduction, subsistence production and self-identification (Kolb and Snead 1997). Yaeger and Canuto (2000, 5–6) emphasise the importance of supra-household interactions and suggest, as a simple definition, that community is the conjunction of “people, place and premise.” Shared premises developed during the course of daily interactions are mobilised in the formation of a common identity which may involve or crosscut kin relationships (see also Pauketat 2000). A sense of place is also central to personal and group membership (on community as a “chain of attachment to place,” see Marcus 2000, 239, citing Rodman 1992, 651). Yaeger and Canuto (2000) have similarly stressed the importance of historical context, which gives meaning to community interaction.

‘Practices of affiliation’ (Yaeger 2000) involve explicit representations of similarity and difference for the purpose of defining membership and creating real and imagined communities for political or other purposes (Marcus 2000, 238–239; Isbell 2000, 258). In the prehistoric record ‘practices of affiliation’ are only likely to be visible, if at all, in material residues, the most durable and abundant of which is usually pottery. In recent years concern with the communicative power of artefacts, and in particular with decorated vessels, has led to significant advances in understanding the role played by material culture in forming social identities (see Costin 2001 and Schortman and Urban 2004 for recent reviews). Coupled with an ever growing literature on style, agency and the individual (Hill and Gunn 1977; Conkey and Hastorf 1990; Hegmon 1992; Stark 1998; Dobres 2000; Dobres and Robb 2000; Knapp and van Dommelen 2008, etc.), this provides a rich theoretical and conceptual basis from which more empirically minded archaeologists have much to gain. This paper examines the ceramic industry of Deneia (Dhenia) in north-west Cyprus during the Middle Bronze Age, and in particular the dominant Red Polished ware, arguing that the vessel forms and decorative motifs which distinguish this assemblage were active symbols which operated meaningfully within their cultural context. It will examine the feedback between social purpose and design innovation in an attempt to identify the dynamics driving diversity and uniformity in ceramic production and explain the singular nature of this settlement. It is offered to Eddie Peltenburg on the occasion of his (semi)-retirement in appreciation of his immense contribution to Levantine prehistory and almost single-handed construction of a social archaeology for the Cypriot Chalcolithic.

21. The ceramic industry of Deneia: Crafting community and place in Middle Bronze Age Cyprus

Deneia in the Middle Bronze Age The cemeteries at Deneia-Kafkalla (henceforth Deneia) and Mali extend over six hectares and are the most extensive on the island. A total of 1,286 tombs have been documented, with the original number likely to have been two to three times higher (Frankel and Webb 2007; Webb 2009; Webb and Frankel 2009). Systematic looting over generations and limited formal excavation (Åström and Wright 1962; Hadjisavvas 1985; Nicolaou and Nicolaou 1988; Webb and Frankel 2001; Frankel and Webb 2007) have produced an enormous number of pottery vessels now scattered across the world. The associated settlement or settlements have not yet been found. The earliest burials date to the Early Bronze Age (c. 2400–2000 cal BC) and are relatively few in number. At least 764 chamber tombs, however, were constructed in the Middle Bronze Age (c. 2000–1650 cal BC). They are considerably larger than contemporary tombs elsewhere and appear to have contained a higher number of burials (Frankel and Webb 2007, 149–151). Estimates of the total Middle Bronze Age mortuary population at Deneia range from 9,000 to 20,000 (Frankel and Webb 2007, 152–154). Natural demographic increase is insufficient to account for the expansion of the cemeteries from the Early to the Middle Bronze Age. Rather, even the lowest figure suggested above implies a major influx of people in Middle Cypriot I. Toward the end of the period, in late Middle Cypriot III, an equally abrupt retraction of settlement led to the abandonment of partly constructed tombs and greatly reduced use of the cemeteries through to the end of the Late Bronze Age. The rapid increase in population at Deneia in the early years of the Middle Bronze Age is likely to have involved the aggregation of groups with links to a widespread network of parent villages and to have required considerable adjustment in many dimensions of social interaction and negotiation. The energy expended in tomb construction may mark attempts by newly arrived families to proclaim ties to non-ancestral burial grounds. Other aspects of material culture are also likely to have been utilised in processes of identity construction and the assertion of rights to place. For subsequent generations family tombs and their contents would have provided highly visible sites for the affirmation of claims to land and privileges as Deneia rose to prominence in a wider regional system.

The Deneia Ceramic Assemblage Middle Bronze Age pottery production at Deneia shows contrasting dimensions of association and identity, of similarity and difference. At the broadest level linkages to other villages are apparent in the general styles of Red Polished and the related but less common Black Polished ware. The closest parallels are with material from Lapithos, a

175

key contemporary site on the north coast. Individual vessels moved from one village to the other, and it has long been recognised that Deneia and Lapithos belonged to the same regional sphere of interaction. Other aspects of the pottery highlight inter-site difference and are typical of, and largely confined to, Deneia. Black Polished ware, for example, used for small, finely decorated vessels, is far more common at Deneia than anywhere else and appears in a number of forms only at this site (Brewster 2007). Typical, too, are large Red Polished vessels with elaborate relief decoration, the use of very distinctive decorative motifs and a range of unusual vessel forms. These belong to an assertive local style and may be seen, like the tombs themselves, as part of a conscious construction of community identity. There is at the same time a counter-current of variability. This is most apparent on the common black-topped Red Polished ware small bowls. Although a limited repertoire of motifs was used to decorate these vessels, of the many hundreds known no two are exactly alike. These different dimensions of ceramic production may be equated with higher level (corporate) and lower level (individual) motif and form variability, each of which is examined in more detail below.

Crafting Corporate Identity A number of specific vessel forms and clusters of stylistic attributes are unique to pottery produced and consumed at Deneia. These include globular bodied jugs with wide cutaway mouths (dubbed ‘monster jugs’ by Stewart 1962, 304; 1988, 30–31) and large thick-walled serving bowls (known as ‘Deneia basins’) probably used for extended gatherings associated with funerary ceremonial (Frankel and Webb 2007, 48–51, 154). Both vessel forms have elaborate decoration covering much of the surface, typically in bold relief with additional incisions or impressions, and involve ‘eye-catching’ differences in size, form and decoration to generically similar vessels from elsewhere (Fig. 21.1). The Deneia cemeteries have also produced a number of eccentric and composite vessels (e.g. Webb and Frankel 2001, figs 12.50, 21.6; Stewart 1992, 33, pl. II.6, 9, unprovenanced but surely from Deneia). Less ostentatious, but equally characteristic of local Red Polished, are a range of jug, juglet, askos and black-topped bowl forms found in large numbers at Deneia, occasionally at Lapithos, and only rarely elsewhere (Frankel and Webb 2007, 51–59, 154–155). Decorative forms and motifs are equally distinctive. These include so-called ‘fretwork’ ornament, ring and ‘wheel’ motifs in relief, multiple pierced lugs and the use of heavily drawn concentric circles (Fig. 21.2). More generally, Deneia Red Polished is distinguished by emphatic relief decoration, often bordering on the baroque (Stewart 1988, 105), and an abundant use of incised motifs

176

Jennifer M. Webb

Fig. 21.1 ‘Deneia basins’ and ‘monster jug’ (after Nicolaou and Nicolaou 1988, fig. 5, Tomb 48.230; Webb 2001, no. 31, Nicholson Museum 47.359; Stewart 1962, fig. LXVI.3, Tomb 1.91)

frequently covering the entire surface. Once established, these Deneia-specific type-varieties appear to have been repeatedly replicated with variability expressed through subsidiary elements such as lugs and feet. While local ceramic ideas are evident elsewhere, this level of idiosyncratic production is not matched at any other Middle Cypriot site. Deneia vessel types, in particular ‘Deneia basins’ and incised black-topped bowls, have been recovered from numerous tombs in both the Kafkalla and Mali cemeteries, attesting a common array of mortuary practices and signalling broad community affiliation rather than sub-group relationships based on individual households or lineages. These elements of stylistic behaviour may be equated with Wiessner’s concept of ‘emblemic style’ (1983) and Sackett’s ‘iconological style’ (1982; 1985; 1986; 1990) in which symbolic variation is actively used to communicate identity and send clear, purposeful, conscious messages aimed at a specific target population, in this case both within and beyond the settlement. The use of style markers as referents of corporate identity does not, however, rule out the additional employment of specific vessels to signify individual status or in competitive visual display. This is likely to have been the case in particular

with eccentric and unique vessel forms, which suggest substantial local investment in ‘promotional technologies’ (Hayden 1995; 1998). Indeed, the link between style and social context is likely to be particularly important in the Deneia assemblage. While many vessels placed in tombs had probably been in prior domestic service, some may have been specifically intended for ceremonial use (both funerary and non-funerary). Community-specific pottery designs are frequently most apparent in ritual deposits (Bartlett and McAnany 2000, 117), and the selection of vessels for mortuary use is likely to reflect activities in which both community and sub-group affiliations were clearly signalled.

Asserting Individuality: Variations on a Theme Incised black-topped Red Polished bowls are a distinctive Middle Cypriot phenomenon at Deneia (Frankel and Webb 2007, 58–59). They are far more common here than elsewhere and are found in large numbers in individual tombs. They show a strong co-variation among attributes of size, shape, colour and surface treatment but great diversity in the configuration of their decorative motifs, as

21. The ceramic industry of Deneia: Crafting community and place in Middle Bronze Age Cyprus

177

Fig. 21.2 Fragmentary vessels from Deneia, showing distinctive decorative style and motifs (after Frankel and Webb 2007, figs 4.5, 4.8, 4.11, 4.14, 4.16, 4.28, Tombs 55.P21, 782.P28–29, P60, P64, 789.P40, P242, P249, P299, P332, P349; Webb and Frankel 2001, figs 9–11, 16, 19, Tombs A.35, 37, 46, C.12, E.1).

illustrated in Fig. 21.3. Although a relatively small number of geometric and curvilinear designs were employed, few bowls display the same selection, combination and placement. This range of variation is beyond that which might be attributed to random effects, suggesting that potters set out to create unique vessels. To quote Morris in reference to similar material, the potters appear to have been “playing a game the rules of which demand that the decoration on every vessel shall be as different as possible from all others, while at the same time sharing with them as many individual elements as possible” (1985, 323).

The Deneia bowls have highly burnished surfaces, black interiors, exterior rim bands and abundant lime-filled incision and depend equally on line, colour and finish. The importance of colour (red, white and black) should not be underestimated in view of the careful manipulation of firing conditions required to produce black tops and interiors and the time involved in applying lime infill after firing. These vessels are also significantly larger than the mostly undecorated knob lug bowls found in contemporary deposits across the island (Fig. 21.4). Other aspects of this assemblage that are unique or unusual within the broader RP III tradition include incised

178

Jennifer M. Webb

Fig. 21.3 Incised black-topped Red Polished ware bowls from Deneia (after Frankel and Webb 2007, figs 4.16, 4.19, Tombs 185.P23, 789.P245, P246, P247; Åström and Wright 1962, pl. VII, 3, 6, 10, Tomb 6.50, 55, 71; Webb 2001, nos 27–28, Nicholson Museum 47.366, 64.468).

decoration on the exterior of the base, concave bases, tripod forms and occasional use of a multiple incising tool. Much of the heterogeneity visible in the Deneia small bowl assemblage operates at the lower or individual level of design variability. Creative choices made by individual potters or potting households no doubt account for much of this internal diversity and the products of such ‘analytical individuals’ (Redman 1977) may be visible with detailed work along the lines of that carried out elsewhere in relation to painted pottery (e.g. Hardin 1977; Hill 1977). Style and stylistic change, however, result from the interplay of personal creativity with constraints imposed by use context, meaning and cultural tradition (Bunzel 1929; Roe 1995, 45). Potters, like all artisans, do not make choices within an infinite field of possibilities. While stylistic behaviour at the broad level may reflect intentional goals, the individual craftsperson walks the “thin wire between creativity and tradition” (Roe 1995, 31). Design innovation in traditional societies frequently involves playing with variations on a theme. Innovation takes place within an accepted traditional framework (e.g. the placement and organisation of a design field) with the addition of new elements at localised junctures within it

(Roe 1995, 48). This gives rise to continuous, relatively lowvisibility variability involving elements such as decorative zone dimensions and design element repetitions, which might be expected to have a clinal distribution reflecting specific personal relationships and sub-groups, or a more random distribution where they reflect individual creativity and levels of skill (Voss and Young 1995). The resulting patterns of variation will be both unique and recognisably traditional, giving rise to a proliferation of subtle differences and distinctions and an array of stylistically similar artefacts in which no two are alike. This lower level of stylistic variability involves both passive elements of transmission (Sackett’s ‘isochrestic’ style, 1977; 1982; 1985; 1986; 1990) and a more active engagement between artisan and product. It is likely to have served multiple purposes, ranging from the expression of personal creativity and identity, to ensuring the recognition of individual vessels, to small-scale messaging among household or lineage groups. Pottery designs also serve consciously or unconsciously to reinforce social values, beliefs and traditions, particularly when used in daily routines such as cooking, eating and drinking (David et al. 1988; Sterner 1989).

21. The ceramic industry of Deneia: Crafting community and place in Middle Bronze Age Cyprus

179

Fig. 21.4 Box and whisker plot showing height and diameter measurements of incised black-topped Red Polished bowls from Deneia and undecorated Middle Cypriot knob lug bowls from Vounous, Karmi and Marki.

Discussion Stylistic behaviour promotes and verifies social integration and at the same time can manifest and validate social differentiation (Wobst 1977, 327). Style, that is, can be used both as an inclusionary device and as one denoting exclusion or difference (Conkey 1989; Wright 1993). Stylistic behaviour at Deneia, expressed through ceramic form and decoration, appears to have operated overtly in both dimensions. The development of a community-wide style served to integrate recently nucleated groups, establish cohesive networks, and differentiate the people of Deneia from their neighbours, strengthening initially disparate communities into a significant socio-political force. The emphasis on lower level heterogeneity suggests an attempt to define individuals and sub-groups and locate them within this specific tradition. Once in place, both forms of stylistic behaviour would have been replicated through enculturation and subject to normative evolutionary processes of stylistic drift (Nieman 1995). While comparative settlement data are lacking, it may be that these dimensions of identity formation were negotiated in particular in the mortuary domain, traditionally an arena for competitive visual display and ceramic elaboration in Cyprus (Keswani 2004; Webb and Frankel 2008). At Deneia burial is likely to have played a particularly critical role in the legitimation of rights to land as incoming groups

sought to establish themselves within the local landscape. This was achieved through the multi-generational use of well-cut tombs, creating a ‘genealogy of place’ with each successive interment (see McAnany 1995, 100). Crafting vessels denoting community, subgroup and personal identity also appears to have been a key mechanism. At the broadest scale of analysis this is visible in the strong pattern of attribute co-variation (vessel form, motif, technique) that distinguishes the Deneia assemblage within the island-wide Red Polished tradition. At the intra-site level smaller-scale social networks, innovation and individual creativity are reflected in the high level of internal diversity, particularly in the frequency, redundancy and variability of small bowl decoration. At Deneia the appearance of an overtly local ceramic signature coincides with a major phase of population aggregation. Pottery styles developed at this time demonstrate internal social linkages and inter-site interaction, as well as being a vehicle for expressing individual identity and a broad affinity to the local area. These functions were shared with an investment in inter-generational family tombs constructed with an eye to establishing a place within a rapidly expanding community and concepts of long-term association and continuity. Both suggest that a deliberate crafting of a corporate identity had been achieved at Deneia by the early years of the Middle Bronze Age. Elsewhere ceramic design styles with limited spatial

180

Jennifer M. Webb

distribution are also thought to have evolved in conjunction with higher population densities and the establishment of localised social networks (Wobst 1977; Plog 1980; 2003, 667; Braun and Plog 1982; Hantman and Plog 1982; Hantman 1984). Similarly, studies of rates of design change in ceramic assemblages suggest that the size of the birth cohort is the most significant factor affecting design variation (Graves 1983; Plog 1990, 67). Rapid population gain or loss within or across a community is likely to result in a marked increase or decrease in design variability as rates of transmission between and among generations of potters fluctuate. This impacts primarily on attributes at lower levels in the design hierarchy and may in the Deneia context explain the proliferation and diversity of motifs and configurations on incised Red Polished black-topped bowls. With the nucleation of migrating groups from different villages, a diverse array of isochrestic information would have entered the system, prompting processes of emulation and borrowing as well as differentiation and innovation.

Conclusions Much has been written about the social meaning and function of artefacts and, more specifically, about style as cultural reproduction. The nature and rate of ceramic design change from the Early to the Middle Bronze Age at Deneia does not, however, follow a simple evolutionary trajectory. On the contrary, observed changes in decorative patterns almost certainly reflect rapid population growth and related changes in communication density and social dynamics. The distribution of crafted pottery attributes and the use of selected vessel forms and motifs as vehicles of ‘symbolic style’ are likely to have been closely linked to identity negotiation at various spatial scales as well as operating as signifiers of place. This generated both homogeneity and heterogeneity, resulting in vessel forms and attribute clusters which are generically and individually unique. The stylistic complexity of Deneia Red Polished ware lasted for over a century before giving way in Middle Cypriot III to a relatively poorly differentiated array of unmarked or less overtly marked vessels. This coincided with a general hybridisation of monochrome fabrics and a simplification of decorative styles and techniques across the island, as well as a decrease in ceramic variation between sites (Frankel and Webb 2007, 106, 138). Middle Cypriot III also witnessed a downturn in the economic and political status of Deneia, followed by partial abandonment of the site prior to the transition to the Late Bronze Age. This was part of a widespread settlement restructuring which led to a retraction of population north and south of the Kyrenia range in favour of emerging settlements on the south, east and north-west coasts (Georgiou 2007, 448–454, 465–468, figs 13.1–3).

The demise of the distinctive Red Polished ware tradition at Deneia is a complex phenomenon deserving of more intensive analysis. The settlement may have reached its maximum rate of expansion and started to slow as opportunities diminished and density-dependent constraints were experienced (Graves 1983, 305). Increased stability and internal cohesion are also likely to have led over time to a decrease in the need for strategies of community affiliation and perhaps to an institutionalisation of internal hierarchies, requiring less overtly competitive signalling and resulting in a reduced investment in mortuary ritual and ceramic elaboration. Movement away from the site, which began at this time, would also have reduced the number of potters and consumers and the intensity of artisan training. While these factors, however, may have played a role at the site level, the ‘aesthetic blunting’ (Morris 1985, 323) of the red monochrome tradition is visible across much of the island at this time and was no doubt more broadly related to the growing popularity of painted fabrics and changes in the context and scale of ceramic production associated with an increasing external demand for Cypriot pottery and, ultimately, the arrival of wheelmade wares. Ceramic production and decoration are likely to have been only one strategy in a larger use of material culture in processes of identity formation and place-making at Deneia in Middle Cypriot I. Concepts of community no doubt included the ancestors, who, by virtue of their burial in the ground, provided an inter-generational link to place. Practices of community and sub-group affiliation may have placed particular emphasis on the mortuary domain, providing a link between newly arrived groups and an imagined ancestral community. Similar emblemic and individualising ceramic styles are occasionally visible elsewhere on the island in similar contexts of population expansion and social transformation (Webb and Frankel 2008). These spatial and chronological discontinuities in the production and consumption of elaborate pottery appear to reflect differences in the intensity with which ancestral links and mortuary practices were used to promote and validate competing claims to authority and place and are likely to have been linked with major episodes of population movement or aggregation.

Acknowledgements The excavations at Deneia upon which much of this paper is based were jointly directed by myself and David Frankel, and funded by the Australian Research Council. As always, the ideas developed here were explored in numerous conversations with David Frankel and owe much to his insights.

