MOBILITY, EXCHANGE, and TOMB MEMBERSHIP in BRONZE AGE ARABIA: a BIOGEOCHEMICAL INVESTIGATION DISSERTATION Presented in Partial F

Home , Ajman, Asimah, Autonomous city, Bahrain National Museum, Dalma Culture, Dibba

MOBILITY, EXCHANGE, AND TOMB MEMBERSHIP IN BRONZE AGE ARABIA: A BIOGEOCHEMICAL INVESTIGATION

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in the Graduate School of The Ohio State University

Lesley Ann Gregoricka, B.A., M.A.

Graduate Program in Anthropology

The Ohio State University

2011

Dissertation Committee:

Clark Spencer Larsen, Advisor

Joy McCorriston

Samuel D. Stout

Paul W. Sciulli

Lesley Ann Gregoricka

2011

ABSTRACT

Major transitions in subsistence, settlement organization, and funerary architecture accompanied the rise and fall of extensive trade complexes between southeastern Arabia and major centers in Mesopotamia, Dilmun, Elam, Central Asia, and the Indus Valley throughout the third and second millennia BC. I address the nature of these transformations, particularly the movements of people accompanying traded goods across this landscape, by analyzing human and faunal skeletal material using stable strontium, oxygen, and carbon isotopes. Stable isotope analysis is a biogeochemical technique utilized to assess patterns of residential mobility and paleodiet in archaeological populations. Individuals interred in monumental communal tombs from the Umm an-Nar (2500-2000 BC) and subsequent Wadi Suq (2000-1300 BC) periods from across the Oman Peninsula were selected, and the enamel of their respective tomb members analyzed to detect (a) how the involvement of this region in burgeoning pan-

Gulf exchange networks may have influenced mobility, and (b) how its inhabitants reacted during the succeeding economic collapse of the early second millennium BC.

Due to the commingled and fragmentary nature of these remains, the majority of enamel samples came from a single tooth type for each tomb (e.g., LM1) to prevent

ii repetitive analysis of the same individual. However, in a few instances, multiple in situ teeth were present and permitted inter-tooth sampling as a means of evaluating temporal shifts in mobility and diet within the lifetime of single individuals.

Stable strontium and oxygen ratios indicate that the Umm an-Nar inhabitants of southeastern Arabia were not highly mobile despite their increasing involvement in regional and interregional trade. However, such patterns do fit with archaeological evidence for an increasingly sedentary lifestyle associated with intensified oasis agriculture and the construction of large, permanent settlements and fortification towers.

Non-local immigrants were interred in small numbers within Umm an-Nar tombs alongside local peoples, perhaps suggestive of some form of fictive kinship, a potential by-product of growing interregional commerce. By relaxing the restraints of tomb membership, a more flexible and complex funerary ideology was adopted and reflects the broader appropriation of kinship in the formation of a multi-ethnic society. In addition, stable carbon isotope ratios suggest the consumption of a broad, mixed C3-C4 diet fitting with the employment of a variety of subsistence strategies, although preference was given to C3-based sources of food.

The dramatic changes in the archaeological record associated with the transition to the Wadi Suq period are not mirrored in isotopic indicators of paleomobility. As in the

Umm an-Nar, the Wadi Suq population does not appear to have been highly mobile despite a decrease in the number, size, and permanence of settlements. Oxygen isotope values do not differ from the preceding period, and while strontium ratios are significantly different, this is likely a reflection of the exploitation of different geographic areas with correspondingly disparate isotope signatures. Stable carbon isotope values

iii indicate a considerable change in subsistence practices involving a greater reliance on C3- based foodstuffs and a more restricted dietary intake, with an emphasis not on marine resources but on oasis agriculture. These data corroborate the strontium and oxygen isotope results and portray a society that was still relatively sedentary and continued to practice cultivation. Moreover, the continued presence of non-local immigrants interred in local tombs suggests that interregional economic relations did not completely break down during this “Dark Age” of purported cultural isolation.

The findings of this study illustrate continuity between the Umm an-Nar and

Wadi Suq periods and call into question how substantial the so-called “collapse” of the early second millennium BC actually was. Furthermore, the presence of non-locals in both Umm an-Nar and Wadi Suq tombs, with no outward expression of foreign identity, reinforces the idea that these immigrants may have readily adopted the practices of their local host community, even in death. It appears that, rather than being subjugated by a hegemonic system controlled by more complex centers like the Indus Valley or

Mesopotamia, the inhabitants of southeastern Arabia lived in relative autonomy. Finally, while local communities were undoubtedly affected by a disintegration of external economic relationships with the larger Gulf, this collapse may have also been partially influenced by internal social dynamics; however, isotopic evidence challenges the validity of an ideological conflict between traditional, kinship-based factions and a growing social elite.

Dedicated to my wonderful parents, Larry and Susan, and to my sisters and best friends, Lindsay and Lauren

ACKNOWLEDGMENTS

This dissertation could not have been completed without the support and assistance of many individuals. Funding for this research was provided by a number of institutions. International travel to the United Arab Emirates was made possible by The

Ohio State University Office of International Affairs’ International Affairs Grant as well as the Alumni Grant for Graduate Research and Scholarship from the OSU Graduate

School. Equipment and supplies were partially funded by a Sigma Xi Grant-in-Aid of

Research. Finally, the Ruggles-Gates Fund for Biological Anthropology, the National

Science Foundation Doctoral Dissertation Research Improvement Grant (BCS-0961932), and the Philanthropic Educational Organization (PEO) Scholar Award supplied financial support for domestic travel to the University of North Carolina-Chapel Hill Isotope

Geochemistry Laboratory as well as laboratory costs associated with stable isotope analysis.

I wish to thank the many museums and people who made my work in the United

Arab Emirates possible. Dr. Daniel Potts of the University of Sydney was unwaveringly generous in helping me to make initial connections with museum directors, curators, and archaeologists across the Emirates, including the patient and ever-helpful Emma

vi Thompson and Michele Ziolkowski. At the Sharjah Archaeological Museum in the

Emirate of Sharjah, finds supervisor Johanna Olafsdotter, Director of Archaeological

Excavations Dr. Sabah Jasim, and conservationist Asma Al Hrmoodi all took time from their busy schedules to assist me in sorting through the Tell Abraq human skeletal collection and in making valuable contacts with local Emirati government and museum officials. Thanks also go to archaeologists Dan Potts and Carl Phillips for permission to sample the Sharjah teeth and to Peter Magee for sharing his in-depth knowledge of the archaeology of Sharjah and the UAE and for his useful suggestions/information relating to this project.

At the Ajman Museum in the Emirate of Ajman, Director Ali Mohammed Al

Matroushi graciously allowed me to study the Mowaihat skeletons, while archaeologist

Ernie Haerinck permitted the sampling of human teeth.

At the National Museum of Ras al-Khaimah in the Emirate of Ras al-Khaimah, resident archaeologist Christian Velde and archaeologist Imke Moellering of the

Department of Antiquities and Museums consented to an analysis of the human and faunal teeth from the Bronze Age sites of Unar 1, Shimal 103, and Shimal 95. Christian,

Imke, and Ahmed Hilal made RAK a home away from home and did not hesitate to provide me with anything that I needed during my multiple stays there.

At the University of Nevada, Las Vegas, bioarchaeologist Deb Martin offered me the Tell Abraq human remains for analysis and welcomed me into her home. Her efforts to get me this material, along with bioarchaeologist Jennifer Thompson, were very much appreciated.

vii At the University of Tübingen, Germany, anthropologists Margarethe and Hans-

Peter Uerpmann were kind enough to ship the Tell Abraq faunal teeth to me, with permission to sample these teeth granted from excavator Dan Potts.

At the Moesgård Museum in Højbjerg, Denmark, Dr. Flemming Højlund was eager to assist this project and enthusiastically offered the skeletal collection of Umm an-

Nar Island as well as comparative material from the Bahrain burial mounds. His hospitality, along with archaeologists Dr. Bo Madsen and Dr. Mohammed Bander, made me feel most welcome there. In addition, at the Zoological Museum at the University of

Copenhagen in Denmark, curator Dr. Kim Aaris-Sørensen and conservator Kristian

Gregersen graciously permitted the sampling of three faunal collections from the Gulf region, including Umm an-Nar Island, Barbar Temple, and Failaka.

At the Peabody Museum at Harvard University in Boston, Massachusetts, Dr.

Richard Meadow opened his laboratory to me and permitted the sampling of comparative faunal material from the Indus Valley sites of Allahdino and Balakot.

Dr. Drew Coleman in the Department of Geology at the University of North

Carolina at Chapel Hill provided unwavering support of my research and took time to personally train me in laboratory methods related to strontium sample preparation, as well as to instruct me on the inner workings of thermal ionization mass spectrometry.

Moreover, Drew’s fantastic graduate and undergraduate students generously assisted me in the lab, including Ryan Mills, Miquela Ingalls, Courtney Beck, Katie Moore, and

Arden Larberg. Emeritus professor Dr. Paul Fullager kindly donated his time towards helping me interpret data and analyze samples. I felt most welcome and am incredibly appreciative of my time there.

viii I am grateful to Shane Whitacre and Franklin Sanders “Sandy” Jones at the Soil

Characterization Laboratory at The Ohio State University for use of their freeze-drying equipment. Thanks go to Dr. Andrea Grottoli and Lab Manager Yohei Matsui of the

Stable Isotope Biogeochemistry Laboratory at The Ohio State University for their assistance and use of their mass spectrometer. Also at Ohio State, Dr. Richard Yerkes graciously offered his comparative collections and assisted in the identification of faunal teeth. In addition, OSU undergraduate student assistant Meg Rood volunteered to make molds of all whole human teeth before enamel sampling.

Appreciation and acknowledgment must be extended to my family and friends, who never stopped supporting me and without whose optimism I could not have finished this work. My parents and sisters were always there to offer words of encouragement and to put things into perspective. Fellow graduate students Amy Hubbard and Gabriela

Jakubowska as well as recent graduates Jaime Ullinger, Robin Feeney, and Tracy

Betsinger were tremendously supportive, and Julia Giblin and Erica Chambers provided me with much methodological assistance and stimulating biogeochemical discussions that improved this dissertation substantially. Jaime Ullinger deserves special thanks for her help with human tooth identification and for her countless hours as a study buddy.

Particular thanks must be extended to Dr. Susan Guise Sheridan at the University of Notre Dame, who introduced me to biological anthropology and whose support throughout my undergraduate and graduate years have meant so much to me and have made me into a better scientist and teacher. I am grateful to Dr. Mark Schurr for initiating me into the world of archaeological chemistry and for his continuing support and assistance throughout my undergraduate and graduate years.

ix My dissertation committee, including Dr. Clark Spencer Larsen, Dr. Joy

McCorriston, Dr. Samuel Stout, and Dr. Paul Sciulli deserves the utmost thanks for their invaluable guidance and support during this process. As always, I am grateful for Paul

Sciulli’s statistical expertise. Joy McCorriston introduced me to the archaeology of this incredible region and provided much time and support in the theoretical development of this dissertation as well as in making contacts with archaeologists working there. Lastly,

I owe a debt of gratitude to my advisor Clark Larsen, whose perspective, guidance, and sage advice were much appreciated throughout my years at Ohio State.

VITA

November 22, 1982………….Born – Owosso, Michigan, USA

2005………………………….B.A. Anthropology Honors, University of Notre Dame

2007………………………….M.A. Anthropology, The Ohio State University

2005-2007……………………Editorial Assistant, American Journal of Physical Anthropology, The Ohio State University

2007………………………….Assistant Editor, American Journal of Physical Anthropology, The Ohio State University

2007-2008……………………Foreign Language and Area Studies Fellow

2008-2010……………………Graduate Teaching Associate, The Ohio State University

2011-present……….…………Presidential Fellow, The Ohio State University

FIELDS OF STUDY

Major field: Anthropology Minor field: Anatomy

TABLE OF CONTENTS

Abstract……………………………………………………………………………………ii

Acknowledgments………………………………………………………………………..vi

Vita………………………………………………………………………………………..xi

List of Tables…………………………………………………………………………….xv

List of Figures…………………………………………………………………………..xvii

Chapter 1. Introduction…………………………………..……………………………..1 Purpose…………………………………………………………………………….1 Expected Outcomes…………………………………………………………….…2 Organization of Dissertation………………………………………………………3

Chapter 2. Biocultural Context of the Oman Peninsula………………………………4 The Late Stone Age (ca. 8000-3100 BC)………………………………………...4 Dispersal into Arabia and the Origins of the Neolithic……………………5 Subsistence and Seasonality………………………………………………9 Mortuary Practices……………………………………………………….16 Interregional Exchange…………………………………………………..27 The Bronze Age (ca. 3100-1300 BC)…………………………………………...32 Hafit Period (ca. 3100-2500 BC)……………………………………….32 Settlement and Subsistence………………………………………33 Mortuary Practices……………………………………………….40 Interregional Exchange…………………………………………..45 Umm an-Nar Period (ca. 2500-2000 BC)……………………………...53 Settlement and Subsistence………………………………………53 Mortuary Practices……………………………………………….60 Interregional Exchange…………………………………………..66 Wadi Suq Period (ca. 2000-1300 BC)………………………………….75 Settlement and Subsistence………………………………………76

xii Mortuary Practices……………………………………………….84 Interregional Exchange…………………………………………..93 Statement of Hypotheses……………………………………………………….97

Chapter 3. Stable Isotopes…………………………………………………………….104 Strontium……………………………………………………………………….106 Oxygen………………………………………………………………………….116 Carbon…………………………………………………………………………..124 Enamel Mineralization and Crown Formation…………………………………133

Chapter 4. Political Economy and Agency…………………………………………..137 World-Systems Theory…………………………………………………………137 Reaction to World-Systems Theory…………………………………………….141 Agency and Social Organization in the Bronze Age…………………………...145

Chapter 5. Materials and Methods…………………………………………………...150 Site Descriptions: United Arab Emirates……………………………………150 UAE: Persian Gulf Coast……………………………………………..152 Mowaihat, Emirate of Ajman…………………………………..152 Shimal 95 and 103, Emirate of Ras al-Khaimah………………..159 Tell Abraq, Emirate of Sharjah…………………………………170 Umm an-Nar Island, Emirate of Abu Dhabi……………………178 Unar 1, Emirate of Ras al-Khaimah…………………………….187 UAE: Gulf of Oman Coast……………………………………………190 Bidya, Emirate of Fujairah……………………………………...191 Dadna, Emirate of Fujairah……………………………………..195 Dibba, Emirate of Fujairah……………………………………..198 Mereshid, Emirate of Fujairah………………………………….199 Qidfa, Emirate of Fujairah……………………………………...201 Comparative Sites……………………………………………………..202 Methods………………………………………………………………………...203 Enamel Extraction………………………………………………………203 Strontium……………………………………………………………….206 Oxygen and Carbon…………………………………………………….207 Statistical Analyses……………………………………………………..209

Chapter 6. Results……………………………………………………………………..212 Geological Setting of the United Arab Emirates…………………………….212 Strontium………………………………………………………………………218 Oxygen…………………………………………………………………………252 Carbon…………………………………………………………………………285

Chapter 7. Discussion…………………………………………………………………323

Chapter 8. Conclusions……………………………………………………………….357

xiii

References……………………………………………………………………………...362

Appendix A: List of Enamel Samples and Sample Identification………………….417

Appendix B: Isotope Data for Faunal Samples……………………………………...431

Appendix C: Isotope Data for Human Samples……………………………………..437

Apendix D: 18O Conversion Data…………………………………………………...445

xiv

LIST OF TABLES

Table 2.1. Dental pathological data of Neolithic sites in the Oman Peninsula based on tooth counts………………………………………………………….14

Table 2.2. Neolithic burials and cemeteries of the Oman Peninsula…...………………..18

Table 3.1. Natural abundances of strontium isotopes in the environment……………...106

Table 3.2. 87Sr/86Sr values of various natural materials on Earth………………………109

Table 3.3. Natural abundances of oxygen isotopes in the environment...... 116

Table 3.4. Summary of oxygen isotopic contributions to environmental water………..118

Table 3.5. Natural abundances of carbon isotopes in the environment………………...125

Table 5.1. Bronze Age tombs from the United Arab Emirates with human and faunal dental enamel sampled in this study……………………….……151

Table 5.2. Comparative Bronze Age human and faunal teeth sampled in this study…..203

Table 6.1. Strontium isotope values from modern water samples across Abu Dhabi, United Arab Emirates…………………………………………...213

Table 6.2. Modern carbonate sediments sampled off the coast of the Emirate of Abu Dhabi……………………………………………………………214

Table 6.3. Mean 87Sr/86Sr ratios in human tooth enamel from Bronze Age sites in the United Arab Emirates……………………………………………241

Table 6.4. Inter-tooth 18O sampling at Mowaihat for six individuals…………………261

Table 6.5. Inter-tooth 18O sampling at Tell Abraq for one individual………………...264

Table 6.6. Inter-tooth 18O sampling at Umm an-Nar Island for five individuals……...269

Table 6.7. Inter-tooth 18O sampling at Unar 1 for three individuals…………………..275

Table 6.8. Inter-tooth 18O sampling at Bidya, Qidfa and Dibba in the Emirate of Fujairah………………………………………………………………279

Table 6.9. Mean 18O ratios in human tooth enamel from Bronze Age sites in the United Arab Emirates……………………………………………….280

Table 6.10. Bronze Age 13C ratios from Tell Abraq…………………………………..286

Table 6.11. Inter-tooth 13C sampling at Mowaihat for six individuals………………..294

Table 6.12. Inter-tooth 13C sampling at Tell Abraq for one individual……………….299

Table 6.13. Inter-tooth 13C sampling at Umm an-Nar Island for five individuals…….302

Table 6.14. Inter-tooth 13C sampling at Unar 1 for three individuals…………………307

Table 6.15. Inter-tooth 13C sampling in the Emirate of Fujairah……………………...313

Table 6.16. Mean 13C ratios in human tooth enamel from Bronze Age sites in the United Arab Emirates……………………………………………….314

Table 6.17. Mean 13C and ranges for comparative sites across the Persian Gulf……..321

Table 6.18. Mann-Whitney (U) comparative statistics for carbon isotope ratios of fauna from the United Arab Emirates and across the Persian Gulf…….322

xvi

LIST OF FIGURES

Figure 2.1. Toothpick groove on a maxillary incisor from individual HD may indicate advanced oral hygiene practices at the Neolithic site of Jebel al-Buhais 18………………………………………………………..16

Figure 2.2. Map illustrating the location of Neolithic cemeteries unearthed in the Oman Peninsula………………………………………………………….17

Figure 2.3. Secondary burial from the site of RH-5, Oman……………………………...20

Figure 2.4. Flexed burial from the fourth millennium BC site of GAS-1 at Wadi Shab, Oman………………………………………………………………22

Figure 2.5. Collective primary burial at Jebel al-Buhais 18 with five articulated individuals………………………………………………………………..24

Figure 2.6. Secondary disarticulated burial at Jebel al-Buhais 18……………………….25

Figure 2.7. Fifth millennium BC collective burial from Umm al-Quwain 2…………….26

Figure 2.8. Bitumen slab from H3, As-Sabiyah, Kuwait………………………………...30

Figure 2.9. Hili 8 plan during period I…………………………………………………...35

Figure 2.10. Typical Hafit-type grave from Jebel Haift, United Arab Emirates………...41

Figure 2.11. Beehive and Hafit-type tomb reconstructions……………………………...42

Figure 2.12. Beehive tombs at Al Ayn in the interior of Oman………………………….43

Figure 2.13. Horde of copper bun-shaped ingots recovered from House 4 at the site of Maysar 1, Oman………………………………………………48

xvii Figure 2.14. Artist’s reconstruction of copper smelting during the Early Bronze Age in the Oman Peninsula………………………………………………49

Figure 2.15. Finely-hewn curved ashlars made of limestone provided the facing of Umm an-Nar tombs…………………………………………………...61

Figure 2.16. Typical Umm an-Nar-type tomb at al-Sufouh, United Arab Emirates…….62

Figure 2.17. A decorated, crescent-shaped ivory comb from Bactria…………………...74

Figure 2.18. Mineral concentrations from 6000-2000 cal. BP as measured from a marine sediment core from the Gulf of Oman…………………………...77

Figure 2.19. Mudbrick platform paving the fortification tower at Tell Abraq…………..79

Figure 2.20. Shimal type above-ground collective tomb from Sharm, Emirate of Fujairah, United Arab Emirates………………………………………….86

Figure 2.21. Ghalilah type above-ground collective tomb from Shimal (SH 103), Emirate of Ras al-Khaimah, United Arab Emirates……………………...87

Figure 2.22. Khatt type above-ground collective tomb from Bithna, United Arab Emirates………………………………………………………………….88

Figure 2.23. Dhayah type subterranean T-shaped collective tomb from Bithna, Emirate of Fujairah, United Arab Emirates……………………………...89

Figure 2.24. Horseshoe-type subterranean collective tomb from Wadi al-Qawr, Emirate of Ras al-Khaimah, United Arab Emirates……………………...90

Figure 3.1. Strontium isotope evolution of the Earth…………………………………...108

Figure 3.2. 87Sr/86Sr in seawater through Phanerozoic time……………………………110

Figure 3.3. A schematic diagram of the isotope fractionation process via evaporation and condensation…………………………………………..120

Figure 3.4. δ13C values for modern grasses and the resulting enrichment in 13C of Apatite following isotope fractionation………………………………...128

Figure 5.1. Map of the northern Oman Peninsula illustrating the location of Bronze Age tombs sites in the United Arab Emirates used in this study….….....151

Figure 5.2. Map of third millennium sites of the United Arab Emirates……………….152

xviii Figure 5.3. Plan of Tombs A and B at Mowaihat………………………………………153

Figure 5.4. Human skeletal material from Mowaihat, Tomb B………………………...156

Figure 5.5. Map of excavated Shimal tombs, including Sh 95 and Sh 103…………….161

Figure 5.6. Tomb plan of Shimal 95……………………………………………………164

Figure 5.7. Articulated forearm with associated beads in Area 4A of the tomb of Shimal 95……………………………………………………………….166

Figure 5.8. Tomb plan of Shimal 103…………………………………………………..167

Figure 5.9. Eastern tomb chamber of Shimal 103 with commingled human remains………………………………………………………………….169

Figure 5.10. Umm an-Nar tomb at Tell Abraq…………………………………………172

Figure 5.11. Human remains from the Umm an-Nar tomb at Tell Abraq……………...173

Figure 5.12. Location of the tomb fields and settlement on Umm an-Nar Island……...179

Figure 5.13. Grave I, Umm an-Nar Island……………………………………………...181

Figure 5.14. Tomb plan of Grave V, Umm an-Nar Island……………………………...183

Figure 5.15. Tomb plan of Grave II, Umm an-Nar Island……………………………...184

Figure 5.16. Unar 1, a circular Umm an-Nar tomb on the Shimal Plain……………….189

Figure 5.17. Partially articulated skeletal segments from the Umm an-Nar tomb of Unar 1………………………………………………………………..190

Figure 5.18. Map of the northern Emirates……………………………………………..191

Figure 5.19. Major archaeological sites in the northern Emirates……………………...192

Figure 5.20. Tomb plan of Bidya 1 in the Emirate of Fujairah………………………...194

Figure 5.21. Location of the sites of Dibba and Dadna, Emirate of Fujairah…………..196

Figure 5.22. Tomb plan from Dadna…………………………………………………...197

Figure 5.23. Map of Middle Bronze Age (Wadi Suq) sites along the eastern coast of the United Arab Emirates, including the tomb at Mereshid…………200

xix

Figure 5.24. Map of the Arabian Peninsula and South Asia, illustrating the location of comparative Bronze Age sites across the Persian Gulf used in this study…………………………………………………………………….202

Figure 5.25. Ovicaprine RM3 from Balakot, Pakistan after removal of outer coating of cementum and enamel abrasion……………………………………...206

Figure 6.1. The geomorphology of the northern Emirates……………………………...215

Figure 6.2. Geologic map of the eastern United Arab Emirates………………………..217

Figure 6.3. Strontium isotope ratio ranges for Bronze Age fauna in the United Arab Emirates, used to define local ranges at each site………………...219

Figure 6.4. Strontium isotope ratio ranges by species for Bronze Age fauna in the United Arab Emirates……………………………………………….220

Figure 6.5. LM1 strontium isotope ratios of individuals interred at Mowaihat, Emirate of Ajman……………………………………………………….222

Figure 6.6. Inter-tooth sampling at Mowaihat for six individuals……………………...223

Figure 6.7. Strontium isotope ratios of subadult and adult individuals interred at Tell Abraq, Emirate of Sharjah…………………………………………225

Figure 6.8. Detailed scale showing local subadults and adults from Tell Abraq (non-locals excluded)…………………………………………………...227

Figure 6.9. Strontium isotope ratios of individuals interred in Tombs I, II, and V at Umm an-Nar Island, Emirate of Abu Dhabi…………………………229

Figure 6.10. Inter-tooth sampling from Umm an-Nar Island…………………………...231

Figure 6.11. Strontium isotope ratios of individuals interred at the Umm an-Nar tomb of Unar 1 and the Wadi Suq tombs of Shimal 95 and 103, Emirate of Ras al-Khaimah……………………………………………..234

Figure 6.12. Strontium isotope ratios of subadult and adult individuals interred at Unar 1, Emirate of Ras al-Khaimah…………………………………….236

Figure 6.13. Inter-tooth sampling for three individuals from Unar 1…………………..237

Figure 6.14. Strontium isotope ratios of individuals interred across the Emirate of Fujairah…………………………………………………………………239

Figure 6.15. Inter-tooth sampling from Wadi Suq tombs across the Emirate of Fujairah…………………………………………………………………240

Figure 6.16. Variance in strontium isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods……………………………..242

Figure 6.17. Mean  1 s.d. in strontium isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods……………………………..243

Figure 6.18. Strontium isotope ratios of individuals interred at the A’ali Mound Field, Bahrain…………………………………………………………...244

Figure 6.19. Strontium isotope ratios of all individuals from the United Arab Emirates, organized chronologically from left to right, compared with local ranges from Bahrain………………………………………………246

Figure 6.20. Strontium isotope ratios of all individuals from the United Arab Emirates, organized chronologically from left to right, compared with local ranges from Tepe Yahya………………………………………….247

Figure 6.21. Strontium isotope ratios of all individuals from the United Arab Emirates, organized chronologically from left to right, compared with local ranges from Failaka Island, Kuwait………………………………249

Figure 6.22. Strontium isotope ratios of all individuals from the United Arab Emirates, organized chronologically from left to right, compared with local ranges from Allahdino (blue) an Balakot (red)…………………...251

Figure 6.23. Stable oxygen isotope ratio (VSMOW) values of modern annual precipitation in Asia…………………………………………………….253

Figure 6.24. Oxygen isotope ratio ranges for Bronze Age fauna in the United Arab Emirates………………………………………………………………...254

Figure 6.25. Oxygen isotope ratio ranges by species for Bronze Age fauna in the United Arab Emirates…………………………………………………..255

Figure 6.26. Ranges of stable oxygen isotope ratios from Bronze Age human enamel across the United Arab Emirates………………………………………..257

Figure 6.27. Stable oxygen isotope ratio (VSMOW) values of modern annual precipitation in Asia…………………………………………………….258

xxi Figure 6.28. Oxygen isotope ratios of individuals interred at Mowaihat, Emirate of Ajman………………………………………………………………..259

Figure 6.29. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from Mowaihat………………………………………….260

Figure 6.30. Oxygen isotope ratios of adult and subadult individuals interred at Tell Abraq………………………………………………………………263

Figure 6.31. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from Tell Abraq…………………………………………264

Figure 6.32. Detailed scale of strontium-oxygen bivariate scatter plot showing local subadults and adults from Tell Abraq (non-locals excluded)…………..265

Figure 6.33. Oxygen isotope ratios of individuals interred at Umm an-Nar Island…….267

Figure 6.34. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from Umm an-Nar Island……………………………….268

Figure 6.35. Oxygen isotope ratios of individuals interred at the Umm an-Nar tomb of Unar 1 and the Wadi Suq tombs of Shimal 95 and 103, Emirate of Ras al-Khaimah………………………………………………………....271

Figure 6.36. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from the Shimal Plain, Ras al-Khaimah………………...273

Figure 6.37. Oxygen isotope ratios for subadults and adults from the tomb of Unar 1, Emirate of Ras al-Khaimah…………………………………….274

Figure 6.38. Oxygen isotope ratios of individuals interred in the Emirate of Fujairah...277

Figure 6.39. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from the Emirate of Fujairah……………………………278

Figure 6.40. Variance in oxygen isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods………………………………..281

Figure 6.41. Mean  1 s.d. in oxygen isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods……………………………..282

Figure 6.42. Oxygen isotope ratios of all human individuals from the United Arab Emirates compared with those from the A’ali Mound Field in Bahrain………………………………………………………………….283

xxii Figure 6.43. Oxygen isotope ratios of all human individuals from the United Arab Emirates compared with those from al-Khubayb in Oman…………….284

Figure 6.44. Carbon isotope ratio ranges for Bronze Age fauna in the United Arab Emirates………………………………………………………………...287

Figure 6.45. Carbon isotope ratio ranges by species for Bronze Age fauna in the United Arab Emirates…………………………………………………..289

Figure 6.46. LM1 carbon isotope ratios of individuals interred at Mowaihat, Emirate of Ajman………………………………………………………………..291

Figure 6.47. Bivariate plot of carbon and oxygen isotope ratios for human enamel from Mowaihat and comparative faunal enamel from Tell Abraq……..293

Figure 6.48. Bivariate plot of carbon and strontium isotope ratios for human enamel from Mowaihat and comparative faunal enamel from Tell Abraq……..294

Figure 6.49. Carbon isotope ratios of adult and subadult individuals interred at Tell Abraq……………………………………………………………………296

Figure 6.50. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from Tell Abraq………………………………………………...297

Figure 6.51. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from Tell Abraq………………………………………………...298

Figure 6.52. Carbon isotope ratios of individuals interred at Umm an-Nar Island…….300

Figure 6.53. Bivariate plot of strontium and carbon isotope ratios for human and faunal enamel from Umm an-Nar Island……………………………….301

Figure 6.54. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from Umm an-Nar Island………………………………………302

Figure 6.55. Carbon isotope ratios of adults and subadults interred at the Umm an-Nar tomb of Unar 1, Emirate of Ras al Khaimah…………………...304

Figure 6.56. Bivariate plot of strontium and carbon isotope ratios for human and faunal enamel from the Shimal Plain, Ras al-Khaimah………………...306

Figure 6.57. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from the Shimal Plain, Ras al-Khaimah………………………..307

xxiii Figure 6.58. Carbon isotope ratios of individuals interred at the Wadi Suq tombs of Shimal 95 and 103, Emirate of Ras al-Khaimah………………………..308

Figure 6.59. Carbon isotope ratios of individuals interred in the Emirate of Fujairah…310

Figure 6.60. Bivariate plot of strontium and carbon isotope ratios for human and faunal enamel from the Emirate of Fujairah……………………………311

Figure 6.61. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from the Emirate of Fujairah……………………………312

Figure 6.62. A comparison of carbon isotope ratios of individuals interred at all sites in the United Arab Emirates……………………………………………314

Figure 6.63. Variance in carbon isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods……………………………………315

Figure 6.64. Mean  1 s.d. in carbon isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods………………………………..316

Figure 6.65. Carbon isotope ratios of all human individuals from the United Arab Emirates compared with a sample of individuals from the A’ali Mound Field, Bahrain…………………………………………………..317

Figure 6.66. Carbon isotope ratios of all human individuals from the United Arab Emirates compared with a sample of individuals from the al-Khubayb necropolis, Oman……………………………………………………….318

Figure 6.67. Carbon isotope ratios of fauna from the United Arab Emirates compared with fauna from sites across the Persian Gulf…………………………..320

xxiv

CHAPTER 1

INTRODUCTION

Purpose

The purpose of this investigation is to assess mobility and how patterns of mobility changed from the Umm an-Nar (Early Bronze Age; 2500-2000 BC) to the Wadi

Suq (Middle Bronze Age; 2000-1300 BC) periods relative to times of economic interregional connectivity and disjunction in southeastern Arabia, as well as the impact of migration on local tomb membership by using stable strontium, oxygen, and carbon isotope ratios extracted from human dental enamel. The Oman Peninsula has until recently been largely ignored by archaeologists and bioarchaeologists as a “peripheral” region, unimportant relative to other major centers in the Near East (Blau 1999, 2001). In fact, the Oman Peninsula sat at the nexus of burgeoning Bronze Age trade networks; by the Umm an-Nar period, coastal sites across southeastern Arabia show evidence of extensive exchange systems with the Indus Valley (Pakistan), Mesopotamia (Iraq),

Dilmun (Bahrain), Central Asia, and Elam (Iran) (Potts 1997). Such evidence is present in the form of numerous foreign artifacts and exchange standards recovered within local tombs, corroborating the emergence of unprecedented communication networks

1 characteristic of a world-system (Frank 1993; Ratnagar 2001). Previous archaeological and bioarchaeological studies have speculated that foreign immigrants were interred in both Umm an-Nar and Wadi Suq communal graves, but no systematic study of skeletal material has been done to support this claim. This research addresses the hypotheses that mobility, the presence of non-locals, and dietary variability increased during the Umm an-Nar in the Oman Peninsula as a result of increasingly complex and widespread interregional exchange networks, but that interregional mobility, the presence of non- locals, and dietary variability decreased during the Wadi Suq as a result of a “collapse” of these pan-Gulf trade systems. In conjunction with addressing the nature of the movements of people in association with these trade systems using stable isotopes, this project also seeks to address broader issues by challenging theoretical assumptions and reconsidering southeastern Arabia’s assignment to the periphery. This research is significant in that it represents the first biogeochemical study in the region and also the first to take on such a broad comparative expanse extending across both the Middle East and South Asia.

Expected Outcomes

Previous archaeological studies in this region have noted the abundant presence of exotic goods in Umm an-Nar tombs reflective of the period’s increasing involvement in interregional trade, while the subsequent Wadi Suq period saw a dramatic decrease in foreign presence. It is expected that non-locals from areas actively engaged in trade with southeastern Arabia were interred in local tombs during the Umm an-Nar Period, and that as a result of these exchange networks, local populations were highly mobile. It is also

2 expected that a sharp economic decline in the Middle Bronze Age resulted in decreased interregional mobility and an absence of non-locals in these Wadi Suq tombs. Associated shifts in dietary variability are also anticipated in light of changes in mobility.

Organization of Dissertation

This dissertation has eight chapters. The second chapter consists of an introduction to the Neolithic and Bronze Age archaeology and mortuary practices of southeastern Arabia. In addition, it includes a brief assessment of the few bioarchaeological studies conducted in the region thus far. This chapter concludes with a statement presenting the hypotheses to be evaluated using biogeochemical techniques.

The third chapter introduces biogeochemical theory and the three stable isotopes utilized in this study: strontium, oxygen, and carbon. It also discusses the process of enamel mineralization, critical to understanding at what ages isotopes become incorporated into enamel and thus serving as a proxy for determining the timing of residential mobility.