21. The ceramic industry of Deneia: Crafting community and place in Middle Bronze Age Cyprus

References Åström, P. and G. H. R. Wright 1962. Two Bronze Age tombs at Dhenia in Cyprus. Opuscula Atheniensia IV, 227–276. Bartlett, M. L. and P. A. McAnany 2000. ‘Crafting’ communities: The materialization of Formative Maya identities. In M. A. Canuto and J. Yaeger (eds) The Archaeology of Communities: A New World Perspective, 102–122. London, Routledge. Braun, D. P. and S. Plog 1982. Evolution of ‘tribal’ social networks. American Antiquity 47, 504–525. Brewster, D. 2007. Black Polished ware. In D. Frankel and J. M. Webb, The Bronze Age Cemeteries at Deneia in Cyprus, 64–67. Studies in Mediterranean Archaeology 125. Sävedalen, Paul Åströms Förlag. Bunzel, R. L. 1929. The Pueblo Potter: A Study of Creative Imagination in Primitive Art. New York, Columbia University Press (reprinted 1972 by Dover Publications, New York). Canuto, M. A. and J. Yaeger (eds) 2000. The Archaeology of Communities: A New World Perspective. London, Routledge. Conkey, M. 1989. Style and information in cultural evolution: Toward a predictive model for the Palaeolithic. In C. C. Lamberg-Karlowsky (ed.) Archaeological Thought in America, 135–152. Cambridge, Cambridge University Press. Conkey, M. and C. Hastorf (eds) 1990. The Uses of Style in Archaeology. Cambridge, Cambridge University Press. Costin, C. L. 2001. Craft production systems. In G. M. Feinman and T. D. Price (eds) Archaeology at the Millennium: A Sourcebook, 273–328. New York, Kluwer Academic/Plenum. David, N., J. Sterner and K. Guava 1988. Why pots are decorated. Current Anthropology 29, 365–389. Dobres, M.-A. 2000. Technology and Social Agency: Outlining a Practice Framework for Archaeology. Oxford, Blackwell. Dobres, M.-A. and J. E. Robb (eds) 2000. Agency in Archaeology. London, Routledge. Frankel, D. and J. M. Webb 2007. The Bronze Age Cemeteries at Deneia in Cyprus. Studies in Mediterranean Archaeology 125. Sävedalen, Paul Åströms Förlag. Georgiou, G. 2007. The Topography of Human Settlement in Cyprus in the Early and Middle Bronze Ages (in Greek). Unpublished Ph.D. thesis, University of Cyprus (Nicosia). Graves, M. W. 1983. Growth and aggregation at Canyon Creek ruin: Implications for evolutionary change in east-central Arizona. American Antiquity 48, 290–315. Graves, M. W. 1998. The history of method and theory in the study of prehistoric Puebloan pottery style in the American Southwest. Journal of Archaeological Method and Theory 5, 309–343. Hadjisavvas, S. 1985. A Late Cypriote II tomb from Dhenia. Report of the Department of Antiquities, Cyprus, 133–136. Hantman, J. L. 1984. Regional organization of the northern Mogollon. American Archaeology 4, 171–180. Hantman, J. L. and S. Plog 1982. The relationship of stylistic similarity to patterns of material exchange. In J. E. Ericson and T. K. Earle (eds) Contexts for Prehistoric Exchange, 237–263. New York, Academic Press. Hardin, M. A. 1977. Individual style in San José pottery painting: The role of deliberate choice. In J. N. Hill and J. Gunn (eds) The Individual in Prehistory: Studies of Variability in Style

181

in Prehistoric Technologies, 109–136. New York, Academic Press. Hayden, B. 1995. The emergence of prestige technologies and pottery. In W. Barnett and J. W. Hoopes (eds) The Emergence of Pottery: Technology and Innovation in Ancient Societies, 257–266. Washington DC, Smithsonian Institute Press. Hayden, B. 1998. Practical and prestige technologies: The evolution of material systems. Journal of Archaeological Method and Theory 5, 1–55. Hegmon, M. 1992. Archaeological research on style. Annual Review of Anthropology 21, 517–536. Hegmon, M. 1995. Social Dynamics of Pottery Style in the Early Puebloan Southwest. Occasional Papers of the Crow Canyon Archaeological Center 5. Cortez (CO), Crow Canyon Archaeology Center. Hill, J. N. 1977. Individual variability in ceramics and the study of prehistoric social organization. In J. N. Hill and J. Gunn (eds), The Individual in Prehistory: Studies of Variability in Style in Prehistoric Technologies, 55–108. New York, Academic Press. Hill, J. N. and J. Gunn, J. (eds) 1977. The Individual in Prehistory: Studies of Variability in Style in Prehistoric Technologies. New York, Academic Press. Isbell, W. H. 2000. What should we be studying. The ‘imagined community’ and the ‘natural community’. In M. A. Canuto and J. Yaeger (eds) The Archaeology of Communities: A New World Perspective, 243–266. London, Routledge. Keswani, P. S. 2004. Mortuary Ritual and Society in Bronze Age Cyprus. London, Equinox. Kolb, M. J. and J. E. Snead 1997. It’s a small world after all: Comparative analyses of community organization in archaeology. American Antiquity 62, 609–628. Knapp, A. B. and P. van Dommelen 2008. Past practices: Rethinking individuals and agents in prehistory. Cambridge Archaeological Journal 18, 15–34. MacDonald, W. K. 1990. Investigating style: An explanatory analysis of some Plains burials. In M. W. Conkey and C. A. Hastorf (eds) The Uses of Style in Archaeology, 52–59. Cambridge, Cambridge University Press. Marcus, J. 2000. Toward an archaeology of communities. In M. A. Canuto and J. Yaeger (eds) The Archaeology of Communities: A New World Perspective, 231–242. London, Routledge. McAnany, P. A. 1995. Living with the Ancestors: Kinship and Kingship in Ancient Maya Society. Austin, University of Texas Press. Morris, D. 1985. The Art of Ancient Cyprus. Oxford, Phaidon. Nicolaou, I. and K. Nicolaou 1988. The Dhenia-Kafkalla and Mali tombs. Report of the Department of Antiquities, Cyprus, 71–120. Nieman, F. D. 1995. Stylistic variation in evolutionary perspective: Inferences from decorative diversity and interassemblage distance in Illinois Woodland ceramic assemblages. American Antiquity 60, 7–36. Pauketat, T. R. 2000. Politicization and community in the Pre-Columbian Mississippi valley. In M. A. Canuto and J. Yaeger (eds) The Archaeology of Communities: A New World Perspective, 16–43. London, Routledge.

182

Jennifer M. Webb

Plog, S. 1980. Stylistic Variation in Prehistoric Ceramics: Design Analysis in the American Southwest. New Studies in Archaeology. Cambridge, Cambridge University Press. Plog, S. 1990. Sociopolitical implications of stylistic variation in the American southwest. In M. W. Conkey and C. A. Hastorf (eds) The Uses of Style in Archaeology, 61–72. Cambridge, Cambridge University Press. Plog, S. 2003. Exploring the ubiquitous through the unusual: Color symbolism in Pueblo Black-on-White pottery. American Antiquity 68, 666–695. Redman, C. L. 1977. The ‘analytical individual’ and prehistoric style variability. In J. N. Hill and J. Gunn (eds) The Individual in Prehistory: Studies of Variability in Style in Prehistoric Technologies, 41–53. New York, Academic Press. Rodman, M. C. 1992. Empowering place: Multilocality and multivocality. American Anthropologist 94, 640–656. Roe, P. G. 1995. Style, society, myth, and structure. In C. Carr and J. E. Nietzel (eds) Style, Society, and Person: Archaeological and Ethnological Perspectives, 27–76. New York, Plenum. Sackett, J. R. 1977. The meaning of style in archaeology: A general model. American Antiquity 42, 369–380. Sackett, J. R. 1982. Approaches to style in lithic archaeology. Journal of Anthropological Archaeology 1, 59–112. Sackett, J. R. 1985. Style and ethnicity in the Kalahari: A reply to Wiessner. American Antiquity 50, 154–159. Sackett, J. R. 1986. Style, function, and assemblage variability: A reply to Binford. American Antiquity 51, 628–634. Sackett, J. R. 1990. Style and ethnicity in archaeology: The case for isochrestism. In M. W. Conkey and C. A. Hastorf (eds) The Uses of Style in Archaeology, 32–43. Cambridge, Cambridge University Press. Schortman, E. M. and P. A. Urban 2004. Modeling the roles of craft production in ancient political economies. Journal of Archaeological Research 12, 185–226. Stark, M. T. (ed.) 1998. The Archaeology of Social Boundaries. Washington DC, Smithsonian Institute Press. Sterner, J. 1989. Who is signalling whom? Ceramic style, ethnicity and taphonomy among the Sirak Bulahay. Antiquity 63, 451–459. Stewart, J. R. 1962. The Early Cypriote Bronze Age. The Swedish Cyprus Expedition IV(1A), 205–401. Lund, The Swedish Cyprus Expedition. Stewart, J. R. 1988. Corpus of Cypriot Artefacts of the Early Bronze Age (Part I). Studies in Mediterranean Archaeology 3(1). Göteborg, Paul Åströms Förlag. Stewart, J. R. 1992. Corpus of Cypriot Artefacts of the Early Bronze Age (Part II). Studies in Mediterranean Archaeology 3(2). Jonsered, Paul Åströms Förlag. Voss, J. A. and R. L. Young 1995. Style and the self. In C. Carr and J. E. Nietzel (eds) Style, Society, and Person: Archaeological and Ethnological Perspectives, 77–99. New York, Plenum. Webb, J. M. 2001. Corpus of Cypriote Antiquities 20. Cypriote Antiquities in the Nicholson Museum at the University of Sydney. Studies in Mediterranean Archaeology 20(20). Jonsered, Paul Åströms Förlag.

Webb, J. M. 2009. Deneia: A Middle Cypriot site in its regional and historical context. In I. Hein (ed.) The Formation of Cyprus in the 2nd Millennium B.C. Studies in Regionalism during the Middle and Late Bronze Age. Proceedings of a Workshop Held at the 4th Cyprological Congress, May 2nd 2008, Nicosia, 27–37. Denkschriften der Österreichischen Akademie der Wissenschaften. Contributions to the Chronology of the Eastern Mediterranean 20. Vienna. Webb, J. M. and D. Frankel 2001. Corpus of Cypriote Antiquities 21. Eight Middle Bronze Age Tomb Groups from Dhenia in the University of New England Museum of Antiquities. Studies in Mediterranean Archaeology 20(21). Jonsered, Paul Åströms Förlag. Webb, J. M. and D. Frankel 2008. Fine ware ceramics, consumption and commensality: Mechanisms of horizontal and vertical integration in Early Bronze Age Cyprus. In R. Laffineur, L. Hitchcock and J. Crowley (eds) Dais: The Aegean Feast. Proceedings of the 12th International Aegean Conference, University of Melbourne, Centre for Classics and Archaeology, 25–29 March 2008, 287–296. Aegaeum 29. Annales d’archéologie égéenne de l’Université de Liège et UT-PASP. Liège, Université de Liège, University of Texas at Austin. Webb, J. M. and D. Frankel 2009. Exploiting a damaged and diminishing resource: Survey, sampling and society at a Bronze Age cemetery complex in Cyprus. Antiquity 83, 54–68. Wiessner, P. 1983. Style and information in Kalahari projectile points. American Antiquity 48, 253–276. Wiessner, P. 1985. Style or isochrestic variation? A reply to Sackett. American Antiquity 50, 160–166. Wiessner, P. 1989. Style and changing relations between the individual and society. In I. Hodder (ed.) The Meanings of Things: Material Culture and Symbolic Expression, 56–63. London, Unwin and Hyman. Wiessner, P. 1990. Is there a unity to style? In M. W. Conkey and C. A. Hastorf (eds) The Uses of Style in Archaeology, 105–112. Cambridge, Cambridge University Press. Wobst, M. H. 1977. Stylistic behaviour and information exchange. In C. E. Cleland (ed.) For the Director: Research Essays in Honor of James B. Griffin, 317–342. Anthropological Papers of the University of Michigan 61. Ann Arbor, Museum of Anthropology, University of Michigan. Wright, R. 1993. Technological styles: Transforming a material into a cultural object. In S. Lubar and W. D. Kingery (eds) History from Things: Essays on Material Culture, 242–269. Washington DC, Smithsonian Institute Press. Yaeger, J. 2000. The social construction of communities in the Classic Maya countryside. In M. A. Canuto and J. Yaeger (eds) The Archaeology of Communities: A New World Perspective, 123–142. London, Routledge. Yaeger, J. and M. A. Canuto 2000. Introducing an archaeology of communities. In M. A. Canuto and J. Yaeger (eds) The Archaeology of Communities: A New World Perspective, 1–15. London, Routledge.

PART 5

INSULARITY, ETHNICITY AND CULTURAL INTERACTION

22 OUTSIDE THE CORRIDOR? THE NEOLITHISATION OF CYPRUS Carole McCartney

Introduction In 1989 Alain Le Brun suggested that the concept of Neolithisation did not apply to Cyprus (1989, 95). Neolithic Cyprus has, of course, developed dramatically and no longer represents an ‘original culture’ plucked from thin air, yet few have looked beyond colonisation to consider a process of becoming Neolithic. Le Brun’s usage of the term Neolithisation was, like Cauvin’s (2000, 81, including references), a view of diffusion initiated in the ‘cradle’ of the Middle Euphrates, the central focal point of the Levantine Corridor. Newly available data from Cyprus, however, permit a different perspective on the term Neolithisation that signifies not a colonisation or diffusion event but a process of change involving shifts in subsistence and social organisation, interaction, ritual, material culture and technological innovation that distinguish the Neolithic period. By looking at this medley of characteristics, we can observe gradual changes during the Cypriot Neolithic that correspond closely with Neolithic developments seen throughout the Near East (see Figs 22.1–2 for the locations of Cyprus and mainland regions discussed in this text). Cyprus, however, is frequently defined with terms of separation where it never truly participates in the Neolithic of the Near East (Özdoğan 2005, 18; Watkins 2004, 32). This removes Cyprus from central discussions of the Near Eastern Neolithic, binding it to insularity, a ‘laboratory’ enclosed by sea walls in which Cyprus is certainly far outside the Levantine Corridor. This separation of Cyprus from the Near East is pre-determined by the model of island colonisation, providing interpretive constraints against understanding a process of Neolithisation. This hampers our ability to evaluate Cypriot evidence in terms of the polycentric, mosaic and interactive models current in Near Eastern research, the very models with which Cyprus

can be viewed as part of mainstream Neolithic diversity. Understandably, Cyprus cannot be part of a polycentric subregional entity yet remain completely isolated from the arena of supra-regional debates. The result is an uncomfortable avoidance wherein researchers accept that Cyprus needs to be mentioned but do so separately rather than as an integrated part of discussions on the wider Near East (e.g. Cauvin 2000; Simmons 2007). Whether we choose to emphasise differences or similarities, the Cypriot Neolithic can neither be fully understood nor inform on the Neolithic of the Near East if it is treated separately. Alternatively, by employing newer models we can accept the island’s evidence as a normal part of variability within the wider Near East, and look beyond the focus on complex villages of the Levantine Corridor model to include small scale Neolithic transitions as well. These small-scale settings can illustrate forager transitions under different constraints and indicate selective assimilation while showing forager interaction with developed centres. The ability of Cyprus to inform on this variable transition to a Neolithic way of life stems not from an assigned insular status, but because change occurred on the island at a different scale and within the reduced selective pressures that existed ‘outside’ the Corridor.

Looking Beyond Colonisation Because it is an island, Cyprus was assumed to provide an easy case for the discussion of the diffusion of farmers from the Levantine core (Guilaine et al. 2000, Peltenburg et al. 2001) although Peltenburg began to question the Levantine Corridor model on the basis of the new evidence. This original questioning, however, sought to expand the

186

Carole McCartney

Fig. 22.1 Map of Cyprus in the Near East.

model’s notional centre, placing Cyprus within the core area of farming invention if not exactly inside the geographical limits of the Corridor. Using the term Cypro-PPNB (retained here), Peltenburg sought to designate a distinct geographic entity within the loosely defined PPNB interaction sphere (Bar-Yosef and Belfer-Cohen 1989; Peltenburg 2004a). Though Peltenburg was clear that the Cypro-PPNB is a chronological rather than a Childean cultural entity (2004b, 72), dissension arose through the debate over Levantine Primacy, and, as Kuijt indicates (2004, 9), the extension of PPNB labelling to Cyprus, which perpetuates the use of totalising models that Peltenburg was seeking to avoid (see also Watkins 2008, 146). Instead, polycentric and inter-actionist models have been proposed to replace the centre-periphery model of diffusion from the Levantine Corridor (cf Cauvin 1989; 2000; Sherratt 2004; various

authors in Neo-Lithics 2004, no. 1; and Watkins 2008). What started as a shift from one centre into many or no centres at all also focussed on interaction and the role played by foragers in the transition to farming (Hole 2003, 33; Gebel 2004, 31). Peltenburg’s ‘integrationist’ model (2004a) represents a step towards such a polycentric sub-regional approach for Cyprus, but it views foragers as a frontier that farmer colonists needed to cross enroute to Cyprus rather than seeing local foragers as the primary actors. The notion of an open border between Cyprus and Syro-Cilicia is, however, an important shift towards increased interaction from the strict separation embodied by the model of island colonisation. Cyprus is not a distinct ‘centre’ of agricultural transformation, but Cypro-EPPNB evidence is ‘precocious’, and the lack of progenitors necessitated a diffusion process

Fig. 22.2 Map of Cyprus showing main sites discussed in text.

22. Outside the corridor? The Neolithisation of Cyprus 187

188

Carole McCartney

that included people (Peltenburg et al. 2000). Requiring the movement of farmers stems not just from Levantine Primacy but is also assumed in the model of island colonisation in which farmers could settle permanently whereas foragers could not (Cherry 1990; Broodbank and Strasser 1991; Peltenburg 2003; but see McCartney et al. in press a). In contrast, Watkins disputes the arrival of farmers to Cyprus, suggesting instead colonisation during the Epi-Palaeolithic with an agricultural transition achieved in relative isolation from the rest of the Near East (2004, 31). Though Watkins is correct to focus on foragers as the prime actors, his suggestion of a quick transition to farming on the island in isolation from mainland developments is not supported, and evidence for interaction is underestimated (see below). Regardless of the timing proposed, the colonisation model involves directionality, which Ammerman suggests (2003, 6–7) excludes frontier integration and mosaic concepts embedded in the term ‘demic diffusion’ (but see Colledge et al. 2004, 46). Peltenburg’s use of both terms simultaneously shows the difficulty of relinquishing the colonisation model despite acknowledging that the data require consideration of assimilation and interactive demic migration (2003, 93). The attempt to find Corridor farmers arriving in Cyprus required special pleading for overland diffusion, employing ‘jump-dispersals’ and ‘ferrymen’ to transport long travelled farmers to Cyprus (Peltenburg et al. 2001, 56). Such discussions overlook proximity to water as a possible factor in ‘precocious’ agricultural development although studies of early seafaring have demonstrated the need to view the sea as a highway not a barrier to interaction (Boardman 2001, 34; Finlayson 2004, 17; Broodbank 2006, 214). It is unnecessary to assume that people who cultivated stayed permanently when Cyprus very likely conformed to existing patterns of mobility (Finlayson 2004, 17–18; Thissen 2005, 29). In the sub-regional scenario suggested here, Epi-Palaeolithic and Neolithic foragers utilised their proximity to the sea to exploit the resources of nearby Cyprus. This requires an alternative perspective of colonisation wherein a colony represents not a distant land to be conquered, but a ‘home away from home’, placing the process of Neolithisation in the hands of foragers, who initiated this process from the adjacent mainland (Boardman 2001, 34; Peltenburg 2003, 94–95). Using this ‘home away from home’ perspective, we can begin to discuss an interactive, sub-regional process of Neolithisation incorporating Cyprus within a wider Near Eastern sphere.

The Neolithisation of Cyprus The Cypriot database is small but rapidly expanding, and the data are not fully digested (Simmons 2004, 16); thus narratives or sequences presented now will need to be revised as new data come to light. Sites are typically eroded

or damaged by ploughing and terracing, making evaluation of relative continuity or discontinuity problematic, while the tendency to focus on type-sites provides a significant barrier to understanding the process of Neolithisation for the island as a whole. Detailed inter-site comparison, though beyond the scope of this limited essay, is needed. Current chronological data are summarised in Table 22.1 and key features, technologies and artifact types outlined in Table 22.2. Primary concerns here are the degree of continuity and settlement permanence and how this relates to evidence for interaction and increasing materiality. At issue is whether we can demonstrate that the groups involved were originally foragers who did not simply add new technologies or changes in subsistence, but who adopted a Neolithic ‘way of life’ (cf Özdoğan 2005, 16–17).

A Forager Transition Current data argue against an isolated Neolithic development from the island’s Late Epi-Palaeolithic to the end of the Neolithic. Recent dates from Asprokremnos reduce the gap between the Late Epi-Palaeolithic and the Neolithic, but a gap nevertheless persists and is supported by a dramatic change of material culture (Tables 22.1–2; see also McCartney et al. 2008; in press b). An additional gap prior to the Cypro-EPPNB indicates discontinuities in Neolithic occupation that similarly imply recurrent rather than permanent occupation. Patterns shown by the radiocarbon calibration curve, when linked to climatic oscillation, may help to explain the punctuated archaeological record, which seems weighted towards favourable periods on Cyprus as on the mainland (McCartney et al. in press a). Notably, early Cypriot phases are shorter and tend to occur towards the end of their mainland counterparts until c. 7500 BC when temporal scales correspond more closely. Only Shillourokambos was occupied through the Cypro-MPPNB while discontinuity is shown by gaps in the sequences of both Mylouthkia and Tenta. These settlement histories are important and need to be considered with evidence changes in interaction and temporality of the built environment (Ingold 1992, 795–796; see below). We can expect a degree of mobility to have carried foragers around the island, but longer term discontinuities, if unaltered by new discoveries, imply not a ‘hiatus’ (cf Guilaine and Briois 2006, 161) but regular return visits to the Levantine mainland. In built environments storage is used to infer sedentism, where seasonal food is stocked as insurance against shortages for hunter-gatherers, eventually enabling farming communities to deal with the creation of a food surplus (Testart 1982; Bar-Yosef and Meadows 1995, 41; Watkins 2008, 155). No botanical remains have been found in any Late Epi-Palaeolithic or Early Neolithic Cypriot context. Grain storage begins in the Cypro-PPNB with ‘granaries’ at Shillourokambos in the Aceramic period and ‘silos’ in

22. Outside the corridor? The Neolithisation of Cyprus

PERIODS (Based on Aurenche et al. 2001) Period 1 Natufian (12500–9500BC) Period 2 PPNA (10500–8800 BC)

Cypriot Phase

EPPNB (9200–8300 BC)

Period 3 MPPNB (8400–7500 BC) Periods 4-5 LPPNB (7600–6400 BC)

Period 6 LN (6400–5600 BC)

189

Assemblages

Akrotiri

Cypriot Dates (all cal BC) (Weighted mean minus outliers) 10500 BC

Early Neolithic

8900–8600 BC

Asprokremnos

Cypro-EPPNB

8400–7900 BC

Mylouthkia 1a Shillourokambos A Tenta 5

Cypro-MPPNB

7900–7500 BC

Shillourokambos B

Cypro-LPPNB

7500–6400 BC

Ais Yiorkis Mylouthkia 1B Shillourokambos m/r Tenta 4–2

Khirokitian

6400–5600 BC

Cape Andreas Khirokitia Ortos Tenta 1

Aetokremnos

Table 22.1 Late Epi-Palaeolithic and Aceramic Neolithic chronology for Cyprus (Dates for sites other than Asprokremnos based on Le Brun 1981; 1991; Todd 1987; Simmons 1996; 1999; 2003; Peltenburg 2003; Guilaine and Briois 2006).

the Ceramic Neolithic (Guilaine and Briois 2001, 45; 2006, 159). Cereal chaff and combustion structures imply local crop processing (Peltenburg et al. 2001, 46; Guilaine 2003a, 7–8), yet cereal collections in wells at Mylouthkia and Shillourokambos or informal pits as at Tenta do not suggest a substantial food supply or surplus (Hansen 2005, 326; Murray 2003, 67; Guilaine et al. 2005, 1014). Ais Yiorkis provides a significant exception, yielding features containing a substantial collection of grain (Simmons 2007, 242). More commonly, evidence of storage in early Cyprus comprises non-perishable items that need not denote increased sedentism (Testart 1982, 523–524). From the start of the Neolithic large deposits of chipped and ground stone, as well as personal and symbolic items, are found at Aprokremnos, Shillourokambos, Mylouthkia and Ais Yiorkis (McCartney et al. 2008; Guilaine and Briois 2006, 167; Peltenburg 2003, 89; Simmons 2007, 242). These features show care and organisation in refuse disposal or material storage, implying extended occupations and delayed or anticipated return by complex foragers moving between well established locations (Ingold 1982, 531; Kelly 1998, 9; Kent 1999, 90; Watkins 2004, 24–27). A process of Neolithisation in Cyprus is illustrated by a gradual investment in the built environment. Ephemeral short-term occupations of early seafarers and inland foragers give way to evidence for built structures during the Neolithic. At Asprokremnos a semi-subterranean shelter

evokes PPNA structures on the mainland (Goring-Morris and Belfer-Cohen 2008, 254; McCartney et al. in press b). In the Cypro-EPPNB overlapping sequences of circular, timber-framed huts imply a palimpsest of reoccupation well documented at Shillourokambos as well as at Tenta Period 5, recalling huts prevalent in mainland sites from the EpiPalaeolithic (Todd 1987, 28, fig. 37; Guilaine and Briois 2006, 162, figs. 2, 4; Goring-Morris and Belfer-Cohen 2008, 242). Construction in temporary materials across more than a millennium suggests repeated engagements with the same places, denoting a lengthy process of Neolithisation rather than a ‘colonisation’ event (Peltenburg 2003, 98, table 11.6; Arroyo-Kalin 2004, 75). The wells at Mylouthkia, which are widely dispersed, and at Shillourokambos show investment that similarly indicates anticipated return (Guilaine 2003a, 4, fig.1; Peltenburg 2003, 87–89, fig. 11.3). By the Cypro-MPPNB at Shillourokambos the appearance of rough stone walls and floor plaster begin to suggest greater permanence (Guilaine and Briois 2001, 44). Later, during the moyen phase (c. 7500 BC), circular stone built houses indicate considerably greater investment as do the stone platforms built at Ais Yiorkis (Guilaine 2003a, 10; Simmons 2007, 242). The first true village of densely packed structures surrounded by a purpose built wall appears during the Cypro-LPPNB at Tenta (Period 4) and at its much enlarged successor, Khirokitia (Todd 1987, 29, fig. 19; Le

Carole McCartney

190 Sites/Dates (cal BC) 10500 BC Aetokremnos

Site type Rock shelter/open air Hearths, pits

8900–8600 BC Asprokremnos

Open air

Funerary

Semi-sub structure Postholes/channels Hearths, pits Lithic dumps

8400–7900 BC Mylouthkia 1A Shillourokambos A Tenta 5

7900–7500 BC Shillourokambos B (Akanthou?)

Open air/settlement Posthole structures Channel palisades Wells, pits Combustion features Lithic dumps

Single inhumations

Open air/settlement Posthole structures Rough stone walls Plastered surfaces Lithic dumps

7500–6400 BC (Akanthou?) Ais Yiorkis Mylouthkia 1B Shillourokambos moyen/recent Tenta 4, Phase 2

Village

6400 – 5600 BC Cape Andreas Khirokitia Ortos Tenta 1

Village

Perimeter wall Circular stone houses Internal pillars, cells Mudbrick Painted walls-floors Circular stone platforms Wells, pits Lithic dumps

Village expansion Perimeter wall Village entrance Circular stone houses Internal pillars, cells Mudbrick Pained walls-floors

Collective pit/well House burials

House burials Grave goods

Material Lunates Microlithic Thumbnail scrapers Picrolite pendants dentalium Shaft-straighteners Lozenge points Clay human head Polished stone axes Unidirectional blade Calcareous bowls Ochre, Picrolite/shell beads Macehead Cathead Baton Clay human head Obsidian Byblos/ovular points Bidirectional bladepreferential/pressure retouch Calcareous bowls Deep lug vessels Obsidian Byblos/Ovular points Bidirectional bladepreferential/ pressure retouch Glossed crescents Picrolite ornament Obsidian (decreased) Marble stone rings Vaiselle blanche Bidirectional blade Tanged blades Large glossed crescents Truncated glossed Simple diabase bowls Picrolite ‘thimbles’ Picrolite ornaments Stone/clay figurines Bone fish hooks Obsidian (decreased) Carnelian beads Butterfly-beads Engraved pebbles Sheep Unidirectional blade Tanged blades Lg. glossed crescents Backed blades Stone/clay figurines Picrolite ornaments Stone/bone fish hooks Decorated vessels

Economy Bird, molluscs, reptiles, fish

Pig, bird, crab

Pig, sheep, goat, cattle, fallow deer, molluscs Domestic cereals

Pig, sheep, goat, cattle fallow deer, molluscs Domestic cereals Pig, sheep, goat, cattle, fallow deer, molluscs Domestic cereals

Pig, sheep, goat, fallow deer, molluscs Domestic cereals

Table 22.2 Outline of Cypriot Late Epi-Palaeolithic and Aceramic Neolithic evidence (based on references in text).

22. Outside the corridor? The Neolithisation of Cyprus Brun 2001, 111, fig. 1). The construction of likely special function structures, such as the S14 complex at Tenta and the D-shaped structure at Khirokitia, occurred synchronously (Todd 1987, 80–81, fig. 33; Le Brun 2003). Generating this increasing temporal durability undoubtedly shaped social practice and helped to foster a new materialism on the island by the Cypro-LPPNB (De Marrais 2004, 15; Helms 2004, 123; Watkins 2004, 26–27). Before that time there is little evidence for permanent settlement on Cyprus, supporting the notion that mobility is preferred by foragers when conditions permit (Benz 2004, 27; Akkermans 1998, 288; Finlayson 2004, 18). In Cyprus these conditions continued to prevail; thus the question is not why early sites show a shorter-term temporality but why, given an absence of evidence for population or territorial pressures, permanence should exist at all and increase significantly after c. 7500 BC. The idea that the use of circular architecture relates to island isolation can be challenged. A general PPNA heritage linked to the use of circular structures by complex foragers does seem likely (Peltenburg 2004b, 75–79, fig. 7.2; Asouti 2006, 118, including references), but the perpetuation of this choice is shared by other regions of the Near East, particularly in arid areas. Shaqarat Mazyad, for example, shows a substantial settlement of circular stone constructions dating to the MPPNB, indicating that this preference was not unique to the island of Cyprus (Hermansen 2004, 37). Similarly, the re-dating of Aswad brings the use of semi-subterranean round houses closer to the CyproPPNB (Cauvin 2000, 39; Stordeur 2004). This choice of architectural form reflects a local negotiation towards a Neolithic way of life that is part of the wider variation in architectural design seen across the Near East (Hermansen 2004, 37; Watkins 2008, 153–160). Burial evidence is correspondingly absent from Late Epi-Palaeolithic and Early Neolithic contexts in Cyprus and continues to be meagre at the start of Cypro-PPNB. A single infant burial in well 116 at Mylouthkia (Period1A) suggests the presence of a woman while an elaborate burial of an adult interred with a cat from Shillourokambos phase A recalls Natufian examples on the mainland (Fox et al. 2003, 43; Vinge et al. 2004). This paucity of human remains also accords with the suggestion of continued residential mobility between Cyprus and the mainland by foragers since hunter-gatherers place less emphasis on formal burials (cf Helms 2004, 119–120). A larger local population is implied from c. 7500 BC in Mylouthkia well 133, which contained three adults and two youths; the collective pit burial at Shillourokambos; 18 burials at Tenta; and the large sample of 248 interments at Khirokitia (Crubézy et al. 2003; Le Mort 2003, 318; Moyer 2005, 2). It has been suggested that sedentism promotes increases in population size and increased infant mortality, which is confirmed at Tenta and especially at Khirokitia (Testart 1982, 524–525; Kelly 1998,

191

20; Belfer-Cohen and Bar-Yosef 2000, 23; Le Mort 2003, 318, fig. 3; Moyer 2005, 2; but see Ingold 1982, 531 for an opposing view). Burial below floors during the CyproLPPNB shows greater continuity of physical proximity with the dead, which, like the use of both human and animal figurines and wall paintings, indicates an altered cosmology in the context of sedentism as documented throughout the Near East (Todd 1987, fig. 39; Le Brun 1994, fig. 104; 2001, 111, 115; Guilaine 2003b, 329–332, fig. 1; Helms 2004, 119–124; Peltenburg 2004b, 75–77; Moyer 2005, 5–6; South and Todd 2005, 312; McCartney et al. 2008, fig. 3). The widely disbursed, small endemic foods available to the earliest broad spectrum foragers required the high mobility shown by small sites on the coast and inland (Simmons 1999, 164–191; Ammerman et al. 2006; McCartney et al. 2006). Marine resources continued to be exploited in coastal locations, and seafaring coastal foragers appear to have co-existed with those becoming farmers (Ammerman in press). Inland foragers at Asprokremnos relied on imported pigs, birds and freshwater crab, confirming the pattern beginning with Aetokremnos and corroborated by two pig populations at Shillourokambos that attest to early stocking of the island with pig for hunting purposes (Guilaine and Briois 2006, 161–162; Horwitz et al. 2004, 36–40; Rowley-Conwy 2004, 87; Reese 1999, 164–167; McCartney et al. 2008). Asprokremnos is not a farmer colonisation as it lacks the ‘package’ of cereals and fauna seen in the Cypro-PPNB (Peltenburg 2003, 94, Guilaine and Briois 2006, 171–172); it also challenges the idea of a rapid transformation to domestication (Watkins 2004, 31). Instead there was a gradual investment in managed animals corroborated by the use of animal pens in the Cypro-EPPNB and the later disappearance of arrowheads on Cyprus (Croft 1991, Horwitz et al., 2004, 37–38; McCartney 2004, 114; Guilaine and Briois 2006, 162, 171; Wasse 2007, 56–57). Wasse (2007, 60–61) suggests that lower population and environmental pressures failed to precipitate the shift to domesticated fauna shown by the mainland LPPNB, but his suggestion that Cyprus remained in a ‘Cypro-EPPNB time warp’ over-emphasises faunal evidence and recalls the notion of the ‘para-neolithic’, a term used elsewhere (e.g. Morrison and Watkins 1974; Kent 1992, 54). Notably, percentages of hunted fauna vary between sites and represent less than 50 per cent of Neolithic assemblages. The balance is made up by domestic fauna, and the rest, if ‘managed’, may not be strictly comparable to free-ranging wild populations on the mainland. Several assemblages in the Khirokitian and Ceramic Neolithic show increased sheep husbandry when population and climatic pressures were felt (Wasse 2007, 63), yet the inhabitants of Khirokitia chose to stay in their village though land was readily available elsewhere (Peltenburg 2004b, 84–85). As the only settlement approaching a ‘megasite’ in Cyprus, Khirokitia clearly persisted for reasons that were not exclusively economic,

192

Carole McCartney

and it would be mistaken to consider the whole of the Cypriot Neolithic only in terms of food production.

Interaction and Materialism Aetokremnos lacks an extensive material culture agreeing with early seafarer/forager mobility and little desire for material elaboration seen elsewhere during the Younger Dryas (Belfer-Cohen and Bar-Yosef 2000, 21). By the Neolithic, material elaboration and interaction is readily apparent. Interaction attributed to foragers who circulated projectiles, lithic technologies, and obsidian can be shown, and it likely promoted the Neolithisation of Cyprus as elsewhere (Binder 2005, 129; Robb and Farr 2005, 25, 33; Asouti 2006, 102; Baird 2008, 15–18; McCartney in press). On the mainland, forager interaction gave way to the formalised networks of the PPNB interaction sphere when increased sedentism impeded forager interaction (Bar-Yosef and Belfer-Cohen 1989, 64; Gamble 2004, 87; Watkins 2008, 139, 146). In Cyprus evidence for non-permanence suggests that forager networks were maintained until the end of the Cypro-MPPNB, after which the character of contact altered, precipitating the isolation argument embedded in island colonisation (Peltenburg 2003, 100; Watkins 2004, 29, 32). However, the people of Cyprus could not have existed in a vacuum, and the inconsistency of the isolation argument in relation to interaction models results in an uncomfortable ‘island paradox’ (e.g. Akkermans 1998, 288; Knapp 2007, 47). Prior to the Cypro-LPPNB, low population and impermanent sedentism suggest that ‘extended’ interaction was not limited only to island sites (Watkins 2008, 151–152). Shared symbols like decorated shaft-straighteners and lozenge points from Asprokremnos (McCartney et al. 2008) strongly suggest forager interaction with the Middle Euphrates, but such interaction does not imply the ‘cultural unity’ suggested for south-east Anatolia and north-east Syria by Stordeur (2004) since contact does not ‘cause’ assimilation (Kent 1992, 55). This interaction extended also to the southern Levant as a baked clay human figurine head, recalling a Gilgal example, and a number of shunera-like points suggest (McCartney in press). Notably, this range of Early Neolithic parallels corresponds with the arc of cereal progenitors, suggesting that these too reached Cyprus through forager contacts that incorporated interaction with both the northern and southern Levant, while conforming with the wide geography needed to obtain the early Cypriot animal ‘package’ (Colledge et al. 2004, 43–46, fig. 5; Horwitz et al. 2004, 43–44; Wasse 2007). The macehead from Mylouthkia and the cathead and baton from Shillourokambos show additional symbolic entrainment (Peltenburg et al. 2001, 58, fig. 2; 2003, 96–97, fig. 11.5; Guilaine 2003b, 329–330, fig. 1; Watkins 2008, 155). We cannot assume that the meanings

interpreted for such objects were the same in Cyprus as on the mainland (Watkins 2008, 157–160, 165), and it should be noted that stone objects in particular could have been altered, yet these portable symbols were selected with care and provide evidence of an interactive environment between Cyprus and the mainland. The appearance of obsidian in the Cypro-PPNB shows that the foragers who used Cyprus had joined the PPNB interaction sphere (McCartney in press). Though absent earlier, obsidian was traded intensively during the Cypro-E/ MPPNB, yet this commodity appears to have lost its intrinsic exotic value after c. 7500 BC (Le Brun 2001, 113; Renfrew 2004, 29). However, obsidian did not disappear and has been recovered at nearly all Cypro-LPPNB and Khirokitian sites, as well as sporadically in the following Ceramic Neolithic period (Le Brun 2001, 113; Simmons 1996, 35; 2007, 242; McCartney and Todd 2005; McCartney 2007, 75, 83). The decline of obsidian during the Cypro-LPPNB is invariably assumed to indicate isolation and lack of participation in the PPNB supra-regional network, despite a synchronous shift from down the line exchange to site specific distribution in the LPPNB on the mainland where some sites received more than others (Asouti 2006, 102–104; Watkins 2008, 165). In Cyprus obsidian decline implies an end to forager networks, which were replaced in the Khirokitian by a new formalised ‘global-network’ that very likely arose in response to sedentary village life (Watkins 2008, 152). This entails not so much a disruption of old networks but the building of new ones out of prior relations (Arroyo-Kalin 2004, 73–77). Engraved pebbles found at both Khirokitia and Ortos have mainland parallels at Ras Shamra, Byblos, Munhata and Sha’ar Hagolan, showing a new focus of interaction oriented toward the southern Levant (McCartney 2007, 77, including references). The introduction of marble stone rings at Tenta and large sheep during the Khirokitian, as well as imported carnelian butterfly beads and a small amount of obsidian, demonstrate the results of such interaction (Dikaios 1953, 303–306; South and Todd 2005, 305–308; Wasse 2007, 63). These, like the vaiselle blanche introduced at Ais Yiorkis and pottery technology in the succeeding Ceramic Neolithic, would not have appeared if Cyprus was isolated after the Cypro-MPPNB (Clarke 2007; Simmons 2007, 242). Due to the assumption of isolation, no one has ever looked for potential Cypriot exports, which could include green stone (picrolite), chert, ochre, polished stone, salt, fish, or even piglets (e.g. Robb and Farr 2005, 28–31; Ammerman in press). Resources like chert do not have to be unique; they may be desired simply for their quality and availability. Mainland areas like the Orontes-Bouqaia basin, where flint outcrops are confined to limited areas, provide examples of a potential mainland demand (HaïdarBoustani et al. 2007, 4). Evidence for chert reduction disproportionate to that of subsistence at Asprokremnos, as well as a dearth of finished tools relative to production

22. Outside the corridor? The Neolithisation of Cyprus debris at sites like Shillourokambos, suggests the possibility that Neolithic foragers were as interested in exploiting the mineral resources of Cyprus as becoming farmers (Guilaine and Briois 2006, 167; McCartney in press). Importantly, Cypro-E/MPPNB production of bidirectional preferential blades, not featured at Asprokremnos, had to be learned by knappers on the mainland. This development does not represent a simple functional parallelism (cf Kuijt 2004, 8) since the main purpose of such an investment in blade production was the manufacture of arrowheads, which were not abundant in the Cypro-PPNB (Abbès 2003; Guilaine and Briois 2006, 168–170). The same can be said for the appearance of pressure retouch. Other new craft technologies, particularly after c. 7500 BC, show an altered perception of objects and material engagement and signify the gradual accumulation of new experiences in a sedentary landscape (Renfrew 2004, 23–24; Gamble 2004, 85–86). The expansion of picrolite carving in the Cypro-LPPNB and the evolution of carved diabase vessels in the Khirokitian represent examples of the process of materialisation in which new relations were rendered in material form, providing elements of a growing regional entity through the increase of shared practice (Arroyo-Kalin 2004, 75; De Marrais 2004, 11–12). However, such shifts were not uniform across Cyprus, and we need to ask why the later Cypro-PPNB assemblages at Shillourokambos, Ais Yiorkis and Akanthou contained a trio of materials, namely obsidian, cattle and picrolite ‘thimbles’, not seen in assemblages elsewhere (Şevketoğlu 2001, 104, fig. 1; Simmons 2007, 242; Guilaine and Briois 2006, 172). Like the range of interactions noted above, such variability illustrates the narrow focus in the idea of a planned colonisation and may explain why no single ‘founder population’ has ever been discovered (Kent 1992, 60).

Conclusion To understand processes of Neolithisation in Cyprus we need to look at the entire island and the entire Neolithic period rather than simply focussing on its beginnings, its subsistence patterns or its presumed island isolation. Geographically distinct traits indicate regionalisation, but they developed from a forager transition that included a wider sub-regional area of the western Levant where forager groups persisted outside of the Levantine Corridor (Bar-Yosef 2001, fig. 4). Lower pressures of population growth permitted the retention hunting longer than on the mainland, yet evidence of increased sedentism, materialism and eventual village development shows the same forager to sedentary village transition (with all of the correlated ritual developments and shifts in social organisation that this implies) that defines the process of Neolithisation throughout the Near East.

193

Abbreviations Aetokremnos Ais Yiorkis Asprokremnos EPPNB Khirokitia LN LPPNB MPPNB Mylouthkia Ortos PPNA PPNB Shillourokambos Tenta

Akrotiri-Aetokremnos Kritou Marottou-Ais Yiorkis Ayia Varvara-Asprokremnos Early Pre-Pottery Neolithic B Khirokitia-Vounoi Late Neolithic Late Pre-Pottery Neolithic B Middle Pre-Pottery Neolithic B Kissonerga-Mylouthkia Kholetria-Ortos Pre-Pottery Neolithic A Pre-Pottery Neolithic B Parekklisha-Shillourokambos Kalavasos-Tenta

Acknowledgements The opinions discussed in this paper have benefited from discussions with numerous scholars, including Albert Ammerman, Paul Croft, Bill Finlayson, Nigel GoringMorris, Trevor Watkins and in particular Eddie Peltenburg, who has never ceased to provide inspiration. Though I thank all for their eloquence and insights, responsibility for the opinions expressed here is mine alone.

References Abbès, F. 2003. Les outillages néolithiques en Syrie du Nord: méthode de débitage et gestion laminaire durant le PPNB. British Archaeological Reports International Series 1150. Oxford, Archaeopress. Ammerman, A. 2003. Looking back. In A. Ammerman and P. Biaji (eds) The Widening Harvest, 3–23. Boston, Archaeological Institute of America. Ammerman, A., P. Flourentzos, C. McCartney, J. Noller and D. Sorabji 2006. Two new early sites on Cyprus. Report of the Department of Antiquities, Cyprus, 1–22. Ammerman, A. in press. The paradox of early voyaging in the Mediterranean and the slowness of the Neolithic transition between Cyprus and Italy. In G. Vavouranakis (ed.) Seascapes in Aegean Prehistory. Monographs of the Danish Institute at Athens. Athens, Danish Institute at Athens. Akkermans, P. 1998. Hunter-gatherer continuity: The transition from the Epi-palaeolithic to the Neolithic in Syria. In M. Fortin and O. Aurenche (eds) Espace naturel, espace habité en Syrie du Nord (10e–2e mill av. J.C.), 281–293. Lyon, Maison de l’Orient. Arroyo-Kalin, M. 2004. An ongoing outcome, a surrounding world: Materiality, agency and history. In E. De Marrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World. Cambridge, McDonald Institute for Archaeological Research.