The fourth chapter provides an overview of prevalent political economic theory and its application to pre-industrial, archaeological populations as a means of testing the hypotheses proposed for this study. In chapter five, a report is given for each of the thirteen Bronze Age mortuary sites located in the United Arab Emirates from which human dental enamel was analyzed. This chapter concludes with a detailed description of methodology related to enamel sampling and stable isotope analysis. Chapter six presents the results of this study. Chapter seven discusses the hypotheses in light of these results while also evaluating economic and mortuary theory, while chapter eight concludes this dissertation.

CHAPTER 2

BIOCULTURAL CONTEXT OF THE OMAN PENINSULA

The Arabian Peninsula sits at the junction between the land masses of Africa,

Asia, and Europe, and has served not only as a gateway from one continent to another, but as an important site of cultural interaction for thousands of years. Until the last few decades, southeastern Arabia had been largely relegated by archaeologists to the periphery, unimportant relative to major centers in Mesopotamia and the Indus Valley.

However, recent archaeological evidence points to the existence of growing interregional trade networks, emerging as early as the Neolithic and evolving into a complex system of exchange by the Early Bronze Age, in which the Oman Peninsula sat at the nexus.

Late Stone Age (ca. 8000-3100 BC)

While this dissertation will focus primarily on the changing mortuary practices and emergent interregional exchange networks of the Bronze Age, it is important to understand the context in which this period developed. Subsequently, an understanding of the Early Holocene and what some have dubbed the ‘Late Stone Age’ (Uerpmann

1989, 1992; Kiesewetter et al. 2000; Uerpmann 2002; Kiesewetter 2003; Jasim et al.

2005) is critical to appreciating resultant Bronze Age systems. The Late Stone Age

4 represents the post-Pleistocene period following the Last Glacial Maximum before the

Bronze Age and covers approximately five millennia; however, this label may be misleading because materials dating to this period were fashioned not only from stone but also from shell and clay (Uerpmann 1992; Potts 1993a). Within this episode of time, the

Neolithic period can be more specifically defined as occurring between ca. 5000-3100

BC.

Dispersal into Arabia and the Origins of the Neolithic

The peopling of Arabia in the Early Holocene and the subsequent origins of the

Neolithic in this region have been subject to much debate. Past arguments emphasized an indigenous development of the Arabian Neolithic by populations already present

(Cleuziou and Tosi 1997; Cleuziou et al. 2002). However, recent zooarchaeological, paleoclimatic, and lithic evidence has shed considerable light on this process and suggests another point of origin – the Levant (Amirkhanov 1996; Uerpmann and

Uerpmann 1996; Uerpmann et al. 2000; Uerpmann and Uerpmann 2003; Drechsler 2007;

Uerpmann et al. 2009).

Evidence for human occupation of the Arabian Peninsula in the Upper Pleistocene is sparse, with a handful of lithic scatters (e.g., Amirkhanov 1996; Uerpmann and

Uerpmann 2003; Wahida et al. 2009) the only indication of a limited human presence.

This lack of archaeological evidence corresponds with severe climatic conditions during this time. The Last Glacial Maximum, which lasted from approximately 68,000 to 8000

BC, represents an episode of extreme desiccation, conditions which would have made any lengthy occupation (and autochthonous development of the Neolithic) unlikely

5 (Glennie et al. 1994; Potts 1997a). Paleolithic assemblage typology suggests that those small, transient groups present may have originated in northeastern Africa (Petraglia

2007; Wahida et al. 2009). However, current evidence indicates no continuity between late Pleistocene and early Holocene populations (Uerpmann et al. 2009).

With the onset of the Early Holocene in Arabia around 8000 BC, a much more favorable, humid climate appeared in both north and south Arabia and persisted for almost two millennia (Drechsler 2007). During this climatic optimum, the Pre-Pottery

Neolithic B (PPNB) culture of the southern Levant thrived, resulting in substantial population growth and expansion, particularly as new pastures were sought for domesticated herds of sheep, goat, and cattle (Uerpmann et al. 2009). It is hypothesized that these mobile hunter-herders rapidly spread across the Arabian Peninsula and into southeastern Arabia, potentially generating social fission between these groups and those remaining in the Levant (Drechsler 2007).

Two lines of evidence support this view of population and cultural diffusion into

Arabia. First, the recovery of Qatar B-type blade arrowheads from Early Holocene sites across Qatar and the Eastern Province of Saudi Arabia, some of the earliest lithics in the region, display striking technological and typological similarity with that of the Levantine

PPNB blade arrowhead technology and date to the eighth and seventh millennium BC

(Kapel 1967; Potts 1993a; Uerpmann et al. 2009). Secondly, while a significant amount of domestic faunal remains dating to the Early Holocene have been found in southeastern

Arabia, the habitats of the wild progenitors of domestic sheep, goat, and cattle fall in the northern and southern Levant (Drechsler 2007). Subsequent morphological and genetic

6 analyses of fauna material from the Arabian Peninsula confirm a Levantine origin for these domesticates (Uerpmann and Uerpmann 2000; Uerpmann et al. 2008).

Between 6500/6200-6000 BC, a deterioration of climatic conditions brought about increasing aridity and cooler temperatures (Gasse 2000; Mayewski et al. 2004;

Drechsler 2007; Uerpmann et al. 2009). Because of the severity of this cooling event,

Uerpmann and colleagues (2009) postulated that any population expansion into Arabia must have taken place before this change in climate. This is corroborated by evidence of late ninth to seventh millennium BC occupations in southeastern Arabia at sites such as

Jebel Faya and Nad al-Thamam (both located in modern-day Sharjah, United Arab

Emirates) (Uerpmann et al. 2009). However, moister conditions returned during the early fifth millennium BC, ushering in a widespread new cultural tradition known as the

Arabian Bifacial Tradition, or ABT. Bifacial, pressure-flaked tools and trihedral points characterized the ABT lithic industry, which persisted throughout the fifth millennium, although many local facies in bifacial production existed within this larger complex

(Uerpmann 1992; Potts 1993a; Uerpmann et al. 2006). Bifacial retouch technology and a flake-based toolkit represented a complete departure from an emphasis on blades seen in

Mesopotamia and in the Levant, but did bear typological similarity to east Africa and might suggest some (as yet unknown) link between the two regions (Carter 2002;

Drechsler 2007; Uerpmann et al. 2009).

Very little architecture existed during the fifth millennium BC, although intentionally placed stones interpreted as housing structures have been found across southeastern Arabia at sites across the United Arab Emirates (1Akab – Mery et al. 2009;

2Dalma Island – Flavin and Shepherd 1994; Beech and Elders 1999; Beech et al. 2000;

7 Popescu 2003; 3Kharimat Khor Al Manahil – Kallweit et al. 2005; Beech et al. 2006;

4Marawah Island – Beech et al. 2005), Oman (5Suwayh SWY-11 – Charpentier et al.

2000; 6Suwayh SWY-1 – Charpentier et al. 2003; Biagi and Nisbet 2006; 7Wadi Shab

GAS-1 – Usai 2006), Kuwait (8H3, As-Sabiyah – Carter et al. 1999; Carter 2002; Carter and Crawford 2003), Qatar (9Ras Abaruk 4b – de Cardi 1978; Frifelt 1989; 10Shagra –

Inizan 1988), Bahrain (11Hawar Islands – Crombe et al. 2001), and Saudi Arabia (12Ain

Qannas – Inizan 1988). Most of these are circular or semicircular in nature.

Furthermore, many sites located in the Oman Peninsula possessed circular and semicircular outlines of postholes, suggesting the former presence of structures (Biagi et al. 1984; Beech 2000; Mery and Charpentier 2002; Biagi and Nisbet 2006). The emergence of more permanent structures, while still relatively uncommon, may point to a transition towards a less nomadic way of life for these populations that involved a longer- term or seasonal occupation of sites. However, it should be noted that the majority of

Neolithic sites do not show evidence of habitation structures.

The beginning of the fourth millennium BC brought with it a climatic shift resulting in increased aridity and decreased moisture, ushering in present-day conditions and forcing the abandonment of seasonal encampments in the desert interior such as Jebel al-Buhais (Drechsler 2007; Boivin et al. 2009). Such changes compelled these groups to settle along the coast where rich maritime resources were still plentiful (Uerpmann and

Uerpmann 2003; Drechsler 2007). Dubbed the “Dark Millennium” (Uerpmann 2002;

Mery et al. 2009), this period is poorly represented archaeologically, particularly along the western coast of the Oman Peninsula (United Arab Emirates), due to freshwater shortages as wadis that had once flowed from the mountains to the sea dried up

8 (Uerpmann 2002). Conversely, however, the eastern coastline of the peninsula (Oman) was not so adversely affected and retained its larger wadis, with increasing numbers of shell midden sites reflecting either a migration of southeastern Arabian populations to this area (Uerpmann 2002) or the rapid growth of an in situ population.

Subsistence and Seasonality

Subsistence during the Neolithic period is represented by remarkably broad strategies that included marine fishing and gathering, herding, foraging, and hunting; importantly, however, agriculture had not yet emerged (Santini 1987; Phillips 2002;

Beech and al-Husaini 2005). Such dietary breadth reflects adaptations to a largely nomadic lifestyle, with time spent both inland and on the coast (e.g., Beech et al. 2006;

Cuttler et al. 2007). The ephemeral nature of most Neolithic sites, with only surface scatters of lithics, imported pottery, and faunal remains, suggests only brief seasonal visits (Potts 1997a; Kallweit 2003). Conversely, a few sites containing cemeteries or stone foundations of houses indicate longer, seasonal occupations that were consistently exploited over long periods of time (Beech 2002).

Coastal shell middens dominate the archaeological record and unsurprisingly show a heavy reliance on maritime resources (Smith 1978; Boucharlat et al. 1991a,

1991b; Vogt 1994a; Phillips 2002; Phillips and Mosseri-Marlio 2002). Small fish like sardines, sea breams, herrings, and groupers comprise the majority of fish remains, easily captured in shallow waters near the shore with either basket traps or nets (Beech 2000;

Mery et al. 2008). Stone net-sinkers recovered from many of these coastal sites affirm this supposition (Uerpmann 1992; Mery and Charpentier 2002). The importance of fish

9 may also be attested by the presence of features associated with fish curing, particularly clusters of firepits lined with burnt stone and stone drying frames containing copious amount of fish scales and bone (Smith 1978; de Cardi 1986; Flavin and Shepherd 1994;

Mery and Charpentier 2002). However, a few sites also display evidence of offshore, deep-sea fishing based on the remains of larger fish such as tuna, grouper, mackerel, jacks, and pompanos; in order to reach these pelagic species, boats would have been required (Biagi et al. 1984; Beech 2000; Beech and Glover 2005). These fish were likely caught using hooks, which have been uncovered at a small number of sites along the Gulf

(Beech 2002; Mery et al. 2008). Deep-sea fishing sites are particularly concentrated along the Gulf of Oman, a result of its deep coastal waters in contrast with the shallow waters of the Arabian Gulf (Beech 2002). Sea turtles and large marine mammals like the dugong, and less frequently dolphins and porpoises, were also hunted (Biagi et al. 1984;

Beech 2000; Uerpmann 2002; Biagi and Nisbet 2006; Mery et al. 2009).

Other strategies for procuring food along the coast included the gathering of shellfish, including molluscs such as oysters and clams as well as crabs (de Cardi 1986;

Frifelt 1989; Flavin and Shepherd 1994; Mery and Charpentier 2002; Carter 2008).

These coastal sites were predominantly inhabited during the fall and winter seasons when such resources would have been most abundant (Potts 1997a; Kiesewetter 2003; Biagi and Nisbet 2006), although a transition to a more arid climate towards the end of the

Neolithic may have prompted these populations to occupy the coast for longer periods of time, as spring and summer occupations have been determined at some sites (Beech and

Al Shaiba 2004; Mery et al. 2008).

10 While maritime resources dominated the subsistence economies of these Arabian coastal settlements, the gathering of wild edible plants also likely took place, although the poor preservation of botanical remains in this arid environment makes the recovery of such evidence rare (Phillips 2002). Nevertheless, a few sites have yielded plant remnants that provide clues into the foraging strategies of these Neolithic populations. At the late sixth/early fifth millennium BC site of DA11 on Dalma Island, located just off the coast of modern-day Abu Dhabi in the United Arab Emirates, the recovery of two carbonized date stones suggests that date fruit (Phoenix dactylifera) was consumed (Beech and

Shepherd 2001; Beech 2003a). While the possibility exists that these fruits were cultivated, it is more likely that dates were simply collected in the wild or traded (Oates et al. 1977; Tengberg 2003a; Beech and Glover 2005).

Other archaeobotanical remains have been unearthed at the fourth millennium BC site of Ra’s al-Hamra (RH)-5 and the fifth millennium BC site of RH-6 along the Gulf coast of Oman, including hundreds of charred stones from the jujube (Zizyphus sp.), a fruit rich in carbohydrates and protein, as well as two wild grass seeds (Setaria sp.) at

RH-5 (Biagi et al. 1989; Biagi and Nisbet 1992). Perhaps the most controversial find in the region, also uncovered at RH-5, was the presence of two fragments of grain classified as sorghum (Sorghum bicolor) (Constantini 1979; Cleuziou and Constantini 1980; Nisbet

1985). However, Biagi and Nisbet (1992) later retracted this assertion after SEM analysis revealed these grains as belonging to the Setaria sp. Indirect evidence in the form of artifacts also permits an assessment of plant exploitation during the Neolithic, with grinding stones found at Suwayh 2, Ra’s al-Hamra 5, and Wadi Shab in Oman, used

11 for the processing of wild cereal grains (Charpentier et al. 1998; Gaultier et al. 2005;

Cleuziou and Tosi 2007).

These economies were also largely based on the herding of domesticated animals, including sheep, goat, and cattle. At some sites, including Jebel al-Buhais, it appears that such herds were guided inland during the spring and summertime to take advantage of more abundant grazing opportunities (de Beauclair et al. 2006); during the summer months, these herds were likely taken to the Hajar Mountains where temperatures would have been significantly cooler (Kiesewetter 2003; de Beauclair et al. 2006). These mobile groups are assumed to have inhabited both coastal and inland sites because of similarities in subsistence, tool kits, herd animals, and the presence of shells, shell jewelry, and net-sinkers found in the Arabian interior (Potts 1997a; Jasim et al. 2005; de

Beauclair et al. 2006).

While the abundance of arrowheads as well as cutting and scraping tools for hide processing at many sites points to a hunter-gatherer way of life, wild game such as gazelle, oryx, wild ass, and ibex are present but did not form the main crux of subsistence

(Mery and Charpentier 2002; Phillips 2002; Kallweit 2003, 2006; Drechsler 2007).

Instead, these animals likely represent supplementary sources of protein, which in concert with maritime resources, would have assisted in preserving the herds and the highly valued secondary products they provided (Uerpmann and Uerpmann 1996; Potts 1997a).

Moreover, hunting may have also been motivated by the need to conserve adequate grazing land for their own herds; by removing competing grazers from the environment, larger herds could be supported, and the security of the group’s on-the-hoof wealth would be maintained (Kallweit 2003).

12 Direct evidence of dietary intake is extremely limited, although a few preliminary studies have been undertaken that may disclose more specific aspects of subsistence in these groups. Trace elemental analysis of human ribs and femora from the coastal site of

Ra’s al-Hamra 5 in Oman produced elevated strontium values indicative of a diet reliant on maritime resources (Palmieri et al. 1994). Similarly, stable carbon isotope analysis of two individuals from Jebel al-Buhais 18 exhibited ratios enriched in 13C (reported as

“around -16‰”), suggesting a potential marine contribution to the diet and supporting the idea that these groups may have been seasonally mobile and perhaps visited the coast on an annual basis (Uerpmann and Uerpmann 2000:48). However, as this sample size is extremely small, these data should be interpreted with caution.

Indirect evidence from the human skeleton also provides information on the diets of these Arabian Neolithic populations, although few bioarchaeological analyses have been performed in the region. An absence of dental caries and low frequencies of antemortem tooth loss in individuals from Ra’s al-Hamra 5, Jebel al-Buhais 18, and

Umm al-Quwain 2 (Table 2.1) corroborate this assessment of the importance of aquatic fauna, particularly shellfish and fish, with an associated lack of cariogenic domesticated plants including the date palm (Macchiarelli 1985, 1989; Mack and Coppa 1992; Coppa et al. 1993; Palmieri et al. 1994; Bondioli et al. 1998; Uerpmann et al. 2000, 2006;

Kiesewetter 2006).

However, a lack of carious lesions visible on tooth occlusal surfaces may be explained by the prevalence of extreme dental wear at all Neolithic cemetery sites examined. One detailed study of dental attrition from this region comes from

Macchiarelli (1989), who scored 600 permanent teeth from 49 individuals interred at the

1 3 Table 2.1. Dental pathological data from three Neolithic sites in the Oman Peninsula based on tooth counts. When tooth counts were unavailable, the number of individuals was included in parentheses. When raw data were unavailable, the descriptive term used by the author (e.g., “low”) was included. Oman UAE Jebel al-Buhais Dental Ra's al-Hamra 5 Umm al-Quwain 2 Pathology 18 n # % n # % n # % 0 Caries nd (144 ind) 0 0% 2499 0 0% nd (41 ind) 0 % 4 n AMTL nd (low) nd nd 2147 4 2% nd (low) d nd 130 94 6 n LEH 1392 8 % 1349 1 5% nd d nd 8 n Calculus nd nd nd 1669 2 4.9% nd d nd *nd = no data **“low” = description of AMTL frequency, no data given

coastal cemetery of Ra’s al-Hamra 5 using methods established by Scott (1979) and

Molnar (1971). He noted heavy wear patterns in both males and females, particularly on the posterior teeth, in older as well as young adults (Macchiarelli 1985, 1989). In order to compare these results with attrition patterns from later Bronze and Iron Age sites,

Littleton and Frohlich (1993) designated Scott’s (1979) severest grades of molar wear, 9 and 10, as corresponding to Brothwell’s (1981) grades 5+ to 7, and found severe wear in

12.3% of individuals.

As at Ra’s al-Hamra 5, dental wear from the small burial ground of Umm al-

Quwain 2 is described as severe (with no associated data) and is also positioned on the coast (Strongman 1993:3). While attrition is prevalent across the Oman Peninsula, it appears more severe among these coastal populations than at Jebel al-Buhais 18

(Kiesewetter 2006). From this inland site, Kiesewetter (2006) evaluated 130 adult

14 individuals (n=2,332 permanent teeth) using methodology outlined by Brothwell

(1981:72) and found only moderate attrition in both sexes, with the most severe wear

(defined as grades 5+ to 7) occurring in only 7.2% of those individuals sampled, and occurring most frequently on the molars. This represents a lower degree of attrition than at coastal Ra’s al-Hamra 5.

Attrition differences between coastal and inland populations may be explained by the geographically-specific foods available at each locale. Foodstuff typically encountered on the coast, including dried fish and shellfish, produced considerable masticatory stress, whereas inland populations would have had decreased access to maritime resources (Macchiarelli 1989; Kiesewetter 2006). All groups were likely affected by the introduction of windblown sand into food, which contributed considerably to the severe wear present in the general populace (Brothwell 1963; Molnar 1971).

Moreover, Kiesewetter (2006) noted that the majority of teeth at Jebel al-Buhais 18 experienced horizontal occlusal surface wear, as opposed to oblique or concave attrition, suggestive of the processing of tougher foods associated with a hunter-gatherer diet

(Smith 1984). This fits with what is known about Neolithic populations in the Oman

Peninsula during this time, as domesticated plants had not yet been introduced.

In addition to dental caries and attrition, the presence of dental calculus at Jebel al-Buhais 18 may suggest the consumption of foods high in protein, supporting the conclusion that these populations relied heavily on both marine resources and herds

(Lieverse 1999; Uerpmann et al. 2006). Low levels of calculus at the site have been attributed to post-depositional processes as well as the potential for advanced hygienic practices, including the use of toothpicks (Kiesewetter 2006) (Figure 2.1).

Figure 2.1. Toothpick groove on a maxillary incisor from individual HD may indicate advanced oral hygiene practices at Neolithic Jebel al-Buhais 18 (from Kiesewetter 2006:219).

Mortuary Practices

The discovery of burial grounds and the recovery of human remains dated to the

Neolithic in the Oman Peninsula are rare; in fact, most Late Stone Age ‘sites’ are represented by simple surface lithic scatters and ‘Ubaid pottery sherds characteristic of ephemeral camps inhabited only briefly by these hunter-herders (e.g., Smith 1978; Edens

1982; Frifelt 1989; Uerpmann 1992; Kallweit 2003; Kallweit et al. 2005; Cuttler et al.

2007). While a handful of isolated human bones and burials have been found scattered across the Arabian Gulf coast, including in Oman (Charpentier et al. 2003; Biagi and

Nisbet 2006), Qatar (Tixier 1980; Midant-Reynes 1985; Tixier 1986; Hublin et al. 1988;

Inizan 1988), and the UAE (Boucharlat et al. 1991b; Vogt 1994b; Beech et al. 2005;

Kutterer and de Beauclair 2008), a number of cemeteries, some with a substantial number of interred individuals, have now been uncovered and illustrate practices associated with treatment of the dead (Figure 2.2; Table 2.2). 16

Figure 2.2. Map illustrating the location of Neolithic cemeteries unearthed in the Oman Peninsula.

Ra’s al-Hamra 5 (RH5)

The early fourth millennium necropolis at Ra’s al-Hamra 5 in northern Oman sat at the edge of a Neolithic coastal settlement and likely contained more than 200 burials, with approximately 171 recovered thus far (Bondioli et al. 1998). The graveyard was probably in use for around 200 years (Biagi and Nisbet 2006). Graves consisted of shallow oval pits and may be classified into four distinctive types based on construction materials (Biagi et al. 1984). Type I graves consisted of simple pits with either no limestone covering or one to two blocks placed over the body; Type II graves were paved with stone from the nearby Wadi Aday and are characterized by the presence of a

17 Table 2.2. Neolithic burials and cemeteries of the Oman Peninsula.

Site Country # of Individuals References Biagi et al. 1984; Coppa et al. 1985; Santini 1987; Coppa et al. 1990; Palmieri et al. 1994; Ra's al-Hamra (RH-5) Oman 171-200+ Salvatori 1996; Bondioli et al. 1998; Biagi and Nisbet 2006 Biagi et al. 1984; Coppa et al. 1985; Santini Ra's al-Hamra (RH-10) Oman 32 1987 Wadi Shab (GAS-1) Oman 14 Tosi 1975; Tosi and Usai 2003; Gaultier et al. 18 2005; Biagi and Nisbet 2006; Usai 2006 Suwayh (SWY-1) Oman ? Charpentier et al. 2003; Biagi and Nisbet 2006 Khor (Khor FPP) Qatar 1 Tixier 1980; Tixier 1986 Khor Qatar ? Inizan 1988:95 Khor Qatar isolated bone Hublin et al. 1988 Khor Qatar ? Midant-Reynes 1985; Tixier 1986 Umm al-Quwain 2 (UAQ-2) UAE at least 42 Strongman 1993; Phillips 2002 Marawah Island (MR-11) UAE 1 Beech et al. 2005 Jebel al-Buhais (BHS-18) UAE 500-1000 Kiesewetter et al. 2000; Kiesewetter 2003; Jasim et al. 2005; de Beauclair et al. 2006 Jebel Faya (FAY-NE15) UAE at least 3 Kutterer and de Beauclair 2008 Jazirat al-Hamra UAE isolated burials Vogt 1994a Al-Madar UAE disturbed burials Boucharlat et al. 1991a

1 tumulus stone cover; Type III graves were composed of enormous limestone coverings; and Type IV pit graves contained wadi pebble fill and were topped with flat rocks (Coppa et al. 1985; Salvatori 1996). A significant amount of faunal material found above each grave may suggest some kind of ritual or funerary meal (Biagi et al. 1984).

Both primary and secondary burials were utilized in disposing of the dead. In primary burials, individuals were placed in a flexed position, predominantly on their right sides (Coppa et al. 1990). Primary burials typically involved a single inhumation, but double and multiple interments within the same pit did occur (Salvatori 1996). No sex or age patterns have been discerned from these multiple burials, with individuals of all ages and sexes buried together (Salvatori 1996). Arms were deliberately positioned, with the right hand often placed directly in front of the face and sometimes found grasping a shell valve or pearl (Santini 1987; Bondioli et al. 1998). Secondary burials represent bodies that were allowed to decompose in primary pit-graves before being removed from the original grave and reburied in a disarticulated fashion (Macchiarelli 1989; Bondioli et al.

1998). In most cases of secondary inhumation, only the largest bones, including the crania and long bones, were represented (Figure 2.3) (Salvatori 1996). Three phases of burial are evident, with burials gradually become collective with increasing frequency over time (Bondioli et al. 1998; Phillips 2002). Regardless of burial type, all individuals were oriented in a northeast-southwest direction, with the head facing either northwest or southeast (Coppa et al. 1985). However, while males, females, and juveniles were well represented in the cemetery, newborns and infants were rare, normally found with an adult female assumed to be the mother (Bondioli et al. 1998).

Figure 2.3. Secondary burial from the site of RH-5, Oman (from Cleuziou and Tosi 2007:76).

Grave goods were present but limited in typology (Salvatori 1996). While everyday objects such as awls, net sinkers, and shell fish-hooks were rarely recovered from graves, personal ornaments dominated these assemblages and consisted of necklaces, bracelets, earrings, and pins made of shell, stone, and bone (Salvatori 1996).

Faunal remains interpreted as ritual food offerings, including molluscs, fish, and green sea turtles, accompanied almost every burial (Bondioli et al. 1998). In particular, a unique relationship existed with turtles, with entire turtle skulls carefully placed on human crania in two separate burials (Biagi et al. 1984; Coppa et al. 1985). Again, no sex or age patterns could be deduced based on the distribution of these goods (Coppa et al. 1990; Salvatori 1996).

20 Ra’s al-Hamra 10 (RH10)

A second cemetery at Ra’s al-Hamra, also dating to the late fifth/early fourth millennium BC, has been designated RH10 and displayed remarkable similarities with that of the larger, contemporary necropolis of RH5 (Santini 1987; Salvatori 1996). The burial ground houses 26 graves containing 32 individuals, with adult males, adult females, juveniles, and infants present, although infants are once again underrepresented

(Santini 1987). Pit graves were uniformly constructed in the Type IV style also seen at

RH5 (Salvatori 1996). All burials were primary inhumations, with each individual laid only on the right side in a flexed position (Santini 1987; Salvatori 1996). As at RH5, upper limb position most commonly entailed the placement of the right hand next to the head (Santini 1987). After placement into the grave, six of these 32 individuals were cremated in situ (Tosi 1981).

While identical to RH5 in terms of typology, grave goods are more elaborate at

RH10 but are found in just 60% of graves (although a lack of grave goods in the other

40% may in part be a product of the perishable nature of some offerings) (Santini 1987).

Once again, two turtle crania were placed in the graves of an adult male and an adult female, here positioned on the chest (Coppa et al. 1985; Santini 1987).

Wadi Shab (GAS-1)

The early fourth millennium BC coastal site of GAS-1 is located near Tiwi, Oman on the Wadi Shab (Tosi 1975). In addition to its three phases of settlement, a burial ground containing 14 individuals has been excavated (Tosi and Usai 2003; Biagi and

Nisbet 2006). Bodies were arranged in a tightly flexed position and placed into small

21 circular pits (Gaultier et al. 2005) (Figure 2.4). These pits were generally covered with stone slabs and less commonly included circular arrangements of stones or pebble/stone fill (Gaultier et al. 2005). Despite evidence of on-site soapstone jewelry manufacturing in all settlement layers, few grave goods accompanied these interments, with a shell fishhook, flint blade, shell, and scanty fish and turtle remains comprising the total funerary offerings for all burials present (Gaultier et al. 2005; Usai 2006). Bodies were variably laid on right and left sides, with heads facing towards the east (Usai 2006). Poor bone preservation made basic paleodemographic reconstructions difficult, but Gaultier and colleagues (2005) identified adults of both sexes, as well as a two-year-old subadult.

While the majority of burials represent primary inhumations, two secondary burials of

Figure 2.4. Flexed burial from the fourth millennium BC site of GAS-1 at Wadi Shab, Oman (from Cleuziou and Tosi 2007:86).

22 three individuals were also recorded; in Grave 2, secondary reburial included only the skull and long bones, but in Grave 3, two individuals retained partial articulation of the lower limb (Gaultier et al. 2005).

Jebel al-Buhais 18 (BHS18)

The vast Neolithic cemetery at Jebel al-Buhais 18 at the foothills of the western

Hajar Mountains dates to the mid-fifth millennium BC (Kiesewetter 2003). Unlike most sites and cemeteries during this time, Jebel al-Buhais was located in the interior of the

Oman Peninsula, approximately 60 km from the nearest coastline (de Beauclair et al.

2006). This cemetery contained between 500 and 1000 individuals interred close to a nomadic spring camp (Kiesewetter 2003; Jasim et al. 2005). As with RH5 and UAQ2, hearths and/or faunal remains scattered throughout the graveyard may be indicative of a funerary, ritual meal, although the placement of food items directly into the graves is not evident (Kiesewetter 2003).

Primary burials entailed the interment of between one and five individuals in shallow pit-graves with no stone coverings. Bodies were placed in a flexed position, normally laid on their right sides, and oriented east-west with heads to the east (Figure

2.5) (Kiesewetter et al. 2000; Kiesewetter 2003, 2006). All individuals from primary graves were adorned with a variety of personal ornaments, including pendants, anklets, headbands, necklaces, and belts made of beads, which in turn were crafted from limestone, shell, coral, chert, carnelian, and agate (Kiesewetter 2003). Secondary burial involved the defleshing of individuals in primary pit-graves before collecting the larger

Figure 2.5. Collective primary burial at Jebel al-Buhais 18 with five articulated individuals (GI, AK, AL, GG, and FO, respectively) who likely died around the same time and were interred together (from Kiesewetter 2006:121).

bones (e.g., long bones, crania, innominates, scapulae) for reburial, and no jewelry was recovered from any of these burial types (Figure 2.6) (Kiesewetter 2003). It is postulated that individuals in secondary burial contexts were those that died away from the site, were buried, and then exhumed before the group’s springtime return to Jebel al-Buhais for final burial (Jasim et al. 2005). As with primary burials, secondary long bone piles and crania were oriented east-west, with skulls facing east (Kiesewetter et al. 2000). The cemetery was represented by equal numbers of males and females, although only around

20% of its occupants were children; however, no age or sex patterns in terms of burial or grave goods were evident (Kiesewetter 2003). 24

Figure 2.6. Secondary disarticulated burial at Jebel al-Buhais 18 (individual BG), represented only by a cranium, long bones, and innominates (from Kiesewetter 2003:37).

Umm al-Quwain 2 (UAQ2)

Located on the northern border of the Emirate of Umm al-Quwain, the fifth millennium BC site of UAQ2 consists of a shell midden containing a cemetery with a minimum of 42 individuals (Strongman 1993; Phillips 2002). All ages and sexes are represented in the sample, and all individuals were positioned in a flexed position and placed variably on either their left or right sides (Strongman 1993; Phillips 2002) (Figure

2.7). Only nine burials were retained as primary, fully articulated inhumations; previously primary burials were exhumed and re-deposited as disarticulated secondary burials dominated by crania and long bones on the periphery of the burial grounds

Figure 2.7. Fifth millennium BC collective burial from Umm al-Quwain 2 (from Cleuziou and Tosi 2007:50).

(Phillips 2002). Hearths encircling these pit-graves, as well as the presence of large amounts of burned faunal bone, may point to ritual meals taking place at the time of interment of the deceased (Phillips 2002).

As is evident from these descriptions, a great deal of similarity exists between these cemeteries in terms of mortuary patterns, consistent with the presence of a single cultural tradition. This idea ties in well with corroborating evidence of lithic affinities witnessed across the Arabian Peninsula and suggests that an integrated, widespread network was already in place by the Neolithic period, albeit with local traditions and facies (Potts 1997a). Moreover, the presence of both sexes and all age categories, similar interment treatments, and no evident patterns in the association of grave goods with particular sex or age groups within these cemeteries suggests, at least in a mortuary

26 context, a relatively egalitarian lifestyle (Benton 2006). Nevertheless, the placement of turtle remains on the bodies of a few individuals from both Ra’s al-Hamra cemeteries may point to the existence or emergence of some social stratification.

Interregional Exchange

Evidence for trade with Mesopotamia during the late sixth and fifth millennium

BC comes from the considerable distribution of ‘Ubaid pottery sherds across the Arabian

Gulf. ‘Ubaid pottery originated in southern Mesopotamia during its ‘Ubaid period (ca.

5300-4000 BC), manufactured at sites including Ur, al-‘Ubaid, and Eridu (Nayeem 1990;

Jasim 1996; Pollock 1999). The ‘Ubaid period of Mesopotamia corresponds with the

Arabian Bifacial Tradition/Neolithic period of southeastern Arabia and may be divided into five phases (‘Ubaid 0-4) as well as a terminal (‘Ubaid 5) phase; of these, contact with the Arabian Peninsula took place during ‘Ubaid 2-4 (Oates 1986; Potts 1993a). In addition to sherds found across the Fertile Crescent in what is now Iraq, Iran, Syria, and

Turkey, ‘Ubaid wares have also been recovered from more than 60 Neolithic Gulf sites in

Bahrain, Kuwait, Qatar, Saudi Arabia, and the UAE (Oates 1986; Carter 2003a, 2006).

Petrographic and neutron activation analyses confirm that these ceramics were not produced locally but were manufactured in southern Mesopotamia (Oates et al. 1977;

Oates 1986; Frifelt 1989; Roaf and Galbraith 1994). No local, Arabian Neolithic ceramic industry has been discovered (Potts 1997a).

Many different hypotheses have been put forth to explain the presence of ‘Ubaid pottery in Arabia. Based on initial excavations in the region, archaeologists including

Masry (1974) and Piesinger (1983) originally speculated on the shared ethnic and cultural

27 origins of Gulf populations with those in Mesopotamia, as evidenced by continued systems of exchange during the ‘Ubaid period. This “common origin” hypothesis has since been dismissed, not only because of substantially different subsistence strategies

(Carter 2002) but also due to the spatial distribution of ‘Ubaid sherds and their relationship to the ABT. While hundreds of Neolithic sites with ABT artifacts have been identified, not all of these sites possess ‘Ubaid wares; conversely, all recovered ‘Ubaid sherds are found in associated with ABT sites, suggesting that this pottery was intrusive

(Uerpmann and Uerpmann 1996).

Following this hypothesis, it was also suggested that coastal sites with ‘Ubaid sherds may have represented ephemeral camps occupied briefly by Mesopotamian seafarers who had traveled to the Lower Gulf, presumably on fishing expeditions (Oates et al. 1977; Potts 1990; Oates 1993). Subsequently, the presence of this pottery was not a reflection of a trade relationship but simply ceramic debris left behind. This hypothesis has recently undergone intense criticism, as it intrinsically assumes that populations along the Arabian Gulf coast lacked any agency and thus played no role in the deposition of these artifacts (Stein 2002a; Carter 2006). Moreover, the occurrence of ABT lithics so different from the blade technology characteristic of Mesopotamia at sites also containing

‘Ubaid pottery casts serious doubt on this assertion. It is extremely unlikely that these fishermen would have transformed their tool kit so drastically; instead, it seems probable that the ABT corresponded to an independent, local source (Uerpmann and Uerpmann

1996).