194

Carole McCartney

Asouti, E. 2006. Beyond the Pre-Pottery Neolithic B interaction sphere. Journal of World Prehistory 20, 87–126. Aurenche, O., P. Galet, C. Régagnon and J. Évin 2001. ProtoNeolithic and Neolithic cultures in the Middle-East – The birth of agriculture, livestock raising and ceramics: A calibrated 14C chronology 12,500–5,500 cal BC. Radiocarbon 43(3), 1191–1202. Baird, D. 2008. Pınarbaşi; From Epipapaeolithic camp-site to sedentarising village in central Anatolia. In M. Özdoğan and N. Başgelen (eds) Turkey in the Neolithic: New Excavations, New Discoveries, 1–27. Istanbul, Arkeoloji ve Sanat Yayinlari. Bar-Yosef, O. 2001. The world around Cyprus: From Epipalaeolithic foragers to the collapse of the PPNB civilization. In S. Swiny (ed.) The Earliest Prehistory of Cyprus, 129–164. CAARI Monograph Series 2. Boston, American Schools of Oriental Research. Bar-Yosef O. and A. Belfer-Cohen 1989. The Levantine ‘PPNB’ interaction sphere. In I. Hershkovitz (ed.) People and Culture in Change: Proceedings of the Second Symposium on Upper Palaeolithic, Mesolithic and Neolithic Populations of Europe and the Mediterranean Basin, 59–72. British Archaeological Reports International Series 508. Oxford, British Archaeological Reports. Bar-Yosef, O. and R. Meadows 1995. The origins of agriculture in the Near East. In T. D. Price and A. Gebauer (eds) Last Hunters, First Farmers: New Perspectives on the Prehistoric Transition to Agriculture, 34–94. Santa Fe, School of American Research Press. Belfer-Cohen, A. and O. Bar-Yosef 2000. Early sedentism in the Near East: A bumpy ride to village life. In I. Kuijt (ed.) Life in Neolithic Farming Communities: Social Organization, Identity, and Differentiation. New York, Kluwer Academic/Plenum. Benz, M. 2004. The emic view: Social questions of the Neolithisation of the Near East. Neo-Lithics 1(04), 27–28. Binder, D. 2005. Autour de l’obsidienne: Un aspect des processus d’interaction entre agro-pasteurs et chasseurs-cueilleurs en Anatolie centrale. In J. Guilaine (ed.) Populations, Économies et Environments, du Néolithique et de l’Age du Bronze, 117–134. Séminaires du Collège de France. Paris, Errance. Boardman, J. 2001. Aspects of colonization. Bulletin of the American Schools of Oriental Research 322, 33–42. Broodbank, C. 2006. The origins and early development of Mediterranean maritime activity. Journal of Mediterranean Archaeology 19(2), 199–230. Broodbank, C. and T. Strasser 1991. Migrant farmers and the Neolithic colonization of Crete. Antiquity 65, 233–245. Cauvin, J. 1989. La Néolithisation au Levant et sa première diffusion. In O. Aurenche and J. Cauvin (eds) Néolithisations, 3–36. British Archaeological Reports International Series 516. Oxford, British Archaeological Reports. Cauvin, J. 2000. The Birth of the Gods and the Origins of Agriculture. Translated by T. Watkins. Cambridge, Cambridge University Press. Cherry, J. 1990. The first colonization of the Mediterranean islands: A review of recent research. Journal of Mediterranean Archaeology 3, 145–221. Clarke, J. 2007. On the Margins of Southwest Asia: Cyprus during the 6th to 4th Millennia BC. Oxford, Oxbow. Colledge, S., J. Conolly and S. Shennan 2004. Archaeobotanical

evidence for the spread of farming in the eastern Mediterranean. Current Anthropology 45, 35–47. Croft, P. 1991. Man and beast in Chalcolithic Cyprus. Bulletin of the American Schools of Oriental Research 282/283, 63–79. Crubézy, E., J. D. Vinge, J. Guilaine, T. Giraud, P. Gérard and F. Briois 2003. Aux origines des sépultures colletives: La structure 23 de Shillourokambos (Chypre, 7500 B.C.). In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie 17–19 Mai 2001, 295–311. Bulletin de Correspondence Hellénique Supplément 43. Athens, École Française d’Athènes. De Marrais, E. 2004. The materialization of culture. In E. De Marrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World, 11–22. Cambridge, McDonald Institute for Archaeological Research. Dikaios P. 1953. Khirokitia. Oxford, Oxford University Press. Finlayson, B. 2004. Island colonization, insularity or mainstream? In E. Peltenburg and A. Wasse (eds) Neolithic Revolution, 15–22. Levant Supplementary Series 1. Oxford, Oxbow. Fox, C., D. Lunt and M. Watt 2003. Human Remains. In E. Peltenburg (ed.) The Colonisation and Settlement of Cyprus: Investigations at Kisonerga-Mylouthkia, 1976–1996, 59–71. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Gamble, C. 2004. Materiality and symbolic force: A Palaeolithic view of sedentism. In E. De Marrais, C. Gosden and C. Renfrew (eds.) Rethinking Materiality: The Engagement of Mind with the Material World, 85–95. Cambridge, McDonald Institute for Archaeological Research. Gebel, H. G. 2004. There was no centre: The polycentric evolution of the Near Eastern Neolithic. Neo-Lithics 1(04), 28–32. Goring-Morris, N. and A. Belfer-Cohen 2008. A roof over one’s head: Developments in Near Eastern residential architecture across the Epipalaeolithic-Neolithic transition. In J. P. Briquet and O. Bar-Yosef (eds) The Neolithic Demographic Transition and its Consequences, 239–286. New York, Springer Science + Business Media. Guilaine, J. 2003a. Parekklisha-Shillourokambos: périodisation et aménagements domestiques. In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie 17–19 Mai 2001, 3–14. Bulletin de Correspondence Hellénique Supplément 43. Athens, École Française d’Athènes. Guilaine, J. 2003b. Objets, “symboliques” et parures de Parekklisha-Shillourokambos. In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie 17–19 Mai 2001, 329– 340. Bulletin de Correspondence Hellénique Supplément 43. Athens, École Française d’Athènes. Guilaine, J., F. Briois, J. D. Vinge and I. Carrère 2000. Découverte d’un Néolithique précéramique ancient chypriote (fin 9e, début 8e millénaires cal. B.C.), apparenté as PPNB ancient/moyen du Levant nord. Earth and Planetary Sciences 330, 75–82. Guilaine, J. and F. Briois 2001. Parekklisha Shillourokambos: An Early Neolithic site in Cyprus. In S. Swiny (ed.) The Earliest

22. Outside the corridor? The Neolithisation of Cyprus Prehistory of Cyprus, 37–53. CAARI Monograph Series 2. Boston, American Schools of Oriental Research. Guilaine, J., F. Briois, J. D. Vinge, T. Perrin, Y. Béliez, I. Carrère, P. Gérard and M. Remicourt 2005. Shillourokambos (Parekklisha): L’Établissement Néolithique Pré-Céramique. Bulletin de Correspondence Hellénique 128/129, 1006–1021. Guilaine, J. and F. Briois 2006. Shillourokambos and the neolithization of Cyprus: Some reflections. Eurasian Prehistory 4(1), 159–175. Haïdar-Boustani, M., J. J. Ibáñez, M. Al-Maqdissi, A. Armendáriz, J. G. Urquijo and L. Teira 2007. New data on the Epipalaeolithic and Neolithic of the Homs Gap: Three campaigns of archaeological survey (2004–2006). Neo-Lithics 1(07), 3–9. Hansen, J. 2005. Flora. In I. Todd (ed.) Excavations at KalavasosTenta, vol. 2, 323–341. Studies in Mediterranean Archaeology 71(7). Sävedalen, Paul Åströms Förlag. Helms, M. 2004. Tangible materiality and cosmological others in the development of sedentism. In E. De Marrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World, 117–127. Cambridge, McDonald Institute for Archaeological Research. Hermansen, B. D. 2004. Supra-regional concepts from a local perspective. Neo-Lithics 1(04), 34–37. Hole, F. 2003. Centers in the Neolithic? Neo-Lithics 2(03), 33–35. Horwitz, L., E. Tchernov and H. Hongo 2004. The domestic status of the early Neolithic fauna of Cyprus: A view from the mainland. In E. Peltenburg and A. Wasse (eds) Neolithic Revolution, 35–49. Levant Supplementary Series 1. Oxford, Oxbow. Ingold, T. 1982. Reply to Testart. Current Anthropology 23(5), 531–532. Ingold, T. 1992. Foraging for data, camping with theories: Hunter-gatherers and nomadic pastoralists in archaeology and anthropology. Antiquity 66, 790–803. Kelly, R. 1998. Foraging and sedentism. In T. R. Rocek and O. Bar-Yosef (eds) Seasonality and Sedentism, 9–23. Peabody Museum Bulletin 6. Cambridge (MA), Peabody Museum of Archaeology and Ethnology, Harvard University. Kent, S. 1992. The current forager controversy: Real versus ideal views of hunter-gatherers. Man 27, 45–70. Kent, S. 1999. The archaeological visibility of storage: Delineating storage from trash areas. American Antiquity 64(1), 79–94. Knapp, A. B. 2007. Insularity and island identity in the prehistoric Mediterranean. In S. Antoniadou and A. Pace (eds) Mediterranean Crossroads, 37–62. Athens, Pierides Foundation. Kuijt, I. 2004. Cyprus as a regional Neolithic entity: Do researchers need to revisit the concept of the Levantine PPNB interaction sphere? Neo-Lithics 1(04), 8–9. Le Brun, A. 1989. La Néolithisation de Chypre. In O. Aurenche and J. Cauvin (eds) Néolithisations, 95–107. British Archaeological Reports International Series 516. Oxford, British Archaeological Reports. Le Brun, A. 1994. Fouilles récentes à Khirokitia. Paris, Éditions Recherche sur les Civilisations. Le Brun, A. 2001. At the other end of the sequence: The Cypriot Aceramic Neolithic as seen from Khirokitia. In S. Swiny (ed.) The Earliest Prehistory of Cyprus, 109–118. CAARI

195

Monograph Series 2. Boston, American Schools of Oriental Research. Le Brun, A. 2003. Idéologie et symbols à Khirokitia: la “fermeture” d’un bâtiment et sa mise en scène. In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie 17–19 Mai 2001, 431–439. Bulletin de Correspondence Hellénique Supplément 43. Athens, École Française d’Athènes. Le Mort, F. 2003. Les restes humains de Khirokitia: particularités et interprétations. In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre: Actes du Colloque International Organisé par le Département des Antiquités de Chypre et l’École Française d’Athènes, Nicosie 17–19 Mai 2001, 313– 325. Bulletin de Correspondence Hellénique Supplément 43. Athens, École Française d’Athènes. McCartney, C. 2004. Cypriot Neolithic chipped stone industries and the progress of regionalization. In E. Peltenburg and A. Wasse (eds) Neolithic Revolution, 103–122. Levant Supplementary Series 1. Oxford, Oxbow. McCartney, C. 2007. Lithics. In J. Clarke (with contributions by C. McCartney and A. Wasse) On the Margins of Southwest Asia: Cyprus during the 6th to 4th Millennia BC, 72–90. Oxford, Oxbow. McCartney, C. in press. The lithic assemblage of Ayia VarvaraAsprokremnos: A new perspective on the Early Neolithic of Cyprus. In E. Hildebrand (ed.) Studies in Technology, Environment, Production and Society. Berlin, ex Oriente. McCartney, C. and I. A. Todd 2005. Chipped stone. In I. A. Todd (ed.) Vasilikos Valley Project 7: Excavations at KalavasosTenta, vol. 2, 177–264. Studies in Mediterranean Archaeology 71(7). Sävedalen, Paul Åströms Förlag. McCartney, C., S. Manning, D. Sewell and S. T. Stewart 2006. Elaborating Early Neolithic Cyprus (EENC), Report of the Department of Antiquities, Cyprus, 39–62. McCartney, C., S. W. Manning, S. Rosendhal and S. T. Stewart 2008. Elaborating Early Neolithic Cyprus (EENC). Preliminary report on the 2007 field season: Excavations and regional field survey at Ayia Varvara-Asprokremnos. Report of the Department of Antiquities, Cyprus, 67–86. McCartney, C., S. W. Manning, D. Sewell and S. T. Stewart in press a. Reconsidering Early Holocene Cyprus within the eastern Mediterranean landscape. In B. Finlayson and G. Warren (eds) Landscapes in Transition. Levant Supplementary Series. Oxford, Oxbow. McCartney, C., P. Croft, S. W. Manning and S. Rosendhal in press b. Preliminary report on the 2008 excavations at Ayia Varvara-Asprokremnos and EENC Field Survey, Report of the Department of Antiquities, Cyprus. Morrison, I. and Watkins, T. 1974. Kataliondas-Kourvellos: A survey of an Aceramic Neolithic site and its environs in Cyprus. Palestine Exploration Quarterly 106, 67–75. Moyer, J. 2005. Human Remains. In I. A. Todd (ed.) Vasilikos Valley Project 7: Excavations at Kalavasos-Tenta, vol. 2, 177– 264. Studies in Mediterranean Archaeology 71(7). Sävedalen, Paul Åströms Förlag. Murray, M. A. 2003. The plant remains. In E. Peltenburg (ed.) The Colonisation and Settlement of Cyprus: Investigations

196

Carole McCartney

at Kisonerga-Mylouthkia, 1976–1996, 59–71. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Özdoğan, M. 2005. Westward expansion of the Neolithic way of life: What we know and what we do not know. In C. Lichter (ed.) How did Farming Reach Europe?, 13–27. Byzas 2. Istanbul, Ege Yayinlari. Peltenburg, E., (ed.) 2003. The Colonisation and Settlement of Cyprus: Investigations at Kissonerga-Mylouthkia, 1976–1996, 59–71. Studies in Mediterranean Archaeology 70(4). Sävedalen, Paul Åströms Förlag. Peltenburg, E. 2004a. Cyprus: A regional component of the Levantine PPN. Neo-Lithics 1(04), 3–7. Peltenburg, E. 2004b. Social space in early sedentary communities of southwest Asia and Cyprus. In E. Peltenburg and A. Wasse (eds) Neolithic Revolution, 71–89. Levant Supplementary Series 1. Oxford, Oxbow. Peltenburg, E., S. Colledge, P. Croft, A. Jackson, C. McCartney and M. A. Murray 2000. Agro-pastoralist colonization of Cyprus in the 10th millennium BP: Initial assessments. Antiquity 74, 844–853. Peltenburg, E., S. Colledge, P. Croft, A. Jackson, C. McCartney and M. A. Murray 2001. Neolithic dispersals from the Levantine corridor: A Mediterranean perspective. Levant 33, 35–64. Reese, D. 1999. The faunal assemblages. In A. Simmons (ed.) Faunal Extinction in an Island Society: Pygmy Hippopotamus Hunters of Cyprus, 153–192. New York, Kluwer Academic/ Plenum. Renfrew, C. 2004. Towards a theory of material engagement. In E. De Marrais, C. Gosden and C. Renfrew (eds) Rethinking Materiality: The Engagement of Mind with the Material World, 23–31. Cambridge, McDonald Institute for Archaeological Research. Robb, J. and H. Farr 2005. Substances in motion: Neolithic Mediterranean ‘trade’. In E. Blake and A. B. Knapp (eds) The Archaeology of Mediterranean Prehistory, 24–45. Malden, Blackwell. Rowley-Conwy, P. 2004. How the west was lost. Current Anthropology 45, 83–113. Şevketoğlu, M. 2001. Akanthou-Arkosyko (Tatlisu-Çiftlikdüzü): The Anatolian connections in the 9th millennium BC. In W. H. Waldren and J. A. Ensenyat (eds) World Islands in Prehistory: International Insular Investigations: V Deia International Conference of Prehistory, 98–106. British Archaeological Reports International Series 1095. Oxford, Archaeopress.

Sherratt, A. 2004. Fractal farmers: Patterns of Neolithic origin and dispersal. In J. Cherry, C. Scarre and S. Shennan (eds) Explaining Social Change: Studies in Honour of Colin Renfrew, 53–63. Cambridge, McDonald Institute Monographs. Simmons, A. 1996. Preliminary report on the multidisciplinary investigations at Neolithic Kholetria-Ortos, Paphos District. Report of the Department of Antiquities, Cyprus, 29–44. Simmons, A. 1999. Faunal Extinction in an Island Society: Pygmy Hippopotamus Hunters of Cyprus. New York, Kluwer Academic/Plenum. Simmons, A. 2004. The Mediterranean PPNB Interaction Sphere? Neo-Lithics 1(04), 16–17. Simmons, A. 2007. The Neolithic Revolution in the Near East. Tucson, University of Arizona Press. South, A. and I. A. Todd 2005. Other ground stone artifacts and human figurines. In I. A. Todd (ed.) Vasilikos Valley Project 7: Excavations at Kalavasos-Tenta, vol. 2, 287–312. Studies in Mediterranean Archaeology 71(7). Sävedalen, Paul Åströms Förlag. Stordeur, D. 2004. New insights and concepts: Two themes of the Neolithic in Syria and south-east Anatolia, Neo-Lithics 1(04), 49–51. Testart, A. 1982. The significance of food storage among huntergatherers: Residence patterns, population densities, and social inequalities. Current Anthropology 23(5), 523–530. Thissen, L. 2005. Coming to grips with the Aegean in Prehistory: An outline of the temporal framework, 10,000–5,500 cal BC, In C. Lichter (ed.) How did Farming Reach Europe?, 29–40. Byzas 2. Istanbul, Ege Yayinlari. Todd, I. A. 1987. Vasilikos Valley Project 6: Excavations at Kalavasos-Tenta, vol. 1. Studies in Mediterranean Archaeology 71(6). Göteborg, Paul Åströms Förlag. Vinge, J. D., J. Guilaine, K. Dubue, L. Haye and P. Gérard 2004. Early taming of the cat in Cyprus. Science 304, 259. Wasse, A. 2007. Climate, economy and change: Cyprus and the Levant during the Late Pleistocene–Mid Holocene. In J. Clarke (with contributions by C. McCartney and A. Wasse) On the Margins of Southwest Asia: Cyprus during the 6th to 4th millennia BC, 43–63. Oxford, Oxbow. Watkins, T. 2004. Putting the colonization of Cyprus into context. In E. Peltenburg and A. Wasse (eds) Neolithic Revolution, 23–34. Levant Supplementary Series 1. Oxford, Oxbow. Watkins, T. 2008. Supra-regional networks in the Neolithic of southwest Asia. Journal of World Prehistory 21, 139–171.

23 CONTEXTUALISING NEOLITHIC CYPRUS: PRELIMINARY INVESTIGATIONS INTO CONNECTIONS BETWEEN CYPRUS AND THE NEAR EAST IN THE LATER NEOLITHIC Joanne Clarke

Introduction It is widely held that connections between Cyprus and the Levant ceased around 7000 BC when the expansive world of the Pre-Pottery Neolithic B (PPNB) contracted and dislocated into a number of smaller ‘worlds’ during the later Neolithic period (all dates referred to in this paper are cal BC). In northern Syria and upper Mesopotamia, a dynamic and continuously changing, east/west sphere of influence can be documented through the post-PPNB/Halaf/Ubaid traditions of the 7th to 5th millennia. Likewise, in the south a largely disconnected, east/west sphere of influence existed, beginning with the Yarmoukian around 6200 BC and continuing into the Wadi Rabah and Chalcolithic traditions of the 6th and 5th millennia. Between these two geographically delineated spheres of influence is the less well documented region of the central Levant, comprising the Syrian and Lebanese coastal zone, the Ansariyah, Zawiyah, Lebanon and Anti-Lebanon mountains, the Homs Gap and the Beqa’a valley (Fig. 23.1). Archaeological research at the site of Arjoune (Parr 2003) and more recently at Tell Ezou and Tell al-Marj (Haïdar-Boustani et al. 2003–2004; 2005–2006) and Shir (Bartl, Haider and Nieuwenhuyse 2006; Bartl, Hijazi and Haider 2006) is beginning to provide new insights into this poorly understood region. In addition, two sites further to the north, Tell Kurdu (Özbal and Gerritsen 2004) and Tell ‘Ain el-Kerkh (Tsuneki et al. 1998; 1999; 2000; Iwasaki and Tsuneki 2003) are supplementing and widening our understanding of the western Levant more generally. My own examination of the disappearance of evidence for connections between Cyprus and the mainland in the aftermath of the PPNB resulted in a paper in 2003 in which I suggested that Cypriot social relations were structured by a general community-wide anomalousness, which in turn was manifested as a particular material culture repertoire

that was both internally homogeneous and lacking in any clear evidence of links with contemporary mainland cultures (Clarke 2003, 212). My view then was that “the absence of any evidence for external influences [in the Ceramic Neolithic period] appears to have been socially prescribed rather than any real absence of contact” (Clarke 2003, 215). As circumstantial evidence for contact between Cyprus and the mainland subsequent to the PPNB continues to accrue (Clarke 2007; see also Erikh-Rose 2004; McCartney 2007) it seems appropriate to return to the question of how Cyprus negotiated its relationships with the outside world and whether human agency, in the form of social identity, was the only factor shaping interaction.

Connections: What Connections? In order for connections between Cyprus and the Levant to have continued beyond the PPNB expansionist period, there needed to have been a perceived benefit to one or both participants. Current evidence suggests that influence was unidirectional (in that Cyprus shows evidence of connections with the mainland, whereas to date the mainland shows no evidence of connections with Cyprus). One way in which connections or contact with the mainland may have taken place, but would have resulted in very low archaeological visibility, is in the form of occasional sojourns to the mainland by Cypriots looking to replenish resources that might have, from time to time, become scarce on the island. This idea is based loosely on a model described by Horwitz, Tchernov and Hongo (2004) for the restocking of animals following initial colonisation during the PPNB. Horwitz et al. (2004, 40) argue that following the first introduction of domestic animals to

198

Joanne Clarke

Fig. 23.1 Map of Syria and Cyprus showing 7th to 5th millennia BC sites mentioned in the text.

Cyprus in the PPNB, there may have been frequent trips to the mainland to restock depleted faunal populations in order to maintain founder herds. At present there is no evidence for replenishment of depleted stocks of plants and animals after the PPNB, but it is unlikely that there would be if only existing species were involved. There is a degree of logic in this line of thinking if environmental conditions on the island are taken into consideration. Although population levels on Cyprus would have remained low throughout prehistory, which would inevitably have lessened pressure on resources, Cyprus lacks standing bodies of water and deep aquifers and therefore would have been completely reliant upon adequate, regular rainfall (while there is ample evidence for well digging in the 8th millennium BC at Parekklisha-Shillourokambos and at Kissonerga-Mylouthkia, no later wells are known on the island). As happens today, if there are extended periods of drought, then vegetation dies off quickly and even hardier trees, such as the olive, can suffer. Herds in these conditions would have been difficult to maintain as food would have become scarce. Broodbank states that “longerrange sea faring seems to correspond to hard, rather than halcyon times, and suggests that the perceived risks, both direct and indirect….long outweighed the benefits under all but the harshest circumstances” (2006, 218). Thus, when environmental conditions were harsh (and recent climatic research indicates that the mid-Holocene was not as benign

as originally believed with at least two, and possibly three short arid intervals; see Brooks 2006), Cypriot communities may have assumed a greater degree of risk, including long-range sea crossings, in order to replenish resources. Therefore, although not reliant on the mainland, it is not hard to imagine instances when Cypriot communities made forays beyond their own borders. Bolger (forthcoming) has considered the evidence for interaction between Cyprus and south-western Anatolia in the 3rd millennium BC, prior to the generally accepted appearance of external influences during the Philia Culture phase. Bolger argues that the limited nature of this interaction had less to do with isolation than with the island’s reception of foreign cultural elements. This argument is persuasive for the Late Chalcolithic period when it was possible that ‘pioneering’ forays to Cyprus from the southern Anatolian coast could have begun many hundreds of years prior to full enculturation during the Philia Culture phase. However, the context for connections during the 5th millennium BC is clearly different from that of the Late Chalcolithic period as there is no evidence of increased interaction either prior to or following the Ceramic Neolithic period. Connections during the 5th millennium must be considered in light of the 1,000 year hiatus in the archaeological record that directly preceded it and the subsequent Early and Middle Chalcolithic periods, during which evidence for interaction with the mainland reaches an all time low. On this basis, the

23. Contextualising Neolithic Cyprus: Preliminary investigations inevitable conclusion should be that links with the mainland ceased with the demise of the Khirokitian, not to be renewed until the 3rd millennium BC. Yet the corollary of this is that the emergence of pottery on the island in the 5th millennium BC happened completely independently of the mainland, a conclusion that this writer is not yet reconciled with.

Direct Evidence for Continued Contact following the Pre-Pottery Neolithic B Probably the most convincing evidence in support of connections between Cyprus and the mainland following the end of the PPNB is the presence, in quantity, of incised flat and conical stones on both the mainland and in Cyprus (Fig. 23.2). Variations of these are found widely throughout the Levant, but the most common are those with incised vertical and horizontal lines, cross hatching, cruciform designs, chevrons and radiating lines (see Erikh-Rose 2004 for a comprehensive review). In Cyprus incised stones occur at three sites: Kholetria-Ortos (Fox 1988, fig. 6.2; Simmons and Corona 1993, fig. 7; Simmons 1996, fig. 2); KhirokitiaVounoi (Dikaios 1953, pl. 138; Cluzan 1984, figs 97–99; Le Brun 1984); and as a surface find at the 5th millennium site of Kalavasos-Kokkinoyia (Clarke 2004, 59, fig. 3.3; McCartney 2007, 78, fig. 6c). Incised stones from Cyprus share with the mainland many of the design preferences listed above. Erikh-Rose argues that they were an early coding system, as no two are alike in design and execution (2004, 159). They are often found in graves and so have a personal association, and they are partly contemporary with and go out of use not long after the introduction of true seals (2004, 159). Whether or not these stones were part of an early system of coding or, as has been previously suggested, models of bread loaves (Dikaios 1953, 291), stamps (Cauvin 1972, 91), animal brands (Bar-Yosef and Belfer-Cohen 1989, 38), gaming stones, lunar calendars (Marshack 1972) or art representations (Stekelis 1972, 44; Gopher and Orelle 1996, 267), their widespread distribution on the mainland and the relatively large numbers found in Cyprus suggest that Cypriot communities had access to the information encoded in the markings on the stones and thus were part of a wider cultural network that included both Cyprus and the mainland, and in which these stones played a part. Their broad contemporaneity, the shared medium of stone and their similar form across the region as a whole tend to support a case for the circulation of shared cultural knowledge. In addition to incised stones are forty carnelian beads and a ‘butterfly bead’ from Khirokitia-Vounoi (Dikaios 1953, 306), white marble stone ring fragments from KalavasosTenta (South and Todd 2005, 305–308; Stanley Price 1977, 82–84; Todd 1986, 21; 2005) and the considerably depleted, but still present, obsidian blades found at many Khirokitian sites (McCartney and Gratuze 2003), including those very

199

occasionally found at 5th millennium BC sites, such as the three blades from Kalavasos-Kokkinoyia and a nugget from Ayios Epiktitos-Vrysi (Peltenburg 1982, 101). Following the Khirokitian, there is a break in the Cypriot archaeological record of approximately 1,000 years. During this time there is no discernable human activity or occupation on the island. This anomaly may be the result of poor archaeological visibility or inadequate dating, but the gap doggedly persists irrespective of new research and excavation. Yet the incised stone and the three well-stratified obsidian blades from Kalavasos-Kokkinoyia suggest more than just chance finds or the possibility of the collection of ‘heirlooms’ and might instead suggest that connections between the mainland and Cyprus were either maintained through the long hiatus in the Cypriot archaeological record (for which there is no evidence) or were re-established for a short time in the 5th millennium BC.

Indirect Evidence for Continued Contact with the Mainland after the 7th Millennium BC By the 5th millennium BC cultural divergences between Cyprus and the mainland were clearly apparent. While intensification and specialisation were by this time features

Fig. 23.2 Incised cross-hatched stone from Kalavasos-Kokkinoyia.

200

Joanne Clarke

of both the northern and southern Levant, reflected in the way in which built space was organised, in the manufacture and use of pottery and stone tools, and in economic practices, on Cyprus, where pressures driving cultural change were largely absent, an early Neolithic way of life persisted. In light of this it is important to examine how pottery might have emerged on Cyprus. Sherds of low-fired and unfired clay were discovered, along with an unbaked clay figurine, in the 7th millennium BC levels at Khirokitia-Vounoi (Dikaios 1953) and, more recently, Peltenburg (personal communication) has confirmed that low-fired pottery occurs in pre-Khirokitian contexts at Kissonerga-Mylouthkia. Yet the full-scale use of pottery did not take hold until the first half of the 5th millennium BC when it appeared in two ware types, a form of Dark Faced Burnished Ware (Cypro-DFBW, also known as Monochrome Burnished Ware) and a Coarse Ware, in the basal levels at Philia-Drakos A (Watkins 1972; 1973), at Klepini-Troulli (Clarke 2007, 99) and in Concentration A at Dhali-Agridhi (Lehavy 1989). Unusually, both Coarse Ware and Cypro-DFBW are manufactured from the same clay, found only in the lower reaches of the Troodos mountains (Clarke, Goren and Boness forthcoming). Although neither of these ware types appears to have had direct antecedents on the mainland, both exhibit broad technological similarities with the much earlier mainland Dark Face Burnished Ware while the restrictive use of clay suggests that potters aimed to create a specific vessel finish. Cypriot Dark Faced Burnished Ware has been distinguished from later wares on the basis of fabric, shape and surface finish (Fig. 23.3). The most characteristic form is a dark, highly burnished surface with a very hard, dark, fine mineral tempered fabric. This can sometimes be found with embossed decoration, such as small ‘nipple-like’ protrusions and wavy and straight cords. At the other extreme it can have a thin red, yellow or buff-coloured burnished slip and a soft crumbly fabric with heavy vegetable temper akin to Coarse Ware. Between these two extremes is a whole range of variations. The dominant shape in Cypriot Dark Faced Burnished Ware are small bowls with rim diameters rarely exceeding 20 cm. All are thin walled and usually highly burnished on both inner and outer surfaces and are almost always hemispherical or, less commonly, deep straight sided small buckets. Bowls usually have flat, omphalos and disc bases and flat or rounded rims. There is also a range of small flasks. Within a very short space of time Cypriot Dark Faced Burnished Ware was replaced by decorated pottery made from paler firing fabrics and painted either in plain red monochrome or as red painted designs on a white ground, a design technique that also replaces monochrome wares and dark faced burnished wares on the mainland. By the last quarter of the 5th millennium, when combed decoration appears in the south-central part of the island, pottery technology in Cyprus is recognisably Cypriot and remains so

until the middle of the 3rd millennium BC when influences from southern Anatolia become evident in the local pottery assemblages in the west of the island (Bolger 2007). Until recently, antecedents for the first pottery in Cyprus have been sought in southern Anatolia, Cilicia, northern Syria and the southern Levant; however, no direct parallels or imported sherds from these regions have been recorded in the Cypriot assemblages (Clarke, Goren and Boness forthcoming). However, the published pottery assemblage from Trench VI at Arjoune in central Syria produced a series of radiocarbon determinations placing it in the second quarter of the 5th millennium B.C. (Gowlett 2003, 27–29) and therefore contemporary with the beginning of the Late Neolithic period in Cyprus. Stylistically, this pottery provides the closest parallels yet with the earliest pottery of Cyprus, that is at a point before it becomes distinctly Cypriot by diversifying into painted and combed varieties and consequently when mainland influences should still be documentable. Arjoune is characterised by a series of pits, which in the 6th millennium BC appeared to be used as habitation or utilisation spaces due to the presence of floors, some of which were furnished with ovens (Marfoe, Parr and Phillips 2003, 18). In the 5th millennium BC, cultural material was restricted to two large shallow pits that utilised natural depressions in the bedrock. Although there was little evidence to indicate how these features were used, at least one possible floor was identified which had a number of heavy stone artefacts lying on, or just above it (Marfoe, Parr and Phillips 2003, 20). The 5th millennium BC pottery from Arjoune is dominated by burnished wares of a form and type that had long ago largely disappeared from most sites in the northern and southern Levant. Mathias notes that “virtually all the small and medium sized vessels, both open bowls and jars, are burnished. This is long after the very ancient tradition of Dark Faced Burnished Ware had ceased to be current in other parts of the northern Levant” (Campbell, Mathias and Phillips 2003, 37). Shapes are simple and confined to hemispherical bowls, holemouth jars and jars with short, wide, straight or flaring necks (Fig. 23.4), all found in Cypriot Dark Faced Burnished Ware repertory. Yet within this restricted repertoire are features, such as omphalos and flanged bases, flattened rims and applied decoration; these are not common in contemporary Levantine assemblages but are characteristic of the earliest Cypriot pottery. Surface finishes, too, are very similar to Cypriot Dark Faced Burnished Ware. These are either slipped and burnished or simply burnished, and are found in a range of shades from black and dark reddish brown (Munsell colours 5YR 2.5/2–7.5YR 2.5/1 to 5YR 4/3). Out of a total of 3,943 sherds from Arjoune VI, only 17 had incised, impressed or punctate design, and only six were painted in the style of pottery traditions found further to the north (Campbell, Matthias and Phillips 2003, 36).

23. Contextualising Neolithic Cyprus: Preliminary investigations

201

Fig. 23.3 Pottery from Trench VI at Arjoune, Syria (drawings by Graham Reed).

The evidence from Arjoune suggests that in central Syria, at least, there existed an independent local tradition that bore little resemblance to pottery traditions further to the north and south. While most regions had adopted painted decoration by the 5th millennium BC, at Arjoune a form of Dark Faced Burnished Ware continued to be made. The similarity between, and contemporaneity with, the earliest Cypriot pottery may be an indication that whatever form contact with the mainland took following the PPNB, it shifted south from the northern to the central Levant. The line of reasoning maintained here is not that central Syria provided Cypriots with the skills and technology for pottery making, but rather that connections between Cyprus and central Syria, in the form of either resource replenishing expeditions or as exchange in perishables, exposed the Cypriot population to pottery making and enabled them to experiment locally and independently with the technology. Unlike the experimentation of the 7th millennium BC, this time it was a success, leading to the wide adoption of pottery across the island in a very short space of time. A slightly different example, but relevant to this line of reasoning, is the late adoption of rectilinear architecture. On the mainland, rectilinear architecture fully replaced continuous-walled curvilinear architecture by the beginning

of the 5th millennium BC and was widely used from the 7th millennium BC onwards. Like pottery, rectilinear architecture was known of on Cyprus in the 7th millennium, but it was not widely adopted until the Bronze Age. Recently, a rectilinear building was uncovered at Khirokitia-Vounoi (Odile Le Brun, personal communication), but to date it is the only example pre-dating the 3rd millennium BC that this writer is aware of. The concept of rectilinear architecture, therefore, was occasionally tried out by Cypriot communities, but it was not widely adopted until many thousands of years after it had fully replaced curvilinear architecture on the mainland. In this instance it is likely that Cyprus had not reached a level of intensification whereby the pressures driving cultural change on the island necessitated the adoption of rectangular architecture, but this does not negate the fact that the knowledge to create rectangular architecture existed. In sum, although Cypriot communities had knowledge of ways of living and technologies emerging on the mainland in the 7th millennium BC, rarely were these adopted in preference to traditional ways of doing things. The reasons for this can no longer be argued to be due entirely to the construction of social boundaries as a way of marking identity although Cypriot communities certainly maintained a degree of anomalousness throughout prehistory (see

202

Joanne Clarke

Fig. 23.4 Cypriot Dark Faced Burnished Ware.

Clarke 2003 for an account of this line of reasoning). Instead they should be imagined as a panoply of environmental, social and demographic factors, which together shaped Cypriot cultural development, including the way in which the island negotiated relationships with the outside world. Two contributing factors, in addition to the maintenance of social boundaries, were most likely de-intensification (Wasse 2007, 62) and failed experimentation (Redding 2005, 42), which considered in combination contribute to a fuller understanding of Cypriot cultural development following

the PPNB and explain why evidence for connections with the mainland remained so limited.

De-intensification In a comprehensive review of the environment and economies of Cyprus, Wasse has characterised the development of Cyprus in bio-evolutionary terms as follows:

23. Contextualising Neolithic Cyprus: Preliminary investigations Within a few hundred years of the introduction of agriculture to Cyprus, during the latter half of the 11th millennium cal. BP, the focus on pigs … seems to have died out… This is not unexpected, as evidence from across the Levant points to a gradual replacement of earlier economic strategies by the new agricultural economies throughout the 10th millennium cal. BP (Moore, Hillman and Legge 2000, 422; Peters et al. 2005, 116; Wasse 2000 and 2002). However, it is at this point, with agriculture seemingly well on the way to becoming established on the island and offering almost unlimited scope for further intensification and specialisation, that the archaeological record of Cyprus begins to diverge from that of the Levantine mainland… The process of economic intensification, which had been gathering pace on the mainland since the beginnings of sedentism in the 15th millennium BP (and which, it could be argued, has continued unabated to this day), appears to have stopped prematurely in Cyprus at the stage it was at when agriculture reached the island during the second half of the 11th millennium cal. BP. This left Cypriot economies frozen in the far more typically transitional phase of the mainland Early PPNB, in which agriculture was practiced as a minor adjunct to hunting and gathering (Wasse 2007, 60).

Wasse (2007, 62) describes this phenomena as “de-intensification” whereby the evolutionary pressures driving cultural change on the mainland were absent on Cyprus. Population levels remained low and so did stress on land and resources. Thus, simply put, Cypriot communities did not need to innovate or indeed to adopt foreign innovations in order to survive; the island provided adequately for the needs of its population.

Experimentation A second way of considering this might be from the perspective of a model recently advanced by Redding (2005) to explain why the faunal record at the transition from hunting and gathering to farming and herding on the mainland displays very little variation. Redding’s theory argues that shifts in human subsistence behaviour were characterised by failed experiments. As archaeologists we observe the successful experiments which go on to become fully-fledged subsistence strategies; the unsuccessful experiments, however, due to their small scale and transient quality, do not survive in the archaeological record (Redding 2005, 42). Redding’s model can be extrapolated to the material culture record. Assuming Cypriot communities continued to be in contact with the mainland throughout the later Neolithic period, from time to time they clearly experimented with mainland economic practices, technologies and materials; yet in discarding these practices they become largely invisible in the archaeological record. The reason for this is because the scale and intensity of use are not detectable by the rather ‘broad brush’ approach of archaeological fieldwork.

203

Cypriot experimentation with pottery and with rectangular architecture in the 7th millennium BC exemplifies Redding’s model. Khirokitia-Vounoi has been continuously excavated since 1975, yet only in 2006 was a rectangular building uncovered. Had excavations ceased earlier, rectangular architecture would remain unknown in the Neolithic of Cyprus. Likewise, pottery dating to the 7th millennium BC has only recently been discovered at Kissonerga-Mylouthkia, and prior to this only a couple of fragments were known from Khirokitia-Vounoi. Low fired, unburnished pottery is typically very friable and does not survive well in the archaeological record. It could therefore be surmised that much more of it existed than the few documented examples but the likelihood of recovery is limited. Thus, not only did Cypriot communities not need to innovate, but when technologies were from time to time experimented with, they were not adopted until it became expedient to do so. The emergence of pottery in the middle of the 5th millennium BC probably happened because the skills needed to make good quality ground stone bowls had been lost during the long hiatus that followed the end of the Khirokitian. Ground stone bowls occur occasionally in Late Neolithic assemblages, but quality and artisanship are greatly diminished.

Geographical Considerations Arjoune is situated at the western end of the Homs Gap, an agriculturally significant zone allowing access between the Syrian hinterland and the coastal plain. To the north of the Homs Gap are the Ansariyah mountains, which separate the narrow coastal plain from the Syrian plateau and the badia to the east. To the south are the Lebanon and Anti-Lebanon mountains. In antiquity, as now, there were two principal routes for movement, a north/south route along the coast between the major maritime cities of Byblos, Tyre, Sidon and Ugarit, and a second, north/south route inland from Damascus, skirting east round the Anti-Lebanon mountains, along the Orontes River, through the Ghab depression and up to the Amuq Plain. It is no coincidence that many major archaeological sites are located along these north-south routes. The view of the present writer is that Arjoune, its larger and more famous neighbour, Tell Nebi Mend, and the coastal site of Tabbat al-Hammam (Hole 1959) are key to elucidating the nature of Late Neolithic sea faring expeditions to and from Cyprus because of their location in the Homs Gap, which is on approximately the same latitude as the southern coast of Cyprus and therefore offers the greatest expanse of Cypriot coastline for sea-faring vessels relying wholly on ocean currents and trade winds in order to reach their destination. Research on the Neolithic assemblages of Arjoune, Tell Nebi Mend and Tabbat al-Hammam is underway as part of the

204

Joanne Clarke

Contextualising Neolithic Cyprus Project, one aim of which will be to determine whether the 6th and 5th millennia BC assemblages resemble those of other central Syrian coastal sites further north and south, or whether they should be grouped within a separate central Syrian sphere of interaction, already identified for the 7th millennium BC, which includes the site of Shir (Nieuwenhuyse in press). There has been a glut of literature concerning the relative difficulty or otherwise of sea crossings between Cyprus and the mainland and this is not the place to discuss this issue again (see Broodbank 2006; Held n.d.), but key to the debate is the relationship between crossing distance and target size, taking into consideration the anti-clockwise currents that circulate around the island. Some would argue that an embarkation point on the southern Turkish coast or on the Amuq Plain might entail a shorter sea crossing, but it would be against the current and the chance of drifting past the island would be increased. Embarking from a point on the coast due west of Arjoune would certainly involve a greater sea crossing, but the current would be working with the craft and the pan handle provides a relatively large target from a south-westerly direction. Thus, in the view of this writer, the central Syrian coastal region would have been as likely a point of origin for travellers from the Levant as would have been Iskenderun (Broodbank 2006, 216) or the region around Mersin or Silifki.

Conclusions During the PPNB, Cyprus was colonised by people from the northern Levant, who brought with them a suite of subsistence practices and material culture elements extensively documented in the archaeological record of the 9th and 8th millennia BC. After the PPNB, evidence for links with the mainland became fewer, and this has created the view that contact stopped. This position does not adequately explain the presence of incised stones and a number of imported objects in 7th millennium BC contexts from several sites in Cyprus. Evidence for links with the mainland are much more tenuous in the 5th millennium BC, but a recent preliminary consideration of the pottery from Arjoune suggests that interaction with the central Levant could have been possible. Interaction, although greatly diminished from the levels recorded in the PPNB, may have continued in the form of irregular visits by Cypriots to the mainland looking to replenish depleted stocks of either plants or animals following bouts of drought. Yet, because the pressures driving cultural change on the mainland were absent on Cyprus, there was no immediate requirement for the island’s population to either intensify interaction or to adopt mainland ways of doing things. Experimentation with foreign technologies and ways of doing things happened, but it rarely led to the replacement of indigenous Cypriot behaviours and technologies.

After the PPNB, Cypriot communities became inward looking, but equally, so too did mainland populations even though the end of the PPNB did not mark the collapse of village life in the Levant. What led Near Eastern societies to become expansive and outward looking during the PPNB, and, conversely, to become inward looking after the PPNB is a question that requires much further examination beyond the scope of this paper.

Acknowledgements I wish to thank Peter Parr for allowing me to use the sherds from Arjoune stored in the Institute of Archaeology, University College London in this paper. I would also like to thank Graham Reed for his illustrations of the Arjoune sherds. Most of all I wish to thank Eddie Peltenburg, without whom my research may not have ventured beyond the shores of Cyprus.

References Bar-Yosef, O. and A. Belfer-Cohen 1989. The Levantine PPNB interaction sphere. In I. Hershkovitz (ed.) People and Culture in Change, 59–72. British Archaeological Reports International Series 508(i). Oxford, British Archaeological Reports. Bartl, A., A. Haidar and O. Nieuwenhuyse 2006. Shir: A Neolithic site in the Middle Orontes Region. Neo-Lithics 1(06), 25–27. Bartl, K., M. Hijazi and A. Haidar 2006. The Late Neolithic site of Shir: Preliminary report of the German-Syrian Cooperation Project 2006. Neo-Lithics 2(06), 15–18. Bolger, D. 2007. Eastern Mediterranean interactions in the 3rd millennium BC: Evidence of ceramics. In S. Antoniadou and A. Pace (eds) Mediterranean Crossroads, 163–188. Athens, Pierides Foundation. Bolger, D. forthcoming. A matter of choice: Cypriot interactions with the Levantine mainland during the late 4th–3rd millennia B.C. In K. Wright (ed.) Ancient Levant. Walnut Creek (CA), Left Coast. Broodbank, C. 2006. The origins and early development of Mediterranean maritime activity. Journal of Mediterranean Archaeology 19(2), 199–230. Brooks, N. 2006. Cultural responses to aridity in the Middle Holocene and increased social complexity. Quaternary International 151, 29–49. Campbell, S., V. T. Mathias and C. S. Phillips 2003. The prehistoric pottery. In P. J. Parr (ed.) Excavations at Arjoune, Syria, 31–70. British Archaeological Reports International Series 1134. Oxford, British Archaeological Reports. Cauvin, J. 1972. Religions Néolithiques de Syro-Palestine. Paris, Maison-neuve. Clarke, J. 2003. Insularity and Identity in Prehistoric Cyprus. In J. Guilaine and A. Le Brun (eds) Le Néolithique de Chypre, 203–218. Athens, Bulletin de Correspondance Hellénique, École Française d’Athènes. Clarke, J. 2004. Excavations at Kalavasos-Kokkinogia and

23. Contextualising Neolithic Cyprus: Preliminary investigations Kalavasos-Pampoules 2002–3. Report of the Department of Antiquities, Cyprus, 51–71. Clarke, J. 2007. On the Margins of Southwest Asia: Cyprus in the 6th to 4th Millennia B.C. Oxford, Oxbow. Clarke, J., Y. Goren and D. Boness forthcoming. A petrographic study of pottery from Neolithic Cyprus. In K. Wright (ed.) Ancient Levant. Walnut Creek (CA), Left Coast. Cluzan, S. 1984. L’outillage et les petits objects en pierre. In A. Le Brun (ed.) Fouilles Récentes à Khirokitia (Chypre), 1977–1981, 111–124. Paris, Éditions Recherche sur les Civilisations. Dikaios, P. 1953. Khirokitia: Final Report on the Excavation of a Neolithic Settlement in Cyprus. Oxford, Oxford University Press. Erikh-Rose, A. 2004. Geometric patterns on pebbles: Early identity symbols? In E. Peltenburg and A. Wasse (eds) Neolithic Revolution: New Perspectives on Southwest. Asia in Light of Recent Discoveries on Cyprus, 145–162. Oxford, Oxbow. Fox, W. A. 1988. Kholetria-Ortos. A Khirokitia culture settlement in the Paphos District. Report of the Department of Antiquities, Cyprus, 29–42. Gopher, A. and E. Orelle 1996. An alternative interpretation for the material imagery of the Yarmukian, a Neolithic culture of the sixth millennium BC in the southern Levant. Cambridge Archaeological Journal 6(2), 255–279. Gowlett, J. A. J. 2003. The AMS radiocarbon dates: An analysis and interpretation. In P. J. Parr (ed.) Excavations at Arjoune, Syria. British Archaeological Reports International Series 1134. Oxford, British Archaeological Reports. Haïdar-Boustani, M., J-J. Ibá ez, M. al-Maqdissi, A. Armendáriz, J. González and L. Teira 2003–2004. Prospections archéologiques à l’ouest de la ville de Homs: Rapport préliminaire campagne 2004. Tempora (Beyrouth) Annales d’Histoire et d’Archéologie Université Saint-Joseph, Beyrouth (14–15), 59–89. Haïdar-Boustani, M., J-J. Ibá ez, M. al-Maqdissi, A. Armendáriz, J. González and L. Teira 2005–2006. Prospections archéologiques à l’ouest de la ville de Homs: Rapport préliminaire campagne 2005. Tempora (Beyrouth) Annales d’Histoire et d’Archéologie Université Saint-Joseph, Beyrouth (16–17), 9–37. Held, S. n.d. Backwater blues: Einstein’s islands and the disharmony of interaction spheres. Paper presented at the American Schools of Oriental Research 92nd Annual Meeting, San Francisco, November 19–24, 1992. Hole, F. 1959. A re-analysis of basal Tabbat al-Hammam, Syria. Syria 36, 149–183. Horwitz, L., E. Tchernov and H. Hongo 2004. The domestic status of Early Neolithic fauna on Cyprus: A view from the mainland. In E. Peltenburg and A. Wasse (eds) Neolithic Revolution: New Perspectives on Southwest Asia in Light of Recent Discoveries on Cyprus, 35–48. Oxford, Oxbow. Iwasaki, T. and A. Tsuneki 2003. Archaeology of the Rouj Basin: A Regional Study of the Transition from Village to City in Northwestern Syria, vol. 1. Al-Shark 2. Ibaraki, University of Tsukuba. Le Brun, A. 1984. Fouilles Récentes à Khirokitia (Chypre), 1977–1981. Paris, Éditions Recherche sur les Civilisations. Lehavy, Y. M. 1989. Excavations at Dhali-Agridhi: 1972, 1974, 1976. In L. E. Stager and A. Walker (eds) American Expedition to Idalion, Cyprus, 1973–1980, 203–32. Chicago, Oriental Institute, University of Chicago.