Consequently, the vast majority of current hypotheses focus on the active trade relations between Mesopotamia and the Gulf coast. As early as 1977, Oates and

28 colleagues proposed a model of seafaring Mesopotamian merchants sailing along the

Arabian coast and trading with local populations there. Others (e.g., Masry 1974;

Haerinck 1991) have theorized that land-based and not maritime exchange took place, with trade occurring between small settlements along land routes so that ‘Ubaid pottery slowly made its way down the coast. However, this would necessitate evidence of the continuous distribution of ‘Ubaid sherds along coastal sites, evidence that has not been recovered thus far. Additionally, the absence of these sherds at inland sites in the Lower

Gulf and the presence of sherds on island sites seems to preclude this possibility (Oates

1986; Uerpmann and Uerpmann 1996).

Other forms of evidence also suggest early maritime trade along the coast. At the site of H3 in As-Sabiyah, Kuwait, fragments of bitumen coated with barnacles on one side and impressions of reeds on the other indicate that this material was actively used to waterproof sea vessels, perhaps utilized to transport goods between Mesopotamia and the

Gulf (Figure 2.8) (Carter 2002, 2003a). The origins of these bitumen slabs were traced not to Mesopotamia but southwards to the site of Burgan, Kuwait, pointing to the emergence of a seafaring culture independent of (but in close contact with) ‘Ubaid

Mesopotamia (Carter 2006). At the same site, representations of seagoing vessels in the form of a ceramic boat model and a painted image of a boat on pottery confirm the presence of maritime technology and may hint at its importance (Carter 2002, 2003a).

The presence of these types of artifacts at Neolithic coastal sites in Arabia illustrates that

Mesopotamia did not completely dominate every aspect of these trade relations; instead,

Arabia likely played an active role in exchange.

Figure 2.8. Bitumen slab from H3, As-Sabiyah, Kuwait, displaying reed impressions on its inner surface (c) and mature barnacles on its outer surface (d) (from Connan et al. 2005:29).

Two hypotheses involving maritime trade remain. Carter (2002, 2003a, 2003b,

2006) argues that trade with the Gulf can only be evaluated by dividing this region into central and lower portions. Because of the significant amount of ‘Ubaid pottery found in the Central Gulf (Saudi Arabia and Kuwait) as compared to smaller quantities recovered in the Lower Gulf (Bahrain, Qatar, and the United Arab Emirates), Carter (2002, 2003a,

2003b) suggests that while a highly developed exchange system was in place between

Mesopotamia and the Central Gulf, this was not the case with Mesopotamia and the

Lower Gulf. Instead, the Central Gulf effectively became a “middleman,” redistributing

‘Ubaid wares to the Lower Gulf in a “down-the-line exchange” manner (Carter 2002:24).

While Uerpmann and Uerpmann (1996) recognize this disparity in the distribution of

‘Ubaid artifacts, they contend that the lesser quantities of traded pottery in the Lower

Gulf may be a result of regular, seasonally-based visits by Mesopotamian seafarers. This hypothesis states that ‘Ubaid traders arrived in the Lower Gulf only during certain times of the year which corresponded with the autumn/winter occupation of the coast by these

30 nomadic hunter-herders (Uerpmann and Uerpmann 1996; Biagi and Nisbet 2006). When these local populations returned inland with their herds, direct trade would be discontinued until the next season.

Many questions also surround the types of goods that were prepared in exchange for ‘Ubaid wares. Pearls and shell jewelry have both been suggested as valuable commodities for trade, and multiple jewelry workshops have been found along the

Arabian coastline (Frifelt 1989; Potts 1993a; Phillips 2002). However, no such items have been recovered from ‘Ubaid Mesopotamia (Carter 2002). Perishable goods may have also represented an important trade product, including dried fish or hides (Frifelt

1989; Beech 2002; Carter 2003a). Furthermore, secondary products from domesticated herds such as milk and butter, as well as the living animals themselves, may have held some value for ‘Ubaid traders (Kallweit 2003). Finally, as Mesopotamia lacked many natural resources, it is possible that minerals or timber may have been sought after (Oates

1993; Carter 2008).

In addition to Mesopotamian wares, other foreign artifacts appear in the archaeological record in the Gulf during this time, including pottery from eastern Iran

(Biagi et al. 1984), obsidian from Anatolia and Azerbaijan (Edens 1982; Frifelt 1989;

Jasim 1996), and carnelian and amazonite from India (de Cardi 1986; Frifelt 1989;

Kiesewetter et al. 2000). While it is unclear if these items were traded directly with other regions, it seems more likely that Mesopotamian traders distributed these goods along with ‘Ubaid pottery (Phillips 2002).

By the early fourth millennium BC, evidence of contact with Mesopotamia abruptly ends with the commencement of the Mesopotamian Uruk period (ca. 4000-3100

31 BC) (Phillips and Mosseri-Marlio 2002). No artifactual evidence of trade of any kind exists in the Oman Peninsula; even in the Central Gulf, only a few sherds of Uruk pottery have been recovered in eastern Saudi Arabia and Bahrain (Crawford 1998). It is unclear why this relationship was severed, although speculation has focused on increasing aridity

(Parker et al. 2006) and/or changes in social organization (Wright and Johnson 1975;

Pollock 2001). During this time, the sociopolitical organization of Mesopotamia was transformed, becoming increasingly complex with the development of a centralized economy (albeit with numerous, relatively independent, competing polities) and resulting in the emergence of urbanization, social and settlement hierarchies, and the first city- states (Nissen 1988; Oates 1993; Pollock 1992, 1999, 2001; Stone 1995; Potts 2004a;

Yoffee 2005). This complexity is further demonstrated by the existence of ‘colonies,’ outposts founded outside of Mesopotamia in Anatolia, Iran, and Syria, whose existence seems related to acquiring natural resources absent in Mesopotamia itself (these colonies will be discussed in further detail in Chapter 4 as part of the trade diasporas model)

(Oates 1993; Stein 1999a). Trade was thus largely limited to the north/northwest of

Mesopotamia and appears to have only rarely reached southwards towards the Arabian

Peninsula (Crawford 1998).

Bronze Age: Hafit Period (ca. 3100-2500 BC)

The Hafit period extends from ca. 3100-2500 BC and marks the beginning of the

Bronze Age in the Oman Peninsula. The name ‘Hafit’ comes from Jebel Hafit, a mountain near the Oasis of Al Ain in the United Arab Emirates, close to the Omani border, where the first excavations of tombs from this era took place. This episode is

32 defined primarily by the emergence of funerary monuments as well as the apparent reinvigoration of trade with Mesopotamia (Potts 1990, 1997a, 2001).

Settlement and Subsistence

As in the Neolithic, settlement sites during the Hafit period (with one possible exception) have not been found (Potts 1997a). Until recently, the preponderance of Hafit cairns found in the inland regions of southeastern Arabia, particularly concentrated around the foothills and gravel plains of the Hajar Mountains across both the United Arab

Emirates and Oman, led archaeologists to conclude that Hafit populations restricted themselves to the interior of the Oman Peninsula (Potts 1990; Hellyer 1998). Recently, however, evidence from radiocarbon dating at a coastal shell midden in the al-Daith region of the Emirate of Ra’s al-Khaimah suggests that the coast was in fact re-occupied during the Hafit period (Parker and Goudie 2007). Two other shell middens at Akab and

Point 81 in the Emirate of Umm al-Quwain also produced radiocarbon dates similar to that of al-Daith (Boucharlat et al. 1991a; Vogt 1994a; Parker and Goudie 2007). No corresponding settlements or burial grounds have been uncovered to pair with these middens.

The scarcity of Hafit discoveries along the Gulf coast may be a product of environmental change. Biogeochemical evidence from sediments and organic material reveals the arrival of increasingly moister conditions with the start of the Early Bronze

Age that continued throughout the Hafit and Umm an-Nar, contributing to a rise in sea level that may have destroyed or concealed coastal middens and/or settlements (Larsen

1983a; Potts 1993a; Hellyer 1998; Cullen et al. 2000; Staubwasser et al. 2003; Parker et

33 al. 2006). Moreover, the end of the third millennium witnessed a significant intensification in aridity beginning around 2100/2000 BC, leading to vegetation loss and dune reactivation that further contributed to the burial of coastal sites under the sand

(Parker and Goudie 2007).

A potential exception to the lack of settlement in southeastern Arabia during this period lies at the site of Hili, located in the Al Ain Oasis of Abu Dhabi, which contains a settlement area comprised of multiple mounds (Cleuziou 1982). Of these mounds, only one, Hili 8, dates to the Hafit, with occupation beginning around 3100 BC (Cleuziou

1980, 1996). The Hafit at Hili 8, with a site-specific designation of Period I, was represented by numerous structures, including Building III, a large square tower constructed of mudbricks, two later additions to this tower (Buildings V and VI), a number of trenches and hearths, a trash pit, and a central well (Cleuziou 1980, 1982,

1989). Potts (1993) has proposed that the tower may have been a defensive structure, protecting the stone-lined well that lay at its center, perhaps foreshadowing the fortress towers of the later Umm an-Nar period that guarded this valuable water supply (Figure

2.9).

However, the assignment of Hili 8 to the Hafit is based on the radiocarbon dating of two pieces of wood charcoal, dates which have been called into question by some because of a tendency by ancient peoples to recycle wood initially used as timber into fuel (Potts 2001). All other dates derived from charcoal at the site occur at least 500 years later, hence dating the site to the ensuing Umm an-Nar period (Potts 1997b).

Moreover, a multitude of circular Umm an-Nar monumental graves just west of Hili 8,

Figure 2.9. Hili 8 plan during period I (ca. 3100-2700 BC), with Well 2 protected by a square tower with rounded corners (from Cleuziou and Tosi 2007:144, Figure 147).

with a notable absence of Hafit mortuary architecture, may lend credence to suspect the validity of these radiocarbon dates in determining time of occupation (Potts 2001;

McSweeney et al. 2008).

Despite the hundreds of Hafit stone cairns present across the Oman Peninsula

(discussed in greater detail below), the identification of only one potential settlement with no associated Hafit cemetery suggests that any residential areas in existence were positioned away from these burial grounds (Potts 1997a). Alternatively, as proposed by

Cleuziou (1989) at Hili 8, settlements may simply have been an assemblage of barastis,

35 structures constructed from perishable materials like palm fronds that would not have survived archaeologically in such an arid climate.

As Hafit settlement data is scarce, little data exists regarding the subsistence strategies of these populations. While the presence of three coastal shell middens found in the Emirates of Ras al-Khaimah and Umm al-Quwain implies that littoral gathering and fishing likely represented an important means of subsistence acquisition during the

Hafit, some information may also be gleaned from the material culture and food remains found at Hili 8. Charred archaeobotanical remnants and impressions found on site included a number of domestic cereals identified as barley, emmer, and wheat; numerous fruits such as the date, jujube, and melon; and other species like the garden pea

(Costantini 1979; Cleuziou and Costantini 1980; Cleuziou 1982, 1992; Tengberg 1998,

2003b). The recovery of hundreds of date stones coupled with the presence of wood from this palm suggests that the date palm played an important role in the subsistence economy at Hili 8 and may be the first evidence of oasis agriculture in the Oman

Peninsula (Cleuziou 1982).

In such an oasis, or palm garden (bustan) system, date palm is utilized not only for its fruits but also its ability to provide shade for other cultivated plants, including melons, jujube, and garden peas at Hili 8 (Cleuziou 1996; Potts 1993a; Tengberg 2003b).

This type of agricultural development represents an adaptation to the harsh, arid climate of this region by the efficient use of space and water (Tengberg 2003a, 2003b). While the Hafit corresponds to a period of considerable humidity and moisture relative to the preceding Neolithic, annual rainfall nevertheless proved inadequate for more intensive agricultural field systems to develop in this region. Subsequently, the localization of

36 these crops onto a discrete area with a steady supply of groundwater generated a viable means of consistently and reliably producing domestic cereals, fruits, vegetables, and legumes (Cleuziou 1996). Because oases possess a relatively high water table, the construction of wells provided a constant water source from which to irrigate crops; moreover, trenches uncovered at Hili 8 may point to a rudimentary irrigation technique reminiscent of later falaj systems (Cleuziou 1982, 1989, 1996, 1997; Wilkinson 1983).

At Hili 8, cereal crops better adapted to more arid conditions appear to have been cultivated in close proximity to the palm cultivars and would have been harvested during the summer months (Cleuziou and Costantini 1982; Cleuziou 1989, 1992). Of particular interest at Hili 8 is evidence of domesticated sorghum from both mudbrick impressions and charred grains dating to the early third millennium BC (Cleuziou and Costantini

1980; Potts 1993a). Initially, this date placed domesticated sorghum in Arabia well before its recovery from the Indus Valley and East Africa, despite persistent claims that sorghum was first domesticated in Africa. While cultivation of wild sorghum seems to have taken place as early as the 9th millennium BC in the Nile Valley, confirmation of domesticated sorghum in East Africa does not occur until the end of the first millennium

BC (McIntosh 1995; Rowley-Conwy et al. 1997; Neumann 1999; Kimber 2000).

However, the identification of domesticated sorghum at Hili 8 has been accepted by few based on morphological indicators (Costantini 1984a, 1985; Willcox 1992, 1994; Potts

1994a; Rowley-Conwy et al. 1997). Nevertheless, while sorghum in its wild or domestic form may not have been present during the Hafit, second millennium grain imprints from the site of Wadi Yanaim in Yemen imply that sorghum did in fact spread into the Arabian

37 Peninsula well before its domestication in East Africa and was likely introduced to the

Oman Peninsula via southern Arabia (Costantini 1984b; Tengberg 2003b).

The origins of the plant domesticates unearthed in southeastern Arabia must also be addressed. It is generally agreed upon that because the date palm is indigenous to the

Oman Peninsula, local domestication took place and cemented its role as a primary cultivar (Potts 1994a; Tengberg 2003a, 2003b). Besides the date palm, however, there is no evidence that wild progenitors of barley, emmer, or wheat were native to this region and no indication that these cereals were domesticated in the area, as the sudden appearance of these grains in domesticated form suggests that these crops were introduced from a neighboring region (Cleuziou and Tosi 1989). At Hili 8, two of the primary types of cultivated domesticated cereals, hulled barley and naked wheat, comprise some of the main crops also present in the Indus Valley at the beginning of the third millennium BC, possibly indicating some form of contact across the Gulf, or more plausibly, along Indo-Iranian trade routes (Tengberg 1999, 2003b). These domesticates were most likely brought to the Oman Peninsula as part of established exchange systems already in place with Mesopotamia, as evidenced not only by the utilization of cereals like emmer in both regions but also by the presence of other imported Mesopotamian goods at Hili 8 and in numerous Hafit tombs, including Jemdet Nasr and Early Dynastic pottery (Renfrew 1984; Potts 1986a, 1994). These interregional relationships are discussed in greater detail below.

Like botanical remains, faunal material is scarce and comes only from the settlement at Hili 8. Hundreds of bones, many from a trash pit associated with the Hafit period, illustrate the importance of domesticated sheep, goats, and cattle (Cleuziou 1989;

38 Potts 1993a; Hellyer 1998). Although not described in detail, these bones were reported as displaying osteological features characteristic of repetitive, monotonous activities that may have included towing well water for irrigation or plowing fields (Cleuziou 1989).

Moreover, the additional presence of a few wild camel, equid, and bird bones indicates that hunting played a small role in contributing to dietary intake. In summary, subsistence data from Hili 8 suggests that a highly complex agro-pastoral society engaged in both oasis agriculture and herding was fully in place by the Early Bronze Age.

Further clues into the daily lives of the individuals occupying the Oman Peninsula during the Hafit may also be obtained from the human skeletons recovered from the tombs. Unfortunately, due to extremely poor preservation in this arid environment, the fragmentary nature of these commingled remains, and frequent tomb re-use, very few analyses have been possible, and little usable data exists from this period (Benton 2006;

Abu Dhabi Culture & Heritage 2010). Additionally, those reports in existence give little

(if any) raw data; rarely mention the number of individuals or dental/skeletal element under consideration; discuss these elements using vague, unusual, or non-standard terminology; sometimes neglect to mention the site under investigation; and make broad generalizations about dietary intake based on inadequate, small sample sizes. With these caveats in mind, this work may shed some light onto the Hafit lifestyle.

Analysis of skeletal remains from the Hafit have been limited to two tombs

(Tombs 1303A and 1317D) from the site of Jebel Hafit, part of the Al Ain Oasis of Abu

Dhabi in the interior of the United Arab Emirates. Dating to the early third millennium

BC, 12 teeth from a total of three individuals displayed evidence of 11 carious lesions

(Højgaard 1985). From this data, Højgaard (1985) maintained that these individuals

39 engaged in oasis agriculture and were more sedentary than previous populations.

According to further review by Kunter (1996), these inland inhabitants consumed cariogenic, domesticated crops like the date palm. Initially, a second tomb complex containing some skeletal material at the site of Maysar in Oman was thought to date to the Hafit period, although more recent interpretations have assigned its skeletons to the subsequent Umm an-Nar period (Weisgerber 1980, 1981; Kunter 1983, 1996;

Macchiarelli 1985).

Mortuary Practices

Major changes in mortuary practices took place at the end of the fourth millennium BC with the start of the Hafit period, differing considerably from the primary and secondary cemetery inhumations characteristic of the Arabian Neolithic. These innovative grave complexes are characterized by above-ground tombs built with unworked limestone piled atop a false dome and contained a single chamber where multiple individuals were interred (Bibby 1965; Potts 1990, 1997a; Nayeem 1996). This limestone was arranged in two to three thick, concentric ring walls, so that while the average diameter of the circular inner tomb chamber was only 1-2.5 m, the outer diameter of the structure reached 7-12 m (Potts 1993a, 1997a). Due to the extensive plundering of these graves in antiquity, assessing the number of individuals placed within them has proven a difficult task. Nevertheless, the characteristic ‘keyhole’ shape of the interior chamber and its access point to the exterior (Figure 2.10) convey the restricted dimensions of the burial chamber and suggests that it would have been difficult to

Figure 2.10. Typical Hafit-type grave from Jebel Hafit, United Arab Emirates. Note the keyhole entrance/chamber design (from Potts 2001:38, after Cleuziou et al. 1978: Pl. 15).

accommodate more than 12 individuals, making these collective burial sites rather small in comparison to later tomb membership (Potts 2001).

In addition to these ‘Hafit-type’ or ‘cairn-type’ tombs, another contemporaneous tomb type, denoted ‘beehive’ graves, was also circular and contained single interior chambers capped by a false dome but were constructed from smaller pieces of limestone, had smaller average external diameters ranging from 8-9 m, and were likely slightly taller than their Hafit-type counterparts (Potts 1990; Hellyer 1998) (Figure 2.11). Beehive tombs were not as common as the Hafit-type tombs, although numerous cemeteries are dispersed throughout the interior of the Oman Peninsula, including the necropolises at

Bat and Al Ayn in Oman. Now UNESCO World Heritage Sites, Bat and Al Ayn contain hundreds of preserved beehive tombs, some of which lay undisturbed and retain their

Figure 2.11. Beehive (d) and Hafit-type (e) tomb reconstructions (from Potts 1990:75).

original height and overall structure (Figure 2.12) (Potts 1990; Orchard and Stanger

1994; Nayeem 1996).

Recognized Hafit- and beehive-type tombs have been identified in Oman at the inland sites of Bat, Al Ayn, and Ibri in the Dhahirah region, at the Oasis of Zukayt in the

Ad Dakhiliyah region, and at Bisya, Wadi Samad, Maysar, Bilad Bani bu Hassan, Tawi

Silaim, Wadi Jizzi, Mazyad, Siya, Shir Jaylah, Ra’s al-Hadd, Ra’s al-Jinz, and Wadi Suq

(Frifelt 1975a, 1975b; Nayeem 1996; Orchard and Stanger 1994; Cleuziou and Tosi

2007; Bortolini and Tosi 2008). In the interior of the United Arab Emirates, these tombs are found at Jebel al-Buhais, Jebel al-Emalah, and Jebel Hafit, although recent surveys have tentatively identified more of these tombs at Qarn bint Saud in the Emirate of Abu

Dhabi; at Dibba, Jebel Wamm, and the Wadi Fa’y in the Emirate of Fujairah; and at

Khatt and Qarn Al Harf in Ra’s al-Khaimah (Potts 1997a; Hellyer 1998; Uerpmann et al.

2006). Despite the minor typological distinctions made between these tomb styles, it is evident that both may be considered variants of a sole architectural form (Vogt 1985).

Figure 2.12. Beehive tombs at Al Ayn in the interior of Oman. From Flickr.com at .

The overarching uniformity and standardized structural characteristics shared by Hafit tombs constructed throughout the Oman Peninsula suggest a highly cohesive, homogeneous population where contact occurred regularly.

While hundreds of tombs in southeastern Arabia have been surveyed and subsequently attributed to the Hafit, excavations have only occurred for just over 100 of these structures (Benton 2006). Poor skeletal preservation and frequent re-use of tombs in later periods, particularly during the Iron Age, makes more comprehensive reconstructions of Hafit mortuary practices difficult, although some of these have escaped disturbance and allow for an examination of funerary customs and postmortem treatment of the body (Potts 1986a; Hellyer 1998; Al Saad and Yasin 2005). Hafit cairns housed collective burials normally containing between two and 12 individuals successively interred over time, although concentrations of up to 30 individuals have been found at well-preserved monuments at Ra’s al-Jinz and Ra’s al-Hadd in the Ja’alan of Oman

(Cleuziou and Tosi 2007). These individuals were variably laid on either their right or

43 left sides but were always initially placed in a flexed position (Frifelt 1975a, 1975b,

1980; Cleuziou et al. 1978; Al Tikriti 1982; Benton and Potts 1994; Potts 1997a). The positioning of the head appears inconsistent, with crania facing east, southeast, west, and north (Frifelt 1975a, 1975b; Al Tikriti 1982; Benton and Potts 1994). When preserved, both hands were situated directly in front of the face (Frifelt 1980; Hellyer 1998).

As deposition of new individuals took place, older remains were pushed against the inner wall of the cairn, as evidenced by elevated concentrations of crania and larger long bones in these areas (Cleuziou and Tosi 2007). Individuals of both sexes and all ages were interred, possibly suggesting the use of a single tomb by one family or kin group over a few centuries (Cleuziou and Tosi 2007; Potts 2009). Unlike the previous

Neolithic and succeeding Umm an-Nar periods, where cremation is evident in some cases, the Hafit shows no indication that burning played a role in mortuary tradition

(Benton 2006).

Few grave goods accompanied the individuals interred during the Hafit period

(Mery 1997). Pins, copper and bronze rivets, awls, needles, faience and stone beads, and biconical ceramic vessels have been found in undisturbed tombs in very small quantities

(Frifelt 1971; Gentelle and Frifelt 1989; Potts 1990, 1997a; Mery 1997; Cleuziou and

Tosi 2007). While most grave artifacts were of local manufacture, the presence of diagonally perforated square bone beads suggests a link to Iran, and carnelian beads imply long-distance acquisition from as far away as Afghanistan or Iran (Whitehouse

1975; Reade 1979). Moreover, a few decorated jars identified as Jemdet Nasr polychrome pottery interred with the bodies originated in southern Mesopotamia and are discussed in more detail below (Frifelt 1980; Potts 1993a, 1993b, 1997a).

44 Interregional Exchange

With the start of the Hafit period in southeastern Arabia, regular contact with

Mesopotamia resumed once more after an apparent absence of interregional exchange during the previous Uruk period (ca. 4000-3100 BC) (Phillips 2002; Phillips and

Mosseri-Marlio 2002). Evidence for such interaction emerges in the form of a new type of painted polychrome pottery known as Jemdet Nasr, named for the Mesopotamian site where they were first unearthed and later defining the Jemdet Nasr period in

Mesopotamia (ca. 3100-2900 BC) (Mackay 1931; Matthews 2002). This brief episode of

Mesopotamian history remains ambiguous, although it is apparent that substantial changes in social organization were taking place that coincided with a period of intense drought from around 3200-3100/3000 BC (Lemcke and Sturm 1997; Bar-Matthews et al.

1999; Cullen et al. 2000; Weiss and Bradley 2001; Parker and Goudie 2007). As a result of this abrupt, severe climate change, trade routes along the Euphrates were abandoned, far-reaching Uruk trading outposts collapsed, and a notable shift towards urban aggregation continued from the Late Uruk (Potts 1993b; Crawford 1998; Pollock 1999).

Compositional analyses of jars recovered in southeastern Arabia confirm their production during the Jemdet Nasr period in southern Mesopotamia and reinforce the belief that the

Oman Peninsula did not engage in local ceramic manufacture until the subsequent Umm an-Nar period (Mery 2000).

Initially, the discovery of Jemdet Nasr jars in multiple early third millennium graves throughout the Buraimi Oasis caused archaeologists to apply this Mesopotamian terminology to the Oman Peninsula and its tombs (During Caspers 1970; Frifelt 1971,

1975, 1979, 1980). However, this label was soon called into question due to both the

45 relative scarcity of Jemdet Nasr ceramics recovered in these cairns as well as the clear cultural and political differences between Arabia and Mesopotamia during this time

(Oates 1976; Potts 1986a). Because of these disparities and an increasing need to define the novel developments occurring during the Early Bronze Age in southeastern Arabia, a group of archaeologists working in the Oman Peninsula assembled in 1981 and christened this period the “Hafit,” after the site of Jebel Hafit in the Al Ain region of the

Emirate of Abu Dhabi (Potts 1986a).

The presence of Jemdet Nasr pottery in the Oman Peninsula signaled a reinvigoration of trade between Mesopotamia and Arabia that followed the collapse of

Uruk exchange routes with Anatolia, Iran, and Syria (Potts 1986a; Crawford 1998). As

Mesopotamia itself possessed few natural resources, it was dependent on obtaining these raw materials from other regions, particularly to support its sizeable metallurgical industry (Weisgerber 1983; Weeks 2003a). The loss of Anatolian copper must have represented a significant blow to Sumerian production, necessitating the discovery of an alternate source of copper for large-scale exploitation (Potts 1986a). The earliest, proto- cuneiform texts from southern Mesopotamia mention this new, incipient source of copper. Known as the Archaic Texts from Uruk, these clay tablets date to the late fourth millennium BC and contain the earliest reference to the land of Dilmun, today recognized as the island of Bahrain and eastern Saudi Arabia (Englund 1983; Crawford 1998; Potts

2001, 2009). These tablets describe the active importation of copper from Dilmun to

Mesopotamia; however, like Mesopotamia, Dilmun lacked the copper desired by the

Mesopotamian city-states (Hellyer 1998; Cleuziou 2000). The nearest source of these copper deposits lay in the Hajar Mountains, whose range extends from the eastern United

46 Arab Emirates into northeastern Oman (Weisgerber 1983). Subsequently, archaeologists speculate that Dilmun acted as a middleman between southeastern Arabia and Sumer, regulating the copper trade that stimulated the development of more complex interregional exchange systems with this region.

Evidence for small-scale copper mining and production, while sparse, appears during the Hafit period in the Oman Peninsula around the same time that Sumerian administrative tablets begin referencing Dilmun copper, lending support to the claim that southeastern Arabia played a small but increasingly important role in larger trade networks outside of the region. Copper slag, a by-product of the smelting process, has been recovered from an early third millennium site along the Hajar Mountains in Oman known as Maysar 1 as well as at Hili 8 and suggests the emergence of a local copper industry in the region (Bibby 1970; Hastings et al. 1975; Weisgerber 1981; Hellyer

1998). Maysar 1 represents the earliest site of copper production in Oman, where furnace fragments utilized in the smelting process are also present and further indicate the importance of metallurgical activity to this community (Hastings et al. 1975; Weisgerber

1983, 1984; Hauptmann 1985; Hauptmann et al. 1988). In addition to these waste products and furnace components, Maysar 1 contains a significant amount of copper ingots (Figure 2.13), masses of copper cast into a specific, standardized form useful for later processing into a finished tool or other object (Hauptmann et al. 1988). Though ingots are commonly seen as a useful means of transporting unworked metals across long distances, Hauptmann (1985) contends that the copper industry at Maysar 1 primarily satisfied domestic needs, organized as a number of small workshops unearthed in Houses

Figure 2.13. Horde of copper bun-shaped ingots recovered from House 4 at the site of Maysar 1, Oman (from Hauptmann 1985: Abb. 61).

1, 4, 6, and 31. Described by Weeks (2003:49) as independent household-based units of production, the copper workshops of Maysar 1 may nevertheless have contributed to regional and interregional markets, albeit in relatively small quantities (Costin 1991)

(Figure 2.14).

While evidence of copper production during this time is limited to only a few sites, the products of this burgeoning copper industry can be found in small quantities within Hafit graves across the peninsula. Copper objects deposited in these tombs included awls, blades, pins, and rivets (Frifelt 1971, 1975b; Cleuziou 1982, 1996; Potts

1993a, 2001; Salvatori 2001); this, in conjunction with Jemdet Nasr pottery, led Frifelt

(1975b) to assert that Hafit populations in the interior of Oman were actively engaged in local copper extraction and interregional trade (whether direct or indirect) with

Mesopotamia. In association with these copper grave goods, Potts (1986:133) observed

Figure 2.14. Artist’s reconstruction of copper smelting during the Early Bronze Age in the Oman Peninsula (from Cleuziou and Tosi 2007:197).

that the geographic placement of many Hafit necropolises fell “along potentially ancient routes of transport from copper-rich inland mining areas of the peninsula to the coast,” foreshadowing the established routes of later, more developed copper-trade complexes

(see also Gentelle and Frifelt 1989).

Copper objects were exploited at the local level as well from the beginning of the

Hafit period. As early as the late fourth millennium BC, copper fishhooks were reported from the coastal shell midden of Ra’s al-Hamra 5 (RH-5), and later, from the early third millennium shell midden at Ra’s al-Hadd 6 (HD-6). These copper hooks rapidly replaced previous shell fishhook technology (Cleuziou 1996). Middens at Ra’s Shaqallah and

49 Wadi Shab (GAS-1), both in Oman, also produced unspecified “copper pieces” (Biagi and Maggi 1990:546; Weeks 2003a). Interestingly, Uerpmann (1992) points out that the initial appearance of copper, the first metal utilized in the Oman Peninsula, was employed not for ornamental purposes, as was seen throughout Europe during this time, but solely for the improvement of tool technology.

Following the Jemdet Nasr period, the Early Dynastic period (ca. 2900-2350 BC) of Mesopotamia represents a time of considerable urban growth and increased social, political, and economic complexity. A plethora of administrative, historical, and royal textual records permits an unprecedented look into the inner workings of Early Dynastic life and the significant social changes that occurred during the Early Bronze Age in this region. Throughout Mesopotamia, rural village life was largely abandoned in favor of the protection offered by large fortified cities, further contributing to the development of city-states, typically comprised of one (but sometimes two) large, walled urban complex enclosed by expanses of agricultural fields and pastures managed by that city

(Baadsgaard 2008). These city centers, including Girsu, Kish, Lagash, and Nippur along the northern Euphrates as well as Ur and Uruk further south, were not linked under a common political system but were instead largely independent, multi-ethnic communities controlled by palace estates and temple households, which possessed their own herds, agricultural fields, and craft specialists (Stone 1995; Yoffee 2005). Escalating social stratification led to a transition from individual household production to a heavy reliance on the temple estates that controlled production, land tenure, storage, and cult centers

(Stone 1995; Yoffee 2005). While economic exchanges and the formation of political and military alliances between these city-states did take place, violent conflict between

50 them was also commonplace and likely explained why these settlements continued to attract a sizeable portion of the population well into the third millennium BC (Adams

1981; Yoffee 2005). Such conflict was not only internal to Mesopotamia but also arose from tensions with the neighboring civilization of Elam, now located in southwestern

Iran (Pollock 1999).

Chronological reconstructions using both specific artifactual and architectural styles have led to the division of the Early Dynastic (ED) into three phases, denoted as

ED I (ca. 2900-2800 BC), ED II (ca. 2800-2600 BC), and ED III (2600-2350 BC)

(Pollock 1999). This chronology is based primarily on successive shifts in temple architecture noted in the northern Diyala region of Mesopotamia, but also includes stylistic changes in pottery, sculpture, jewelry, weapons, cylinder seals, and textual records (Frankfort 1934, Porada et al. 1992). However, intense criticism of these divisions has emerged in recent years and has centered around two arguments. First, the

Diyala sequence has been liberally applied to the city-states of southern Mesopotamia, many of which do not share the characteristic features associated with its northern counterpart, particularly in the case of ED II (e.g, Hansen 1971). Secondly, extensive comparisons of ED II architecture and other forms of material culture in Diyala with earlier and later phases have revealed no distinct ED II period; instead, these can be attributed simply as an extension of ED I (Evans 2005, 2007; Baadsgaard 2008). While recognition of this debate is important, this terminology persists in the literature and will similarly be used here.

While Magan is not mentioned in Mesopotamian cuneiform records regarding trade until the late third millennium BC, a few Sumerian artifacts found at sites across the

51 Oman Peninsula during the Hafit period attest to the ongoing relationship between these two regions (Nayeem 1990; 1996). In addition to Jemdet Nasr vessels, small amounts of

Early Dynastic pottery have been recovered from Hafit tombs at Bat and Jebel Hafit

(Frifelt 1970, 1975a). At Hili 8 (Period I), buff ware jars of the ED I-II tradition also point to ties with Mesopotamia, while two black-on-red ware sherds are associated with types recovered from southeastern Iran (Potts 1990). Exchange with Mesopotamian city- states also took place further north along the Persian Gulf; for instance, off the coast of the Eastern Province of Saudi Arabia, Tarut Island has yielded graves dating to the early third millennium BC containing ED I-II ceramics identical to those at Ur, as well as sculptures reminiscent of Early Dynastic finds from Diyala (Frankfort 1954; Potts 1986a,

1990). Intensive trade with the strategically-placed Dilmun culture did not appear until the second half of the third millennium and will be discussed further below, although based on meager archaeological evidence, the island of Bahrain was largely uninhabited during the Mesopotamian Uruk, Jemdet Nasr, and ED I-II periods and did not engage in interregional, commerce-driven relationships (Larsen 1983b; Crawford 1998).

Consequently, it appears that the resumed contact between southern Mesopotamia and southeastern Arabia during the Hafit period, as evidenced by both the presence of

Jemdet Nasr /Early Dynastic pottery in Hafit cairns and the emergence of the copper industry in the Oman Peninsula, may have been motivated by a Sumerian need for copper

(Peake 1928; Meadow et al. 1976; Potts 1986a, 2001). This need likely reinvigorated trade with the coastal populations of the Arabian Peninsula and would eventually lead to a complex system of exchange in the following Umm an-Nar period.

52 Bronze Age: Umm an-Nar Period (ca. 2500-2000 BC)

Following the Hafit, the Umm an-Nar period (ca. 2500-2000 BC) comprised the second half of the third millennium BC and continued the Early Bronze Age in southeastern Arabia. The name Umm an-Nar (translation: Mother of Fire) is derived from Umm an-Nar Island off the coast of the Emirate of Abu Dhabi, where the first tombs and a settlement from this era were identified in the late 1950s. This period witnessed a number of major transformations in economy, sociopolitical organization, subsistence, ceramic and metallurgical industries, monumental architecture, and mortuary practices, drastically altering the cultural landscape in this region.