205

Marfoe, L., P. J. Parr and C. S. Phillips 2003. The site and its excavations. In P. J. Parr (ed.) Excavations at Arjoune, Syria, 11–22. British Archaeological Reports International Series 1134. Oxford, British Archaeological Reports. Marshack, A. 1972. Routes of Civilization. New York, McGrawHill. McCartney, C. 2007. Lithics. In J. Clarke (ed.) On the Margins of Southwest Asia: Cyprus in the 6th to 4th Millennia B.C., 72–90. Oxford, Oxbow. McCartney, C. and B. Gratuze 2003. The chipped stone. In E. Peltenburg (ed.) Lemba Archaeological Project, Cyprus II(1): The Colonisation and Settlement of Cyprus. Investigations at Kissonerga-Mylouthkia 1976–1996, 11–34. Studies in Mediterranean Archaeology 70(4). Sävedalen: Paul Åströms Förlag. Moore, A. M. T., G. C. Hillman and A. J. Legge 2000. Village on the Euphrates: From Foraging to Farming at Abu Hureyra. Oxford, Oxford University Press. Nieuwenhuyse, O. in press. The Late Neolithic ceramics from Shir: A first assessment. Forthcoming in Zeitschrift für OrientArchäologie 2. Özbal, R. and F. Gerritsen 2004. Tell Kurdu excavations 2001. Anatolica 30, 37–75. Parr, P. J. (ed.) 2003. Excavations at Arjoune, Syria: Rural Occupation of the Halat-Ubaid and Late Persian Periods in the Orontes Valley. British Archaeological Reports International Series 1134. Oxford, Archaeopress. Peltenburg, E. J. 1982. Vrysi: A Subterranean Settlement in Cyprus. Excavations at Prehistoric Ayios Epiktitos-Vrysi 1969–73. Warminster, Aris & Phillips. Peters, J., A. von den Driesch and D. Helmer 2005. The upper Euphrates-Tigris basin: Cradle of agro-pastoralism. In J.-D. Vigne, J. Peters and D. Helmer (eds) The First Steps of Animal Domestication: New Archaeozoological Approaches, 96–124. Proceedings of the 9th ICAZ Conference, Durham 2002. Oxford, Oxbow. Redding, R.W. 2005. Breaking the mould: A consideration of variation in the evolution of animal domestication. In J.-D. Vigne, J. Peters and D. Helmer (eds) The First Steps of Animal Domestication: New Archaeozoological Approaches, 41–48. Proceedings of the 9th ICAZ Conference, Durham 2002. Oxford, Oxbow. Simmons, A. 1996. Preliminary report on the multidisciplinary investigations at Neolithic Kholetria-Ortos, Paphos District. Report of the Department of Antiquities, Cyprus, 29–44. Simmons , A. H. and R. F. Corona 1993. Test excavations at Kholetria-Ortos, a Neolithic settlement near Paphos. Report of the Department of Antiquities, Cyprus, 1–10. South, A. and I. Todd 2005. Other ground stone artefacts. In I. Todd (ed.) Vasilikos Valley Project 7: Excavations at Kalavasos-Tenta Volume II, 287–311. Studies in Mediterranean Archaeology 71(7). Sävedalen, Paul Åströms Förlag. Stanley Price, N. 1977. Colonisation and continuity in the early prehistory of Cyprus. World Archaeology 9, 27–42. Stekelis, M. 1972. The Yarmoukian Culture of the Neolithic Period. Jerusalem, Magnes. Todd, I. 1986. The foreign relations of Cyprus in the Neolithic and Chalcolithic periods: New evidence from the Vasilikos Valley. In V. Karageorghis (ed.) Acts of the International Symposium:

206

Joanne Clarke

Cyprus Between the Orient and Occident, 13–28. Nicosia, Department of Antiquities. Tsuneki, A., J. Hydar, Y. Miyake, S. Akahane, M. Arimura, S. Nishiyama, H. Sha’baan, T. Anazeki and S. Yano 1998. First preliminary report of the excavations at Tell el-Kerkh (1997), northwestern Syria. Bulletin of the Ancient Orient Museum 18, 1–40. Tsuneki, A., J. Hydar, Y. Miyake, M. Hudson, M. Arimura, O. Maeda, T. Odaka and S. Yano 1999. Second preliminary report of the excavations at Tell el-Kerkh (1998), northwestern Syria. Bulletin of the Ancient Orient Museum 19, 1–40. Tsuneki, A., Hydar J., Miyake, Y., Hudson, M., Arimura, M., Maeda, O., Odaka, T., Tanno, K. and A. Hasegawa 2000. Third preliminary report of the excavations at Tell el-Kerkh (1999), northwestern Syria. Bulletin of the Ancient Oriental Museum 20, 1–32.

Wasse, A. M. R. 2000. The Development of Goat and Sheep Herding During the Levantine Neolithic. Unpublished Ph.D. thesis, Institute of Archaeology, University College London. Wasse, A. 2002. Final results of an analysis of the sheep and goat bones from ‘Ain Ghazal, Jordan. Levant 34, 59–82. Wasse, A. 2007. Climate, economy and change: Cyprus and the Levant during the Late Pleistocene to mid Holocene. In J. Clarke (ed.) On the Margins of Southwest Asia: Cyprus in the 6th to 4th Millennia B.C., 43–63. Oxford, Oxbow. Watkins, T. 1972. Cypriot Neolithic chronology and the pottery from Philia-Drakos A. In Acts of the First International Congress of Cyprological Studies, 167–174. Nicosia, Society of Cypriot Studies. Watkins, T. 1973. Some problems of the Neolithic and Chalcolithic period in Cyprus. Report of the Department of Antiquities, Cyprus, 34–61.

24 WAS ÇATALHÖYÜK A CENTRE? THE IMPLICATIONS OF A LATE ACERAMIC NEOLITHIC ASSEMBLAGE FROM THE NEIGHBOURHOOD OF ÇATALHÖYÜK Douglas Baird

Preface I am pleased to dedicate this paper to Professor Eddie Peltenburg, who first stimulated my interest in Near Eastern archaeology and has set us fine examples of the studies of networks of prehistoric communities, especially in Cyprus, where I had my first experience of field projects neatly designed to address key research questions.

Introduction In the 8th–7th millennia cal BC a number of large communities with sites of the order of 10–20 hectares emerged across the Near East. It remains unclear why such large communities appeared at this time. Equally it is unclear whether such large communities were political or economic centres (often associated with definitions of urbanism) or distinct from their contemporaries in other ways (Hole 2000; Baird 2006; Düring 2007). These questions are obviously key to broader understandings of the development of urbanism or putative social complexity in prehistoric southwest Asia. Therefore one of the aims of the Konya Plain Survey is the investigation of the relationship between Çatalhöyük and other Neolithic sites within the south-west part of the Konya plain. There is a sense that communities of the evident relative scale of Çatalhöyük, comprising several thousand individuals (Cessford 2005, 326), were a minority in the Neolithic, and that particular factors, not operating in the majority of communities, contributed to their growth. Thus significant questions remain about a Neolithic community the size of Çatalhöyük. These include the factors and mechanisms involved in the large scale growth of the Neolithic community at Çatalhöyük in the first place (Todd

1976, 125; Baird 2006). Is it the case that a favoured setting promoted population growth within one community or did particular social practices encourage such growth? Did a number of pre-existing communities amalgamate when Çatalhöyük was first occupied, making it precociously large, or did communities or elements of communities shift to join a pre-existing community? Based on analogy with later urban communities, which are relatively large compared to their contemporaries, it has been suggested that Çatalhöyük may have served the role of a centre to surrounding communities (Bartel 1972; Todd 1976, 130; Düring 2007). In terms of the possible role of Çatalhöyük as a centre, a number of questions are raised by these hypotheses (Todd 1976, 130). What evidence is there for the nature of communities contemporary with Çatalhöyük at its greatest extent or at early stages in its growth? A number of a priori types of relationship could be posited between Çatalhöyük and contemporary local communities. These include firstly rather generalised and non-specific relationships. Secondly, we can consider close but asymmetrical inter-relationships in which Çatalhöyük provided goods or service unobtainable at the smaller sites and thus in which Çatalhöyük was an economic centre (Todd 1976, 130), a classic model for urban communities. To address this possibility we may be able to consider whether Çatalhöyük had a different role in procurement networks for materials, a possibility considered in this paper. A third alternative might suggest unequal relationships in social, ritual (Mellaart 1967, 71) or ‘political’ terms where contemporary smaller communities were dwarfed and disempowered by the proximity of a large neighbour. Fourthly, we might consider asymmetrical relationships in which the community at Çatalhöyük suppressed the development of contemporary communities, demanding a

208

Douglas Baird

consideration of reasons for such. Fifthly, close symmetrical and balanced relationships may have existed with proximate communities, in which Çatalhöyük was not an economic, social, political, or ritual centre. I have made some suggestions based on evidence of the local settlement history for a process of population aggregation at Çatalhöyük (Baird 2006). In this paper I explore them further in light of detailed evidence from the neighbouring Neolithic site of Sancak. Although it is notable that others have attempted to raise some of the issues about the possible economic roles of Çatalhöyük and whether it might be seen as central in such terms, only evidence from Çatalhöyük itself has been deployed. Thus evidence of lithic specialisation at Çatalhöyük is mooted (Conolly 1999), and this raises the possibility of a central production role for Çatalhöyük. However, the implications concerning who might have been receiving such specialist products is not thoroughly explored. Conolly (1999, 798–799) suggests that such specialists may have provisioned a very local network of households on site (Düring 2007, 160), and Todd (1976, 130) proposes a supply to local neighbours. In addition, the role for Çatalhöyük as a centre in the obsidian trade has been suggested, initially by Mellaart (1967). Largely dismissed on the general grounds that its distance from source means it could not have controlled the trade (Düring 2007, 160), such an objection does not deal with the possibility that Çatalhöyük may have been a central node in a more regional distribution of particular products produced at the site or might have had unequal access to materials. In addition, Todd (1976, 129) has argued that large ‘central’ sites like Çatalhöyük may have controlled extraction/workshop sites or provided ‘gateways’ for obsidian entering certain regions. These more specific points need to be addressed in order to establish whether Çatalhöyük may have played a role as an economic centre. Of course, it seems likely that material and products may have been exchanged and circulated along with people and that exchanges may have been key elements in maintaining social relationships whether local or longer distance. Therefore such putative economic centrality is unlikely to have been divorced from key social roles for the site if its position as an economic centre in these terms can be established. The first step in such analyses is the identification of Neolithic sites in the vicinity of Çatalhöyük. The second step is the analysis of material that might bear upon the questions of relationship. These must include questions concerning chronological relationship, but also whether there are indications of differential access to resources and other features that might allow us to determine how large sites like Çatalhöyük developed and whether they performed the role of centres.

Regional Context To understand what we know of earlier and contemporary settlement in relation to Çatalhöyük we must consider the nature of the survey evidence yielding information of ancient settlement and the nature of the geomorphological evidence that bears upon an understanding of that survey evidence. The fundamental factor to consider is the alluvial nature of the setting of Çatalhöyük (Fig. 24.1). This is of significance in understanding the Neolithic landscape, along with information from the excavations, particularly in terms of factors that might reduce ancient settlement visibility. The methodology of the Konya Plain Survey (directed by the author) was designed to take account of this and is therefore relatively intensive (Baird 1996; 2002; 2006). Survey methods were designed to detect not just visible mound sites but areas where sites have been ploughed up and vegetation and soil features indicate buried sites. Buried sites are visible in satellite imagery and in canal sections. Methods thus included intensive field walking, canal walking and visiting features identified as possible sites on the satellite imagery. The reoccupation of many sites was common. Intensive artifact collection was carried out at sites to yield material from the earliest as well as later deposits (Baird 1996). This was particularly successful as indicated by our recovery of microliths showing late Pleistocene/ early Holocene activity at some sites. In a seasonally flooded environment such as that of the alluvial fans of the south-west Konya basin, dry raised areas, natural marl or sand bumps, and previous occupations were premium site locations and continue to jut through alluvium. These factors allow us insight into the nature of the earlier settlement record (Baird 1996; 2002). Deep alluvium c. 3 m around Çatalhöyük tails off to 1 m or less in extensive areas of the alluvial fan, and it is at its deepest in a relatively small area at the apex of the fan. An earlier phase of alluvium over most of the alluvial fan was deposited before and during the occupation of Çatalhöyük in the Neolithic (Roberts et al. 1996, 39; Boyer et al. 2006) and Early Chalcolithic. Sites in the central area of the fan are perhaps surrounded by 2 m deep alluvium; in many areas this is less than 1 m. Away from the fan, on sand ridges or marls, there is no alluvium, so significant alluvial burial affects a restricted area. Sites with Neolithic occupation have been noted both in the central area and in areas of shallower alluvium. There is little evidence of disparities in observed site density between the area of deeper alluvium and surrounding areas at any period. It is clear that sites representing short-term activity and occupations of limited area, whether short-term settlements or task group activity areas, will be poorly represented in the settlement record produced by the survey. Many of these factors mitigate problems of dealing with alluvial burial. This is demonstrated by the recovery of many

24. Was Çatalhöyük a centre? The implications of a late Aceramic Neolithic assemblage

209

Fig. 24.1 Neolithic settlements on the Çarşamba alluvial fan, Konya Plain, Turkey (site 141, Sancak; site 170, Çatalhöyük). Sites shown in relation to soil type boundaries.

small Early Chalcolithic sites occupied whilst early alluvial material was still being deposited. These sites have been recovered in areas where up to 1 m of later alluvium of 2nd–1st millennium BC date was deposited. We have found and know of 16 Early Chalcolithic sites, some as small as 0.5 hectares (Baird 2002, 144). We have found four sites in addition to Çatalhöyük within the alluvial area that have unequivocal evidence of Neolithic activity and another two in the wider area that may have such evidence (Fig. 24.1). This strongly indicates the ability of our methods to detect such sites. Two are small in terms of height and area while the size of two others cannot be estimated because they are components of large multi-period mounds. We can find relatively small Neolithic sites in this landscape. It may be, of course, that other settlement sites remain buried; these are likely to be very low and small-scale sites, small, short-term settlements, encampments of mobile communities, and activity locales away from settlement. We can conclude that settlement earlier than or contemporary with Çatalhöyük was of low density and either very rare or consisting of a number of small communities occupying locales on a short-term basis. Either scenario suggests dramatic distinctions between Çatalhöyük, at least in the ceramic Neolithic period, and any putative contemporary settlement.

What can we divine more specifically from chipped stone assemblages and other evidence of the relationships between these sites and Çatalhöyük? Firstly, no Neolithic ceramics have been recovered to date from these survey sites. It is possible that these sites are aceramic and are either earlier than or contemporary with the earlier levels at Çatalhöyük. However, the rarity of ceramics in Levels VIII and VII at Çatalhöyük, and in some cases the level of burial of Neolithic deposits at the survey sites, suggest caution and demand a closer look at the lithics before we accept such a view. In one case, Sancak, later occupations are not as substantial as those at other sites, and the absence of Neolithic ceramics is more likely to be a genuine phenomenon. A further complication in relation to this site is the presence of a Late Chalcolithic occupation, and lithics from that period of occupation will be part of the assemblage. Lithics are not densely distributed at other Late Chalcolithic sites in the survey area. This and the morphology of the bulk of the chipped stone suggest that the majority of the survey chipped stone from Sancak is Neolithic. Sancak has the best represented chipped stone assemblage of all of the survey sites and will allow us to determine whether it may be contemporary or earlier than Aceramic Neolithic levels at Çatalhöyük and to consider the implications deriving therefrom.

210

Douglas Baird

Fig. 24.2 Proportions of flakes and blades, including blanks, at Sancak.

Fig. 24.3 Proportion of flake and blade tool blanks at Sancak.

Raw material

Debitage and Tool Blanks

Obsidian forms 98% of the assemblage.

Flakes dominate the debitage assemblage whether tool blanks are included or not, forming 67% of the debitage and blanks (Fig. 24.2). The dominance of flakes presumably indicates the combination of flake production relating to core preparation and rejuvenation measures, retouching and pressure flaking, as well as flake production for tool blanks and flake use. The limited proportions of elements with natural surfaces preserved in the obsidian assemblage seem to indicate that most cores reached the site in substantially prepared form, but obviously further preparation and reworking would have produced significant quantities of flakes. Nevertheless, given a reduced role for initial shaping, it is likely that the significant proportion of flakes indicates that unstructured flake production, with flakes as intended products, was an important component of reduction. This, of course, is suggested directly by reworked blade cores and flake cores but probably implies other flake production strategies as well. Fully 29% of the tools are flake tools (Fig. 24.3). Nevertheless, amongst retouched tools it appears that blade blanks were preferentially selected since blades far exceed the proportions in which they occur in the overall assemblage. Seventy-one percent (of tool blanks identifiable to debitage category) are blade tools (Fig. 24.3). Amongst the obsidian tool blade blanks, pressure blades are relatively rare. Only 7.4% of the total blade blanks appear to be pressure blade products. Given the possibility that they may have been a significant part of the Late Chalcolithic component of the chipped stone from the site, the proportions of Neolithic chipped stone pressure blades may be even lower. Obsidian pressure blade production was not a common reduction strategy in this Neolithic assemblage if it existed at all. Fully 25% of the blade blanks were recognisable as belonging to opposed platform cores. Since this is clearly a minimum (many opposed platform products are likely to go unrecognised as such), opposed platform blade production was clearly a common set of strategies (strategies 3 and 4). The highest proportion of

Reduction strategies Çatalhöyük, Pınarbaşı and other sites suggest that different reduction strategies characterise flint and obsidian from the Neolithic to the Early Bronze Age in the Konya Plain, so I will describe and discuss these aspects separately.

Obsidian Cores include opposed platform prismatic blade cores. There is also evidence for further reduction of such cores as flake cores on occasion, perhaps reflecting raw material conservation given distance from sources. Flake cores on large flakes indicate a separate reduction sequence for flake products not reusing blade cores. Debitage and tool blanks suggest between four and seven reduction strategies when allied with the core evidence: 1) single platform blade production from prismatic cores; 2) pressure blade production from single platform prismatic cores; 3) opposed platform blade production from prismatic cores; 4) production of blades from cores that would have been reminiscent of naviform or were naviform cores; 5) further use of prismatic blade cores for unstructured flake production; 6) removal of large flakes from a flat extensive core surface from a number of directions, possibly a part of initial core shaping, but producing distinctive and planned products; 7) unstructured flake production.

Flint Flint provides a smaller sample, so it is difficult to know if the full range of reduction strategies is represented: 1) single platform prismatic blade production; 2) relatively ad hoc flake production.

24. Was Çatalhöyük a centre? The implications of a late Aceramic Neolithic assemblage

211

Fig. 24.5 Sizes of pièces esquillées (length:width).

blades show single platform scars, suggesting that single platform blade production was probably the most important strategy overall.

Technique A similar range of technique related evidence is found on butts, which is evidence of platforms and bulbs from both single platform and opposed platform core products. Both sets of reduction strategies saw significant degrees of platform preparation involving both grinding and flaking of the edge of the platform on the removal surface, faceting of the platform, and isolation of the point of impact. Filiform and punctiform butts show such isolation of, and precision in, applying force to the point of impact, and perhaps softer hammer techniques applied in both strategies. Crushed and small plain platforms in a range of blades almost certainly deriving from both strategies suggest other techniques as well.

Representation of Reduction Strategies on the Site As already indicated, the rarity of cortical flakes suggests that raw material for most strategies reached the site as preforms.

Fig. 24.4 Proportions of tool types at Sancak.

Obsidian Strategies 1) Reworked cores may be present. 2) This is rare, so absence of working evidence is hardly surprising. 3) Irregular opposed platform prismatic blade production – evidence is definitely present. 4) Naviform – no evidence. 5) Reusing of blade cores is definitely present. 6) Production of large flakes; there is no evidence of this strategy, but these could have been produced from cores further worked down especially in flake production. However, this was probably not the case because a number of these are ‘cortical’ flakes otherwise rare in the assemblage, suggesting they were imported as blanks or tools. 7) Unstructured flake production was clearly carried out on site.

Douglas Baird

212

Fig. 24.7 Proportions of retouch types on non-formal tools in the Sancak assemblage.

Fig. 24.6 Chipped stone tools from Sancak.

Flint Strategies Strategies 1 and 2 (above) are represented by on-site production evidence. It seems likely that the reworking of naviform cores would leave distinctive material, and therefore within the limits of the sample size it seems likely that large flake scrapers and projectile points and some other blade tools were imported as blanks or tools.

Tools The most common ‘tool’ category comprises pièces esquillées, which form 16% of the assemblage (PE in Fig. 24.4). It is plausible that some of these were cores, but given the small size of many and the very small products resulting

from these pieces, as well as the presence of retouch on the edges of some away from the crushed areas, it is likely that many were tools, as proposed by Conolly (1999, 46). Four types of this category were distinguished (Fig. 24.5). Three of the four separate neatly on the basis of their size; this may partly relate to the blanks used for these items (Fig. 24.5). Type 1, amongst the smallest, consists of very regular rectangular items made from blade segments and with retouch along some edges separate from crushed areas. Type 3 is lozenge shaped and thick in cross-section. Type 4 is generally the largest and appears to be produced from flake blanks. Type 2 is the most varied although it is generally smaller than 3 and 4 and rather irregular in shape. The next most common categories are burins (Fig. 24.4) although these embrace a number of different burin types and it is unlikely that all tools so classified served a single function. Angle burins on breaks and truncations dominate the burin category. Points and scrapers are equally common and form 5% each of the tool assemblage (Fig. 24.4). Reduction strategy 6 seems to have been aimed particularly at the production of large flake blanks that were turned into large flake scrapers often with broadly circular or oval outline (Fig. 24.6). Two transverse arrowheads were recovered that I would suggest are Late Chalcolithic although, of course, transverse arrowheads are found as early as the Ceramic Neolithic levels of Çatalhöyük, but elsewhere in the Near East. Four varieties of point are distinguished. There are two varieties of tanged point, the first of which has relatively unpronounced shoulders and is retouched all along its edges with direct semi-abrupt to abrupt retouch. The second variety has more distinct shoulders and invasive-covering retouch on the dorsal surface. Two varieties of leaf shaped point were also distinguished (Fig. 24.6). The first was distinguished by its invasive-covering pressure retouch over the dorsal surface, the second by invasive but not covering semi-abrupt non-pressure retouch on the dorsal surface. These types fell

24. Was Çatalhöyük a centre? The implications of a late Aceramic Neolithic assemblage within the same size range, 40–60 mm in length 10–16.5 mm in width. Most were between 40 and 50 mm long. It is worth noting the lack of microliths in this assemblage. Whilst this may be due to the fact that it is a surface assemblage and we were unfortunately not allowed to sieve the surface material from this site, our intensive collection methods have nevertheless yielded microliths on other sites, and not just from sieving. Analysis of non-formal tool categories suggests the following salient features (Fig. 24.7). Most common are areas of semi-abrupt retouch on 27% of all tools. In contrast, abrupt retouch is only 3% and direct marginal low angled only 4%. Truncations are also relatively common on 17% of all tools. A significant proportion of tools (12%) have bilateral retouch where both edges are characterised by the same sort of retouch. Notches and alternating retouch are also moderately common. Pressure flaking occurs on 3% of tools excluding points. Bifacial retouch is rare. A significant proportion of the non-formal tool category shows evidence of significant design or significant regularity in use. The bilateral, alternating, alternate, pressure flaked and truncations have this character. The many tools with semiabrupt retouch suggest that a significant proportion of tools may have been hafted, may have seen heavy-duty use or may have been designed for such. The significant proportions of truncations may suggest regular modifications of inserts to fit hafts and retooling of tool elements.

Comparative Study Reduction Strategies The very limited evidence for obsidian pressure blade production and other blade production strategies suggests similarities with Çatalhöyük pre-XII and XII–VIII. The opposed platform type production and single platform percussion blade production also suggest this (Carter et al. 2005, 223; Conolly 1999, 75). Likewise, the high proportion of flakes in the debitage categories suggests much stronger relationships with the assemblages documented in the earlier levels at Çatalhöyük than with those in later levels (Conolly 1999, 75). These features also suggest close relationships with Can Hasan III where these strategies are also represented by on-site production together (Ataman 1988, 68–9). The importance of opposed platform production is notable at Sancak, and like Çatalhöyük the naviform-like component is not documented by cores on site (Carter et al. 2005, 223; Conolly 1999, 71).

Technique At Can Hasan III (Ataman 1988) and Çatalhöyük there is evidence for similar technique indicators in terms of butt

213

features and especially preparation features, including grinding and other preparation of the edge of the platform on the removal surface.

Products Unlike Çatalhöyük XII–VIII as published by Conolly (1999) but like Can Hasan III, flakes dominate, but blade blanks are preferred for tools (Ataman 1988, 73). This is also suggested to be the case for the opposed platform production of pre-XII at Çatalhöyük as documented by Carter et al. (2005, 272–274).

Points The size of the points at Sancak suggests similarities with the points from the earlier levels at Çatalhöyük. The larger points of levels VI and later at Çatalhöyük are not present. However, there are intriguing differences as well. For example, the Sancak points do not bear close comparison with those types defined by Conolly (1999, 39–40) from Çatalhöyük i.e. from levels XII and later. In particular, bifacial retouch is very limited in extent, usually to the tang area, and the points are smaller. A number of points seem most similar to the typical Can Hasan III (Ataman 1988, 115–118) and Musular (Özbaşaran 1999, 152) points and indeed to the predominant types from Suberde (Bordaz 1968, 53). These are the oval points with invasive or covering retouch on the dorsal surface and little ventral retouch, usually in the tang area. The size range overlaps very closely with those from Can Hasan III where most points are 24–45 mm long (Ataman 1988, 115–118). Points from Çatalhöyük pre-XII.C and B published by Carter et al. (2005) are like the Can Hasan points and are thus close to this category of Sancak points. One contrast between Sancak and both Can Hasan III and Suberde is the significant place for tanged points, rather like Byblos points, which are almost as common at Sancak as leaf shaped examples but are rare at Can Hasan III and Suberde. These are not like the tanged points at Çatalhöyük XII and later. However, it would appear that stemmed Byblos-like points are more common in the earliest phases, specifically pre-XII.C–D, recently excavated at Çatalhöyük (Carter et al. 2005, 274). Thus it may be that there is a distinctive localised late Aceramic tradition in the south-west Konya basin. Already at Çatalhöyük preXII.B, bifacially pressure retouched points become quite common alongside Can Hasan types. Alternatively, these Byblos points may indicate slightly earlier phases of the late Aceramic and suggest that Sancak, where bifacially retouched points are absent, is contemporary with such early phases or indeed is slightly earlier than the pre-XII.D occupation at Çatalhöyük.

214

Douglas Baird

Other Tools Many specific tools are similar between Can Hasan III and Sancak, including scrapers, tools with bilateral retouch, and angle burins (Ataman 1988, 119). The flake scrapers are generally similar from Aşıklı and Musular to Can Hasan III and Çatalhöyük. Like Sancak at Çatalhöyük, abrupt retouch is rare (Conolly 1999) but more common than at Sancak. When comparison is made with retouch attributes on non-formal tools from Level XII upwards at Çatalhöyük, as analysed by Conolly (1999, 48–56), there are obvious differences. However, at Çatalhöyük semi-abrupt retouch is very common as at Sancak but occurs in even greater proportions than at Sancak. At Sancak there is a very different relationship between proportions of alternate, alternating and bifacial retouch compared to Çatalhöyük. In the tool assemblages the high proportions of pièces esquillées are similar to those documented through the sequence at Çatalhöyük XII and later (Conolly 1999, 43–47), as well as at Can Hasan III (Ataman 1988). The pièces esquillées at Sancak (Fig. 24.5) can be divided into more and less regular and robust types as at Çatalhöyük XII and later, but there are differences. Although they are similar in terms of length, almost all of the Sancak examples are wider, suggesting differences in blanks or blank selection. Sancak is different from all sites in its relatively high proportions of burins (Fig. 24.4). Even the specific burin types from Çatalhöyük XII and later published by Conolly (1999, 50) appear to be different. However, the Can Hasan III burins are angle varieties (Ataman 1988, 119) like Sancak. This, along with other distinctive features of the retouch assemblage, suggests some distinctive practices at Sancak, whether distinct activities or different ways of carrying out the same activities. Microliths are not present at Can Hasan III or Musular (Özbaşaran 1999, 152). Can Hasan III appears to be the most similar assemblage although here one or two microliths may be present, as indicated by photographs. Generally, then, this provides a contrast with Suberde (Bordaz 1968) and Çatalhöyük pre-XII (Carter et al. 2005, 273) where microliths are present. However, Suberde stratigraphy is not the most securely based due to small trenches and a number of problems of disturbance. It should also be noted that the Çatalhöyük pre-XII microliths are very like those of 9th millennium BC Pınarbaşı. It is possible that the Çatalhöyük microliths are residuals from an earlier occupation somewhere under this extensive mound. The earliest Later Aceramic occupation of such a mound at Çatalhöyük could easily have derived relatively fresh microlithic material from such an earlier site at the Çatalhöyük location. The absence of microliths from Sancak may reflect the fact that it is a survey assemblage. However, the Konya Plain Survey has gathered a significant number of microliths

from its survey sites, even at some of the sites where it was not able to use sieving. Alternatively, this absence may just underline the similarity of the Sancak assemblage with Can Hasan III and Musular. Conclusions are thus that Sancak has its strongest similarities with Can Hasan III and probably pre-XII Çatalhöyük, specifically pre-XII.C–D. Further detailed comparisons with the very earliest material from Çatalhöyük may indicate whether Sancak is contemporary or earlier.

Comparison in Terms of Representation of Production on Site Much of the chipped stone industry at Çatalhöyük preLevel VI, especially pre-XII, has common features with Can Hasan III and Sancak in terms of the ways in which material is procured, accessed and processed. Of course, all are dominated by obsidian. At each site there are two products for which there is no on site production evidence, that is naviform-like products for projectiles, possibly arriving as blanks and/or tools, and large flake scrapers, often circular, arriving in one or both of those ways. Cores from which these types were produced may have been reused, but various factors suggest that this was not the case. These sites share single and opposed platform blade production strategies of greater and less systematisation and often show much reduced cores, attesting to on-site production using these strategies. Given the absence of cortical material, it is likely the material for these was introduced as preformed cores. In addition, less structured flake production strategies were executed on site reusing blade cores and distinctive flake cores, e.g. discoidal cores. A significant contrast exists with one site of comparable date (Musular) where the naviform strategy is represented by on site production (Özbaşaran 1999, 152). All of these Konya Plain sites seem to have similar access to and procurement and processing strategies for obsidian regardless of the disparate sizes of communities and sites involved.

Implications Sancak seems to be contemporary with the earliest excavated levels at Çatalhöyük, Pre-XII.C–D or even possibly earlier than the pre-XII sequence, as revealed in the South area. Sancak may therefore be a direct predecessor or a contemporary settlement to late Aceramic Neolithic Çatalhöyük. It seems that there were a number of other late Aceramic Neolithic settlements in the area around Çatalhöyük (Baird 2002). In this situation it is plausible that a number of originally distinct, albeit related Neolithic communities amalgamated to contribute to the distinctive growth of the large community at Çatalhöyük; or perhaps

24. Was Çatalhöyük a centre? The implications of a late Aceramic Neolithic assemblage one community was successful at the expense of others (Baird 2006). If Çatalhöyük pre-XII and Sancak were contemporary, it is difficult to know if they were communities of disparate sizes because it is currently impossible to make a realistic assessment of the extent of the early occupation at Çatalhöyük although there are hints it may have been relatively large. This is inferred because pre-XII deposits representing settlement edge activities are found in the south-west of the main mound, and other activities contemporary with this are found in the KOPAL area, an offsite area on the northern fringes of the settlement actually buried by alluvium beyond the northern limits of the currently visible mound. Late Aceramic remains may well have run much of the north-south length of the later settlement. This brings into even starker contrast the situation in which sedentary Ceramic Neolithic communities contemporary with Çatalhöyük were absent from the Çarşamba and May fans or very sparsely distributed and small. It seems Çatalhöyük was not necessarily the only obvious site location since on a relatively flat alluvial plain there are likely to have been many suitable settlement locations, essentially those raised spots that existed in the landscape. So why are there not many significant contemporary communities, and why didn’t communities in other settings on the fan blossom in the way we see at Çatalhöyük? I would suggest that a number of factors led to the suppression of those other communities. The obsidian procurement evidence suggests that economic differentiation between communities in terms of raw material acquisition or access is not likely to be significant. Rather I would suggest factors involved in the possible amalgamation of communities and development of local mating/marriage networks are probably significant. Small communities of less than 500 persons are almost certainly exogamous. This may have involved complex, long distance relationships that may have been difficult or complicated to negotiate. Communities larger than 500 can be endogamous. The development of endogamous communities may well suppress the growth of small communities in the immediate neighbourhood who may be at a disadvantage in acquiring marriage partners. Finally, there is some evidence that the community at Çatalhöyük exploited a wide range of landscapes to meet its ideologically-contexted food, dietary and other requirements. Wall paintings involving hunting of woodland and steppe animals, acquisition of oak and juniper suggest this. The exploitation of a widely defined territory may have suppressed the development of other contemporary communities. Intriguing questions arise as to why this changed dramatically in the Early Chalcolithic in the 6th millennium cal BC (Baird 2002). It is notable that the overall strategies for accessing and procuring raw material were not distinct between small communities at Sancak and Can Hasan III, perhaps

215

contemporary or slightly later and larger Çatalhöyük, and certainly later and larger Ceramic Neolithic Çatalhöyük. Size did not change procurement and processing strategies, so these did not develop specifically to meet the needs of a particularly large community. Thus neither larger sedentary communities nor those sedentary communities like Can Hasan III, nearer raw material sources in Konya basin terms, had a specific, distinctive role in distribution. Differences are apparent rather between small sites close to source and the Konya basin constellation than between larger and smaller sites in the Konya basin. This does not rule out such a specific role in obsidian procurement and distribution for other communities nearer sources such as Musular where different elements of production are attested for some strategies. Nor does it affect our view of the potential role of intermediary mobile communities, which may have been specialised producers and exchangers perhaps attested by workshops, or corporate or kin based task groups from the Konya basin settlements themselves, perhaps embedding procurement in other hunting and herding activities. Indeed the entry of material in blank or tool form alongside preforms points to multiple avenues of obsidian acquisition for these communities. In terms of specific questions about Neolithic Çatalhöyük there is much evidence to suggest it was networked in a similar way to earlier communities and smaller communities in the plain, and there is little evidence to suggest it had a distinctive role in terms of the distribution of chipped stone. In addition, the absence of small sites in its vicinity makes it unlikely that it was an economic centre supplying goods to local communities. Different emphases in production strategies and outcomes that may have existed, e.g. the absence of naviform-like production at sites like Çatalhöyük and Sancak, were balanced by complementary specialisations elsewhere. These may have had more to do with the social networks through which material flowed around the landscape and between source area and consumption zones, than with centralised production.

References Ataman, K. 1988. The Chipped Stone Assemblage from Can Hasan III: A Study in Typology, Technology, and Function. Unpublished Ph.D. thesis, Institute of Archaeology, University College London. Baird, D. 1996. The Konya Plain survey: Aims and methods. In I. Hodder (ed.) On the Surface: Çatalhöyük 1993–1995, 41–46. Çatalhöyük Research Project 1. Cambridge, McDonald Institute for Archaeological Research and London, British Institute of Archaeology in Ankara. Baird, D. 2002. Early Holocene settlement in central Anatolia: Problems and prospects as seen from the Konya Plain. In F. Gérard and L. Thissen (eds) The Neolithic of Central Anatolia, 139–160. Istanbul, Ege Yayinlari.

216

Douglas Baird

Baird, D. 2006. The history of settlement and social landscapes in the Early Holocene in the Çatalhöyük area. In I. Hodder (ed.) Çatalhöyük Perspectives, 55–74. Çatalhöyük Project 6. Cambridge, McDonald Institute for Archaeological Research. Bartel, B. 1972. The characteristics of the Çatal Hüyük supracommunity. American Journal of Archaeology 76, 204–205. Bordaz, J. 1968. The Suberde excavations, southwestern Turkey: An interim report. Türk Arkeoloji Derigisi 17(2), 43–61. Boyer, P., N. Roberts and D. Baird 2006. Holocene environment and settlement in the Konya Plain, Turkey: Integrating geoarchaeology and field survey. Geoarchaeology 21(7), 675–698. Carter, T., J. Conolly and A. Spasojevic 2005. The chipped stone. In I. Hodder (ed.) Changing Materialities at Çatalhöyük: Reports from the 1995–1999 Seasons, 221–284. Çatalhöyük Research Project 5. London, British Institute of Archaeology in Ankara. Cessford, C. 2005. Estimating the Neolithic population at Çatalhöyük. In I. Hodder (ed.) Inhabiting Çatalhöyük: Reports from the 1995–1999 Seasons, 323–326. Çatalhöyük Research Project 4. Cambridge, McDonald Institute for Archaeological Research. Conolly, J. 1999. The Çatalhöyük Flint and Obsidian Industry:

Technology and Typology in Context. British Archaeological Reports International Series 787. Oxford, J. and E. Hedges. Düring, B. 2007. Reconsidering the Çatalhöyük community: From households to settlement systems. Journal of Mediterranean Archaeology 20(2), 155–182. Hole, F. 2000. Is size important? Function and hierarchy in Neolithic settlements. In I. Kuijt (ed.) Life in Neolithic Farming Communities, 191–207. New York, Kluwer. Mellaart, J. 1967. Çatal Hüyük: A Neolithic Town in Anatolia. London, Thames & Hudson. Özbaşaran, M. 1999. Musular: A general assessment on a Neolithic site in central Anatolia. In M. Özdoğan and N. Başgelen (eds) Neolithic in Turkey, the Cradle of Civilization: New Discoveries, 147–155. Istanbul, Arkeoloj ve Sanat Yayinlari. Roberts, N., P. Boyer and R. Parish 1996. Preliminary results of geoarchaeological investigations at Çatalhöyük. In I. Hodder (ed.) On the Surface: Çatalhöyük 1993–1995, 19–40. Çatalhöyük Research Project 1. Cambridge, McDonald Institute for Archaeological Research and London, British Institute of Archaeology in Ankara. Todd, I. 1976. Çatal Hüyük in Perspective. Menlo Park (CA), Cummings.

25 THE BIRTH OF ETHNICITY IN IRAN: MESOPOTAMIAN-ELAMITE CROSS-CULTURAL RELATIONS IN LATE PREHISTORY Andrew McCarthy

Introduction Can we as archaeologists legitimately create a taxonomy that differentiates ethnic groups in the archaeological and historical records? Can we identify whether an ancient people had their own distinctions that we can define as ethnicity? Any archaeological discussion of ethnicity must try to reconcile the extent to which the material observations we make are representative of emic social manifestations in the past rather than simply an etic nomenclature created by us to help us deal with past human interactions. The concept of ethnicity can be used in interpretation and description as a means to identify social behaviours and trajectories of social change, but in prehistory this is especially difficult. Even in historical periods, we must be careful not to fall into the trap of normativity to which a simplistic reading of ethnic markers can lead. Ethnicity is at once the product of social interaction (and can be observable in material remains) and a factor involved in social change. These two qualities, both passive and active, are present in stylistic attributes that can be observed in material culture. Style in material culture can be thought of as an active form of communication because in its intended cultural context it would have helped to define the very social distinctions to which it was referring. But material culture is also a relatively passive frame of reference that can be sought as a standard static symbol even if the interpretations made from this referent may have changed through time. The lines between these two aspects are blurred, however, as they often co-exist in the same objects (see Wobst 1977; Wiessner 1983; 1990; Hodder 1990). Material objects also often have a functional nature, and the very fact that they are made of materials means that they are constrained by natural laws. Therefore, alongside what can be regarded as style, material objects take on what

can be called isochrestic attributes that can vary according to differing procedures of technical manipulation (Sackett 1985; Wiessner 1985). Isochrestic variation comes from a prescribed procedure for the manipulation of material objects. An aspect of this concept is better known as chaînes opératoires. In the present study, we shall be dealing with craft specialisation, in particular seal making, in the 4th and 3rd millennia cal BC in what is today south-western Iran. The end of the 4th millennium in Iran has been termed proto-historic as it marks the first writing in clay in the indigenous protoElamite script. The bureaucratic and technical changes seen at this time are accompanied by a sharp change in political and economic organisation often referred to as ‘early state formation’. The shift from pre-state to state-level society is poorly understood in Iran, however, and is often either equated with or explained by Mesopotamian (Sumerian) developments. What we can observe in the archaeological record at Susa during this period is important and should be discussed in its own right in order to track the social trajectories that led to the eventual expression of a political state. One observation that can be made is that from the late 4th to the beginning of the 3rd millennium a change occurred in the way cylinder seals were made. These changes can be regarded as isochrestic variations related to dual ethnic identities that co-existed at the single site of Susa. In the period of early state formation a stylistic preference of one of these ethnically charged variants was exploited to serve as an emblem for the newly forming state-level identity. Pierre Amiet has argued a case for a mixed ethnic population co-existing at Susa in the late 4th millennium based on a stylistic analysis of stratified seals and impressions (1979). The present analysis of the Susa seals from this period both supports his argument and offers a

218

Andrew McCarthy

rare glimpse into what were probably ethnically defined craft traditions and expressions of ethnic identity. The methodology used here follows from a study which involved a general survey of stamp and cylinder seals from the protohistoric Near East, where it was argued that expressions of newly emerging state-level identities are reflected in, among other things, the metrical styles of cylinder seals, i.e. their heights, diameters and proportionalities (McCarthy 2003). Interestingly, while Amiet’s assessment was based largely on the qualitative differences in seal designs, the present analysis is based on quantifiable metrical qualities that allow us to observe what can be viewed as different hands at work, variations in manufacturing techniques, and different traditions and preferences. The two groups of seals approached here are what Amiet has termed “de style mésopotamien” and “contemporaine des tablettes proto-Elamite” (1972), and henceforth these will be referred to respectively as Mesopotamian Jemdet Nasr and Piedmont Jemdet Nasr styles. Chronologically, these two styles overlap and can roughly be thought to have co-existed. The Mesopotamian Jemdet Nasr seals are referred to by Frankfort as ‘Proto-literate c’ (Sin Temple II from Khafajah; see Frankfort 1955) while the Piedmont Jemdet Nasr seals are designated as ‘Proto-literate d’ (Sin Temple IV from Khafajah; see Frankfort 1955). This differentiation seems to have been an attempt to control the architectural evidence at Khafajah rather than representing a real diachronic distinction in the material culture, and the overall Jemdet Nasr material culture group at present does not have enough chronological control to argue anything beyond synchronicity (this sentiment is mirrored by those who invented the term ‘Proto-literate’; see Delougaz and Lloyd 1942, 8 and n. 10). Therefore, while Amiet draws parallels with Khafajah in his analysis of the Susa seals, both of these groups can be said to be contemporary. Le Breton (1957) reflects this in his evaluation of the glyptic evidence, and he designates seals no. 1074 and no. 1086 (Amiet 1972), both Piedmont Jemdet Nasr style seals, to Susa Cc alongside seals such as the pigtailed figure seal no. 723 assigned to Mesopotamian style according to Amiet (Le Breton 1957, 107–108; Amiet 1972). Because the evidence of object distribution at Susa (of which we have limited knowledge) does not imply a spatial division based on ethnicity, the human population at Susa can likewise be thought to have co-existed. Algaze (2001) has given general support for a non-exploitative relationship between Susa and southern Mesopotamia by including Susiana in his definition of the ‘core’ Uruk culture. The term ‘core’ is problematic, however, and not useful for the present purposes. It is sufficient to say that the situation of an enclave of economically separate ethnic Mesopotamians, as argued by Stein for Hacınebi Tepe (2001), does not appear to be tenable for Susa. Susa should be thought of as at least a peer region to southern Mesopotamia.