Settlement and Subsistence

Unlike the previous Neolithic and Hafit periods, the Umm an-Nar represented a time of widespread sedentism with the appearance of agriculturally-based settlements across southeastern Arabia, not only in the interior of the peninsula (e.g., Bat, Hili,

Maysar, Wadi Munay’i) and along the coast (Al-Sufouh, Bidya, Kalba, Mowaihat,

Shimal, Tell Abraq) but also on small coastal islands (Umm an-Nar Island) (al-Jahwari

2009). Settlement location was based primarily on finding reliable access to fresh water, difficult in an arid climate similar to conditions seen today (Potts 1990). Inland settlements were built in oases with a relatively high water table suitable for cultivation, while coastal inhabitants chose locales with underground aquifers that could be reached by means of a well (Blau 1999a). In addition to water requirements, habitation sites located interiorly were sometimes placed in close proximity to sources of copper, reflective of the growing demand for copper in Mesopotamia (Benton 2006).

53 The most archaeologically visible settlements were those with large fortification towers made of either mudbrick or stone (Potts 2009). These monumental buildings were circular and often dwarfed their associated mortuary structures, reaching external diameters of up to 40 m (Hellyer 1998). With a base constructed of a solid raised platform as high as 8 m from the ground surface and a complex interior of thick crosswalls creating chambers filled with small stones, these formidable structures were fortified still further by an outer wall and surrounding ditches (Potts 1993a, 2001). Most of these massive towers, including those at Bidya, Hili 1 and 8, Kalba, and Tell Abraq, encircled a well that likely served the agricultural needs of the entire community (Blau

1999a). Subsequently, it appears that these fortress towers acted as a means of protecting valuable natural resources like water (Potts 2001).

These fortifications may have served a defensive purpose as well. Akkadian cuneiform records from Mesopotamia make reference to several third millennium military campaigns against the “lords of Magan,” implying that larger political centers with some degree of hierarchical organization were occasionally invaded and required some degree of fortification (Potts 1986b, 2001). Interestingly, despite the seemingly defensive nature of these structures, little evidence of violent skeletal trauma exists during this period, although this may be a somewhat misleading conclusion, in part because of poor skeletal preservation, commingling, and bone fragmentation (Potts 2000;

Mery et al. 2004; Blau 2007). Smaller, more transient domestic structures and workshops surrounded these towers and may also hint at emerging social stratification (Crawford

1998; Mery and Tengberg 2009). However, with the exception of monumental architecture, evidence for hierarchy among the inhabitants of southeastern Arabia is

54 generally lacking, particularly when examining the lengths these communities went to ensure equality in mortuary contexts. Instead, then, a largely heterarchical system may have been in place, where these fortification towers and their surrounding communities represented multiple, relatively independent complexes as part of a decentralized cultural landscape (see, for example, Crumley 1995 and Levy 1995). If such a heterarchy of unranked centers existed in the Oman Peninsula, these Umm an-Nar towers may not have represented a defensive so much as a symbolic structure displaying group collectivity through communal labor.

In addition to these highly visible settlements, it is important to recognize that most of the population of the Oman Peninsula likely lived in smaller coastal or inland oasis villages without monumental architecture. Remains of stone or mudbrick foundations, postholes, and hearths at sites like Al-Ayn 2 in Oman attest to the existence of these small-scale communities (Blin 2007). These rural villages depended primarily on agriculture, and despite a previous emphasis on copper mining as the primary force propelling economic growth in the Oman Peninsula, the majority of these hamlets did not engage in the exploitation of copper (al-Jahwari 2009). Larger yet unfortified settlements have also been reported at al-Sufouh, Ghanadha, and Umm an-Nar Island (Al Tikriti

1985; Frifelt 1991; Benton 1996). Potts (2009) contends that pastoralists practicing transhumance existed in conjunction with larger, centralized oasis and coastal settlements as well as small agricultural units and engaged in seasonal migrations between the interior and coast.

While the relative absence of settlements during the Hafit contributes to a dearth of faunal and botanical material in the archaeological record from this period, Umm an-

55 Nar settlements provide a more holistic picture of subsistence strategies across southeastern Arabia. Nevertheless, poor preservation of biological materials at many sites continues and must be considered in any evaluation of these remains. Evidence from the majority of Umm an-Nar sites indicates a predominantly sedentary lifestyle characterized by a mixed economy, in which inhabitants engaged in agriculture, herding, hunting, and in littoral environments, the exploitation of maritime resources. Domestic sheep, goat, and cattle dominate faunal assemblages at these sites and likely comprised the majority of protein consumed (Al Tikriti 1985; Potts 1998). As the environmental needs of cattle are higher in comparison to sheep or goats, the large body size of cattle coupled with the sizeable assemblages of cattle bone uncovered at sites like Tell Abraq implies that these herders were adept at husbandry and herd management (Uerpmann

2001). Hunting wild mammals like gazelle and oryx also played a small but important role, and wild birds like the cormorant were commonly trapped at Tell Abraq and Qala’at al-Bahrain (Uerpmann and Uerpmann 1994; Uerpmann 2001).

Although less common, canine skeletal remains have been recovered from a handful of sites; however, treatment appears to differ significantly within the Oman

Peninsula. At the late third millennium site of Unar 2 in the Emirate of Ras al-Khaimah, a fully articulated adult dog skeleton was found interred with a human adult female, suggesting that these animals were treated with respect as potential human companions

(Blau and Beech 1999). Conversely, at Ra’s al-Hadd in Oman, dog bones display evidence of butchering and cooking (Frohlich 1986; Blau and Beech 1999).

Coastal inhabitants took advantage of shallow waters, lagoons, and mangrove forests concentrated along the west shoreline of the peninsula, both in the collection of

56 shellfish, crabs, and molluscs, but also fish and sea mammals (Aspinall 1998). Larger mammals, including dolphins, dugongs, and whales, were hunted alongside green turtles and sizeable fish such as stingrays and multiple species of sharks found in shallow waters

(Hoch 1979; Beech 2003b). The presence of shell and metal hooks, stone net sinkers, and fishing spears in both graves and settlements across southeastern Arabia reflects the importance of marine hunting (Frifelt 1995; Beech 2003b). Interestingly, unlike most sites found on the mainland, the settlement on Umm an-Nar Island exhibited a high frequency of large marine mammals in tandem with few domestic ruminants, a product of its offshore location (Hoch 1979). On the other hand, at the site of Tell Abraq, fishing activities do not appear particularly important relative to agriculture and animal husbandry, despite it coastal setting (Uerpmann 2001). In addition to exploiting shallow waters, deep-sea fishing of pelagic species, including barracudas, groupers, and jacks, took place, would have necessitated watercraft, and are thus indicative of the types of

Umm an-Nar maritime transport available (Uerpmann 2001; Beech 2002). Offshore hunting was especially prevalent on the east coast of the Peninsula surrounding the northern Emirates and extending into the Gulf of Oman due to the close proximity of deeper waters to the shore.

Most Umm an-Nar settlements were dependent on agriculture, either in the form of inland, oasis-based bustan gardens or cultivated plots surrounding coastal hamlets irrigated by wells dug to reach underground aquifers (Berthoud and Cleuziou 1983; Potts

2001). These gardens never took the form of extensive field systems due to inadequate rainfall and a lack of permanent freshwater sources (Mery and Tengberg 2009).

Agriculture was completely dependent on the date palm, not only for consumption

57 purposes but also to shade other, less resilient crops that would otherwise be unable to grow in such a harsh environment (Blau 1996, 1999, 2007; Potts 2001). Thousands of charred date stones and stone impressions have been recovered from sites across southeastern Arabia, including at Bat, Hili, Maysar 1, Ra’s al-Jinz 2, Tell Abraq, and

Umm an-Nar Island, and speak to the importance of this dietary mainstay (Weisgerber

1981; Cleuziou 1989; Willcox 1995; Cleuziou and Tosi 2000; Costantini and Audisio

2000; Potts 2000; Frifelt 2002; al-Jahwari 2009). Recently, dates were discovered in a mortuary context at a collective Umm an-Nar tomb at Hili N (2200-2000 BC), perhaps representing a ritual offering (Mery and Tengberg 2009).

Other Bronze Age cultivars grown alongside the date palm included cereals such as wheat and barley, jujube fruit, and a variety of vegetables and pulses (Potts 2001; Blau

2007). Grinding stones attest to the processing of cereals at sites like Tell Abraq (Potts

2000). Poor organic preservation in the Oman Peninsula precludes a more comprehensive assessment of these gardens, and unfortunately, this paucity of data has caused much of the archaeological literature to ignore this important aspect of Umm an-

Nar daily life (Costa and Wilkinson 1987; al-Jahwari 2009).

Skeletal remains from Umm an-Nar tombs can also lend evidence to subsistence practices during this time. Although more skeletal material exists for the Umm an-Nar as opposed to the preceding Hafit period, little has been published, and of these few publications, many are now considered questionable in methodology, simplistic, and even sexist (Blau 2001a). Soren Blau’s work (e.g., Blau 1996, 1998, 1999a, 1999b, 1999c,

2001a, 2001b, 2001c, 2007) across the Emirates, Debra Martin’s research (e.g., Cope et al. 2005; Martin 2007; Baustian and Martin 2010) at Tell Abraq, Kathleen McSweeney’s

58 analysis of the site of Hili in Abu Dhabi (McSweeney et al. 2008, 2010), and Judith

Littleton’s focus on Bahrain (Littleton 1987, 1990, 1997, 1998a, 1998b, 1999, 2007;

Littleton and Frifelt 2006; Littleton and Frohlich 1989, 1993) represent the most reliable and proficient work on the subject conducted thus far.

Dental wear during the Umm an-Nar period is severe across the Oman Peninsula, likely a result of the introduction of coarse grit into food from the use of grinding stones, which increased with the more widespread adoption of cultivars like wheat and barley

(Højgaard 1980, 1981; Kunter 1983, 1991; Blau 2007). Contrastingly, rates of carious lesions are relatively low, despite the intensified utilization of domesticated cereals and the date palm (Højgaard 1980, 1981, 1985; Kunter 1991; Blau 2007); however, this pattern is probably misleading, in that patterns of dental attrition are known to display a genenerally inverse relationship with those of dental caries. Højgaard (1980) has suggested that low caries rates may be a consequence of continued marine exploitation with a significant contribution of terrestrial animal protein, particularly at coastal sites.

Interestingly, Blau (2007) has teased out subtle differences in the frequency of carious lesions between a number of sites in the United Arab Emirates. She found that at al-

Sufouh and Mowaihat, situated in the southern portion of the coast, dental caries were almost absent; this, in conjunction with metallurgical evidence of differential bronze alloy composition in tools from al-Sufouh, prompted Blau to conclude that site hierarchies may have existed, in part a product of geography as well as diminished access to trade routes and/or partners, which may have led to differential access to food resources.

59 Other dental indicators of subsistence lend credence to the growing importance of agriculture during the Umm an-Nar. Dental calculus has been recorded at all sites in the

Emirates with human teeth present (Højgaard 1980, 1981; Blau 2007). These deposits are generally indicative of a rising dependence on plant domesticates, and although Blau

(2007) noted an overall decline in calculus from the preceding ‘Ubaid period, an increase in calculus rates is seen at Tell Abraq, where tooth preservation far surpasses that of most other sites in this area. As such, poor dental preservation may bias our view of these trends at other sites (Blau 2007). High levels of antemortem tooth loss have also been found and fit with mounting evidence for an increasing reliance on carbohydrates like date fruit and cultivated cereals (Kunter 1991; Blau 2007; McSweeney et al. 2008)

Mortuary Practices

Continuity can be seen between Hafit and Umm an-Nar burial structures. During the Umm an-Nar period, a circular tomb shape persisted, and limestone building materials were once again exploited in tomb construction (Potts 1993a). However, major differences distinguished these tombs from the previous period. In particular, tombs were fashioned in monumental tower form, resembling the fortress towers (16-40 m diameter) that served as the center of many Umm an-Nar settlements (Al Tikriti and Mery 2000;

Blau 2001b; Potts 2001). External diameters for these graves ranged from 4-14 m (Potts

1997a). Although none survive today, it appears that some of the graves were two stories high (Cleuziou and Vogt 1983; Al Tikriti and Mery 2000).

Changes in the overall makeup of the ring wall also took place. While the interior layer continued to employ unworked limestone and rough marine stones, carefully-fitted

60 limestone ashlar lined the visible, external face of the tomb (Al Tikriti 1989a) (Figure

2.15). Raised reliefs of animal and human figures have been found decorating these dressed stones at the sites of Hili (Garden Tomb 1059), Mleiha, Umm an-Nar Island

(Grave IV), and Unar 2 (Al Tikriti 1982; Potts 1990; Frifelt 1991; Benton 2006). Despite the external standardization evident in the above-ground, circular stone structure of Umm an-Nar tombs seen across the Oman Peninsula, internal tomb organization possessed no standardized design, consisting of varying arrangements of crosswalls dividing the interior into between 2-10 chambers of assorted size and shape (Figure 2.16) (Blau

2001b). Some walls divided these tombs completely in two, with no access between each half, necessitating two entrances (McSweeney et al. 2008). Other, freestanding walls simply divided the interior while providing passageways between different chambers

(Potts 2001).

Figure 2.15. Finely-hewn curved ashlars made of limestone provided the facing of Umm an-Nar tombs, concealing a second, inner wall of unworked stone (adapted from Velde and Moellering, nd). 61

Figure 2.16. Typical Umm an-Nar-type tomb at al-Sufouh, United Arab Emirates. Note the construction of multiple crosswalls and dual, opposing entrances (from Potts 2001:40).

Umm an-Nar tombs were used for multiple generations (likely between 200-300 years) and contained hundreds of individuals (Al Tikriti and Mery 2000; McSweeney et al. 2008). No selective burial appears to have taken place, as individuals of all ages and of both sexes were interred in these tower tombs. The vast majority of these human remains are extremely fragmentary and disarticulated, a product of arid environmental conditions, plundering, but most importantly, intentional mortuary practice (Blau 2001b).

In a few instances, such as at Unar 2 in Ras al-Khaimah (Sahm 1988; Blau 2001b), al-

Sufouh in Dubai (Benton 1996), Tell Abraq in Sharjah (Blau 2001c), and Tomb A at Hili

North as well as Grave I, Umm an-Nar Island in Abu Dhabi (Vogt 1985; Frifelt 1991;

McSweeney et al. 2008), one or more fully articulated and unburnt individuals were buried in a flexed position and variably laid on either their right or left sides. These

62 individuals were consistently found at the lowest levels of the tomb’s foundations and point to a funerary strategy consisting of an initial primary interment, with whole bodies placed into these graves until a need for space required that those already buried were pushed aside in order to make more room (Blau and Beech 1999). Articulated limbs and other appendages recovered from tombs including Unar 1 (Schutkowski 1988) and Hili N

(Mery et al. 2004) lend credence to this interpretation. In some cases, it appears that small fires were lit within the tomb to clear additional human debris, and at Hili, postmortem cut marks suggest deliberate disarticulation (Al Tikriti and Mery 2000;

McSweeney et al. 2008).

In contrast with the controlled, in situ fires uncovered in Tomb N at Hili that caused approximately 20% of bone within the grave to be burned, cremation appears to have been a further tactic occasionally used throughout southeastern Arabia to create additional space within tombs (Blau 2007; McSweeney et al. 2008). While cremation practices are not evident in all Umm an-Nar graves, some experienced a significant proportion of intentionally burned bodies. For instance, at Unar 2, 90.7% of all human bone display evidence of burning, with most (70.1%) bone completely calcined, indicative of burning at intense, high temperatures at or above 800°C (Shipman et al.

1984; Blau 2001b; Walker et al. 2008). At the multi-grave site of al-Sufouh, 85% of bone was cremated, with the remaining 15% found only in Tomb I (Benton 2006).

Subsequently, it seems that after primary burial in Tomb I, remains were later removed for cremation before final interment in either subterranean pits (Tombs II-IV) or back into Tomb I (Benton 2006). In each case, cremation did not take place within the tomb itself.

63 Few instances of faunal remains recovered from within Umm an-Nar tombs exist, and most of these depositions appear to have been natural and not intentional intrusions

(Blau and Beech 1999). However, at the floor level of the grave at Unar 2, an almost completely articulated dog skeleton was buried next to an articulated adult human female, possibly suggesting a change in human-animal relationships by the end of the third millennium (Blau and Beech 1999).

A second, anomalous grave type also existed during the Umm an-Nar period and has only been uncovered at two sites, Hili N and Mowaihat Tomb B. Rectangular/ovoid in shape, these graves are completely subterranean, consisting of a simple pit with a roof entrance covered by stone slabs (Haerinck 1991; Crawford 1998; Al Tikriti and Mery

2000). Both of these pit-graves were found within a few meters of typical Umm an-Nar circular tower tombs (Al Tikriti 1989a; McSweeney et al. 2008). As with these traditional graves, the rectangular pit-graves held hundreds of individuals, showed no signs of selective burial in terms of age or sex, and predominantly contained heavily fragmented and disarticulated human remains (Al Tikriti and Mery 2000). However, out of an estimated 300 individuals interred within the rectangular pit-grave at Hili N, 31 primary inhumations from the floor level of a single compartment remain intact (Mery et al. 2004; McSweeney et al. 2008). Artifacts characteristic of Umm an-Nar material culture found in these pit-graves support the dating of these tombs to the latter half of the third millennium BC, although at Hili N, a few Wadi-Suq-type objects may signal this tomb’s use at the close of the Umm an-Nar period at the emergence of the Wadi Suq period (Al Tikriti and Mery 2000).

64 While grave goods placed within both circular and rectangular tombs resemble their Hafit counterparts, the major difference between this period and the Umm an-Nar lies in the sheer quantity of the artifacts included (Mery 1997). Some tombs, like that of

Hili A, include more than 600 ceramic vessels alone, enough for each of the approximately 300 individuals inside to claim two pots (Mery 1997). In addition to pottery, rings, beads, stone vessels, seals, weights, copper objects, and shells were included in the assemblage (Potts 1997a). Although local products dominate all assemblages, foreign artifacts comprise a significant proportion of grave goods as well and will be discussed with regards to interregional exchange further below (Cleuziou and

Vogt 1983).

When compared against the preceding Hafit, the Umm an-Nar period exhibited remarkable and novel cultural developments in mortuary customs, subsistence strategy, settlement organization and aggregation, and social structure. Despite clear distinctions between these two phases, the relatively uniform nature of mortuary architecture and associated material culture deposited in tombs suggests that the Umm an-Nar period itself represented a time of considerable cultural homogeneity throughout southeastern Arabia

(Tosi 1975). Nevertheless, distinct site-specific disparities existed, particularly with regards to treatment of the dead (e.g., inhumation, cremation) as well as variation on the typical circular tomb type (e.g., rectangular tombs paired with circular tombs at Hili and

Mowaihat, and grave pits surrounding a circular grave at al-Sufouh), and may point to continued sociocultural changes taking place at the end of the third millennium with the transition to the Wadi Suq period.

65 Interregional Exchange

A major intensification of trade and the development of a pan-Gulf interaction sphere by the mid-third millennium BC instigated a shift in the economy of southeastern

Arabia in which production and exchange played a key role (Bondioli et al. 1998; Hellyer

1998; Carter 2003c; Potts 2009). Such changes coincided with a variety of technological innovations in Umm an-Nar material culture, including, for the first time, the appearance of a local ceramic tradition, the production of high-quality soft stone vessels, and a broad range of manufactured metal objects (Potts 2001, 2009). The emergence of large settlements, both inland (e.g., Bat, Hili) and coastal (e.g., Tell Abraq), as well as the spatial distribution of local goods, suggest that a regional exchange network throughout the Oman Peninsula operated within a larger interregional system involving

Mesopotamia, Dilmun, the Indus Valley, Elam, and Central Asia. The frequency with which such local communications took place explains the high degree of cultural homogeneity across the peninsula, particularly with regards to mortuary practices as well as the standardization of manufactured items like pottery (al-Jahwari 2009).

Communication and trade between southeastern Arabia and Mesopotamia persisted well into the third millennium with the start of the Umm an-Nar period, and a brief history of the changing political landscape of Mesopotamia during this time may be useful here. Following the Early Dynastic I-III periods, which witnessed a substantial population migration from rural hamlets to large urban centers and intensified social stratification, administrative and political changes in Mesopotamia accompanied the rise of the Akkadian dynasty (Pollock 1999). Akkadian rule was established ca. 2334 BC after the conquest of both north and south Mesopotamia by Sargon the Great, who

66 proceeded to unify the region’s relatively autonomous city-states (Edens 1992; Pollock

1999; Oates 2008). Plagued by internal rebellions, this short-lived empire quickly collapsed within a century, facilitating a return to a decentralized, tumultuous political landscape in which city-states reverted back to independent rule (Pollock 1999).

However, by approximately 2100 BC, imperial unification under the authority of general/governor Ur-Nammu marked the beginning of the Third Dynasty of Ur (Ur III)

(Oates 2008). As with the Akkadian dynasty, Ur III lasted only about 100 years, although its impact on the Sumerian cultural sphere was substantial, as evidenced by widespread building projects in most city-states, the emergence of ziggurats, improvements in irrigation and agricultural systems, and the resuscitation of the

Sumerian language (Oates 2008). By the end of the third millennium BC, attacks carried out by both Amorite and Elamite invaders cut off communication between major city- states, leading to the abandonment of farmlands and resulting in famine and the collapse of Sumerian economic exchange networks (Oates 2008).

At the close of the Early Dynastic (ED) III period, increasing Mesopotamian demands for copper coupled with power struggles among the temple and palace elites continued to fuel trade relations with southeastern Arabia (Possehl 1996). Lacking indigenous mineral deposits but possessing rich alluvial soils conducive to agriculture,

Mesopotamia aggressively sought to import copper from the Oman Peninsula and elsewhere (During Caspers 1972; Ratnagar 1981; Oates 1993; Carter 2008). Trace element and lead isotope analyses of copper artifacts uncovered across Mesopotamia, both temporally and spatially, verify that much of the metal used to manufacture these objects originated in Magan (Berthoud and Cleuziou 1983; Begemann and Schmitt-

67 Strecker 2009; Begemann et al. 2010). Furthermore, numerous third millennium copper mining and smelting sites have been identified in the interior of the peninsula along the

Hajjar Mountains – including at al-Bayda, Arjah, Assayab, Bilad, Maysar 1, Muaidin,

Tawi Ubaylah, Wadi Salh, and Zahra – with sizeable slag mounds suggestive of industrial levels of production (Hauptmann 1985; Costa and Wilkinson 1987; Carter

2003c). Nevertheless, it is important to note that southeastern Arabia was likely only one of many copper-producing regions utilized by Mesopotamia (Carter 2003c).

Additionally, cuneiform tablets from southern Mesopotamia mention Magan numerous times and in various contexts, confirming its importance with Akkadian rulers and the later city-states of the Third Dynasty of Ur (Lawler 2010). These texts record the maritime transportation of Mesopotamian exports, including barley, textiles, oil, and perfumes, to the Oman Peninsula in exchange for copper and other regional products

(Leemans 1960; Crawford 1973; Ratnagar 1981; Nayeem 1996). One particular merchant mentioned by name, Lu-Enlila, represented the Nanna-Ningal temple at Ur and arranged expeditions to Magan to exchange hides, sesame oil, textiles, and wool for copper (Potts 2000). Beyond these administrative records, political documents dating to the rule of Sargon of Akkad (2370-2316 BC) and later Ur-Namma (ca. 2100 BC) refer to boats from southeastern Arabia, the Indus Valley, and Dilmun docking at Mesopotamian harbors (During Caspers 1972; Hellyer 1998; Potts 2000).

Other forms of archaeological evidence indicative of contact between these two regions corroborate this copper exchange, particularly with regards to ED III pottery sherds, found throughout the Oman Peninsula within latter third millennium tombs such as Hili 8, Tomb A at Hili North, Tomb N at Hili, Kalba, Ras Ghanadha, Ra’s al-Aysh,

68 Ra’s al-Hadd, Ra’s al-Jinz, Tell Abraq, and Umm an-Nar Island (Al Tikriti 1985; Vogt et al. 1989; Cleuziou et al. 1994; Frifelt 1995; Mery and Schneider 1996; Mery 1997;

Hellyer 1998; Al Tikriti and Mery 2000; Potts 2005; McSweeney et al. 2008). Some of these sherds, including those from Hili and Umm an-Nar Island, were confirmed to be of

Mesopotamian origin using neutron activation analysis (Piesinger 1983). These ceramic vessels were likely used as storage containers for perishable Mesopotamian liquids, including oil, and their inclusion in Umm an-Nar mortuary (and less commonly, in domestic) settings suggests that local inhabitants associated some value with them

(Hellyer 1998). A Mesopotamian seal impression has also been recovered on Umm an-

Nar Island (Frifelt 1995). In addition to copper, a small number of Arabian wares were transported north as well, including a soft-stone vessel originating from Oman unearthed at the southern Mesopotamian city-state of Ur, and indicate that reciprocal trade did in fact occur (Potts 2009).

Despite the consistent presence of Mesopotamian pottery in many Early Bronze

Age tombs, the quantity of sherds recovered in southeastern Arabia remains small relative to the amount and range of goods originating from the Indus Valley (Carter

2003c). By the middle of the third millennium, while Magan continued trade relations with Mesopotamia, it was also strengthening its exchange ties with others, particularly with the Harappan civilization (Blackman et al. 1989; Potts 1990). While ED III imports in the early Umm an-Nar period reflect an ongoing relationship with Mesopotamia, this affiliation waned during the Akkadian dynasty, as attested by the scarcity of

Mesopotamian wares in the peninsula after this time. With the possible exception of Tell

69 Abraq and Umm an-Nar Island, direct ties with the Umm an-Nar culture were effectively severed (Mery 1997; Potts 1992, 1993, 1997a; Al Tikriti and Mery 2000; Carter 2003c).

The abrupt ending of this relationship was likely a consequence of the shifting political and imperialistic atmosphere of Mesopotamia after the founding of the Akkadian dynasty. Edens (1992) argues that the aggressive campaigns undertaken by Akkadian rulers against the Oman Peninsula may have irrevocably damaged ties between

Mesopotamia and Magan, causing the inhabitants of southeastern Arabia to instead intensify exchange networks with the Indus Valley. Textual descriptions of the military exploits of the Akkadian ruler Manishtushu (ca. 2269-2255 BC) brag of his invasions of

Magan in order to acquire diorite, while the spoils of a subsequent campaign led by

Akkadian leader Naram-Sin (ca. 2254-2218 BC) were commemorated by engraving objects with the phrase “booty from Magan” (Hellyer 1998; Potts 1986b, 2000).

Nevertheless, the subsequent Third Dynasty of Ur reinvigorated direct communications with southeastern Arabia and reinstated shipments of Magan copper until its collapse in approximately 2004 BC (Oppenheim 1954; Potts 2009).

Indus material began to appear throughout the Oman Peninsula towards the end of the third millennium BC with increasing regularity at both coastal and inland sites.

Unlike Mesopotamia, whose presence was dominated by pottery sherds, imported

Harappan assemblages consisted of a wide variety of non-perishable goods, including pottery, metal objects, silver and etched carnelian beads, seals, and weights.

The Harappan weights represent an important aspect of Indus Valley contact with the Arabian Peninsula that deserve special mention here. These weights, cubical in form, were made from materials quarried in the Indus Valley and were utilized as part of a

70 larger standardized system of measurement within the Gulf (Potts 2000). Only three

Harappan stone weights associated with Umm an-Nar culture have been unearthed in the

Oman Peninsula, all of which come from Tell Abraq. All three conform to known, fundamental Harappan units of weight and were fashioned from jasper and a banded chert (Potts 1993c; Potts 2000). These materials suggest that the weights were in fact imported objects from Harappa, objects that signify a highly developed exchange relationship involving some degree of centralized economic control. While weights outside of the Indus Valley have also been discovered in Susa in modern-day Iran (Amiet

1986) and at Qala’at al-Bahrain on the island of Bahrain (Potts 1985; Bibby 1986), only one other weight has ever been uncovered in southeastern Arabia, from the site of

Shimal, but dates to the later Wadi Suq period (de Cardi 1989).

The Harappan civilization also integrated ceramics into their regional, and later, interregional exchange networks, with vessels serving as a means of measurement as well as transport (Kenoyer 1991). These vessels likely contained dairy products and have been unearthed throughout the Oman Peninsula at Hili 8, Tell Abraq, Wadi ‘Asimah,

Shimal 6, Maysar 1, Ras Ghanadha, Ra’s al-Jinz, Ra’s al-Hadd, and Ra’s al-Junayz

(Mery 1988; Potts 1990; Edens 1993; Crawford 1998; Potts 2000, 2001). In fact, approximately 13% of ceramics from the settlement of Ra’s al-Jinz were Harappan (Mery

1988); similarly, at ‘Asimah in the modern-day Emirate of Ras al-Khaimah, 20% of all storage jars originated from the Indus Valley, while no Mesopotamian pottery was present by the late third millennium (Hellyer 1998). Harappan inscriptions have also been engraved on jars from Tell Abraq, Ra’s al-Junayz, and Maysar 1 (Potts 1990,

1993d; Edens 1993). Umm an-Nar items have been recovered from the Indus Valley

71 (e.g., a soft-stone vessel from the Oman Peninsula at Mohenjo-Daro), and while few in number, indicate some eastern movement of goods as well (Potts 2009).

Such an extensive array of imported Harappan goods may suggest that, like

Dilmun, Magan acted as a sort of entrepôt, providing a series of commercial outposts for maritime traders that not only offered access to southeastern Arabia but beyond to

Dilmun and even Mesopotamia (Possehl 1996). While textual evidence confirms that direct contact between Mesopotamia and the Indus Valley was maintained during the ED

III and Akkadian dynasty, trade of Indus wares in the latter part of the third millennium, during the Third Dynasty of Ur, was at least partially regulated by Magan (During

Caspers 1972). Geographically, the Oman Peninsula provided an ideal, centralized location for merchants following sea trade routes, particularly along the calm, shallow waters of its western shoreline, where the majority of Umm an-Nar sites congregate (Al

Tikriti 1982). As discussed previously, bitumen and the remains of reed boats date to as early as the sixth millennium BC in the Persian Gulf and suggest the continued use of marine transportation during the Umm an-Nar (Carter 2006).

While Mesopotamia and the Indus Valley are often the focus of Bronze Age interregional exchange networks in the Persian Gulf, these cultures were by no means the only major players in this complex interaction sphere. Located in present-day Bahrain and the eastern shores of Saudi Arabia, Dilmun emerged as a thriving commercial center that would eventually replace Magan as the primary entrepôt in the Gulf with the commencement of the Wadi Suq period in the early centuries of the second millennium

BC (During Caspers 1972). At the close of the third millennium, Dilmun and southeastern Arabia enjoyed fruitful trade relations with one another, as evidenced by

72 Dilmunite vessels at Tell Abraq, Unar 2, Khatt, and Kalba (Potts 1993d, 2000; de Cardi et al. 1994; Mery et al. 1998; Carter 2003c). Umm an-Nar ceramics and soft-stone vessels consistently found in Early Type burial mounds and settlements at Saar, Qala’at al-Bahrain, and Tarut also testify to this productive partnership (Zarins 1989; Srivastava

1991; Højlund and Andersen 1994; David 1996; Carter 2003c; Laursen 2009). However, with the start of Period IIa (ca. 2050 BC), significant shifts in Dilmun socioeconomic organization took place – including monumental building projects, a drastic rise in population size, the emergence of a more stratified social hierarchy and elite burial status, the local manufacturing of stamp seals, and an economic boom – all coincide with the

“decline of Magan” in the early second millennium (Laursen 2008, 2009). This will be discussed in further detail in the Wadi Suq portion of this chapter.

Southeastern Iran and the Indo-Iranian borderlands also played a key role in trade with the Oman Peninsula. This culturally diverse and chronically understudied region had developed well-established trade relationships, taking advantage of both inland and maritime routes, with Mesopotamia, the Indus Valley, and particularly with Magan, forming what Possehl (1996) has dubbed the ‘Middle Asian Interaction Sphere’ (During

Caspers 1992, 1994; Potts 2009). Several foreign objects found in Umm an-Nar tombs appear to have originated from Middle Asia that demonstrate such an affiliation, including several ivory combs from Bactria at Tell Abraq (Figure 2.17) (Potts 1993e), daggers similar to those from Susa and a pedestal chalice reminiscent of forms from

Baluchistan at Wadi ‘Asimah (Vogt 1994b; Hellyer 1998), alabaster vessels suggestive of jars from Iran placed in nine Umm an-Nar tombs, including at Umm an-Nar Island and

Figure 2.17. A decorated, crescent-shaped ivory comb from Bactria (now northern Afghanistan and south-central Asia) found in the Umm an-Nar tomb of Tell Abraq (from Potts 1993e:592).

Tell Abraq (Potts 1993d; Frifelt 1995), and Iranian grey and fine red ware at Tomb A at

Hili North, Hili 8, and Umm an-Nar Island (Frifelt 1995; Mery 1997; Al Tikriti and Mery

2000; Potts 2005). Furthermore, while southeastern Arabia was rich in copper, the manufacturing of bronze requires a tin additive; Middle Asia, and particularly Bactria, possessed rich tin reserves, and it appears that this valuable commodity was shipped in large quantities to sites like Tell Abraq (Weeks 1999; Potts 2000).

Exchange was evidently mutual, with vessels from the Oman Peninsula found in

Iran (Tepe Yahya, Tul-e Peytul, Susa) (Potts 2009) and well into Central Asia at Gonur

Depe North in Turkmenistan (Potts 2008) by the latter third millennium BC, reflecting a considerable geographic range of communication and distribution during the Umm an-

Nar. With substantial copper reserves in Iran, this area also maintained close ties to

74 Mesopotamia in the third millennium BC, and tin from Bactria may have been transported to Magan via the Indus Valley, indicating the circulation of Bronze Age goods over a substantial geographic area (Potts 2000; Carter 2003c). Conspicuous technological and stylistic similarities between the Umm an-Nar local ceramic tradition and pottery from southeastern Iran have even led to speculation that immigrants from

Iran were the actual founders of the Umm an-Nar ceramic industry and will be discussed further in Chapter 4 (Potts 2005). However, by the end of the third millennium, the presence of Iranian pottery in late Umm an-Nar tombs diminished considerably, potentially reflecting the peninsula’s deteriorating interregional relationships that would eventually characterize the subsequent Wadi Suq period (Cleuziou 1989; Laursen 2009).

Bronze Age: Wadi Suq Period (ca. 2000-1200 BC)

The Wadi Suq differed considerably from the preceding Umm an-Nar period in southeastern Arabia, witnessing not only notable shifts in funerary traditions but also in subsistence, mobility, settlement organization, and external trade relations. The name

‘Wadi Suq’ stems from an Omani site of the same moniker excavated in the early 1970s by Karen Frifelt, where new stylistic forms of ceramics dating to this time were first reported (Frifelt 1975a). Similar forms unearthed at Hili 8 by Cleuziou (1979) confirmed the widespread occurrence of this distinct style across the Oman Peninsula, and the Wadi

Suq was formally recognized as a discrete archaeological period soon afterwards

(Cleuziou 1981).