In the second half of the 4th millennium, Susiana appears to have had such close ties to southern Mesopotamia that the two cultures have sometimes been equated or lumped into a single category. This categorisation overlooks the fact that while the Uruk style material culture was popular and enjoyed a widespread usage in the Near East, Susian culture was distinct and discrete, albeit with strong similarities to the material culture in southern Mesopotamia. Pittman, however, convincingly argues that the true ‘home’ of what is referred to here as the Piedmont Jemdet Nasr style can be found in Susa and the surrounding area that would later be known to us as Elam (1994; see also Le Breton 1957, 108). This lends credence to the idea that this is indeed a distinct proto-Elamite style complete with an understood form of signification even though the Piedmont cylinder seals and impressions are found in a much wider area. The distinction between Susian and southern Mesopotamian seal styles are sometimes difficult to define, however, especially when the designs and imagery are identical. The analysis undertaken for this study offers a way to observe this distinction independent of designs that can easily be copied and/or emulated. The benefit that metrical analysis gives is that the actual craftsmanship put into the manufacture of the seal is being observed. Art historical analysis of designs, while useful, only considers the qualitative artistic elements that can easily be copied and spread. The traditions of the lapidary craftspeople, however, can only be maintained through training and repetition (see Porada 1977 for discussion of professional vs. amateur lapidary craftsmanship), and the skills subsequently acquired may therefore be considered to be derived to some degree through traditional behaviour practised through time. The results from the statistical analysis of the metric qualities of the Susa seals support this claim. The Mesopotamian Jemdet Nasr style seals as a group overlap metrically with the Piedmont Jemdet Nasr style seals. To a modern observer it would therefore be very difficult to draw a distinction between these two styles of glyptic based solely on their size and proportionality. At the same time, statistical analysis of these two groups of seals draws a line so clearly between them that it is argued here that the differences could only have come about through contemporary but completely separate craft traditions. This, coupled with the design styles that were subsequently used to distinguish ethnicity during nascent statehood in the early 3rd millennium, provides strong support for Amiet’s assertion that the divergent craft traditions are evidence of ethnic duality at Susa in the late 4th millennium. The novelty of this study is not in the conclusion being drawn. Rather, the method of analysis allows us to approach the concept of ‘ethnicity’ in a more refined and precise way. Observing ethnicity in the archaeological record, while important (Emberling and Yoffee 1999), has been at best a difficult task; even the concept of ethnicity itself has been

25. The birth of ethnicity in Iran: Mesopotamian-Elamite cross-cultural relations in late prehistory

Fig. 25.1 Scatter plot of the height to diameter ratios of both Mesopotamian Jemdet Nasr and Piedmont Jemdet Nasr style seals.

219

Fig. 25.2 Scatter plot of height to diameter ratios of Mesopotamian Jemdet Nasr style seals.

challenged (see Jones 1997). Therefore there will also be some discussion below on how the data recovered from Susa fits into an overall understanding of ethnic identity.

Analysis For this study, metrical information on all of the cylinder seals identified by Amiet (1972) as Mesopotamian Jemdet Nasr and Piedmont Jemdet Nasr have been gathered although only seals with complete heights and diameters have been used. When put into a height vs. diameter scatter plot, they form a more or less continuous and overlapping range of metric forms (Fig. 25.1). When put into scatter plots separately (Figs. 25.2–25.3), however, it becomes clear that the general trends of the two styles are different. While outliers somewhat skew regression lines, it is nonetheless clear that the Mesopotamian style seals tend to be blockier and squatter proportionally, and that the proto-Elamite style tends to be more elongated. At the same time, the two populations overlap, and seals such as no. 917 (Mesopotamian Jemdet Nasr, ht. 2.9 cm, diam. 2.4 cm; see Amiet 1972, vol. 2, pl. 112, no. 917) and no. 1042 (Piedmont Jemdet Nasr, ht. 2.6 cm, diam. 1.5 cm; see Amiet 1972, vol. 2, pl. 99, no. 1042) could easily belong to either metric population (see Fig. 25.1). These two seals were chosen in particular because as objects they are of similar proportions and they look quite similar in construction, despite their engraved designs. They both have suspension loops integral to their construction, making their similarity even more pronounced. Statistical analysis, however, can give us greater resol-

Fig. 25.3 Scatter plot of height to diameter ratios of Piedmont Jemdet Nasr style seals.

ution than our naked eye allows, especially as we are unable to translate many of the stylistic messages being conveyed. A particularly enlightening statistical analysis has been the Ttest, which has produced some surprising results. A T-test is a statistical procedure that determines the likelihood that, in this case, the objects from one group might belong in another group. In other words, a T-test can determine whether one population is distinct from another population, measured by a degree (0–100%) of similarity. The heights, diameters and proportionalities of cylinder seals have been studied from the Mesopotamian style and proto-Elamite style seals. Grouped according to these classifications, T-tests were run to see the

220

Andrew McCarthy

t-Test: Two-Sample Assuming Unequal Variances Mean Variance Observations Hypothesized Mean Difference df t Stat P(T<=t) one-tail t Critical one-tail P(T<=t) two-tail t Critical two-tail

Variable 1 1.58125 0.126958 16 0 30 0.099619 0.460655 1.69726 0.92131 2.04227

Variable 2 1.56875 0.124958333 16

Table 25.1 ‘Dummy’ T-test, showing margin of error of a small sample (sixteen objects).

likelihood that seals from one group might fit in the metrical range of the other group. The result was startlingly clear: the probability that the two populations of objects are the same or even similar is infinitesimal. Because each of these groups is so homogenous and clearly defined in the T-test, it is argued that these artefacts were not merely variations within a single group of craftspersons but were the products of parallel ethnic craft traditions. Given that such overlapping occurs between these two populations in design style, size and proportionality, it was surprising to see that they exhibit such a stark statistical distinction. For this reason, a margin of error for the test was determined by running a T-test on two identical populations of objects (the example in Table 25.1 being a sample of only 16 objects). The similarities between the two groups were strong as would be expected. The 0.92131 P(T<=t) two tail analysis shows that these two groups were approximately 92% similar, which illustrates the margin of error of a very small sample. With the larger sample sizes used in the analysis of the two styles studied here (176 Mesopotamian Jemdet Nasr and 297 Piedmont Jemdet Nasr), the ‘dummy tests’ resulted in 100% similarity, indicating a very small margin of error. The results of the T-tests comparing Mesopotamian Jemdet Nasr and Piedmont Jemdet Nasr cylinder seals from Susa are displayed in Tables 25.2, 25.3 and 25.4. Again, only cylinder seals with complete heights and diameters have been used for this analysis, and no seal impressions were used due to the possibility of metrical inaccuracy. The Piedmont Jemdet Nasr style has close metrical affinities with the later Presargonic style of the 3rd millennium. The corpus of extant Piedmont Jemdet Nasr seals and impressions numbers approximately 300, while the number of extant Presargonic glyptic is unfortunately only about 16. Therefore, although a T-test result shows an approximately 8% similarity between the groups (which is statistically significant—similarity with the Mesopotamian Jemdet Nasr style seals is still less than 1%), the difference

in population sizes means that this is not a reliable figure. One can presume that with a bigger sample of Presargonic seals there would be a greater correlation between the protoElamite and the 3rd millennium seals, but at present this link is better established through comparison of designs. The statistical results show very strong differences between the two groups. Although they would have co-existed at the site of Susa, they were distinct, and these tests suggest that they were of completely different populations. In particular, the probability (highlighted in the above tests as P(T<=t) two tail) that these two groups are of the same population is actually so small that it is shown here in scientific notation (Table 25.2 shows that the two groups are only 6.69%×10-5% similar in the height category; Table 25.3 shows 8.35% ×10-17% similarity in the diameter category; and Table 25.4 shows 8.27% ×10-34% similarity in their proportionalities). This relatively simple and straightforward analysis forcefully indicates that even though there was some overlap in the actual metric styles, they were almost without doubt different populations of objects. By contrast, in another study using this method, T-tests were run comparing Mittanian style seals from the 2nd millennium cities of Enkomi (Cyprus) and Alalakh (Turkey) that exhibited 98% similarity (McCarthy 2008). Comparison of Enkomi with Ugarit (Syria), however, displayed only 40% similarity, which has been interpreted as representative shifting political alliances in the Late Bronze Age (McCarthy 2008). This information strongly supports the argument that in the late 4th millennium the inhabitants of Susa had no opposition to including two distinct styles of seals in their cultural repertoire. This acceptance of divergent craft styles changes in the 3rd millennium, and when the inhabitants of Susa and other Iranian settlements came to create an emblemic style, they excluded the ‘foreign’ elements and focussed on producing seals that helped to solidify protoElamite identity.

25. The birth of ethnicity in Iran: Mesopotamian-Elamite cross-cultural relations in late prehistory

t-Test: Two-Sample Assuming Unequal Variances Mean Variance Observations Hypothesized Mean Difference Df t Stat P(T<=t) one-tail t Critical one-tail P(T<=t) two-tail t Critical two-tail

Variable 1 1.492045 0.230451 176 0 305 5.076786 3.35E-07 1.649864 6.69E-07 1.967774

221

Variable 2 1.276541096 0.143820641 292

Table 25.2 Diameters: Mesopotamian versus Piedmont seals.

t-Test: Two-Sample Assuming Unequal Variances Mean Variance Observations Hypothesized Mean Difference df t Stat P(T<=t) one-tail t Critical one-tail P(T<=t) two-tail t Critical two-tail

Variable 1 1.916477 0.507898 176 0 422 -9.28895 4.17E-19 1.648473 8.35E-19 1.9656

Variable 2 2.601027397 0.743194817 292

Table 25.3 Heights: Mesopotamian versus Piedmont seals.

t-Test: Two-Sample Assuming Unequal Variances Mean Variance Observations Hypothesized Mean Difference df t Stat P(T<=t) one-tail t Critical one-tail P(T<=t) two-tail t Critical two-tail

Variable 1 1.348587 0.250516 176 0 459 -13.6413 4.14E-36 1.648179 8.27E-36 1.965145

Variable 2 2.126973725 0.535117457 292

Table 25.4 Height/diameter ratio of Mesopotamian versus Piedmont seals.

Interpretation The subtle differences between the seals on the margins of the metrical ranges may or may not have been perceptible to the ancient inhabitants of Susa. What does seem clear, however, is that the seal engravers who crafted the seals with Piedmont Jemdet Nasr designs were trained to create seals of

certain metrical dimensions. Likewise, the engravers of seals with Mesopotamian Jemdet Nasr style designs followed a certain manufacturing procedure that resulted in seals with another metrical canon. Hypothetically, one presumably could not go to a ‘proto-Elamite’ craftsperson and ask for a Mesopotamian style seal because the product would be

222

Andrew McCarthy

made by hands trained to construct seals in a certain manner. Even if the design styles were being copied or emulated, the unconscious manufacturing technique would have betrayed an isochrestic variation linked to tradition. The distinction between the two sets of objects seems far too subtle to assume that the differences were deliberate variations by a single group. In the late 4th millennium, therefore, there co-existed two distinct groups, each maintaining its own identities through the processes of tradition and craft training, as well as selection and use of the seals. In the 4th millennium there were people who expressed a Mesopotamian-like material culture living side by side with people who expressed a proto-Elamite material culture. At the beginning of the 3rd millennium, the proto-Elamite culture began to be expressed as symbolic of the state-level identity forming in this region. Amiet (1979) has argued that the people asserting their identity at the state-level were ethnically ‘proto-Elamite’ rather than ‘southern Mesopotamian’ or ‘Sumerian’, in part substantiated by the use of the proto-Elamite language and writing but also through social, economic and political organisation different from that of Sumer. Amiet argues this point through the perspective of the seal designs and other methods of interpretation. The Presargonic seals of the later 3rd millennium continued to exhibit the more elongated metrical style consistent with their precursor in the Piedmont Jemdet Nasr style seals observed in the present analysis. This preference is maintained even into the 2nd millennium when the emerging Middle Elamite Kingdom of Anshan and Susa used designs similar to Mittanian seals but kept the more elongated metrical style (McCarthy and Hill 2009). What this apparently means is that when the time came to consolidate the Susian population according to a newly emerging state-level identity, the preference for the ‘local’ proto-Elamite elongated style becomes clear: the shorter and squatter Mesopotamian Jemdet Nasr seals drop out of the assemblage entirely. The groups that had co-existed in the 4th millennium exhibit stylistic preference to the point of exclusion in the 3rd millennium. Power relations, identity agglomeration and ideology may all have been factors leading to the eventual division. What the present study suggests is that the line where the division was drawn was based on distinct group traditions. It may be that these groups were an important or even defining factor in the expressions of state-level identity in the 3rd millennium.

Ethnicity We can confidently say that there were different craft traditions co-existing at Susa. We can also confidently say that the expressions of these differences in material culture served as way to signify a newly emerging state-level identity in the 3rd millennium. We know from 3rd millennium textual

sources that the identity that solidified around this new state was at least externally (in Mesopotamia the proto-Elamite script is not yet fully translated) perceived of as ‘Elamite’, and that this was both a cultural and a linguistic distinction (although this identity may not have been completely homogenous). What, then, does this tell us about ethnicity and its origins? Can we reasonably project a category that we call ‘ethnicity’ onto a prehistoric time period that does not explicitly acknowledge this concept in the same way that we do? In other words, is ethnicity defined by objective traits or by a subjective function within a society? These questions are related to the differences between etic taxonomy and emic self-definition. This problem is to some degree related to another disparity in defining ethnicity: from where does it come? Is there a primordial imperative to ethnicity (that is, is ethnicity a natural and unavoidable, possibly biological trait?) or is it a “dynamic and situational form of group identity embedded in the organisation of social behaviour and also in the institutional fabric of society” (Jones 1997, 72)? This problem often overlaps with the debate concerning the primacy of the individual actors or the primacy of “social structures and cultural norms” (Jones 1997, 72). As we have seen, it appears as though a mechanical isochrestic procedure is sufficient to explain the variation in the seal repertoire of the 4th millennium. In the 3rd millennium, however, when we begin to see the formation of state-level identities in the Near East, a valuation occurs: the proto-Elamite style seals are preferred as emblems of the newly forming identity. This does not necessarily mean that people who had at one time expressed Mesopotamianstyle material culture no longer lived in Elam; rather, it simply means that the emblemic expressions in the 3rd millennium Elamite seal repertoire grew out of what was once an isochrestic variation. As these isochrestic traits would have been virtually imperceptible, we can assume that the preference was largely for the craft tradition rather than exclusively a choice of design styles. In other words, the people who created the Piedmont Jemdet Nasr seals were selected, and the people who created the Mesopotamian Jemdet Nasr seals were rejected. During the 3rd millennium we know that the isochrestic distinction was between the ethnically Mesopotamian and the ethnically Elamite because we are told of this distinction in the Mesopotamian literature, and it is supported by other corroborating information. This last point is important because in the 4th millennium we see isochrestic variation in material culture, but we cannot confidently say that this distinction is linked to emic ethnic differentiation. Even in the 3rd millennium it is problematical to say this as the concept of ethnicity has a normative taxonomic definition derived from modern usage. That being said, even as a normative term ethnicity is a relative trait. In other words, ethnicity must be defined in

25. The birth of ethnicity in Iran: Mesopotamian-Elamite cross-cultural relations in late prehistory terms of inclusion and exclusion: there are people who are ethnically the same and people who are ethnically different. Therefore, if we are to use the term ‘ethnicity’ in regard to the people involved in the Mesopotamian and proto-Elamite craft traditions, we may surmise that in 4th millennium Susa, where we see these contemporary isochrestic variants, there was at least an emerging ethnicity although it was latent as a political expression. As archaeologists, we can see the distinction in the material culture, but only from an etic perspective. It was not until the 3rd millennium that these distinctions were used for control of power, exclusivity, and social and political manipulation. At this time we see a change from isochrestic variation to a preference of one stylistic variant, which corresponds to the visibility of a probable emic ethnic distinction. All of the elements for ethnic duality were in place in Susa in the 4th millennium, but can we truly speak of ethnicity at all until this distinction was overtly acknowledged? Ethnicity must often be explicitly and materially expressed in order for ethnic solidarity to occur. The communication of this identity need not, however, exist in the form of written or spoken speech. There are many ways in which objects can be read, and no doubt this is what we should be looking at in the material record. We can accept Amiet’s argument that in the design styles that we see in the 4th millennium there is one style that looks more ‘local’ and one that appears to be ‘foreign’ (specifically, Mesopotamian). The fact that we can observe isochrestic variation in the seals of this period provides a strong argument for ethnic duality, even if in the 4th millennium it was only in its very early stages. In this case isochrestic variation does not seem to have carried a cultural value with it during the 4th millennium. If indeed it is the case that this isochrestic variation is indicative of ethnic differentiation, it was most likely a peaceful coexistence of ethnic duality in pre-state Susa. When early state formation occurred, power was consolidated through expressions of one of these pre-existing differentiations which was then to serve as the emblem for state-level identity.

Acknowledgements Professor Eddie Peltenburg has been a guiding influence in much of my work. Early in his career he worked in Iranian archaeology, and years later he published an important catalogue of the Burrell Collection (1991), which includes cylinder seals. It seems fitting that a volume dedicated to him would include a fusion of these two aspects of his wide and varied career. I would like also like to express my thanks to Professor Trevor Watkins for his help with the statistics and his comments regarding this study; all errors remain solely mine. Many thanks go to the University of Edinburgh Development Trust, who provided travel expenses to attend

223

the First International Conference on the Ancient Cultural Relations Between Iran and Western Asia in Tehran 2003 at which an earlier version of this paper was delivered.

References Algaze, G. 2001. Initial social complexity in southwestern Asia: The Mesopotamian advantage? Current Anthropology 42(2), 199–233. Amiet, P. 1972. Glyptique Susienne: des origines à l’époque des perses achéménides. Cachets, sceaux-cylindres et empreintes antiques, vols. 1–2 (text and plates). Mémoires de la Délégation archéologique en Iran 43. Paris, Paul Geuthner. Amiet, P. 1979. Archaeological discontinuity and ethnic duality in Elam. Antiquity 53(209), 195–204. Delougaz, P. and S. Lloyd 1942. Pre-Sargonid Temples in the Diyala Region. Oriental Institute Publications 58. Chicago, University of Chicago Press. Emberling, G. and N. Yoffee 1999. Thinking about ethnicity in Mesopotamian archaeology and history. In H. Kühne, R. Bernbeck and K. Bartl (eds) Fluchtpunkt Uruk. Archäologische Einheit aus Methodischer Vielfalt. Schriften für Hans Jörg Nissen, 272–281. Rahden (Westfahlen), Marie Leidorf. Frankfort, H. 1955. Stratified Cylinder Seals from the Diyala Region. Chicago, University of Chicago Press. Hodder, I. 1990. Style as historical quality. In M. W. Conkey and C. A. Hastorf (eds) The Uses of Style in Archaeology, 44–51. Cambridge, Cambridge University Press. Jones, S. 1997. The Archaeology of Ethnicity: Constructing Identities in the Past and Present. London, Routledge. Le Breton, L. 1957. The early periods at Susa: Mesopotamian relations. Iraq 19, 79–124. McCarthy, A. 2003. Signs of Shared Complexity: Glyptic and Largescale Social Transformations in the 4th–3rd Millennia Near East. Unpublished Ph.D. thesis, University of Edinburgh. McCarthy, A. 2008. Island alliances: The workshops of Ugarit and the Cypriot ‘common-style’ seals. In A. McCarthy (ed.) Island Dialogues: Cyprus in the Mediterranean Network. Proceedings of the Postgraduate Cypriot Archaeology Conference (POCA) Edinburgh 2006, 195–209. Edinburgh, University of Edinburgh Archaeology Occasional Papers 21. McCarthy, A. and I. Hill 2009. Emblems of ethnicity: Hurrians, Elamites and the Mittanian ‘common-style’ glyptic. In D. Owen and G. Wilhelm (eds) Studies on the Civilization and Culture of Nuzi and the Hurrians 18, 299–322. Bethesda (MD), CDL Press. Peltenburg, E. 1991. The Burrell Collection: Western Asiatic Antiquities. Edinburgh, Edinburgh University Press. Pittman, H. 1994. The Glazed Steatite Glyptic Style: The Structure and Function of an Image System in the Administration of Protoliterate Mesopotamia. Berliner Beiträge zum Vorderen Orient 16. Berlin, Dietrich Reimer. Porada, E. 1977. Of professional seal cutters and nonprofessionally made seals. In M. Gibson and R. D. Biggs (eds) Seals and Sealing in the Ancient Near East, 7–14. Malibu, Undena. Sackett, J. R. 1985. Style and ethnicity in the Kalahari: A reply to Wiessner. American Antiquity 50(1), 154–159.

224

Andrew McCarthy

Stein, G. 2001. Indigenous social complexity at Hacınebi (Turkey) and the organization of Uruk colonial contact. In M. S. Rothman (ed.) Uruk Mesopotamia and Its Neighbors: Cross-Cultural Interactions in the Era of State Formation, 265–306. Santa Fe, School of American Research. Wiessner, P. 1983. Style and social information in Kalahari San projectile points. American Antiquity 48(2), 253–276. Wiessner, P. 1985. Style or isochrestic variation? A reply to Sackett. American Antiquity 50(1), 160–166.

Wiessner, P. 1990. Is there a unity to style? In M. W. Conkey and C. A. Hastorf (eds) The Uses of Style in Archaeology, 105–112. Cambridge, Cambridge University Press. Wobst, H. M. 1977. Stylistic behavior and information exchange. In C. E. Cleland (ed.) For the Director: Research Essays in Honor of James B Griffin, 317–342. Anthropological Papers 61. Ann Arbor, Museum of Anthropology, University of Michigan.