For many years, archaeologists linked the significant disparities delineating the

Wadi Suq from Umm an-Nar culture with the domestication of the camel, which

75 seemingly explained a sudden return to a more mobile lifestyle (Potts 1993a, 2001).

However, more thorough zooarchaeological analyses revealed that camel remains dating to the Bronze Age represented a wild form of dromedary, and that domestication did not take place until the Iron Age (Uerpmann and Uerpmann 1997, 1999, 2002). While this revelation, along with improved excavation methods, has taken the once relatively unknown Wadi Suq out of the so-called “Dark Ages” (Cleuziou 1981), much of this period remains poorly understood.

Settlement and Subsistence

The gradual disintegration of Umm an-Nar life and the emergence of Wadi Suq culture at the beginning of the second millennium BC may in part be explained by shifting environmental conditions at that time. After enjoying relatively moist conditions and a high water table during the Umm an-Nar period, inhabitants of southeastern Arabia witnessed considerable environmental desiccation and extreme regional aridity (Figure

2.18) (Cleuziou 1989; Parker et al. 2006). Accompanying this climatic change were dramatic modifications of the landscape, including vegetation loss, dune reactivation, the extension of sabkha (salt flats) along the coast, and a major decline in sea level and in local water tables (Evans et al. 1969; Cleuziou 1989; Hellyer 1998; Parker and Goudie

2008). Aridification also took place across much of this part of the world, as evidenced by geologic and isotopic analyses in North and East Africa (Halfman and Johnson 1988;

Gasse and Van Campo 1994; Cheddadi et al. 1998), the Levant (Frumkin 1991; Bar-

Matthews et al. 1997), Turkey (Lemcke and Sturm 1997), and Yemen (Wilkinson 1997).

Figure 2.18. Mineral concentrations from 6000-2000 cal. BP as measured from a marine sediment core from the Gulf of Oman, illustrating a period of intense aridification beginning ca. 4025 BP, corresponding to the commencement of the Wadi Suq period in the Oman Peninsula (from Cullen et al. 2000:381). Note also the brief period of aridity around 5200 BP, characterizing the severe drought ending the Neolithic before a return to wetter conditions and the emergence of Hafit culture.

Coinciding with this environmental (and, as will be discussed below, economic) shift, the population of southeastern Arabia during the early second millennium appears to have steadily diminished, with some areas no longer capable of supporting the local population (Hellyer 1998; Parker and Goudie 2008). Settlements decreased in number, became smaller and more ephemeral, and likely persisted only in areas with adequate water supply (Carter 1997; Crawford 1998; Hellyer 1998; Blau 2007; Potts 2009). In

77 addition to freshwater availability, potentially enabling a continuation of small-scale agriculture, access to a steady food source would have also played a critical role in determining where people congregated into settlements. This would have made maritime resources particularly attractive, and may explain why the majority of Wadi Suq settlements were positioned along the northern coast (Potts 1997a; Carter 2003c; Parker and Goudie 2008).

Settlements dating to the Wadi Suq are scarce, and their initial absence during excavations in the 1970s and 80s led many archaeologists to assume that the population had reverted back to a predominantly nomadic way of life (Potts 1993d; Vogt 1998).

Current evidence of second millennium settlements, while still limited, has shed light on how Wadi Suq communities were organized. A few large Umm an-Nar settlements, including at Tell Abraq and Kalba, continued to be occupied full time in the Wadi Suq period. At Tell Abraq, the third millennium fortress-tower underwent some modification, including the addition of a mudbrick pavement at the top of the structure as well as the addition of walls both outside and inside the tower (Figure 2.19) (Potts 1990, 1993c,

2001). The abundant presence of foreign ceramic wares from Bahrain, as well as small amounts of material from Middle Elam, a Post-Harappan Indus Valley, and Old

Babylonian Mesopotamia, suggest that interregional ties were not completely severed

(Potts 1990, 2009). At Kalba, a second fortified tower dating to the Umm an-Nar was also paved with mudbrick during the Wadi Suq and was further enhanced with large stone and mudbrick walls outside, inside, and on top of the mound (Carter 1997; Hellyer

1998). Additionally, Hili 8 displayed continuity into the second millennium BC during

Figure 2.19. Mudbrick platform paving the fortification tower at Tell Abraq (Potts 2000:25).

its period III, and again, its large, late third millennium round structure appears to have been utilized during the very first part of the Wadi Suq (Cleuziou 1978-9; Potts 1990).

Evidence of a newly dug well that was later extended further into the ground likely speaks to the lowered water table in the region and may explain why the site was abandoned soon afterwards (Cleuziou 1989).

Other settlements appear for the first time during the Wadi Suq. At the unexcavated sites of Nud Ziba and Bida’a, surface surveys have identified Wadi Suq pottery sherds atop large mounds (45 and 20 m, respectively) (de Cardi et al. 1994;

Kennet and Velde 1995; Crawford 1998; Hellyer 1998). A small sample trench at Nud

Ziba unearthed a massive structure with thick walls, suggestive of a substantial settlement

(de Cardi et al. 1994). Two other known, excavated settlements come from Shimal and 79 Tawi Sa’id. At Shimal, in the Emirate of Ras al-Khaimah, a modest settlement with the remnants of several rectangular stone foundations as well as more transient structures indicate its inhabitants did not permanently occupy the site (Potts 1990; Crawford 1998;

Vogt 1998). Given the enormous Wadi Suq necropolis associated with the settlement, the small size of this settlement is surprising, and implies a less sedentary population than in the Umm an-Nar (Hellyer 1998). Finally, at Tawi Sa’id in Oman, trench excavations revealed a series of second millennium mudbrick walls and platforms, although no associated Wadi Suq pottery or foreign materials were recovered (de Cardi et al. 1979;

Potts 1990; Velde 1991; Carter 2003c). Other sites, including Maysar and near coastal

Jumeirah, have produced small sherd scatters dated to the Wadi Suq (Crawford 1998;

Hellyer 1998).

The profusion of copper mining and smelting sites during the Umm an-Nar fit well with Mesopotamian textual records describing well-developed exchange networks with southeastern Arabia as well as an abundance of slag recovered from mining areas

(Carter 2003c). However, the Wadi Suq lacks evidence of copper working, seemingly suggesting that the system of large-scale copper production characteristic of the Umm an-

Nar had collapsed by the early second millennium (Weeks 1997; Carter 2003c).

Increasing regional aridification would have made living in the interior, without direct access to coastal resources, a challenge, and while a few tombs are found dispersed on the foothills of the Hajar Moutains, no settlements (with the possible exception of Tawi

Sa’id) have been found (Carter 2003c). Once again, this points to a shift away from a completely sedentary way of life and the adoption of a more mobile lifestyle for most areas of the Oman Peninsula.

80 Such major social and structural changes were expectedly accompanied by corresponding shifts in subsistence strategies and diet. While very little zooarchaeological research has been performed on sites from the Wadi Suq, Beech

(2003) examined fish remains from a variety of sites throughout the Bronze Age, and found that while few changes in the exploitation of fish species took place between the

Umm an-Nar and Wadi Suq, a fundamental shift in the environments where fishing occurred (from predominantly deep-water to shallow-water species) was evident. Beech

(2003) speculated that this strategic shift to shallow-water fish might reflect the more impoverished lifestyles of these communities. Conversely, however, pelagic varieties of mackerel, tuna, jacks, and trevallies from Kalba point to a differential strategy employed on the east coast of the peninsula, which possesses much deeper waters closer to shore relative to its western counterpart (Beech 2003b).

At Shimal, extensive shell mounds both in and around the settlement signify an intensified exploitation of maritime resources, particularly shellfish, relative to the Umm an-Nar (Grupe and Schutkowski 1989; Potts 1990; von den Driesh 1994; Glover 1998;

Vogt 1998). Fish comprised almost 90% of the total faunal assemblage, with terrestrial domestic (sheep, goat, cattle) and wild (e.g., camel, gazelle, equid) animals contributing very little to midden deposits (Vogt and Franke-Vogt 1987; Potts 1990). A small number of grinding stones have been recovered from Shimal, although no evidence for date agriculture is present, indicative of a dietary shift from agro-pastoralism towards marine resources (Potts 1990; Hellyer 1998). Conversely, at Tell Abraq, abundant terrestrial faunal remains illustrate that pastoralism was not completely abandoned during the Wadi

Suq, and that despite environmental desiccation throughout much of southeastern Arabia,

81 sites with sufficient resources were able to maintain subsistence practices developed in the Early Bronze Age (Potts 1990). Large numbers of dates and grinding stones suggest continuity in agricultural practices as well (Potts 1990). Still, relative to third millennium finds, those residing at Tell Abraq during the Wadi Suq became increasingly dependent on maritime resources (Potts 1995, 1997a).

Limited skeletal evidence also points to changing subsistence patterns over time in the Oman Peninsula. At Shimal, trace element analysis of skeletal material from three tombs dating to the early, middle, and late Wadi Suq display a gradual shift from a more varied diet with contributions from both terrestrial and marine to a diet dominated by marine resources, consistent with the massive shell middens in the area (Grupe and

Schutkowski 1989). Grupe and Schutkowski (1989) interpreted this as reflective of a trend towards a more nomadic existence. However, it should be noted that as bone is extremely susceptible to diagenesis, and as many trace elements have been shown to be unreliable indicators of diet (e.g., Ezzo 1994), these results must be approached with caution. The Shimal skeletons were also evaluated for dental wear (Wells 1984;

Schutkowski and Herrmann 1987; Littleton and Frohlich 1993) as well as dental caries and antemortem tooth loss (Schutkowski and Herrmann 1987; Littleton and Frohlich

1993). Minimal wear, coupled with copious fish remains on site, led the authors to conclude that Wadi Suq peoples consumed a diet heavily reliant on shellfish and fish and that cereals played only a minor role, while low caries frequencies suggest a decreasing dependence on carbohydrate-rich foods like dates, a staple food during the Umm an-Nar

(Wells 1984; Schutkowski and Herrmann 1987; Littleton and Frohlich 1993).

82 Later, more comprehensive biological anthropological studies of Umm an-Nar and Wadi Suq skeletons encompassing a variety of sites show similar but more nuanced results (Blau 1999b, 2001, 2007). In general, while the highest frequencies of infectious disease, traumatic injury, DJD, and dental wear appear during the Umm an-Nar in conjunction with the emergence of large fortified settlements, monumental building projects, and a mixed economy including agro-pastoralism and marine resource consumption, all of these trends decrease during the Wadi Suq, potentially indicative of population dispersal from most sizeable habitation sites, declining physical labor associated with a loss of large-scale construction projects, and a general change in subsistence strategy and diet from terrestrial to marine resources (Blau 2001a, 2007).

Interestingly, only the prevalence of linear enamel hypoplasias (LEH), antemortem tooth loss (AMTL), and cribra orbitalia increased during the Wadi Suq. Both LEH and cribra orbitalia reflect stress during early childhood, and Blau (2001a, 2007) proposed that with a shift from a broad Umm an-Nar diet (with contributions from a variety of terrestrial and maritime sources) to a more narrow Wadi Suq diet dominated by seafood, such skeletal indicators may signify poor nutrition. Moreover, AMTL occurred with increasing frequency while dental caries and wear simultaneously decreased in number, leading

Blau (2001a) to conclude that abscesses may have facilitated AMTL. Given an overall lack of evidence for excessive date and grain consumption during this period, the relative absence of dental caries is not surprising.

83 Mortuary Practices

Major changes in funerary traditions also marked the divide between the Umm an-Nar and Wadi Suq. These are of particular interest as so few settlement areas exist during this period, compelling archaeologists to rely more heavily on interpreting the

Wadi Suq through a mortuary lens. Unlike habitation sites, second millennium burials and cemeteries are well represented across southeastern Arabia, although many congregate in the northern portion of the peninsula and are especially prevalent in the

Emirate of Ras al-Khaimah (Vogt 1998). Wadi Suq tombs have been identified at al-

Qusais, Bidya, Bithna, Dhayah, Dibba, Ghalilah, Idhn, Jebel al-Buhais, Khatt, Khor

Fakkan, Nizwa, Qarn al Harf, Qattarah, Qidfa, Rafaq, Samad, Sharm, Shimal, Wadi

‘Asimah, Wadi Sunaysi, and Wadi Suq (Frifelt 1975a; Donaldson 1984, 1985; Vogt and

Franke-Vogt 1987; de Cardi 1988; Hellyer 1998; Vogt 1994b, 1998; Yule 2001; Velde

2003). The large necropolis at Shimal stretches over 2.5 km and contains hundreds of these graves – at least 240 have been recorded – while at Khatt, a minimum of 60 tombs have been confirmed (de Cardi et al. 1994; Vogt 1998). Such an extensive mortuary distribution, not only at Shimal and Khatt but also throughout the region, points to the continued existence of a sizeable population during the Wadi Suq, despite speculation of a rapid depopulation throughout the region (Potts 1997a; Carter 2003c; Parker and

Goudie 2008; Laursen 2009).

Considerable variations in tomb construction characterize Wadi Suq monuments, as opposed to the more standardized forms of the Umm an-Nar (Potts 1990; Jaism 2006).

Unlike the fine ashlar limestone that lined the exterior of Umm an-Nar burial structures,

Wadi Suq tombs were built with large, unprepared stones and were typically ovoid in

84 shape with a single, narrow chamber (Potts 2001). In some cases, stone robbed from nearby Umm an-Nar structures was used to build Wadi Suq tombs (Cleuziou 1979). No less than six architectural types have been identified, many of which are named for the site where tomb type was first formally described, and are discussed below:

a. Shimal: The most common Wadi Suq grave found to date, Shimal-type tombs

are characterized by their above-ground construction and a single oblong chamber

containing the remains of multiple individuals (Figure 2.20) (Hellyer 1998).

These enclosures ranged from 9-29 m in length and are between 3-5 m in width, a

sizeable area covered by a series of enormous stone slabs to form a roof (Potts

1990, 2001). Outer walls were faced with unworked stones, while an inner layer

was constructed of more massive stones (Potts 1990). Entrance(s) into the

chamber were positioned along the length of one side of the tomb (Potts 1997a).

At the site of Shimal, approximately 70% of Shimal- (and Ghalilah-) type tombs

were oriented north-south (Vogt 1998). Examples of Shimal-type graves can be

found throughout the Emirates, including at Bidya, Dibba, Dhayah, Ghalilah,

Khatt, Sharm, and Shimal (Al Tikriti 1989b; Hellyer 1998; Vogt 1998; Riley and

Petrie 1999; Potts 1990, 2001).

b. Ghalilah: Broader than their Shimal counterparts, Ghalilah-type tombs are

architecturally unique because of a centrally placed crosswall(s) dividing the

chamber’s interior (Figure 2.21) (Potts 1990). These internal partitions supported

large capstones balanced between the crosswall and the outer and inner layers of

the exterior stone wall (Potts 2001). Reaching up to 11 m in length and 6 m in

Figure 2.20. Shimal type above-ground collective tomb from Sharm, Emirate of Fujairah, United Arab Emirates (from Riley and Petrie 1999:183). Arrows refer to entrance placement.

width and constructed of large, unworked stone, Ghalilah-type graves were placed

above-ground and built as collective monuments (Potts 1990, 1997a). As with the

Shimal-type graves, approximately 70% of Ghalilah-type tombs were oriented

north-south at the vast second millennium BC cemetery of Shimal (Vogt 1998).

Such tombs have been reported at Ghalilah, Kalba, and Shimal (Potts 1990;

Hellyer 1993, 1998). 86

Figure 2.21. Ghalilah type above-ground collective tomb from Shimal (SH 103), Emirate of Ras al-Khaimah, Emirate United Arab Emirates (from Velde nd:305). Note the distinct outer and inner layers of the exterior wall.

c. Khatt: The above-ground, collective Khatt-type tombs bear a striking

resemblance to Shimal-type graves in that each consists of an oval stone structure

surrounding a single, long burial chamber (Figure 2.22) (Potts 1997a). The major

difference lies in the addition of an outer ring-wall encircling its ovoid interior,

creating a passageway between the two (Hellyer 1998; Potts 2001). Hellyer

(1998) speculates that this secondary ring-wall may not have been part of the

original tomb plan, but was instead a later addition to accommodate more

individuals, while Vogt (1998) believes that the rarity of these tombs (e.g., only

three are reported from the 200+ present in Shimal) in combination with their 87 (a) (b)

Figure 2.22. Khatt type above-ground collective tomb from Bithna, United Arab Emirates (from Potts 1990:238 (a) and Hellyer 1998:73 (b)).

unique structure may reflect a reserved resting place for a small subset of Wadi

Suq society. Khatt-type tombs have been described as the most “monumental” of

all Wadi Suq collective burials but have only been recognized at Bithna, Khatt,

and Shimal (Potts 1990; de Cardi et al. 1994).

d. Dhayah: Very little has been written on the Dhayah-type tombs, in part because

so few have been discovered. These T-shaped graves take on a more rectangular

shape than other Wadi Suq graves, and while these represent collective burials,

are semi-subterranean in construction (Figure 2.23) (Potts 2001). Of more than

Figure 2.23. Dhayah type subterranean T-shaped collective tomb from Bithna, Emirate of Fujairah, United Arab Emirates (from Potts 2001:45).

200 monuments at the Shimal necropolis, only two are representative of the

Dhayah-type form, and as with Khatt-types, have led some to suggest that these

may have housed a special subset of the population (Vogt 1998). Examples of

Dhayah-type tombs are known from Bithna, Dhayah, and Shimal (Kästner 1991;

Corboud et al. 1990; Corboud et al. 1996; Vogt 1998).

e. Horseshoe: Another rare burial form, horseshoe-shaped tombs are scattered

along both the east and west coasts of the United Arab Emirates but remain few in

number (Figure 2.24) (Hellyer 1998). These subterranean second millennium

burials have been recognized on the Al Madam Plain in the Emirate of Sharjah as

well as at Dibba, Jebel al-Buhais, Qidfa, and Wadi al Qawr (Phillips 1987;

Hellyer 1998; Potts 2001).

Figure 2.24. Horseshoe-type subterranean collective tomb from Wadi al-Qawr, Emirate of Ras al-Khaimah, United Arab Emirates (from Phillips 1987:Fig. 3).

f. Circular: Unlike the vast majority of Wadi Suq burial monuments, which were

built in the shape of an oval, a few retained the circular tomb shape previously

seen in the Umm an-Nar period. Circular graves have been noted at Ghalilah,

Masirah Island, and Shimal, which alone has 28 tombs of this type (Potts 1990;

Vogt 1998).

g. Single cists: Single burials dating to the Wadi Suq have also been recovered

throughout the Oman Peninsula but are rarely described in detail. Subterranean

chambers lined with stone characterize these individual (and rarely, double)

inhumations (Potts 1990, 2001). In some cases, above-ground tomb markers such

as stones or low earthen mounds delineate the position of these burials (Potts 90 1990). At Shimal, individual cist burials far outnumber collective tombs, and

each contained an articulated, flexed skeleton (Vogt 1998). At Maysar 9 in

Oman, a subterranean rectangular chamber was recorded with dimensions up to

3.1 by 1 m (Potts 1990). More unusual individual interments, including a single

rockshelter tomb from Shimal, also appear in the Wadi Suq (Vogt and Franke-

Vogt 1987). Other examples of single cist graves come from al-Qusais, Khudra,

Wadi Salh, and Wadi Suq (Taha 1982-3; Potts 1990, 1997a). Second millennium

sites from Oman tend to produce graveyards characterized more by these single

interments than in the UAE (Vogt 1998).

Continuity with the Umm an-Nar period is evident in many of the burial practices of the Wadi Suq. Grave architecture, while different in shape, is similar to third millennium collective tomb plans, particularly with features including double-layered walls and entrance construction. Communal tombs persist in the Wadi Suq, although those interred number less than in the previous period (Hellyer 1998). While single inhumations increased markedly in the Wadi Suq, their presence is not unheard of in

Umm an-Nar contexts, as individual burials from the late third millennium at Wadi

‘Asmiah illustrate (Vogt 1994b, 1998). As in the Umm an-Nar, the dead were originally placed in a flexed position before being disturbed by later burials, contributing to commingling, although cremation is no longer performed (Vogt 1998). Likewise, tomb membership includes individuals of all ages and both sexes (Wells 1984; Schutkowski and Herrmann 1987; Potts 1990).

91 Although tomb forms display a significant amount of variation, the deposition of relatively homogeneous grave goods within numerous Wadi Suq burials suggests persistent, widespread communication and a cohesive cultural ethos throughout southeastern Arabia. Nevertheless, major differences with the preceding Umm an-Nar in material culture set the Wadi Suq apart. Considerably more grave goods were placed in

Wadi Suq tombs than found in Umm an-Nar graves, and local wares dominated funerary assemblages (Vogt and Velde 1987). These tomb artifacts display continuity with the

Umm an-Nar period but underwent substantial technological and stylistic changes (Potts

1997a). Interestingly, at Shimal, Wadi Suq pottery assemblages differ between domestic and funerary contexts, with medium wares characterizing the settlement and fine, painted wares typical of mortuary goods (Vogt and Franke-Vogt 1987). This distinction has been attributed to temporal differences in ceramic production, with early second millennium tombs displaying parallels with the predominance of fine wares also seen in the preceding

Umm an-Nar period in contrast to the coarses wares of the later, mid-second millennium

Shimal settlement (Velde 1991). This supposition is supported by corresponding early

Wadi Suq fine wares uncovered from settlement contexts at Hili 8, Tell Abraq, and Tawi

Sa’id (Potts 1990; Mery 1991; Velde 1991).

The most perceptible change was the introduction of new and innovative forms of metal weaponry into second millennium tombs, including the bow and arrow, the long sword, and a socketed spearhead, diverging sharply from the simple daggers and spears of the Umm an-Nar (Vogt 1998; Weeks 2000; Potts 2001). Additionally, the manufacture of local soft-stone vessels increased substantially, accompanied by novel changes to overall shape and design (Haser 1990, 1991). Gold and electrum pendants in

92 the shape of animals found at Bidya, Dhayah, and Qattarah may indicate affluence among some members of the population of Magan, possibly related to the copper trade with

Dilmun (Hellyer 1998; Potts 2001). Moreover, the impressive technical competence exhibited by Umm an-Nar potters was seemingly lost, shown by a notable decline in the both the manufacturing and artistic quality of Wadi Suq ceramics (Laursen 2009; Potts

2009). Even here, however, transitional ceramic forms between these two periods, recovered at both Tell Abraq and Nud Ziba, suggest that this shift was “evolutionary, rather than revolutionary” (Kennet and Velde 1995; Potts 1997a:52). The presence of foreign imports from Mesopotamia, Middle Asia, the Indus Valley, and Dilmun witnessed a dramatic decline and will be discussed further below.

Interregional Exchange

The substantial discontinuities seen during the Wadi Suq cannot entirely be attributed to the regional aridification event that took place at the end of the third millennium BC. Instead, the impact of sociopolitical and economic factors must also be considered. In particular, an apparent collapse in interregional exchange networks across the Gulf played a critical role in the marked decline of southeastern Arabia.

While third millennium trade relations between the Oman Peninsula and

Mesopotamia were briefly restored during the short-lived Ur III dynasty (ca. 2100-2004

BC), its collapse brought about the subsequent Isin-Larsa (ca. 2000-1800 BC), Old

Babylonian (ca. 1800-1600 BC), and Kassite (ca. 1600-1200 BC) periods, which saw the disintegration of commercial exchanges between these two regions (Crawford 1998;

Oates 2008; Potts 2009). Second millennium Mesopotamian cuneiform records no longer

93 referenced Magan in administrative or economic dealings; instead, this focus had been entirely transferred to Dilmun, which had come to dominate Gulf trade during the early

Wadi Suq (Crawford 1998; Carter 2003c). Mesopotamian wares in southeastern Arabia occurred infrequently, and while a few sherds limited to Tell Abraq and Kalba were recovered, no Wadi Suq graves contained Mesopotamian ceramics (Potts 1992; Carter

1997, 2003).

However, like Mesopotamia, Dilmun possessed no natural sources of copper and would have been forced to acquire it elsewhere before engaging in trade with

Mesopotamia. While it is possible that copper was transported from southeastern Arabia to Dilmun (Potts 1990), little evidence for copper mining and smelting activities exists in the Oman Peninsula during this time (Weeks 1997; Carter 2003c). Additionally, with the exception of a stamp seal from Mazyad, a few sherds from Kalba 4, and a sizeable horde of Barbar pottery dating to the late Umm an-Nar/early Wadi Suq at Tell Abraq, surprisingly few second millennium Dilmunite materials have been found in southeastern

Arabia that would indicate continued trade relations (Carter 1997; Crawford 1998; Potts

2009). Correspondingly few Wadi Suq vessels were recovered in Bahrain; no materials were identified at either Qala’at al-Bahrain or on the Dilmun colony of Failaka Island, while Saar possessed only three sherds (Carter 2003c). Carter (2003c) thus argued that

Dilmun attained copper from a variety of locales outside of southeastern Arabia, most probably to the east, where rich stores of copper from eastern Iran and the Indo-Iranian borderlands are known today. These suspicions were supported by an analysis of copper from second millennium contexts at Saar, which showed that not all Dilmun copper originated from Arabia (Weeks and Collerson 2005; Laursen 2009).

94 The conclusion of the Mature Harappan phase and the resultant cultural breakdown of the Indus Valley civilization around 1900 BC also had a significant impact on the peoples of the Oman Peninsula. With the ushering in of the Late/Post-Harappan period (ca. 1900-1300 BC) in the early second millennium, once-reliable communication and exchange networks with Magan and elsewhere broke down, and the previously dominant role played by the merchants of the Indus Valley in late third millennium trade faded as the entrepôt of Dilmun rose to prominence in the Gulf (Kenoyer 2000; Carter

2003c; Reade 2008; Potts 2009). This so-called ‘collapse’ of the Indus civilization was previously attributed to foreign invasion by the Aryans of central Asia, although archaeological and bioarchaeological evidence has since discredited this assertion (Dales

1964; Kennedy 1994; Kennedy et al. 2000). Additionally, the simplistic view of urban center abandonment and subsequent population dispersal was likely a more nuanced process of decentralization as a result of environmental and economic decline (Kenoyer

1991). As in Magan, few Post-Harappan settlements have been identified.

Despite the disruption of these trade routes, exchange between southeastern

Arabia and the Indus Valley was not completely severed. Evidence of a few Post-

Harappan pottery sherds and seal impressions at Tell Abraq illustrate a potentially ongoing, albeit declining, exchange relationship (Potts 2001, 2009). Furthermore, the unusual presence of both a Harappan jar and a cubical Indus-type weight within Shimal

Tomb 6 prompted de Cardi (1989) to assert that an Indus merchant might have been interred there. Nevertheless, these finds indicate that while the Oman Peninsula briefly maintained contact with the Indus Valley just after the collapse of Mature Harappan

95 culture, a period of isolation throughout southeastern Arabia soon followed (Carter

2003c).

A brief mention must also be made regarding eastern Iran and the Indo-Iranian borderlands of Middle Asia. The strong association between Middle/Central Asia and the

Oman Peninsula that existed during the Umm an-Nar also abruptly ended at the latter part of the third millennium, with depopulation and urban collapse plaguing the Indo-Iranian borderlands as well as areas and cities further east, including Makran and Kerman (Carter

2003c).

Consequently, soon after the commencement of the Wadi Suq period, southeastern Arabia seems to have fallen into a period of cultural isolation, segregated from the once productive interregional exchange networks of the Persian Gulf. Local products dominated the assemblages recovered from tombs and settlements throughout the peninsula during the early second millennium BC. The loss of major trading partners in both Mesopotamia and the Indus Valley likely dealt a significant blow to the economy of the region, particularly with regards to the copper trade, and this, together with increasingly arid environment, forced the Wadi Suq peoples of the Oman Peninsula to adapt to new physical and socioeconomic surroundings. This may have prompted the apparent population dispersal from larger settlements, a shift to a more mobile lifestyle, and modifications in subsistence strategy, although a few large fortified sites (e.g., Tell

Abraq, Kalba) continued to successfully employ a balanced, mixed economy established during the Umm an-Nar.

96 Statement of Hypotheses

This study will focus on two main hypotheses, as well as multiple sub-hypotheses, that seek to assess temporal changes in mobility and diet, and from this, address broader questions relating to political economy and social identity throughout the Early and

Middle Bronze Age. To accomplish such an evaluation, this project brings together archaeological (particularly mortuary), biogeochemical, and theoretical approaches to more fully conceptualize (a) the role that mobility played amongst the local population of the Oman Peninsula, (b) the nature of social interaction framed in an interregional context, and (c) how tomb membership played a pivotal role in negotiating identity in the context of the so-called “peripheral” locale of southeastern Arabia. In addition to these hypotheses, reactions to Wallerstein’s world-systems theory, Stein’s distance-parity and trade-diaspora models, and Cleuziou’s social dynamics model will be given in light of archaeological coupled with isotopic evidence.

Hypothesis I: The Umm an-Nar (2500-2000 BC) population of the Oman Peninsula would have been highly mobile as a result of increasingly complex and widespread interregional exchange networks.

Monumental collective tombs housing hundreds of individuals appear for the first time in southeastern Arabia during the Umm an-Nar period, coinciding with a peak in interregional trade with civilizations in Mesopotamia, the Indus Valley, Dilmun, Elam, and Central Asia. As inhabitants of this region became progressively more engaged with this ‘global’ community, mobility should correspondingly increase as local regional networks expanded and became more interactive to meet the growing foreign demand for local products, especially copper. It is expected that in an analysis of stable strontium,

97 oxygen, and carbon isotopes, individuals interred in Umm an-Nar graves will exhibit highly variable isotopic signatures reflective of developing regional and interregional economic relationships. Such elevated mobility is expected despite an increasingly sedentary lifestyle associated with the adoption of cultivars via date palm gardens. If these individuals did not become more mobile under these changing socioeconomic conditions, it is expected that local isotope ranges will be narrow and that little variability will be demonstrated between tomb members.

Hypothesis Ia: Non-locals will be present within Umm an-Nar communal tombs.

A major shift in mortuary practices in the mid-third millennium BC is evident in the archaeological record of southeastern Arabia, as small, (likely) family-based cairns of the preceding Hafit (3100-2500 BC) period were replaced by large, communal Umm an-

Nar tombs. As highly visible structures on the landscape, monumental tombs such as these are believed to represent markers of territoriality legitimized by the tomb members themselves, often assumed to be ancestors once part of the local community. However, as commerce became increasingly important during the Umm an-Nar period, particularly at trading hubs represented by large settlements and exotic wares, this mortuary ideology

– and more specifically, definitions of kinship and tomb membership – may have gradually become more flexible to better meet the needs of the local community and those they interacted with. It is therefore hypothesized that non-local individuals were interred alongside local community members in these monumental funerary structures. If this is the case, it is expected that non-local isotope signatures will be present and will identify individuals whose ratios deviate significantly from locally defined ranges.

98 Should no foreigners be present, isotope ratios should be more constricted in value, clustering within local ranges.

Hypothesis Ib: Umm an-Nar tombs associated with large settlements will contain greater numbers of non-locals than tombs with no associated permanent settlement.

Many Umm an-Nar tombs are closely associated with large settlements (e.g.,

Umm an-Nar Island) or fortress towers (e.g., Hili, Tell Abraq), some of which are positioned only a few meters away. These sites possess abundant evidence of foreign contact, particularly amongst grave goods, suggesting that bigger settlements acted as centers of trade. Such hubs of interregional exchange would have increased contact between locals and non-locals, attracting foreign merchants, ambassadors, or immigrants and encouraging exogamy. However, a lack of permanent settlement associated with

Umm an-Nar tombs at Mowaihat and Unar 1 is reflected in the few foreign wares recovered from both funerary structures. As a result, it is expected that more non-locals will be interred in tombs linked to large settlements and/or towers. This hypothesis will be tested by evaluating whether a greater number of non-local isotope ratios are found in

Umm an-Nar tombs associated with permanent public architecture. If no relationship exists between the presence of non-locals and settlement size, there should be no trend associating isotopic outliers with sizeable, permanent settlements.

Hypothesis Ic: Diet was extremely variable during the Umm an-Nar period.

The Umm an-Nar inhabitants of southeastern Arabia exploited a variety of environments and resources to obtain food. Past subsistence strategies, including

99 hunting, fishing, and pastoralism, continued to be employed, while new methods of food production, particularly small-scale garden agriculture, also played an increasingly important role. Both wild and domestic ruminants likely grazed primarily on C4 plants, while Bronze Age cultivars (e.g., dates, wheat) would reflect a preference for C3-based food sources. Moreover, cuneiform records from Mesopotamia speak to the products exported to Magan in the third millennium BC, including sesame oil and barley (Leemans

1960; Potts 2000). It is thus hypothesized that diet was extremely variable during the

Umm an-Nar period, particularly relative to the later Wadi Suq period (see Hypothesis

IIb). This hypothesis will be tested by examining stable carbon isotope variability, both at the site and regional level. If diet was not broad during this time, carbon isotope value ranges should be limited in scope.

Hypothesis II: Interregional mobility decreased during the Wadi Suq (2000-1300 BC) in the Oman Peninsula as a result of a “collapse” of interregional exchange networks.

At the beginning of the second millennium BC, monumental collective tombs were replaced by small communal or single interments, settlements decreased dramatically in size and number as a purported shift to a more nomadic lifestyle was undertaken, and an apparent collapse of interregional exchange networks across the

Persian Gulf threw southeastern Arabia into a period of relative cultural isolation.

Subsequently, interregional mobility should decrease as these economic ties break down.

It is expected that in an analysis of stable strontium, oxygen, and carbon isotopes, individuals interred in Wadi Suq graves will exhibit more limited isotopic variability

(relative to the preceding Umm an-Nar period) reflective of these radical socioeconomic,

100 mortuary, and subsistence changes. While regional mobility among locals may have increased as these shrinking population became more nomadic, their movements would still fall within a local geographic expanse that would not likely produce substantial isotopic differences.

Hypothesis IIa: Non-locals will be absent from Wadi Suq communal tombs.

A major shift in mortuary practices with the commencement of the second millennium BC is evident in the archaeological record of southeastern Arabia, as large, communal Umm an-Nar tombs housing hundreds of individuals were replaced by (a) small communal tombs of variable construction type containing just a few individuals, or

(b) single interments. The majority of these tombs were semi-subterranean and would have been more difficult to see on the landscape surface relative to the highly noticeable

Umm an-Nar tower tombs. The loss of major trading partners in both Mesopotamia and the Indus Valley dealt a significant blow to the economy of the region, particularly with regards to the copper trade, forcing the Wadi Suq peoples of the Oman Peninsula to adapt to new socioeconomic surroundings. It is hypothesized that interregional contact with non-locals would have decreased dramatically, resulting in an absence of non-local tomb members in Wadi Suq burial structures. If this is the case, it is expected that only local isotope signatures will be present among the individuals interred in these tombs. Should foreigners be present, non-local isotope signatures will identify individuals whose ratios deviate significantly from locally defined ranges.

101 Hypothesis IIb: Dietary variability decreased during the Wadi Suq period.

Inhabitants of the Oman Peninsula were forced to adapt not only to changing socioeconomic circumstances, but also environmental conditions as their surroundings became increasingly arid. Wadi Suq settlements were small and scarce, and likely reflect a reduced population that gradually returned to a more mobile lifestyle. Subsistence strategies were correspondingly modified, with archaeobotanical and zooarchaeological assemblages indicative of a dietary shift from agro-pastoralism towards marine resources.

In addition, the diminished availability of imported foodstuffs would have produced a more restricted, less variable diet during this time. It is hypothesized that diet was more narrowly focused and that dietary variability decreased in the Wadi Suq relative to the

Umm an-Nar. This hypothesis will be tested by examining stable carbon isotope ratio variability, both at the site and regional level. If diet was not limited but instead broader during this time, carbon isotope value ranges should be more variable in scope.

***

In summary, major transitions in mortuary practices, settlement organization and architecture, subsistence, and external trade relations are evident throughout the Bronze

Age. The Hafit period is predominantly represented by hundreds of stone cairns found across the Oman Peninsula and is notable for the reappearance of interregional exchange networks, particularly with Mesopotamia as its demand for copper increased. By the

Umm an-Nar period, the cultural landscape of the region had changed dramatically, evidenced by the presence of large, agriculturally-based settlements with fortification towers and monumental tombs containing hundreds of individuals, the commencement of the first local ceramic tradition in the region, and the presence of a considerable number

102 of non-local goods from across the Gulf, including Mesopotamia, the Indus Valley,

Bahrain, Iran, and Central Asia. However, an apparent breakdown of these trade relations, first with Mesopotamia and later the Indus Valley, instigated a period of supposed cultural isolation that was accompanied by population decline, increasingly variable mortuary practices, and shifts to a more mobile lifestyle. For this study, hypotheses examine the role that mobility played in southeastern Arabia among the local population in light of interregional interactions as well as the role of tomb membership in negotiating identity in this purportedly peripheral setting.

103

CHAPTER 3

STABLE ISOTOPE ANALYSIS: THEORETICAL CONCEPTS AND APPLICATION

Biogeochemical analysis of bone and teeth by means of stable isotopes can provide critical insight into past ways of life, particularly in terms of explicating mobility and subsistence patterns over time. Isotopes exist as atomic variations of elements whose nuclei possess the same atomic number. Subsequently, all isotopes of an element contain the same number of electrons orbiting the nucleus as well as the same number of protons within the nucleus (Hoefs 2004). However, due to the diverse numbers of neutrons in their nuclei, isotopes of the same element vary in atomic mass (Schwarcz and

Schoeninger 1991). Those with a greater number of neutrons weigh more and can subsequently be distinguished from their lighter counterparts. Isotopes are characteristically denoted as mE, where E refers to the particular element under discussion, while m indicates atomic mass (Hoefs 2004). Atomic mass is determined by the number of protons and neutrons in the nucleus, whereas electrons weigh so little that their mass is negligible and is not considered.

Isotopes may be classified as either stable or radioactive (unstable). Stable isotopes do not undergo radioactive decay and thus retain the same atomic configuration

104 over time. Some 300 stable isotopes have been identified, with most elements possessing at least two stable isotopes (although 21 ‘pure elements,’ or those elements with only one stable isotope, do exist) (Hoefs 2004). In many cases, the natural abundance of various isotopes of the same element differs considerably; typically, one isotope dominates the natural environment, while its other variants are present in much smaller quantities.

Stable isotope values are expressed as a ratio, measuring the deviation of the actual sample from an internationally accepted standard of constant value. The delta symbol (δ) preceding the heavier isotope denotes this ratio (Schoeninger 1985). For oxygen and carbon stable isotopes:

δ = [(Rsample – Rstandard) / Rstandard] x 1000

So that:

13 13 12 13 12 δ Csample = [( C/ Csample / C/ Cstandard) – 1] x 1000

18 18 16 18 16 δ Osample = [( O/ Osample / O/ Ostandard) – 1] x 1000

Different substances (biological or inanimate) possess disparate oxygen and carbon isotopic values due to processes of fractionation. This process involves a partial,

‘fractional’ separation between lighter and heavier isotopes during the chemical reactions that occur in predictable ways as a result of a transition from, for example, one trophic level to another, as with carbon, or from one state of matter to another, as with oxygen during evaporation or condensation (Hoefs 2004).

105 Stable carbon isotope values are reported relative to the Vienna Pee Dee

Belemnite carbonate standard (VPDB), while oxygen ratios are reported relative to the

Vienna Standard Mean Ocean Water (VSMOW). Values measured are expressed in parts per thousand, or parts per mil (‰). On the other hand, because the strontium isotope ratios of 87Sr to 86Sr do not demonstrably change as they move from bedrock through the food chain into human consumers, their values are simply expressed in this ratio of

87Sr/86Sr, reported to five decimal places (Bentley and Knipper 2005; Bentley 2006).

This project seeks to examine the geographic origins of individuals interred in

Bronze Age tombs across the Oman Peninsula by analyzing stable strontium, oxygen, and carbon isotopes. These three isotopes are described below.

Strontium

Strontium (Sr) isotopes provide a new way of approaching migration studies, acting as useful geochemical signatures in evaluating population movement or immigration within an individual’s lifetime. Strontium occurs naturally in the geologic environment as four different stable isotopic variants, including 88Sr, 87Sr, 86Sr, and 84Sr, within igneous rock (Faure 1986; Ezzo 1994) (Table 3.1). A minimum of 16 unstable

Table 3.1. Natural abundances of strontium isotopes in the environment.

Strontium Isotope Natural Abundance # of Protons # of Neutrons 88Sr 82.53% 38 50 87Sr 7.04% 38 49 86Sr 9.87% 38 48 84Sr 0.56% 38 46

106 (radioactive) strontium isotopes are also known to exist, including 90Sr, a product of nuclear reactions, but are not discussed here (Langham and Anderson 1957; Hodell et al.

2004). Only 87Sr is both stable and radiogenic, a product of decay from the alkali metal rubidium-87 (87Rb), which possesses a half life (4.7 x 1010 years) so gradual that almost no change in the ratios of strontium isotopes in minerals can be detected over the past several hundred thousand years, making a comparison of values from prehistoric archaeological materials appropriate (Faure 1986; Bentley 2001; Ezzo and Price 2002;

Dickin 2005). Because the abundance of 87Sr varies across the Earth’s crust by (a) mineral type and (b) mineral age, each kind of rock possesses a distinctive strontium isotope signature based on the degree of rubidium decay that can be measured against a constant baseline, the non-radiogenic 86Sr (Ericson 1985; Budd et al. 2000; Bentley

2006).

(a) Mineral Type: Minerals each have distinct 87Sr/86Sr values in part because of their

initial Rb/Sr values during formation. This is primarily determined by the amount

of potassium (K) in a particular mineral. Rb1+ substitutes for K1+ within the

mineral lattice due to similarities in the size of their ionic radii (measured in

ångströms, or Å, Rb = 1.52 Å versus K = 1.38 Å), and as each mineral type

contains differential amounts of K relative to calcium (Ca), Rb/Sr ratios are

distinct for different minerals (Ericson 1985; Bentley 2006).

(b) Mineral Age: Age plays a key role in the amount of Rb/Sr in a particular mineral

(Dasch 1969). Granite and other older rocks formed more than 100 million years

ago (mya) originally had high Rb levels as a result of their origins in the Earth’s

crust, where Rb is particularly abundant, and thus possess elevated 87Sr/86Sr

107 values (Wright 2005) (Figure 3.1). This is because of the inverse relationship

between 87Sr and 87Rb; as the amount of 87Rb decreases through decay over time,

87Sr increases (Bentley 2006). Conversely, younger volcanic rocks generated by

magma in the mantle of the Earth less than 10 mya, such as basalts, have a

significantly lower Rb content, resulting in lower 87Sr/86Sr values (Bentley 2006)

(Table 3.2).

Figure 3.1. Strontium isotope evolution of the Earth, with evolution of high Rb/Sr crust created at 3.8 billion years ago (bya) and the evolution of a mantle being continuously depleted. After Bentley (2006: Fig.1) and White (2007:Fig.8.7). = Assumed initial 87Sr/86Sr of the solar system

Strontium is found in bedrock but is released into local ecosystems by weathering processes and passed from groundwater into the plants and animals consumed by humans

(Price et al. 1994a). Because different regions can be characterized by distinctive isotopic ratios based on the relative amounts of minerals present in rock, subsequent strontium uptake in the skeleton reflects the specific geologic locality where isotopic values were ingested (Ericson 1985; Price et al. 2002). Due to strontium’s high atomic

108 Table 3.2. 87Sr/86Sr values of various natural materials on Earth.

87Sr/86Sr Reference Material Characteristics References Value General Sr variation in Bentley et al. 2006; 0.700-0.750 - environment Price et al. 2002 Bentley 2006; Basalt 0.703-0.704 Relatively invariant Ericson 1985; Price et al. 2002 Younger rock <0.706 Low original Rb/Sr Bentley 2006 Seawater - (Late Cretaceous) 0.7074 Hodell et al. 2004 Value dependent on Marine sedimentary Bentley 2001; 0.707-0.709 seawater during time rock Wright 2005 of formation Hodell et al. 2004; Seawater (Modern) 0.70923 - Sealy et al. 1991 Marine plants/animals 0.70923 Reflect seawater Sealy et al. 1991 (Modern) Large natural Bentley 2001, Granite 0.710-0.740 variation 2006; Ericson 1985 Older rock >0.710 High original Rb/Sr Bentley 2006

mass, its isotopes do not undergo any measurable amount of fractionation as they pass through various trophic levels and into the skeleton, so that the ratio of 87Sr to 86Sr remains constant (Hurst and Davis 1981; Graustein 1989; Beard and Johnson 2000). This should not be confused with elemental strontium, which does experience biopurification, its Sr/Ca ratios diminishing as it travels through trophic systems (Elias et al. 1982; Ezzo

1994).

Strontium is also present in seawater, its ratios an averaged product of the weathering of continental crusts across the planet; consequently, its values fall midway between young and old rocks (Bentley 2001) (see Table 3.2). Strontium ratios in the oceans are thus homogeneous throughout the world at any given time, although its

109 87Sr/86Sr ratios have fluctuated between 0.707-0.709 over the course of prehistory

(Bentley 2006) (Figure 3.2). Subsequently, 87Sr/86Sr ratios of any substances formed in or living in seawater, including marine plants and animals, shells, carbonates, and marine sedimentary rock, will reflect those of the water during that time (Sealy et al. 1991;

Wright 2005).

1.6 Tertiary mya 66 Cretaceous mya 138 Jurassic mya 205 Triassic 240m20 Permian 5ya 290mya Pennsylvanian 330 Mississippian 360 Devonian 410mya Silurian 435mya Ordovician mya 500 Cambrian mya 570 mya Period mya

Figure 3.2. 87Sr/86Sr in seawater through Phanerozoic time determined from analysis of phosphate and carbonate fossils (from White 2007:326). Dashed line shows the composition of modern seawater.

110 It is important to keep in mind that rocks are often composed of multiple minerals, each of which has distinct strontium values that can vary considerably from one another (Bentley 2006). Strontium content of bedrock is thus quite variable, and such variability is further enhanced by differential weathering between different mineral types

(Price et al. 2000; Hodell et al. 2004; Bentley 2006). Accordingly, bedrock may possess

87Sr/86Sr ratios unlike those found in plants and animals in the same region and cannot be relied upon to accurately predict ratios in the food chain.

Other sources of strontium input exist as well and may skew interpretations of its ratios, particularly in an evaluation of human values. For instance, while weathered minerals largely determine the isotopic value of water, the potential integration of much older groundwater from deep in the Earth will alter this modern ratio (Jorgensen et al.

1999). Moreover, sea spray and atmospheric conditions such as precipitation may contribute strontium to an ecosystem (Probst et al. 1992, 2000; Vitousek et al. 1999;

Whipkey et al. 2000). Fertilizers may also affect strontium concentrations in soil (Négrel and Deschamps 1996; Böhlke and Horan 2000; Bentley 2006). While these contributions are negligible in most cases, and while regional geology mostly dominates 87Sr/86Sr ratios in a given area, local environments must be seen as an amalgam of different inputs that may not be reflective of the 87Sr/86Sr ratios present in the geologic substrate, a substrate comprised of minerals often isotopically dissimilar (Price et al. 2002). Subsequently, it becomes necessary to distinguish between geological and biologically available strontium values in order to assess what it means to be local vs. non-local (Sillen et al. 1998).

How can such environmental heterogeneity be overcome and bioavailability measured effectively? Plant metabolism acts to average much of the variability inherent

111 in soils and water, and as such, plants display ratios more representative of the strontium available from the environment (Bentley 2006). However, this “averaging effect”

(Bentley 2006:150) is limited to a single plant’s exposure to a very confined area and may not reflect strontium bioavailability in the total environment; thus, it becomes even more efficient to move through higher levels of trophic systems into animals (Burton and

Price 1999). Animals like herbivores typically consume different types of plants over a wider area, generating strontium ratios that more broadly reflect available strontium in that locale (Sealy et al. 1991; Sillen et al. 1998). The averaging effect is bolstered by the nature of bone formation and remodeling. Because bone is continuously remodeled, ingested strontium isotopes are constantly incorporated into hydroxyapatite, such that strontium ratios in faunal remains represent averages generated over longer periods of time and by a variety of foods (Price et al. 2002; Bentley 2006).

Due to this high degree of homogenization, faunal averages consistently display low standard deviations (s.d.), making it preferable to analyze the 87Sr/86Sr ratios of local animals metabolizing foods from the same environment as humans at a specific site; this will determine the bioavailability of local strontium levels and hence serve as a baseline from which human isotope ranges will be defined (Koch et al. 1992; Sillen et al. 1998;

Price et al. 2002; Bentley et al. 2004; Turner et al. 2009). In this way, local and non-local individuals can be distinguished from one another. After determining the mean 87Sr/86Sr ratio from faunal remains, a recommended ±2 s.d. from this mean is applied to define

‘local’ human values, while anything outside of these limits is considered nonlocal or immigrant (Grupe et al. 1997).

112 The use of several different species of fauna is preferable, again to ensure adequate sampling of the strontium available in that environment. While present-day samples have been utilized in some cases to determine the 87Sr/86Sr ratios of a particular region and applied archaeologically to delineate local values (e.g., Bentley et al. 2004;

Hodell et al. 2004), the skeletons of these modern animals may contain contaminants from pollutants including fertilizers or may simply misrepresent the ratios of local strontium if fed imported diets made of non-local products (Böhlke and Horan 2000;

Price et al. 2002; Bentley and Knipper 2005). This makes the use of fauna dating to the same period as human samples a better option in establishing local strontium bioavailability.

The inorganic portion of enamel and bone consists primarily of a biological apatite (bioapatite) known as calcium phosphate hydroxyapatite, or Ca10(PO4)6(OH)2 .

Those elements structurally similar to calcium, including strontium, barium (Ba), and lead (Pb), may actually replace it in trace amounts as it becomes fixed in the hydroxyapatite crystal lattice (Ezzo 1994). The intestine selectively absorbs calcium during digestion; however, because the ionic radius of strontium (1.12 Å) closely resembles that of calcium (0.99 Å), a small amount of dietary strontium is absorbed as well, substituting for calcium in the formation of bone and enamel hydroxyapatite

(Nelson et al. 1986; Aufderheide 1989; Price et al. 2002; Knudson 2004; Bentley 2006).

The vast majority of absorbed strontium in the body is in fact stored in hydroxyapatite

(Grupe et al. 1997).

In the burial environment, the skeleton is at risk of diagenesis, or biogeochemical changes that can cause substantial shifts in isotopic composition. While hydroxyapatite

113 dominates the composition of both bone and enamel, bone hydroxyapatite is predisposed to diagenetic alteration because of the nature of its crystal lattice, which is extremely porous and is comprised of relatively small crystals that easily undergo recrystallization in the post-depositional environment (Koch et al. 1997). Despite numerous methodological attempts to rid bone hydroxyapatite of diagenetic strontium, most involving leaching in weak acid baths, exogenous strontium from surrounding soils or groundwater contaminates bone by way of elemental exchange with the crystal surfaces of biogenic hydroxyapatite (Hoppe et al. 2003; Bentley 2006). Even in buried human bone as young as twenty years old, Beard and Johnson (2000) found considerable strontium diagenetic signals and were unable to isolate biogenic hydroxyapatite in spite of the application of acid leaching techniques. Lastly, as a specialized form of connective tissue, bone is partially made up of an organic phase (~30%), making bone hydroxyapatite more easily influenced by diagenetic processes (LeGeros 1981; Price

1989).

Conversely, while dental enamel also contains both an organic and inorganic phase, hydroxyapatite dominates its composition at around 96% after maturation and mineralization are fully complete (Nanci 2008). Only approximately 1% of mature enamel belongs to the organic phase; although this may seem insignificant, its benefit for stable isotope analysis lies in the fact that this organic phase is so tightly bound to mineral, conferring a considerable degree of protection against diagenesis and hence preserving biological apatite (Hillson 1996; Kohn et al. 1999). Additionally, enamel possesses comparatively larger crystallines (>1µm) and is significantly less porous than bone, creating a dense structure much more resistant to diagenetic alteration (Kolodny et

114 al. 1996; Hillson 1996; Kohn et al. 1999). While Horn and Muller-Suhnius (1999) and

Grupe et al. (1999) contended that it should be possible for some in vivo elemental uptake to take place in adulthood on the surface of teeth and at the enamel-dentin junction, no evidence for diagenesis at these sites has been recovered. Furthermore,

Budd and colleagues (2000) point out that no isotopic enrichment has ever been experimentally verified (see also Montgomery et al. 1999).

While stable strontium isotope studies have been established as a compelling means of examining past human mobility, particularly in the Americas (e.g., Price et al.

1994b, 2000, 2006; Ezzo and Price 2002; Buikstra et al. 2003; Hodell et al. 2004;

Knudson et al. 2004, 2005; Knudson and Buikstra 2007; Knudson and Price 2007; Quinn et al. 2008; Andrushko et al. 2009; Knudson and Blom 2009; Turner et al. 2009) and

Europe (e.g., Price et al. 1994a, 1998, 2001, 2004; Grupe 1995; Grupe et al. 1997;

Schweissing and Grupe 2000, 2003a, 2003b; Hoogewerff et al. 2001; Bentley 2006,

2007; Bentley et al. 2002, 2003, 2004, 2005; Budd et al. 2004; Montgomery et al. 2003,

2005; Bentley and Knipper 2005; Evans et al. 2006a, 2006b; Price and Gestsdotti 2006;

Bickle and Hofmann 2007; Giblin 2009), few studies have been performed in Africa

(e.g., Cox and Sealy 1997), Asia and the Pacific Islands (e.g., Bentley 2004; Bentley et al. 2005, 2007; Haverkort et al. 2008; Valentine et al. 2008), and the Near East (Buzon et al. 2006; Tafuri et al. 2006; Cooper et al. 2007, Perry et al. 2008). No isotope studies involving human mobility have been conducted in the Arabian Peninsula.

115 Oxygen

Stable oxygen (O) isotope ratios act as ‘tracers’ of geographic residence in much the same way as strontium, thereby possessing the potential to explicate the impact of trade on mobility patterns in both local and interregional contexts. The most abundant element on the planet, oxygen occurs naturally in liquid, solid, and gaseous compounds as three different stable isotopic variants: 18O, 17O, and 16O (Table 3.3) (Hoefs 2004).

Because the 17O isotope is scarce, the ratio of 18O/16O is typically measured and reported as 18O. Oxygen isotopes are incorporated into the hydroxyapatite of bone and dental enamel by means of two oxygen-containing compounds, phosphate (PO4) and carbonate

(CO3) (Sponheimer and Lee-Thorp 1999). Hydroxyapatite normally contains phosphate

(Ca10(PO4)6(OH)2), but carbonate substitutions within hydroxyapatite at hydroxide (Type

A) or phosphate (Type B) sites are not uncommon (LeGeros 1991; Barralet et al. 1998).

Table 3.3. Natural abundances of oxygen isotopes in the environment.

Oxygen Isotope Natural Abundance # of Protons # of Neutrons 18O 0.1995% 8 10 17O 0.0375% 8 9 16O 99.763% 8 8

Enamel and bone carbonate may be classified as either structural or labile/ adsorbed. ‘Structural’ refers to carbonate more securely incorporated into hydroxyapatite and that which is ideally measured during isotope analysis, while ‘labile’ represents those carbonates integrated into immature, poorly-organized bone or adsorbed onto surface apatite crystals (Betts et al. 1981; Rey et al. 1989, 1991). As a result, labile carbonate 116 more readily undergoes diagenetic modification and recrystallization, but is generally more soluble than its structural counterpart and thus more readily dissolved with acidic treatment as part of the sample preparation process before isotope analysis (Garvie-Lok et al. 2004).

Oxygen isotope values in the human skeleton are in equilibrium with those of body water at a constant, normal body temperature of approximately 37C, or 98.6F despite external fluctuations in temperature (Luz et al. 1984; Turner et al. 2009). The oxygen isotope signature of an individual’s body water is determined by a variety of inputs and outputs. Inputs consist of ingested water and food as well as the intake of atmospheric oxygen during respiration, while outputs include the discharge of sweat, urine, and the release of carbon dioxide (CO2) during the exhalation process (Longinelli

1984; White et al. 2004). Despite these multiple isotopic contributions to bone and teeth, oxygen values in these materials largely reflect drinking water consumed from the local environment (Luz et al. 1984; Luz and Kolodny 1985; Dupras and Schwarcz 2001;

Knudson and Price 2007). Such ‘environmental water’ is in fact composed of two primary sources: 1) meteoric water, or atmospheric water precipitated as rain or snow, and 2) surface water, which includes groundwater as well as standing, recycled bodies of water like lakes, rivers, and springs (Dansgaard 1964). Because oxygen ratios in water depend on the unique geomorphology and meteorological cycles of a particular region, isotopes in the skeleton reflect the sources of water ingested in a specific area during enamel and bone mineralization (Dupras and Schwarcz 2001).

Distinct environmental variables facilitate an assessment of an individual’s place of origin and later migration to another, isotopically distinct geographic region (White et

117 al. 2000). Those variables influencing the isotopic composition of meteoric water and hence local drinking water, including precipitation, air temperature, latitude, altitude, distance from the coast, and aridity, are summarized in Table 3.4 and described below:

Table 3.4. Summary of oxygen isotopic contributions to environmental water.

Variable Direction Result Water Vapor Decrease in δ18O Hydrological Cycle Precipitation Increase in δ18O Increasing Increase in δ18O Air Temperature Decreasing Decrease in δ18O Increasing Decrease in δ18O Latitude Decreasing Increase in δ18O Increasing Decrease in δ18O Altitude Decreasing Increase in δ18O Distance from Increasing Decrease in δ18O continental coastlines Decreasing Increase in δ18O Increasing Increase in δ18O Aridity Decreasing Decrease in δ18O

(a) Precipitation: During the process of evaporation over bodies of standing water,

16 16 fractionation takes place as lighter O isotopes within water (H2 O) evaporate

with greater frequency than their heavier 18O counterparts, which are

preferentially left behind in liquid water (Bentley and Knipper 2005). Because of

this, water vapor possesses lower 18O values relative to those found in standing

water (Figure 3.3). As water vapor condenses in the atmosphere into

liquid precipitation, a second fractionation event occurs, in which heavier 18O

118 isotopes present in water vapor are preferentially concentrated into liquid water

18 due to the lower vapor pressure of H2 O (Dansgaard 1961; Yurtsever and Gat

1981). Consequently, precipitation contains elevated isotopic values in relation to water vapor but still reduced compared to standing water.

(b) Air Temperature: As temperatures become cooler, water vapor in the atmosphere condenses in the form of precipitation. As discussed previously, the condensation process preferentially removes heavier 18O isotopes from water vapor into precipitation, leaving behind lighter 16O isotopes in vapor; thus, subsequent precipitation from progressively 18O-depleted vapor becomes increasingly depleted itself (Figure 3.3). Consequently, decreasing temperatures produce lower stable oxygen isotope values in meteoric water (Yurtsever and Gat 1981).

(c) Latitude and Altitude: 18O values are dependent on precipitation and temperature; as a result, 18O values at various latitudes and altitudes are a direct reflection of these variables (Knudson 2009). 18O values are characteristically lower at higher altitudes than lower altitudes because of cooler temperatures found at increasing elevations (Dansgaard 1964). Correspondingly, latitude, which refers to the distance from the Earth’s equator, can be conceived as a measure of the amount of sunlight and hence heat that a particular locale receives.

The angle of incidence, or the angle at which the sun’s rays strike the Earth’s surface, is negligible at the equator, resulting in higher temperatures and higher

18O values in the tropics but lower 18O values with movement in latitude towards the poles.

119

Figure 3.3. A schematic diagram of the isotope fractionation process via evaporation and condensation. Note that waters are lighter when they evaporate and are relatively heavier when condensed in the form of precipitation (from Bruckner 2009).

(d) Distance from the continental coastlines: Above standing bodies of water and

near the continental coastlines, water vapor as a product of evaporation as well as

ensuing condensation and precipitation is relatively enriched when compared with

the values of water vapor and precipitation further inland. Despite a preferential

uptake of 16O during evaporation, coastal locations have a constant supply of 18O

from the ocean that is also carried into the atmosphere, albeit less frequently; as a

result, 18O values closer to these coastlines will reflect this augmented 18O

source. Conversely, with increasing distance from the coast, cycles of

evaporation, condensation, and precipitation persist, but because of the continuous

removal of the heavier 18O isotopes from water vapor during evaporation and a

120 corresponding lack of substantial 18O sources from which to draw upon, these

inland sites display increasingly negative 18O values that continue to decrease as

one moves further into the interior (Bentley and Knipper 2005).

(e) Aridity: Arid environments such as those in the Levant and Arabia experience

little rainfall and acute evaporative processes, resulting in meteoric values

relatively enriched in 18O, particularly along the coast (Gat and Dansgaard 1972;

Bajjali and Abu-Jaber 2001; Perry et al. 2009). Given these severe conditions, it

is not surprising that water consumed by populations living in arid environments

originates from multiple sources, including rainfall, underground wells/springs,

and wadis or seasonal rivers (Knudson 2009). For populations living on the coast,

water from these wadis originated from higher altitude precipitation with lower

18O values; nevertheless, the high degree of evaporation that commenced as

water was transported from higher altitudes to the coast resulted in elevated 18O

values (Gat and Dansgaard 1972; Gat 1996).

Additional fractionation of oxygen isotopes occurs during breastfeeding. As water is ingested by humans and becomes body water, a subsequent enrichment in 18O

16 16 isotopes takes place, with lighter O isotopes preferentially lost as water vapor (H2 O) is exhaled (Wright and Schwarcz 1998; White et al. 2004; Williams et al. 2005). Because body water acts as the source from which breast milk is created, breast milk correspondingly possesses elevated 18O values relative to drinking water that is passed on to infants until weaning (i.e., the cessation of breastfeeding) is complete (Wright and

Schwarcz 1998). Subsequently, throughout the duration of the weaning process, subadult 121 18O values gradually decrease with the inclusion of environmental water sources until adult 18O ratios are reached (Dupras and Tocheri 2007). However, unlike the more extreme fractionation of nitrogen isotopes as a result of breastfeeding, which results in an elevation of 15N values in the bone collagen of breastfeeding infants by approximately

2.0-3.6‰ (Katzenberg et al. 1996; Schurr 1997, 1998; Dupras et al. 2001; Schurr and

Powell 2005), 18O values appear elevated by only 0.5-0.7‰ (Wright and Schwarcz

1998; White et al. 2000; Dupras and Tocheri 2007).

Oxygen is incorporated into bone and enamel hydroxyapatite via both phosphate and carbonate. Initially, phosphate was considered a much more reliable measure of oxygen due to its ability to resist postmortem diagenesis relative to carbonate, a result of the strong covalent bond between phosphorus and oxygen atoms (Luz and Kolodny 1989;

Sponheimer and Lee-Thorp 1999; Dupras and Schwarcz 2001). On the other hand, carbonate has been shown to undergo significant diagenetic alteration, possessing a weaker chemical bond susceptible to diagenesis in the porous and poorly crystalline structure of bone (Lee-Thorp et al. 1989; Wang and Cerling 1994; Lee-Thorp and

Sponheimer 1999). However, enamel proves much more resilient to these postmortem ionic exchanges because of its higher mineral composition, larger crystallines, and fewer pores (Kohn et al. 1999; Hoppe et al. 2003; Prowse et al. 2007). Subsequently, the carbonate component of enamel retains its original oxygen isotope signature and may be used to assess the geographic location in which individuals resided when enamel formation took place.

The utilization of enamel carbonate in measuring 18O values is beneficial for two reasons. First, sample preparation of carbonate is considerably less time consuming and 122 less technically difficult than the preparation of phosphate for stable oxygen isotope analysis (Dupras and Schwarcz 2001; Bentley and Knipper 2005). Secondly, while phosphate will provide only 18O values, carbonate allows for a simultaneous analysis of both oxygen and carbon isotopes, making this portion of hydroxyapatite more data- and cost effective (Sponheimer and Lee-Thorp 1999). Furthermore, a consistent disparity of

8.7‰ of between phosphate and carbonate 18O values means that these hydroxyapatite components can be easily evaluated against one another (Bryant et al. 1996; Iacumin et al. 1996). Such an evaluation is made possible through the use of conversion equations, which permit further, albeit tentative comparisons between observed 18O enamel

18 18 carbonate ( Oc) values and modern meteoric water ( Omw) values (e.g., Coplen et al.

1983; Bryant et al. 1996; Iacumin et al. 1996; Luz et al. 1984; Daux et al. 2008; but see also Pollard et al. 2011). The utilization of conversion equations will be discussed further in Chapter 7.

While some studies take advantage of modern hydrological data and present-day oxygen isotope values (e.g., Turner et al. 2009), which may provide a very basic isotopic guide for a particular region, such values cannot be definitively mapped onto the past, as isotopic signatures can be altered significantly from prehistoric to modern times (Bentley and Knipper 2005). These changes are the result of substantial variation in mean 18O values in precipitation over time (Bentley and Knipper 2005). Additionally, because interregional oxygen isotope variability occurs as a result of environmental variables, strontium and oxygen stable isotopes act as independent lines of evidence for determining geographic residence that can be compared against one another, as strontium relies on local geology while oxygen correlates with local hydrological systems. 123 Despite the potential for oxygen isotopes to explicate mobility in the past, interpreting these data can be problematic due to the sheer number of variables that influence 18O values within a given geographic space. Reservoir effects cause bodies of water (e.g., lakes, ponds), and even water stored in containers, to possess elevated 18O values relative to water vapor and precipitates, creating variability within a single environment (Price et al. 2010). Additional inconsistencies may be introduced with flowing water, which can bring in non-local 18O, as well as variations in seasonal and annual rainfall, cooking and beverage preparation, and diet, all of which have the ability to produce broad 18O values within skeletons at a single site (Rozanski et al. 1993;

Knudson 2009; White et al. 2007; Price et al. 2010). Further variation may be expressed both between teeth and within a tooth from a single individual, particularly since enamel of the permanent dentition typically forms over a number of years, allowing the potential for considerable 18O variability during the incorporation of these isotopes (Fricke and

O’Neil 1996; Wright and Schwarz 1998; Weidemann et al. 1999). Finally, although unconfirmed in humans, a preliminary analysis of bone apatite from mice with sickle-cell disease suggests that some pathological conditions may also affect oxygen isotope fractionation (Reitsema and Crews 2011). Subsequently, oxygen isotope ratios used to evaluate migration in the past must be assessed with caution.

Carbon

Stable carbon (C) isotope ratios enable a general evaluation of dietary intake based on the unique photosynthetic pathways employed by different types of plants.

Carbon is present in all living organisms and occurs naturally as three different stable 124 isotopic variants, including 14C, 13C, and 12C (Table 3.5). δ13C values correspond to the ratio of 13C to 12C; however, because of the overwhelming natural abundance of 12C as opposed to the much smaller biologically available quantities of 13C, the vast majority of organisms display negative δ13C values (Schoeninger and Moore 1992). 14C, a radioactive form of carbon utilized in radiocarbon dating, is not discussed here.

Table 3.5. Natural abundances of carbon isotopes in the environment.

Natural Carbon Isotope # of Protons # of Neutrons Abundance 14C 0.0000000001% 6 8 13C 1.11% 6 7 12C 98.89% 6 6

Plants discriminate against the heavier 13C isotopes in different ways, producing differential fractionation rates that permit various plant types to be grouped into three categories, or photosynthetic pathways: C3 (Calvin-Benson), C4 (Hatch-Slack), and CAM

(crassulacean acid metabolism). C3 plants capture carbon from atmospheric CO2 by means of ribulose bisphosphate carboxylase, an enzyme that helps to transform it into the three-carbon compound phosphoglycerate (PGA), during the Calvin-Benson cycle within the mesophyll cells of the plant (Wong et al. 1979; Boutton et al. 1984; DeNiro 1987;

13 13 Hoefs 2004). Because this enzyme strongly discriminates against C ( CO2) in favor of

12 13 lighter C isotopes, the values of C3 plants are extremely depleted in C, with

13 δ Cap(VPDB) values averaging at around -27‰ and ranging from -35‰ to -20‰ (Smith and Epstein 1971; DeNiro 1987). Common C3 plants include most fruits and vegetables,

125 dates, nuts, legumes, tubers, wheat, barley, and rice, as well as most trees and shrubs

(DeNiro 1987; Lee-Thorp et al. 1989).

12 Alternatively, C4 plants preferentially incorporate C into the mesophyll cell layer through a different enzyme (phosphoenolpyruvate (PEP) carboxylase), fixing atmospheric CO2 into four-carbon aspartic or malic acids as part of the Hatch-Slack pathway (Boutton et al. 1984; DeNiro 1987). As a result, discrimination against 13C is

13 not as extreme, producing higher (but still negative) δ Cap(VPDB) values between -14‰ and -9‰, with an approximate average of -13.5‰ (Whelan et al. 1973; O’Leary 1981;

DeNiro 1987; Tieszen and Chapman 1992). These acids are then sent to bundle sheath cells, where carbon is fixed a second time via the Calvin cycle. C4 plants consist of tropical grasses, including maize, sorghum, millet, and sugarcane (Schwarcz et al. 1985;

DeNiro 1987; Lee-Thorp et al. 1989).

Finally, crassulacean acid metabolism (CAM) plants fix carbon by means of enzymes use by C3 (ribulose bisphosphate carboxylase) or C4 (phosphoenolpyruvate carboxylase) plants. The utilization of either enzyme is determined primarily by the aridity of a particular environment. In humid climates, CAM plants employ ribulose bisphosphate carboxylase, while CAM plants in more arid environments make use of the enzyme phosphoenolpyruvate carboxylase (DeNiro 1987). This photosynthetic flexibility produces δ13C values that range from -27‰ to -12‰, values that overlap considerably with both C3 and C4 plants (Kluge and Ting 1978; Boutton et al. 1984).

Specifically adapted for water storage, CAM plants, including succulents like cacti as well as agave, pineapple, prickly pear, and yucca, rarely contribute substantially to dietary intake in humans, particularly in the Middle East (Ambrose and Norr 1993).

126 All terrestrial plants derive environmental carbon primarily from atmospheric

13 CO2 (Katzenberg 2000). While modern-day atmospheric carbon displays a δ C value of

-7‰, these values differ from those found in prehistory because of the relatively recent introduction of fossil fuels into the environment with the start of the Industrial Revolution

(DeNiro 1987; Schoeninger and Moore 1992). As burning fossil fuels releases carbon with values of less than -26‰ into the air, atmospheric values in the past (pre-19th century AD) were approximately 1-1.5‰ more elevated than those in the present day

(Peng et al. 1983; Marino and McElroy 1991). Consequently, plant δ13C values are a product not only of their respective photosynthetic pathway but of the atmosphere as well.

An additional fractionation of approximately +5‰ takes place between diet and human collagen as proteins are incorporated into bone tissues, with little if any (+1‰) subsequent fractionations taking place during ingestion by secondary consumers (DeNiro and Epstein 1978; van der Merwe and Vogel 1978; Ambrose and Norr 1993).

13 Subsequently, humans feeding on a diet dominated by C3 foods possess collagen δ Cco values around -19‰, while individuals consuming a predominantly C4-based diet exhibit values around -5‰ (Dupras et al. 2001). Human diets consisting of both plant types display intermediate δ13C values between these two extremes.

Fractionation ∆a-d between bone/enamel apatite (a) and diet (d) also results in an enrichment in 13C. In humans, the precise value of this enrichment remains unknown, although estimates have been proposed based on experimental studies on both small and large mammals. A value of +9.5‰ for ∆a-d was found in laboratory rats with controlled

127 13 diets (Ambrose and Norr 1993), similar to δ Cap enrihcment resulting in values elevated by +9.5-9.8‰ in mice (DeNiro and Epstein 1978). In large mammals, Cerling et al.

(1997) observed apatite values elevated by +14.3‰ relative to diet, regardless of preference for C3 or C4 plants (Figure 3.4), that has also been shown elsewhere (e.g., Lee-

Thorp and van der Merwe 1987; Wang et al. 1994). Generally, then, animals consuming

13 a 100% C3-based diet possess apatite δ Cap values of averaging around -13‰, while

Figure 3.4. δ13C values for modern grasses and the resulting enrichment in 13C of apatite following isotope fractionation (modified from Cerling et al. 1997:154, Figure 1).

128 those partaking of a diet comprised solely of C4-based foods display mean values around

+1‰ (Sullivan and Krueger 1981; Lee-Thorp and van der Merwe 1987; Cerling et al.

1997). As mentioned previously, diets consisting of both types will exhibit δ13C values intermediary to these endpoints. However, a loss of methane from the mid-gut of many of these ruminants results in a slight depletion of 12C, meaning that a fractionation value of approximately +14‰ for large mammals may be elevated relative to values expected for human fractionation (Hedges et al. 2000; Prowse et al. 2004). In humans, then, an estimate of the difference between plant δ13C values and apatite (enamel and bone) is

13 approximately +11-12‰, so that humans with a diet consisting of C4 plants have a δ C

13 value around -1‰, while those human diets dominated by C3 plants will display δ C values approximating -12‰ (Krueger and Sullivan 1984; Passey et al. 2005; Dupras and

Tocheri 2007).

Marine plants and animals derive carbon principally from the relatively 13C- enriched dissolved carbonate in seawater (~0‰), although other, minor contributions also exist, including terrestrial debris carried in by rivers as well as dissolved atmospheric

CO2 (Schoeninger and Moore 1992). The observed 7‰ difference between terrestrial and marine organisms due to these disparate sources of carbon is also present in humans consuming foods from these environments, enabling a distinction to be made between various foodstuffs (Katzenberg 2000). However, because the elevated δ13C values found in marine plants mimic those found in a mixed C3/C4 diet, it can be difficult to distinguish what ecosystems humans may have exploited in the past (Price et al. 1985).

Nevertheless, the relative absence of a C4 contribution to human diet makes a distinction between terrestrial C3 and marine diets possible.

129 Agriculture in southeastern Arabia was dependent on the date palm, not only for consumption but as necessary shade for other crops that otherwise would be unable to grow in harsh, arid conditions; subsequently, this C3 plant became a dietary mainstay and may facilitate distinctions to be made between the Oman Peninsula and other regions less reliant on this crop (Nelson et al. 1999; Beech and Shepherd 2001). Subsequently, stable carbon isotopes in enamel offer yet another alternate means of measuring differences in childhood geographic location by examining variations in dietary intake between individuals, specifically the differential contributions of certain edible plants. These societies were also presumably reliant on maritime resources, especially coastal and island settlements like Tell Abraq, Umm an-Nar Island, Kalba, and Bidya, which may enable a intra- and inter-site comparison of terrestrial C3 versus marine diet.

Carbon isotopes are incorporated into both the organic and inorganic portions of the skeleton, including bone collagen and the hydroxyapatite of bone and dental enamel via carbonate (CO3) (DeNiro and Schoeninger 1983; Katzenberg et al. 1993; Sealy et al.

1995; Koch et al. 1997; Wright and Schwarcz 1999; Fuller et al. 2003; Dupras and

Tocheri 2007). Collagen, a structural protein, is the dominant component of the organic phase of bone. Hydroxyapatite, the major inorganic constituent of bone and enamel, ordinarily includes phosphate (Ca10(PO4)6(OH)2), but carbonate substitutions within hydroxyapatite at phosphate sites occur with some regularity and facilitate the measurement of δ13C values (Barralet et al. 1998).

Diagenetic alterations of carbon may affect both the organic and inorganic phases of bone. In the organic fraction, a lack of exchange between collagenous and environmental carbon ions makes collagen an effective medium in analyzing stable

130 isotopes in bone (Ambrose 1987; Grupe et al. 1989). Nevertheless, contaminants such as nitrogen compounds and humic acid (a product of decaying plant matter) may all influence the isotopic signature of bone. Additionally, further degradation may occur with the introduction of bacteria and fungi into organic material (Ambrose 1990;

Schwarcz and Schoeninger 1991; Schoeninger and Moore 1992). In order to remove these diagenetic materials and other, non-collagenous proteins, lipids, and hydroxyapatite, a variety of cleaning, acid washing, and other treatment techniques may be applied (Ambrose 1990). Moreover, the efficacy of such procedures in eliminating diagenetic matter may be evaluated through analytical means, including extraction yields,

%C and %N retained in each sample, and atomic C:N ratios (DeNiro 1985; Ambrose

1990).

Bone hydroxyapatite is even more susceptible to diagenesis than collagen.

Carbonate ions on the surface of the apatite crystal lattice are readily exchanged with ions in the post-depositional environment (e.g., soils, groundwater) (Krueger 1991; Ambrose and Norr 1993). As trabecular bone possesses considerably more surface area than cortical bone, it is exceptionally vulnerable to diagenesis, making cortical sampling preferable (Ezzo 1994). Diagenetic carbonate is a much more soluble material than biogenic apatite, and as a result, some have argued that acid washes can be an effective methodological tool in ridding bone of diagenetic material (LeGeros and LeGeros 1984;

Sealy et al. 1991; Price et al. 1992). Nevertheless, it remains unclear whether or not the physiologically-bound, biogenic apatite is left intact after such treatment and relies on the supposition that diagenetic carbonate does not irreparably undergo ionic exchange with biogenic carbonate (Budd et al. 2000). In addition to exogenous ionic replacement,

131 widespread recrystallization into carbonate hydroxyapatite takes place postmortem in bone as a consequence of the small size of its crystals, thereby bringing in external, diagenetic carbonate from the environment (Koch et al. 1997).

As discussed previously with stable strontium isotopes, enamel hydroxyapatite offers a favorable alternative to bone because of its resistance to diagenetic alteration, primarily because the large size of its crystals offers much less surface area for ionic absorption (Lee-Thorp and van der Merwe 1987, 1991; Koch et al. 1990, 1997; Hoppe et al. 2003). This, coupled with its low organic content, makes enamel an ideal material for stable isotope analysis, particularly as it survives so long after burial.

Differences pertinent to the measurement of stable carbon isotope ratios exist with regards to the macronutrients comprising skeletal tissues. Diet can be broken down into three major macronutrients – carbohydrates, lipids, and proteins – each of which possesses specific isotopic signatures. Collagen is constructed primarily of amino acids and thus reflects the protein component of diet (Krueger and Sullivan 1984; Wright and

Schwarcz 1999). On the other hand, hydroxyapatite (specifically, apatite carbonate) is indicative of total diet, or dietary energy, representative of carbohydrates, lipids, and proteins consumed (Tieszen and Fagre 1993). This is because the formation of apatite carbonate occurs in tandem with blood bicarbonate, a direct product of metabolic processes associated with the synthesis of carbohydrates, lipids, and proteins that is incorporated into connective tissue (Krueger and Sullivan 1984).

Turnover rates between collagen and enamel hydroxyapatite differ considerably as well. Bone collagen turnover is a relatively slow process, with complete replacement taking place over a period of at least 10 years (Chisholm 1989). Subsequently, samples

132 of human collagen taken for isotopic analyses represent the decade before age-at-death of dietary consumption and will signify a mixture of old and new collagen, seasonal variability, dietary complexity, and metabolic changes, all of which will average out over time (Schoeninger and Moore 1992). Accordingly, collagen cannot detect short-term dietary events (Chisholm 1989). The type of bone sampled makes a difference as well.

While denser cortical bone requires at least 10 years to undergo complete replacement, trabecular remodeling may occur at a much faster rate at approximately five years (e.g., rib and iliac crest trabeculae) (Jowsey 1961; Mulhern and Van Gerven 1997; Hill 1998;

Bentley 2006).

As with collagen, bone hydroxyapatite undergoes continuous inorganic phase remodeling, with complete replacement occurring every 7-10 years to 10-20 years of life, depending on bone type (Jowsey 1971; Lowenstam and Weiner 1989; Schutkowski et al.

2001). In contrast, enamel hydroxyapatite formation in permanent teeth occurs in childhood and does not undergo subsequent remodeling during adult life, despite later changes in diet or locale (Kohn et al. 1998; Hillson 2005). This permits an assessment of childhood geographic residence and diet in humans that may be compared with the isotopic signatures of local fauna in order to identify discrepancies indicative of the presence of non-locals.

Enamel Mineralization and Crown Formation

Enamel mineralization, both in its timing and duration, is a complex transformation that continues to be poorly understood (Montgomery and Evans 2006).

Nevertheless, discrete stages of the mineralization process have been identified by

133 Fincham and colleagues (1997, 1999). Secretion of amelogenins by ameloblasts leads to the formation and extension of fine enamel crystallites, deposited in a sequential manner, and represents only around 10% of the final weight of enamel (Hillson 1996; Fincham et al. 1999; Bentley 2006). During ensuing maturation, ameloblasts break down the large organic phase of enamel while crystallites simultaneously expand due to an influx of mineral ions, eventually resulting in an inorganic component of 96% seen in adults

(Hillson 1996; Fincham et al. 1997). Subsequently, while the initial secretion and formation of crystallites likely reflect daily incremental growth, mineralization during the maturation phase not only takes place in multiple directions entailing several consecutive mineralization fronts, but also occurs throughout a time span of unknown duration – possibly, upwards of five years – that varies between tooth types (Boyde 1989; Sasaki et al. 1997). This is particularly important for interpreting biogeochemical data from teeth, as isotopes continue to be incorporated into enamel during this final maturation stage, negating any simple, sequential association between the isotopes of incremental dental microstructures and the timing of enamel mineralization (Bentley 2006; Montgomery and

Evans 2006). Thus, for human teeth, bulk (and not micro-) sampling of enamel is suggested (Montgomery and Evans 2006).

Because enamel microsampling remains an unreliable means of assessing temporal changes in mobility or diet in humans, the analysis of multiple teeth from a single individual permits an evaluation of changing isotopic patterns throughout childhood. While the majority of teeth utilized in this study were commingled and recovered from isolated contexts, a few in situ molars remained within jaw fragments and were associated with other teeth, allowing for sequential stable isotope analysis.

134 Consequently, the timing of crown formation (including both intiation and completion) in permanent molar teeth must also be addressed. Permanent first molars commence formation in utero and possess completed crowns by approximately 2.5-3.0 years old

(Hillson 1996). Permanent second molars begin to form around ages 2.5-3.0 and are completed approximately between 7-8 years of age (Hillson 1996). Finally, the initiation of crown formation in permanent third molars takes place at approximately 7-10 years of age, with crown completion occurring around ages 12-16 (Hillson 1996).

***

In summary, strontium and oxygen isotope analysis of dental enamel represents an effective means of assessing mobility and reconstructing geographic residence patterns in archaeological populations. Because strontium isotope ratios differ among minerals and rock types, different regions possess distinctive isotopic ratios, so that strontium uptake in the skeleton reflects the geological locality where these isotopes were ingested.

Stable oxygen isotope values in enamel depend on the unique hydrological cycle of a particular region and can similarly be used to identify childhood geographic residence. In addition, stable carbon isotopes assist in reconstructing paleodiet by distinguishing between plants with different photosynthetic pathways. These three stable isotopes will be utilized as part of a larger research plan to evaluate the sociopolitical contexts of interregional exchange while simultaneously examining tomb membership at a local level. General isotopic patterns across tomb sites will be tested against various models of interregional interaction in the hopes of gaining a better appreciation of the sociopolitical power dynamics between various regions, while local contexts will also be examined

135 using isotopic data at the level of the individual to evaluate tomb membership (locals vs. non-locals) in terms of identity and kinship-based social frameworks.

136

CHAPTER 4

POLITICAL ECONOMY AND AGENCY

The nature of interregional exchange networks in the Bronze Age during the third and second millennium BC, its impact on local communities in Arabia and the larger

Near East, and the theoretical approaches utilized to better evaluate the emergence and subsequent decline of social, political, and economic complexity in the Oman Peninsula continue to be debated today. Theoretical approaches to political economy, including world-systems theory, the trade diasporas model, the distance parity model, and internal social agency are discussed here. These models are later assessed in light of an investigation of the skeletons of the people of southeastern Arabia themselves using stable isotope analysis.

World-Systems Theory

World-systems theory represents a post-Marxist characterization of large-scale interregional interactions, focusing on multiple cultures and polities as a united economic entity (Wallerstein 1974). Initially developed by Immanuel Wallerstein, world-systems theory can be viewed as a reaction to modernization theory, which viewed current economic systems within the narrow field of a single political unit while ignoring the

137 dynamic social history of worldwide exchange and the development of globalization

(Stein 1999a). Alternatively, Wallerstein maintained that the modern-day, capitalism- based world-system was a direct product of an economic history stretching back to the late 15th century AD, as Europe began extending its reach beyond its borders (Wallerstein

1974). This European expansion triggered the formation of interdependent networks of long-distance exchange, generating a world-system.

One of Wallerstein’s central ideas dealt with the notion of division of labor within a world-system. While other models categorized this division as merely ‘functional,’ defined as a result of occupation, Wallerstein saw the unequal distribution of economic tasks as inextricably connected with geographic location (Wallerstein 1974). Obviously, this is in part due to a reliance on local ecology and environmental productivity with regards to raw materials, natural resources, and agricultural success; however,

Wallerstein also attributed to geography a critical role in the historical exploitation inherent in the social organization of a world-system in the form of a hierarchical structure consisting of core and peripheral economic regions (Wallerstein 1974). The core represents developed, capital-intensive areas characterized by high skill levels, a powerful military, and elevated social complexity (Stein 1999a). On the other hand, the periphery corresponds to regions with a weak political organization, often providing labor-intensive, low-skill work (Wallerstein 1974).

Inequalities develop with increasingly asymmetrical interactions between these areas. The core dominates the periphery, primarily by controlling the majority of the surplus produced by the world economy as a whole, but also politically and ideologically

(Chase-Dunn and Hall 1993). Moreover, the periphery is limited by the production

138 structure set in place by the core, reinforcing their inferior position within the world economy and constraining development (Chase-Dunn and Grimes 1995). Peripheral production for use by the core usually takes the form of human labor and the extraction of raw materials (but not the actual manufacturing of valued goods from such materials)

(Wallerstein 1974). Despite this inherent structural subordination, diffusion from the core to the periphery does take place and includes the spread of advanced technology, instigating the rise of peripheral elites; however, Wallerstein (1974) contends that this merely creates a superficial image of developing peripheral complexity, masking the increasing economic disparity between the periphery and the core. Finally, it is important to keep in mind that there is no single core from which the world-system arises; instead, multiple political and economic centers exist, with each controlling particular aspects of economic flow from the periphery to the core (Wallerstein 2000). Nonetheless, change can occur by means of cycles of both economic expansion and regression.

Wallerstein (2000) was hesitant to apply world-systems theory to societal organization before AD 1500, speculating that world economies would necessarily have been uncommon and poorly organized, if they existed at all. He classified societies throughout history as falling into one of three types: mini-systems, world empires, and world economies (Wallerstein 1974). Mini-systems include small but independent polities that largely employ a reciprocal exchange system, such as bands, tribes, and chiefdoms (Stein 1999a; Wallerstein 2000). World empires refer to a state level of organization and may encompass multiple cultural groups. For the first time, a core region emerges and begins to take advantage of the periphery in the form of taxes or tribute (Wallerstein 1974). Wallerstein’s (2000) final unit of analysis, a world economy,

139 is comprised of many different polities, each with a distinct economic and political system. World economies imply a high degree of trade between polities that ultimately creates a power structure exhibited by the core/periphery, ushering in a world-system

(Stein 1999a).

Some have argued that a world-systems analysis can in fact be applied to pre- capitalist exchanges, although adherence to Wallerstein’s criteria varies considerably between authors. Jane Schneider (1977) challenged Wallerstein’s assumption that world economies developed only with trade in staple crops or raw materials, arguing instead that luxury goods also have the ability to prompt drastic ideological, technological, and structural transformations. Because of this, archaeologists can utilize world-systems theory as a means of interpreting interregional interaction in prehistoric and early historic times. Andre Frank (1993) also adopted this point of view, making a case for a single world-system that came into existence during the Bronze Age that has persisted until the present day. To support his point, Frank drew extensively on archaeological evidence from Mesopotamia, the Levant, Anatolia, the Arabian Peninsula, Egypt, and the Indus

Valley, illustrating relatively simultaneous cyclical economic expansions and downswings experienced by both the core and the periphery (see Edens and Kohl 1993 as well). Guillermo Algaze (1993) concurs, adding that any initial peripheral growth due to trade with core societies will eventually serve to weaken the periphery as overspecialization contributes to increasing underdevelopment without creating additional employment. Consequently, archaeologists tend to adopt world-systems theory in a more heuristic sense as a means of explaining both interregional

140 communication networks and the economic consequences of such interaction on both the core and periphery, and may not follow a strict Wallersteinian approach to such analyses.

Reaction to World-Systems Theory

Not every historian or archaeologist utilizes some form of world-systems theory to tease apart the economic complexities of the past, and others altogether reject our ability to apply Western conceptualizations of economic organization onto both pre- industrial and past societies (Ratnagar 2001; Lamberg-Karlovsky 2009). For instance,

Lamberg-Karlovsky (2009) vehemently disassociates world-systems theory with the economic complexities of the ancient Near East and rejects this and other universal explanatory models as naïvely disregarding the multifaceted and ever-changing face of these interregional relationships. Additional theoretical frameworks of interregional interaction, generated by analysts dissatisfied with the application of world-systems to prehistory, offer new suggestions for more accurately portraying exchange in the past. In particular, archaeologist Gil J. Stein (1999a) notes two alternative models to world- systems theory: trade-diasporas and distance-parity.

The trade diasporas model seeks to explain interregional trade networks in which representatives from one group are sent to live in the communities of their trading partners in order to more closely regulate and control the majority of stages involved in the exchange process (Stein 1999a). Originally developed by Cohen (1971) to address modern-day trading relationships between various tribes in west Africa, trade diasporas can be defined as:

141 interregional exchange networks composed of spatially dispersed specialized merchant groups that are culturally distinct, organizationally cohesive, and socially independent from their host communities while maintaining a high level of economic and social ties with related communities who define themselves in terms of the same general cultural identity (p.266-7).

By living with ‘host’ communities, these specialized foreign intermediaries effectively gain a monopoly over particular aspects of trade (Stein 1999a). Trade diasporas normally emerge in response to obstacles in the exchange process, including when environmental conditions make transportation or general communication problematic, or in situations where economic security cannot be guaranteed to those involved in trade over long distances (Stein 1999a). Trade diasporas thus represent a strategy aimed at more tightly controlling this system.

The foreign enclaves taking up residence in host communities are part of a larger diaspora community who act within a cohesive communication network in order to keep trade systems running smoothly (Stein 2002a). However, this organizational framework is only part of what connects these enclaves. Trade diasporas deliberately express themselves as a distinct social group, clearly and visibly defining their cultural identity within the larger host community (Stein 2002a). In doing so, all enclaves within a particular communication network remain connected to one another while simultaneously defining themselves as separate from the community in which they reside.

Although any number of possible relationships may exist between trade diasporas and host communities in terms of power differentials, three relations stand out. First, a trade diaspora may be marginalized by the host community and is only permitted to remain because of the enclave’s usefulness in regulating exchange relations (Stein

142 2002b). Secondly, a trade diaspora may be highly independent of its host community

(Stein 2002a). A local ruler may grant autonomous status to the foreign enclave, a privileged position again resulting from the usefulness of the trade diaspora to that leader

(Stein 1999). Finally, in rare situations, the trade diaspora may actually control its host

(Stein 2002b).

Stein (1999a, 2002a, 2002b) has further developed this model as an alternative to world-systems theory and as a more appropriate paradigm to apply to the archaeological record. Like Cleuziou (2007), Stein (2002b) believes that external forces, including climate and interregional exchange, have been overemphasized as instigators of change, and that a new, socially-based perspective which focuses on internal socio-political and economic forces as prime movers is needed. Subsequently, Stein (1999a) looks to the relationships between local interactions with larger exchange systems via the trade diaspora.

The identification of trade diasporas in the archaeological, and more specifically, mortuary record can be a challenging endeavor, principally because it requires the recognition of the ethnicity of a foreign enclave within the host community (Stein 2002a).

However, as Stein (1999a) stresses, archaeologists cannot simply assume that complex interregional exchange and the presence of foreign goods necessarily entails a trade diaspora; these goods may simply reflect the existence of internally controlled exchange networks without the presence of foreign merchant groups. As such, he outlines basic criteria for archaeologically detecting trade diasporas within settlements (Stein 1999a).

First, foreign ethnic identity would have been stressed both in foreign residential architecture as well as publicly through ceremony involving a ritual structure that

143 architecturally reflected foreign style and/or symbolism (Stein 1999a). Such private and public structures should be visible archaeologically. Secondly, while the archaeological record often does not preserve many markers of ethnic identity, including distinct clothing and language, mortuary practices can be utilized to identify the ethnic differences that separate foreigners from the rest of the community (Stein 1999a). These criteria convey a critical assumption of the trade diasporas model: that differences between the enclave and host are greatly emphasized as a means of defining a distinct social identity, an identity that must be reproduced through the visible perpetuation of these differences.

Stein’s (1999a) second framework, the distance-parity model, claims that the dominance of the core over the periphery and the resultant inequality and underdevelopment in this relationship will decrease with distance; in other words, the further away the periphery is from the core, the less power the core has over it.

Practically, the cost of transportation to and from the core increases as peripheries become increasingly remote. Even with differences in technological advancement between core and peripheral areas, a diminished hegemonic structure will occur, resulting in lessened economic pressure on the periphery as a whole (Stein 1999a). However, the core would continue to influence peripheral elites to a certain extent. In addition, no major changes in local production or social complexity should occur in the periphery.

Once again, unlike world-systems theory, power dynamics are variable between the core and the periphery, and may even be manipulated to work for the periphery’s advantage

(Stein 1999a).

144 Shereen Ratnagar (2001), who takes a more literal approach to applying

Wallerstein’s work to past societies, rejects the notion of a world-system existing in the

Bronze Age altogether. Like Stein (1999a), Ratnagar voices concern over the both the visibility of inequality in the archaeological record as well as the existence of peripheral underdevelopment in general. Even in instances of less technologically advanced peripheries in comparison to the core, Ratnagar (2001) contends that such inequality may not be the result of interregional exchanges systems. Accordingly, a lack of technological diffusion may simply be a product of time or distance and not hegemonic restriction on the periphery.

Agency and Social Organization in the Bronze Age

A pioneer in the prehistory and archaeology of Arabia, Serge Cleuziou dedicated his final years to examining social complexity and the evolution of societal dynamics in the Oman Peninsula in light of the economic, political, and social upheaval that took place with the transition from the Umm an-Nar to Wadi Suq period at the end of the third millennium BC. Cleuziou was particularly interested in the apparently stunted development of the Oman Peninsula in relation to the state-level hierarchies and social complexity of its trading partners, including Mesopotamia and the Indus Valley.

This is not to say that political and socioeconomic changes did not take place in southeastern Arabia. Neolithic subsistence strategies of nomadic herding and fishing- gathering gave way to an increasingly sedentary way of life, as indicated not only by the development of palm garden and oases agriculture to complement marine and domestic ruminant resources but also the appearance of monumental tombs during the Hafit period.

145 Cleuziou (2007) estimated that more than 100,000 of these burial structures exist throughout the Oman Peninsula after a detailed survey in the eastern Ja’alan of Oman revealed over 2500 tombs alone, speaking to a dramatic rise in population over the course of only a few hundred years (for a discussion of demographic estimates based on mound surveys, see also Charles 1992, 1995). Highly visible on the landscape, these funerary monuments marked important areas of resource exploitation, whether coastal (fishing) or inland (oases or copper mining), thereby delineating the surrounding territory as belonging to a particular lineage group and limiting access to its resources (see also

Cleuziou 2002b). The assumption that these collective tombs, typically utilized for only two or three generations and with no indication of selective burial, represent lineage or

‘extended family’ burials is not new (e.g., Potts 1990; Salvatori 2001; Cleuziou 2002).

By the Umm an-Nar period, this cultural system had gained considerable complexity, evident with the appearance of fortification towers and settlements, increased tomb monumentality and membership, the emergence of a local ceramic tradition, and a highly interconnected system of domestic and interregional trade networks. In particular,

Cleuziou (2007) highlighted the relatively rapid changes to these collective graves from the Hafit to the Umm an-Nar, such as an increase from a few individual interments to the inclusion of hundreds of people within these structures, as well as a change in the actual placement of tombs, now near or within settlements. Cleuziou (2007) viewed these modifications as consistent with the continued development of a kinship-based societal organization, but with further emphasis on “higher levels of lineage units” as the population size increased. Larger kin groups were compensated for by the construction of larger but still all-inclusive tombs serving as landmarks that would have strengthened

146 lineage ties within the community. As in the preceding period, no indication of status during life is evident in mortuary practice for any individual in death.

Major social transformations took place at the beginning of the second millennium BC with the apparent collapse of the Umm an-Nar cultural system, coinciding with dramatic changes in mortuary tradition, material culture, trade relations, population size, subsistence, and land use. While the typical archaeological accusations against climate change and foreign influence have been mounted to account for this collapse, Cleuziou (2007) proposed that we should view these changes using an alternative perspective, one that emphasizes internal social dynamics and individual agency as prime movers of this cultural system. Specifically, Cleuziou viewed the breakdown of societal organization in southeastern Arabia as the result of a fundamental conflict between traditional kin/tribal systems and developing social hierarchies.

A growing elite subset of the population would be expected to arise as increasingly complex exchange networks emerged, requiring a reorganization of social control over economic transactions. Maintaining and improving commodity exchanges on both a regional and interregional level would have taken on an increasingly important role with continued population growth in the region, necessitating that settlements obtain food and other resources to sustain such growth as local reserves were gradually depleted.

The growing presence and influence of these hierarchies were reflected in mortuary practices and settlement fortification, with larger, more monumental construction projects requiring an increasingly organized labor force. Settlement hierarchies are also evident in the archaeological record, with some settlements designated by simple compounds with small, inland oases gardens or coastal middens representative of extended family

147 units, while others possessed substantial settlements and associated fortification towers and were likely comprised of multiple family lineages joined by political alliance.

Additionally, cuneiform textual records describing the invasion of Magan and the defeat of its “lords” by the Akkadians around 2300 BC point to the existence of a non- egalitarian way of life in the Oman Peninsula, although the possibility exists that these so-called lords may have been temporarily appointed in times of crisis (Cleuziou 2007).

According to Cleuziou, traditional tribal organization and kinship-based systems rooted in cooperation, redistribution, and egalitarianism would have been increasingly challenged as these elites gained power. This conflict, creating considerable stress and opposing ideological forces, rapidly propelled the Oman Peninsula away from its path towards a state-level hierarchical system (of the kind seen in Mesopotamia) and into collapse. Such a collapse, marking the commencement of the Wadi Suq period in the early second millennium BC, exposes an Arabian economy that could not be sustained as a result of a lack of resolution between these two societal factions (Cleuziou 2007). With interruptions and even cessations of trade networks once controlled by this burgeoning elite, the growing population of southeastern Arabia could not be sustained, leading to a dramatic decrease in population size across the peninsula as evidenced by significant shifts in mortuary practices coupled with the sudden disappearance of large, fortified settlements (Cleuziou 2007).

***

In summary, various theoretical models have been proposed as a means of explicating political economy amongst archaeological populations. World-systems theory, which emphasizes large-scale interregional and asymmetrical interactions

148 between hegemonic cores and subordinate peripheries, has controversially been applied to pre-capitalist societies, including those of the Persian Gulf. Various reactions to this theory include the trade diasporas model, which expects core representatives to live in local communities and visibly express their foreign identity while regulating trade in more peripheral locations, and the distance parity model, which anticipates that the dominance of the core over the periphery and the resultant inequality inherent in this relationship will decrease with distance. More recent, alternate viewpoints attribute greater internal agency to the substantial socioeconomic transitions in the Bronze Age of

Arabia. Cleuziou viewed the so-called breakdown of societal organization in the Oman

Peninsula as the result of a fundamental conflict between traditional kin/tribal systems and developing social hierarchies that came to a head by the end of the third millennium

BC. All of these approaches will be assessed in conjunction with biogeochemical evidence from the dental enamel of those individuals involved in constructing the economy of southeastern Arabia.

149

CHAPTER 5

MATERIALS AND METHODS

Introduction

The purpose of this chapter is to describe the archaeological sites from which enamel was sampled for isotopic analysis. I begin by describing the sites upon which this dissertation focuses – those located in the United Arab Emirates – and have organized these according to their locations on either the Persian Gulf or Gulf of Oman coast. Next,

I describe those sites from which comparative isotopic data has been taken. These are listed in alphabetical order according to country and site name, respectively.

Secondly, I outline the methodology associated with enamel sample extraction, sample preparation, and stable strontium, oxygen, and carbon isotope analysis. All human and faunal enamel samples were collected, prepared, and analyzed by the author.

Site Descriptions: United Arab Emirates

Individuals from 13 tombs across the United Arab Emirates were utilized for this project (Figure 5.1, Table 5.1). Six of these burial monuments – from Mowaihat, Tell

Abraq, Umm an-Nar Island Tombs I, II, and V, and Unar 1 – date to the Umm an-Nar

150

Figure 5.1. Map of the northern Oman Peninsula illustrating the location of Bronze Age tombs sites in the United Arab Emirates used in this study (images from Google Earth).

Table 5.1. Bronze Age tombs from the United Arab Emirates with human and faunal teeth sampled in this study. Numbers in parentheses refer to the number of individuals.

Site Country Emirate Human n Faunal n Bidya UAE Fujairah 3 (2) 0 Dadna UAE Fujairah 1 0 Dibba UAE Fujairah 4 (3) 1 Mereshid UAE Fujairah 1 0 Mowaihat UAE Ajman 19 (13) 0 Qidfa UAE Fujairah 2 (1) 1 Shimal 95 UAE Ras al-Khaimah 2 0 Shimal 103 UAE Ras al-Khaimah 7 0 Shimal Settlement UAE Ras al-Khaimah 0 9 Tell Abraq UAE Sharjah 30 (29) 12 Umm an-Nar Island UAE Abu Dhabi 38 (33) 15 Unar 1 UAE Ras al-Khaimah 28 (25) 1

151 period, while seven collective graves – from Bidya, Dadna, Dibba, Mereshid, Qidfa,

Shimal 95, and Shimal 103 – date to the Wadi Suq period.

United Arab Emirates: Persian Gulf Coast

Mowaihat, Emirate of Ajman

In the Emirate of Ajman along the western coast of the Oman Peninsula, the late third millennium site of Mowaihat was home to Umm an-Nar funerary structures and a possible settlement (Figure 5.2). While the site now sits approximately seven kilometers

Figure 5.2. Map of third millennium sites of the United Arab Emirates, including Mowaihat (from Phillips 2007:5). 152 from the coast, it would have been positioned only 500 m from shore during the Bronze

Age (Al Tikriti 1989a). Excavations first took place at the site in February 1986, led by

Dr. Walid Yasin Al Tikriti of the Department of Antiquities and Tourism in Al Ain, after construction workers uncovered the tombs while mechanically digging a trench (Figure

5.3) (Al Tikriti 1989a; Haerinck 1991). A second season in November 1986-January

1987 resulted in the completion of excavations of the first tomb (Tomb A), while Tomb B excavations purposefully remained incomplete until 1990, when a Belgian archaeological

Figure 5.3. Plan of Tombs A and B at Mowaihat (from Phillips 2007:3). Note the two foot-wide trench dug through both tombs, inflicting considerable damage to Tomb B. 153 campaign headed by Dr. Ernie Haerinck resumed excavations and lifted all remaining skeletal material (Haerinck 1991; Phillips 2007).

Two tombs, both of which date to the end of the third millennium BC, have been found at Mowaihat. Tomb A is representative of a typical circular Umm an-Nar tomb with a diameter of 8.25 m (Al Tikriti 1989a). As with other burial structures of this period, two layers of stone comprise the exterior wall, including an outer ring constructed of finely-hewn and fitted ashlar and an inner lining consisting of unworked rock (Al

Tikriti 1989a). This rough stone was also utilized in building the crosswalls within the tomb itself, forming multiple chambers (Al Tikriti 1989a). Based on surviving foundations, Al Tikriti (1989a) estimated that the tomb might have risen three meters above ground. A single entrance located on the western tomb wall was evident based on the presence of a blocking stone and a lintel, and while many Umm an-Nar tombs possess dual entrances opposite of one another, the area directly across from the known entrance of Tomb A had been destroyed during construction of the trench (Al Tikriti 1989a).

Grave goods from Tomb A are scarce, in part because of the damage inflicted by the trench but also due to significant looting in antiquity (Al Tikriti 1989a). A few locally made ceramic sherds, two rings made of silver and copper, and a handful of carnelian and metal beads represent the remaining assemblage from the tomb (Al Tikriti

1989a). Similarly, scant evidence of human remains exist, represented only by small fragments in two of the inner tomb chambers as well as a small scatter of splintered bone around the tomb’s exterior that suggest that the grave was collective (Al Tikriti 1989a).

The lone exception to this poor skeletal preservation was a skull protected from past plundering by an interior crosswall (Al Tikriti 1989a). A Wadi Suq burial of an infant

154 placed in an urn and buried immediately outside the tomb is unusual, both in the large size of the jar and the way in which this subadult was interred, remaining the only urn burial to date in the United Arab Emirates (Al Tikriti 1989a).

Located just three meters northwest of Tomb A, Tomb B is an extremely unusual

Umm an-Nar funerary structure both in terms of structure and contents. Measuring 3.9 m in length, 1.9-2.1 m in width, and 85-95 cm in depth, this rectangular grave was built underground and constructed of rough, unworked limestone and a flat, stone-lined floor

(Al Tikriti 1989a; Haerinck 1991). The absence of a traditional entrance on the sides of the tomb suggests that the dead were placed into it from the surface, which was covered by a roof of flat stones (Al Tikriti 1989a; Haerinck 1991). The extreme damage caused by the trench digger through the entirety of this tomb split this monument into northern and southern portions, generating a makeshift division used by the archaeologists at the site.

Unlike Tomb A, many well-preserved grave goods were recovered from Tomb B, including late Umm an-Nar ceramic and stone vessels (some of which were complete and not fragmented), two steatite seals, over 4000 beads, and various shell and copper ornaments and tools (Al Tikriti 1989a; Haerinck 1991). While no vessels of foreign origin were found, beads made of carnelian and frit may point to external relations with the Indus Valley and Mesopotamia, respectively (Al Tikriti 1989a).

A considerable quantity of commingled human skeletal material was recovered from Tomb B (Figure 5.4). Using crania, Al Tikriti (1989a) estimated that some 120 individuals had been interred in this subterranean tomb; later, Haerinck (1991) proposed an MNI of 160-180 individuals to account for the damage caused by the trench, which

155 destroyed nearly one-third of the tomb and its contents. A small number of articulated limbs, a thorax, and an incomplete female skeleton are far outnumbered by disarticulated bone, suggesting that this structure was predominantly used to house secondary burials

(Al Tikriti 1989a; Haerinck 1991). Based on these patterns of deposition, both Al Tikriti

(1989a) and Haerinck (1991) suggested that deceased individuals might have first been placed in the large round Tomb A until the structure was completely full. The need for additional space to accommodate the newly dead would have required either (a) the construction of a new circular, above-ground tomb, a time-consuming endeavor that would have necessitated a considerable labor force consisting of at least some skilled

Figure 5.4. Human skeletal material from Mowaihat, Tomb B (from Haerinck 1991:22).

156 workers to dress the outer limestone ringwall; or (b) a less intensive construction project involving digging a simple rectangular pit lined with rough, unworked marine stone

(Haerinck 1991). The inhabitants of Mowaihat chose the latter, and with no discernable layers of deposition within Tomb B, it appears that a mass transfer of skeletal material from Tomb A to B occurred all at one time (Haerinck 1991).

Tomb B may represent a transitional grave type between the Umm an-Nar and

Wadi Suq periods, particularly as grave goods within the tomb date it to the late third millennium (Al Tikriti 1989a). This might explain the sudden appearance of ovoid and rectilinear graves so characteristic of the Wadi Suq at sites such as Bidya and Shimal (Al

Tikriti 1989a, 1989b). More recently, however, it has been suggested that perhaps more of these subterranean, secondary repositories may be associated with Umm an-Nar tombs, but have simply not been found yet. In fact, the presence of an adjacent below- ground tomb is not an isolated occurrence but has also been reported at Hili (Haerinck

1991; Phillips 2007). Just 2.4 m northwest of the circular Umm an-Nar funerary structure

Tomb E at Hili, a second, rectangular-shaped, underground tomb called Tomb N measured 6.5 m long, 2 m wide, 2.5 m deep, and contained the remains of hundreds of individuals of all ages and of both sexes (Al Tikriti 1989a; Haddu 1989; Al Tikriti and

Mery 2000; Mery et al. 2004, 2008; McSweeney et al. 2008, 2010).

As at Mowaihat, Hili Tomb N seems to date to the late third millennium based on both Umm an-Nar wares as well as apparently transitional vessel forms between Umm an-Nar and Wadi Suq types (Al Tikriti 1989a; Al Tikriti and Mery 2000; Mery et al.

2008). However, unlike Mowaihat Tomb B, out of an estimated 700+ individuals interred within the rectangular pit-grave at Hili N, hundreds of articulated skeletal regions

157 and 31 primary inhumations from the floor level of a single compartment remain intact

(Mery et al. 2004; McSweeney et al. 2008, 2010). This, along with the presence of many tiny bones in the commingled assemblage, indicates that primary inhumations took place within Hili Tomb N, with commingling and fragmentation a probable result of pushing older remains aside to make room for new bodies (McSweeney et al. 2010).

Consequently, while not unique, Mowaihat Tomb B was nevertheless an uncommon burial structure that may point to shifting mortuary traditions in funerary architecture at the end of the Umm an-Nar period. Brief pathological assessments of skeletal material from Martin (1996) and Blau (2007) show few caries, dental abscesses, and calculus, although a high degree of dental wear does suggest that grit may have been introduced into the diet, possibly through the grinding of cereals. Nineteen human molars from

Mowaihat Tomb B were sampled for this study.

In addition to the two collective graves uncovered at Mowaihat, a small, contemporary settlement located approximately 300 m northeast of the tombs was likely home to the tomb-builders (Haerinck 1991). An initial magnetometry survey of the area by Carl Phillips failed to locate any stone foundations, although the instrusive modern trench revealed the presence of multiple hearths (Al Tikriti 1989a; Haerinck 1991).

Additionally, shells and potsherds of Umm an-Nar date were scattered across the surface of the site, along with a very few number of turtle shell fragments and a piece of vertebra from a camel (Haerinck 1991; Phillips 2007). This scarcity of material culture, food remains, and permanent domestic structures led Phillips (2007) to conclude that living spaces within the settlement were likely built of ephemeral plant materials similar to the

‘arish or barasti date palm huts seen in modern times in the region (Phillips 2007).

158 Shimal 95 and 103, Emirate of Ras al-Khaimah

As the first recorded second millennium BC site in southeastern Arabia, Shimal represents one of the most significant Middle Bronze Age centers in the northern

Emirates. The site is located near the modern village of Shimal, where the site gets its name (referring to the shamal, or northwesterly winter winds), and is eight kilometers east of the capital city of Ras al-Khaimah (Vogt et al. 1989). The vast cemetery and associated settlement, situated at the foothills of the Jabal Qasr al-Dhaba, extend approximately 2.5 km from north to south and are today bordered to the west by palm gardens (Vogt et al. 1989; Velde nd-a). Based on reconstructions of the ancient coastline,

Shimal sat between 2.0-3.5 km from the sea during the Bronze Age and was in close proximity to a variety of rich environmental zones, permitting easy access to freshwater wadis, fertile agricultural soils and pasturelands, and marine foods (Vogt and Franke-

Vogt 1987; Glover 1991).

Shimal possesses one of the longest excavation histories in the Oman Peninsula.

In 1968, a preliminary survey conducted by British archaeologists Beatrice de Cardi and

Brian Doe revealed a large cemetery with hundreds of stone funerary monuments (de

Cardi and Doe 1971; Vogt et al. 1989). During two seasons in 1976 and 1977, Peter

Donaldson led the excavation of two tombs (Sites 1 and 6) in the southern portion of the cemetery (Donaldson 1984, 1985). These were followed by three campaigns undertaken by Burkhardt Vogt and the German Archaeological Mission between 1985-1987, with an additional survey done by Vogt in 1988, which reported some 250 tombs throughout the cemetery area, although excavations focused primarily on five tombs (Shimal 99, 100,

101, 102, and 103) as well as sections of a settlement (Vogt and Franke-Vogt 1987;

159 Schutkowski 1988; Haser 1991). The most recent survey took place in 1998 under the direction of de Cardi and German archaeologist Imke Moellering, who sought to measure, describe, and identify each tomb by type (e.g., Shimal, Ghalilah, Khatt, Dhayah, circular) (Velde nd-a).

The cemetery at Shimal is comprised of a dense collection of both single and collective burials indicative of a once-substantial population residing in the region (de

Cardi 1989). This extensive site is partitioned into three gravel fans by three great wadis running from the mountains westward towards the coast (Figure 5.5) (Vogt et al. 1989).

These areas, known respectively as Shimal North, Shimal Middle, and Shimal South, house predominantly above-ground, stone-built, collective tombs with either single or double chambers, although subterranean and semi-subterranean stone cists containing single and multiple interments are also present in lesser numbers (Vogt 1998). These cairns generally follow the foothills and wadis of the landscape, and a loss in tomb density from north to south may be indicative of the cemetery’s chronology, developing in a southerly direction (Vogt 1998; Velde nd-a).

With the exception of two circular tombs dating to the Umm an-Nar period (Unar

1 and 2), the cemetery is clearly dominated by Wadi Suq funerary architecture (Vogt

1998). Single and double interments (n=106), whether over-ground, subterranean, or semi-subterranean, outnumber collective burials of all types and vary considerably in orientation (Vogt 1998). Collective graves are represented by a number of tomb types, including Shimal (n=78), circular (n=28), Ghalilah (n=19), Dhayah (n=2), and Khatt

(n=3) types (Vogt 1998). Nevertheless, these graves were not uniformly distributed amongst the three portions of the site. Shimal-type tombs dominate all collective burial

160 161

Figure 5.5. Map of excavated Shimal tombs, including Sh 95 and Sh 103 (highlighted) (adapted from Velde nd-a). monuments, are preferentially placed in the foothills and not along the wadis, and increase in size from northern to middle to southern portions of the cemetery (Vogt 1998;

Velde nd-a). Circular-type tombs, on the other hand, seem concentrated along the margins of the site, primarily in the northern section (Vogt 1998; Velde na-a). Tombs of

Ghalilah-type are more consistently distributed throughout the cemetery, with the largest of these predominantly converged along the central wadi around Shimal Middle and the southern portion of Shimal North (Vogt 1998; Velde nd-a). Dhayah- and Khatt-types occur infrequently in the cemetery, and may have signified a special group interred on site during the second millennium (Vogt 1998).

Velde (nd-a) suggests that subdivisions within these extensive mortuary grounds were likely, and that each sector of the cemetery may have been associated with its own settlement. Such settlements beyond that unearthed in Shimal Middle have not yet been found. Nonetheless, surface scatters of pottery may support this hypothesis; while northern and middle sections of Shimal contained sherds from the Umm an-Nar, Wadi

Suq, and Iron Age, only Wadi Suq ceramics were recovered from Shimal South (Vogt

1998). Moreover, a higher concentration of tombs in Shimal North, both single and collective, may suggest that interment within this section of the cemetery was more highly valued (Vogt 1998).

A large settlement sits in the central section of Shimal and dates from the latter part of the second millennium to the beginning of the Iron Age. The site appears to have been approximately 200 m in length and contains the remains of stone foundations as well as a variety of shell middens, which also date primarily to the second millennium and are scattered throughout the northern and central portions of the area (Vogt et al.

162 1989; Velde 1991; Velde nd-a). Excavations in the settlement began in 1985 with the opening of trenches SX and SY, and it is estimated that this area represents the core of a larger domestic precinct (Vogt et al. 1989). Unlike the massive fortification towers thought to be the center of Umm an-Nar settlements, this Wadi Suq settlement seems to have been composed of dwellings made not only of stone, but also dried vegetation to create thatched roofs (Potts 1998; Blau 2001a). However, modern intrusions in the form of agriculture, construction projects, Islamic cemeteries, and the present-day village of

Shimal have largely destroyed any traces of the domestic sphere (Vogt et al. 1989, 1998).

Interestingly, in stark contrast to its associated tombs, the settlement contained very few beads, pendants, and even tools (Glover 1991). Further, ceramic assemblages on site differed considerably from those recovered in funerary contexts, with settlement pottery characterized by a lack of unpainted and fine wares (Mery 1991; Velde 1991). Faunal material was also unearthed within the settlement, and nine ovicaprine molars were included in the present analysis.

While the majority of tombs at Shimal seem to have been placed away from areas of habitation (e.g., shell middens), the settlement itself was built over the early second millennium tomb of Shimal 95. The tomb was not visible during the initial surface reconnaissance of the area and was only uncovered during the 1987 excavation season of the settlement in Shimal Middle, which prompted a return to the tomb and the commencement of formal excavations in 1988 (Velde nd-c). While settlement intrusions partially destroyed the southern portion of the tomb, its overall excellent preservation – including instances of in situ human remains and grave goods – can be attributed to a

Late Bronze Age midden deposited after the tomb fell into disuse (Velde nd-c). The

163 tomb is unique in that its northern extremity was cut into the rock face of the foothills, although the rest of the tomb followed more traditional Wadi Suq construction strategies

(Figure 5.6) (Velde nd-c). No concrete dates have been put forward for this tomb, but based on subsequent deposits during settlement occupation in the latter Wadi Suq and

Iron Age, Shimal 95 was occupied in the early second millennium BC Wadi Suq.

Figure 5.6. Tomb plan of Shimal 95. Note that the northern portion of the tomb has been cut into a natural stone ridge (from Velde nd-c:179). 164 The Shimal 95 grave may be categorized as a Shimal-type tomb and consisted of a single chamber approximately 13.1 m in length, 1.7 m wide, and oval in shape (Velde nd-c). Two rows of large, paralleled stones made up the southern perimeter wall of the tomb and were strengthened by a fill of smaller pebbles placed between them; however, the reuse of stones during settlement construction by the later Bronze and Iron Age inhabitants of the site have largely destroyed the external wall layer (Velde nd-c). At least one tomb entrance was detected in the western wall, as evidenced by a partially preserved doorframe and threshold stone, and remarkably, the grave’s original flat-stoned pavement remains preserved throughout the entire chamber (Velde nd-c). In addition to the overlying settlement midden, collapsed roof slabs protected some areas of the chamber from looting, making Shimal 95 incredibly important in more accurately deciphering Wadi Suq mortuary practices (Velde nd-c).

Human skeletal material was badly fragmented and, for the most part, commingled (Hummel 1988). No evidence of cremation exists. Bodies were initially placed in a flexed position, although later interments coupled with looting contributed to the commingled state of human remains (Hummel 1988). A minimum number of 15 individuals are estimated by Hummel (1988), with all ages present and both sexes equally represented. Antemortem tooth loss occurred with some regularity (reported at 15.8%), while vertebral osteophytosis was recorded on only one lumbar vertebra (Hummel 1988).

One articulated skeleton, that of a young adult female, was laid flexed on her right side in a north-south orientation parallel to that of the tomb’s axis (Hummel 1988; Velde nd-c).

A silver earring, bronze bowl, chlorite vessel, and small jar all accompanied her remains, and a partially articulated male was buried nearby on his left side (Hummel 1988; Velde

165 nd-c). Additionally, eight articulated limbs and other skeletal segments reveal a high degree of ornamentation and artifact association with in situ skeletal material, including a male skull surrounding by four spearheads and a razor blade, as well as two forearms encircled with carnelian beads (Figure 5.7) (Hummel 1988; Velde nd-c). A wealth of metal artifacts from the undisturbed areas of this grave gives insight into the original deposits of many of these collective burials before looting. At Shimal 95, these included rings, razor blades, spearheads, a dagger, and a bronze bowl (Velde nd-c). Finally, as part of this research program, two human molars underwent isotopic analyses.

Figure 5.7. Articulated forearm with associated beads in Area 4A of the tomb of Shimal 95 (from Velde nd-c:183).

Also located in the Shimal Middle portion of the cemetery, the collective tomb of

Shimal 103 was situated on a gravel fan around 200 m south of Shimal 95 (Vogt and

Velde 1987). Local and foreign artifacts associated with skeletal remains date this early

166 Wadi Suq grave to approximately 1800 BC (Vogt and Velde 1987). Although poorly preserved, this Ghalilah-type tomb still maintained a visible external diameter of around

10.5-11.0 m and width of 5.6 m, while each of its two internal chambers reached ca. 8.5 m in length and 1.3 m in width (Figure 5.8) (Velde nd-b). These chambers were of equal size and spatially divided by two central partition walls, both of which open at either end

Figure 5.8. Tomb plan of Shimal 103 (from Velde nd-b).

167 of the tomb as well as in the center, directly across from the western entrance (Vogt and

Velde 1987). As at Shimal 95, the perimeter wall of Shimal 103 consisted of a double facing of larger stones with a mixed earth and gravel fill centrally reinforcing these two layers (Vogt and Velde 1987; Velde nd-b).

As a result of significant looting activity in antiquity, none of the surviving grave goods (or skeletons) are thought to remain in situ (Velde nd-b). Although fragmentary,

Shimal 103 contained relatively homogeneous local painted pottery, including jars (both spouted and miniature) and beakers, as well as soft stone vessels, beads made of a variety of materials (shell, silver, chalcedony, quartzite, lead, carnelian), and chlorite containers

(Vogt and Velde 1987; Haser 1991). Four foreign vessels found within the tomb are also of special interest. Clearly of Dilmun origin, these jars represent the highest concentration of Dilmun vessels recovered from a single tomb during the Wadi Suq period (Velde nd-b). Of these four, three were found clustered in Square 6 and may have been intentionally placed with either an individual or a familial group interred in close proximity to one another (Velde nd-b). Velde (nd-b) has suggested that this individual(s) was either possibly (a) involved in interregional trade with Dilmun, or (b) from Dilmun, acting as a non-local trader at the port of Shimal.

Similar to other collective tombs from the Wadi Suq period, human skeletal material from Shimal 103 was commingled and extensively fragmented (Figure 5.9)

(Vogt and Velde 1987). However, this tomb contained significantly more human remains than its Shimal 95 counterpart, with an estimated MNI of at least 50 individuals

(Schutkowski and Herrmann 1987). This grave housed individuals of all ages and possessed a relatively balanced sex ratio (Schutkowski and Herrmann 1987; Vogt and

168

Figure 5.9. Eastern tomb chamber of Shimal 103 with commingled human remains (from Velde nd-b).

Velde 1987). Cribra orbitalia was reported for one individual, and post-cranially, osteoarthritis of a proximal radius as well as seven thoracic and three lumbar vertebrae was noted in low frequencies (Schutkowski and Herrmann 1987). Dental caries were present in some individuals (data is not given and numbers not specified), corresponding to a relatively high frequency of antemortem tooth loss (21.9%), primarily with regard to premolars and molars (Schutkowski and Herrmann 1987). Calculus deposits were rare, and dental attrition described as “slight to moderate” (Schutkowski and Herrmann 169 1987:62), perhaps indicative of a softer, marine-based diet. Of the remaining molars, seven were analyzed for strontium, oxygen, and carbon isotope ratios.

Tell Abraq, Emirate of Sharjah

Tell Abraq, located in the United Arab Emirates on the borderline between the

Emirates of Sharjah and Umm al-Quwain, was a large Bronze Age community in the

Arabian Peninsula (Potts 2000). Although a test trench was dug in 1973, the tell was not revisited until 1989, with five excavation seasons carried out between 1989 and 1998 headed by archaeologist Daniel Potts (Baustian and Martin 2010). Excavations recommenced in January 2007 by a team from Bryn Mawr University led by Peter Magee

(Magee et al. 2009).

Unlike other sites on the Oman Peninsula, Tell Abraq shows evidence of continuous occupation from the Bronze Age (2200-1200 BC) through the Iron Age

(1200-300 BC), making it an important settlement in terms of understanding both internal site and external interregional changes over time (Potts 1989). A rectangular-shaped mound four hectares in size, Tell Abraq was a coastal site during the Bronze Age, approximately 200-300 m from the water’s edge, although its current location lies a few kilometers from the Gulf (Potts 1993c). An enormous Umm an-Nar-period fortification tower dominated the settlement, with radiocarbon dates placing its construction at around

2200 BC (Potts 2000). This circular structure is unique both in its usage of dual construction materials (stones make up the exterior walls while mud bricks line the interior of the building) as well as its size, spanning a diameter of 40 m (Potts 1993d).

Moreover, this fortress-tower still stands eight meters high, making it the best-preserved

170 third millennium fortification in the Oman Peninsula (Potts 1994b). In contrast, other surviving fortresses, including those from Bidya and Hili 8, spanned diameters of just 16-

25 m and possessed only the foundations of these once-massive structures (Potts 1993a).

Like other fortifications of its time, the Tell Abraq tower contained a well at its center and was probably built to protect and/or control access to this valuable underground freshwater resource (Potts 1994b, 2000). While individuals of higher social status may have actually resided on the tower with ready access to this water, the majority of the population likely inhabited the area immediately surrounding the fortress, as evidenced by a large number of postholes encircling this structure (Potts 2000). The general populace may thus have lived in ephemeral structures built of palm fronds similar to traditional ‘arish housing referenced in texts as early as the Abbasid period (ca. AD

750-1258) and seen as late as the twentieth century AD in the UAE (King 2001). Other evidence of occupation around the tower includes large amounts of food refuse, grinding stones, ceramic sherds, and metal artifacts (Potts 2000).

The circular tomb at Tell Abraq, placed just 10 m west of the fortification, would have been another visible marker on the coastal landscape (Figure 5.10) (Potts 2000). Six meters in diameter, the collective tomb was paved with flat stones laid down underneath an internal dividing wall that partitioned the interior into two chambers, although a small passage (approximately 50 cm wide) between these chambers was present on the southern end (Potts 1993d). Two external ring-walls surround the tomb, including an outer wall of finely masoned ashlar blocks and a second, inner wall of rough stone (Potts

1993d). Based on both radiocarbon dates and artifacts associated with human skeletal

171

Figure 5.10. Umm an-Nar tomb at Tell Abraq. Note the paved floor, central dividing wall, and two outer ring walls composed of an external facing of fine ashlar stones and an inner layer of rough stone (from UAEInteract: Tell Abraq; http://www.uaeinteract.com/ history/e_walk/con_4/con4_18.asp; accessed 1/23/11).

material, the tomb was only in use from ca. 2200-2000 BC and thus dates to the end of the Umm an-Nar period (Potts and Weeks 1999; Baustian and Martin 2010). This mortuary structure is unique in that it was used for a limited period of time and remained undisturbed; conversely, the nearby fortress-tower was in use from the Umm an-Nar period until the end of the Iron Age (Potts 1993d).

The tomb contained the commingled, fragmentary remains of at least 286 adults and 127 subadults (Figure 5.11) (Baustian 2010). Individuals of both sexes and all ages were interred with no apparent differential treatment (Potts 2000). Due in part to their relocation to the United States, the skeletal material from Tell Abraq has been studied to a greater extent than that of any other Bronze Age tomb in southeastern Arabia. Physical anthropologists Debra Martin and Alan Goodman oversaw tomb excavations over a 172 period of four field seasons beginning in 1993 (Potts 1993d). A preliminary study on activity patterns using the tali and calcanei showed that these individuals led active lifestyles involving strenuous activities like walking and/or squatting, while metatarsals revealed that a small portion of the population practiced repetitive movements involving kneeling (Blau 1996). Associated archaeological evidence suggests that the site’s inhabitants may have engaged in food preparation, the processing of grains, and various maritime-related activities, all of which may have entailed such movements that led to markers on the skeleton (Blau 1996). Adult metacarpal and carpal bones possessed a high degree of osteoarthritis and overall robusticity, indicative of a population who

Figure 5.11. Human remains from the Umm an-Nar tomb at Tell Abraq (from Baustian 2010:3).

173 routinely utilized their hands in biomechanically demanding work (Cope et al. 2005;

Cope 2010).

Preliminary paleopathological reports suggest a high frequency of nonspecific infections indicative of stress, including periostitis, osteomyelitis, and porotic hyperostosis, as well as the presence of healed fractures and osteoarthritis in a considerable number of adults (Potts 1993d). In addition, one exceptional interment belonging to a single articulated female skeleton, aged 18 years old and placed in a flexed position within the passage connecting the eastern and western compartments of the tomb, shows evidence of chronic inactivity and neuromuscular deformities possibly associated with poliomyelitis (Potts 1993d; Anonymous 1994). Her skeleton had to have been protected in some way from the recurring deposition of the dead into the tomb, where deceased individuals were consistently pushed aside to make room for new bodies, thus contributing to the disarticulated state of the remains inside. The disparity between this interment as opposed to the mortuary treatment given to the other individuals may imply that the community held this woman in high regard (Baustian and Martin 2010).

Dental pathology at Tell Abraq has also been reported. Severe dental wear was common (Potts 1993d), which may explain why dental caries were only noted in 6.8% of the 628 teeth examined by Blau (2007). Enamel hypoplasias were also present on 7% of teeth, relatively low compared to other nearby sites like Mowaihat (Blau 2007). Further,

65.6% of 29 available mandibles display evidence of antemortem tooth loss, a likely result of the introduction of agriculture, particularly domesticates such as dates and other cereals, into the region during the Umm an-Nar period (Blau 2007). Calculus deposits

174 were also relatively high with a frequency of 14.2% (n=628), suggestive of an increased dependence on agricultural domesticates (Blau 2007).

A preliminary assessment of biological affinity has also been conducted on the skulls of those interred at Tell Abraq by Dr. Richard Wright. Using cranial measurements, multivariate analysis “suggests that the tomb was used by individuals originating in more than one population” (Potts 1993d:121). This, in conjunction with gross anatomical variation noted by initial skeletal evaluations, led Wright to conclude that Tell Abraq represented a mixed population, unsurprising given the artifactual evidence of a high degree of foreign exchange on site (Potts 1993d).

In addition, a paleodemographic evaluation of the 413 individuals placed in this tomb revealed an extremely high number of pre-term fetuses, neonates, and subadults under age 2 (Baustian 2010). This may suggest a higher susceptibility to infection and other environmental factors at younger ages (Baustian 2010; Baustian and Martin 2010), although Baustian (2010) also highlights the importance of cultural traditions, including early marriage age and consanguinity, both of which may have contributed to increased subadult morbidity and mortality. The presence of a considerable number of subadults may also indicate higher fertility levels (e.g., Buikstra et al. 1986; Wood 1992; Larsen

1997; Wright and Yoder 2003; Bocquet-Appel and Naji 2006). Conversely, very few skeletons represent individuals aged 6-19, which may point to increased survivability during this period of life history at Tell Abraq (Baustian 2010).

While environmental conditions in the Gulf prevented the survival of plant remains, abundant faunal material gives key insights into subsistence practices at Tell

Abraq. Domestic animals, including sheep, goat, and cattle, represent a considerable

175 portion of discarded faunal remains, and while sheep and goat outnumber cattle, their larger size probably permitted greater dietary contributions than the smaller domesticate ungulates on site (Potts 1993d; Stephan 1995). Wild fauna were also present, but in small quantities, making up less than 5% of total meat consumed (Potts 1993d).

Interestingly, birds, particularly cormorants, comprised a significant portion of bones recovered, a pattern not seen at other sites that perhaps suggests hunting at a nearby nesting ground (Uerpmann 2001). Large marine animals like dugongs, whales, dolphins, and turtles were also hunted but provided only a negligible amount of meat to overall consumption patterns (Stephan 1995). In addition, while fish were present but relatively unimportant in Umm an-Nar levels on site, a marked increase in fish, shellfish, dugong, and turtle is evident in Wadi Suq levels, indicative of a dramatic shift in which at least

50% of dietary intake was reliant on marine resources (Potts 1993d; Uerpmann 2001).

This likely points to diminishing terrestrial outputs, fitting with regional climatic changes and an increasingly arid environment beginning at the end of the third millennium BC

(Uerpmann 2001).

As at other Umm an-Nar sites in the Oman Peninsula, the vast majority of wares found at Tell Abraq were made from local materials (Cleuziou 2003). Nevertheless, foreign artifacts maintain a consistent presence at the site. For instance, while no local ceramic production is apparent at Tell Abraq, a substantial number of Harappan jar fragments were found on site (Potts 1993d). In addition, some rims sherds displayed incised Harappan signs (Hellyer 1998; Potts 2000). Moreover, three Harappan stone weights imported from the Indus Valley speak to the importance of Tell Abraq as an ancient trade port in the Persian Gulf during the third millennium (Potts 1993d, 2000).

176 During the Umm an-Nar period, the Oman Peninsula also supplied a thriving copper market in Mesopotamia, likely exporting raw metals that were subsequently reworked by specialists in these regions. However, many finished bronze and copper products were discovered at Tell Abraq as well and appear to be of local origin, as the workmanship is not comparable to either the Indus Valley or Mesopotamia and Dilmun

(Potts 1993d; Crawford 1998). A large amount of metallic waste in the form of copper droplets, or slag, was found scattered throughout the site, suggesting that metalworking took place here; significantly, some remained attached to ceramic fragments, indicative of a furnace involved in metals manufacturing (Potts 1993d; Weeks 1997). Furthermore, a pyramid-shaped ingot made from copper adds to the argument that Tell Abraq was involved in copper production (Potts 1993d; Weeks 1997).

An exception to the locally manufactured metal goods at Tell Abraq is the presence of a large copper axe with a distinctive smooth metal coating covering the blade, a technique practiced specifically at Mohenjodaro in the Indus Valley (Potts

1993c). While it is uncertain as to whether this axe was a direct Harappan import or if the technique was simply copied by local practitioners (Potts 1993c), its presence in the

Oman Peninsula speaks to the commercial ties between these two regions.

While local bead production occurred throughout southeastern Arabia, etched carnelian beads, engraved using an alkali solution, were specific to the Harappan manufacturers of the Indus Valley (Benton 1996; Potts 2000). More than 600 of these beads were recovered from within the tomb at Tell Abraq, all of which can be traced to the Harappan civilization (Potts 2000). In addition, one of the few articulated burials of a

177 woman within the Tell Abraq tomb possessed many necklaces and bracelets composed of flat, silver beads typical of hoards found at Mohenjodaro (Potts 1993d).

Tell Abraq did not limit itself to contact with Mesopotamia and the Indus Valley.

The tower tomb at Tell Abraq produced an unusual, crescent-shaped ivory comb decorated with flowers surrounding dotted circles (Figure 2.18) (Potts 1993e). Its motif appears almost identical to floral decorations found on vessels from Bactria (now southern Uzbekistan and northern Afghanistan), particularly its three petals and serrated- edged leaves (Potts 1993e, 2000). Such a striking parallel led Potts (2001) to conclude that the Tell Abraq comb was imported from Central Asia and belonged to an individual of elite status.

Umm an-Nar Island, Emirate of Abu Dhabi

Situated just 200 m off the western coast of the Emirate of Abu Dhabi, Umm an-

Nar Island is one of the most famous Umm an-Nar period sites in the Oman Peninsula, not only because of the almost 50 cairns scattered across the island, but also because of its role as the first Umm an-Nar site to be recorded and excavated, thereby giving this

Early Bronze Age period its name. Umm an-Nar Island is part of an archipelago along the southwestern coast of the Emirates, stretching only 3 km in length with an area of 4.5 km2 comprised of limestone rock and calcite sands (Hoch 1979; Potts 1990). The Umm an-Nar tomb fields on the island cluster into two distinct groups atop two plateaus partitioned by a wide expanse of sand (Figure 5.12) (Frifelt 1991). The southern plateau houses Tombs I-IV, while its northern counterpart contains Graves V-XLVIII; a forty-

178

Figure 5.12. Location of the tomb fields and settlement on Umm an-Nar Island (Frifelt 1991:15).

ninth grave constructed in isolation is present at the northeastern tip of the island (Frifelt

1991).

While stone monuments on the island had been known to area residents for some time, their initial ‘discovery’ was made in 1958 by T. Hillyard (Glob 1959; Bibby 1964).

179 Local enthusiasm to learn more about the pre-Islamic past of Trucial Oman (later gaining independence and forming the United Arab Emirates in 1971) led to the funding of an expedition to the island beginning in 1959 after an invitation was extended by the ruler of

Abu Dhabi to the Danes, already known for their work in Bahrain (Blau 2004; Benton

2006). This archaeological campaign lasted for six field seasons from 1959 to 1965 and remained largely unpublished until the early 1990s with the issuing of Karen Frifelt’s reports on the tombs (1991) and associated settlement (1995).

During the first season in 1959, the Danish Gulf Expedition initiated excavations in three areas, including two large mounds (Tombs I and II) and a nearby settlement

(Frifelt 1991, 1995). However, the unexpected complexity of the mortuary structures led archaeologists to halt work on Tomb II and the settlement in order to focus on Grave I

(Frifelt 1991). It was not until a second season in 1960 that excavations on this first cairn were completed. Like subsequent Umm an-Nar tombs excavated across the peninsula,

Tomb I possessed an exterior double ringwall approximately 1 m in thickness, with an inner band constructed of unworked limestone and an outer ring of carefully fitted, dressed ashlar (Frifelt 1991). 11 m in diameter, this circular grave surrounded numerous crosswalls segregating the tomb into eight chambers, as well as a central passage flanked by two median dividing walls and oriented north-south (Figure 5.13) (Frifelt 1991). This arrangement mirrored the two parallel entrances on both the north and south sides of the tomb (Frifelt 1991). Flat stones paved the floor of the tomb, and large slabs found caving into the tomb’s interior suggest a stone roof (Frifelt 1991).

180

Figure 5.13. Grave I, Umm an-Nar Island (Frifelt 1991:15).

Tomb I enclosed both grave goods and human skeletal material. Around 40 clay and alabaster vessels, small copper objects and fragments, grinding slabs, and hundreds of beads made of various materials were recovered from all chambers of the tomb. These artifacts suggest a date of approximately 2500-2300 BC (Benton 2006). An MNI of 21 individuals has been estimated from this burial monument, including nine males, five females, 1 individual of indeterminate sex, and six subadults (Højgaard 1980; Kunter

1991). While poor preservation precluded any conclusive diagnoses, two healed cranial fractures and three healed Parry’s fractures were reported in conjunction with a high prevalence of osteoarthritis of the mandibular condyles (Kunter 1991). However, these assessments have been heavily criticized regarding the lack of information given on methods of age and sex estimation as well as MNI and pathological identification

181 (Martin, personal communication). In this study, five molars belonging to four individuals from Umm an-Nar Tomb I were examined.

The second Danish campaign on Umm an-Nar Island in 1960 also witnessed the excavation of five small graves (IV, V, VI, VII, and VIII); here, 15 teeth from 14 individuals interred in Grave V were analyzed. Smaller relative to Graves I and II, the circular, limestone-built Grave V had a diameter of only 6.5 m, but like its larger counterparts, possessed a double ringwall with shaped, smoothed blocks facing its exterior (Frifelt 1991). Two entrances on both the north and south walls of the tomb also correspond to a N-S oriented crosswall separated from the ringwall by two short passages

(Figure 5.14), with a second, complete dividing wall perpendicular to this creating four inner chambers (Frifelt 1991). As with Tomb I, the floor of Tomb V was stone-paved and originally roofed with large pieces of flat stone (Frifelt 1991).

An abundance of material culture and skeletal remains occupied Tomb V, with particularly thick deposits accumulating in the two southern chambers. Approximately

60 clay vessels, including one depicting a humped oxen similar to motifs found at Tepe

Yahya, were found, as well as quernstones, alabaster vessels, and 17 copper awls, daggers, fish-hooks, pins, rivets, and other fragments (Frifelt 1991). Over 5000 beads from all chambers of the tomb were also unearthed (Frifelt 1991). Based on these grave goods, Grave V was likely in use from ca. 2700-2500 BC (Benton 2006). Heavily fragmented and disarticulated human remains were scattered throughout the tomb, representing at least 37 individuals (12 males, 8 females, 10 indeterminate, and 7 subadults) (Kunter 1991) and up to 49 persons (Højgaard 1980); four burials lining the exterior ringwall of the tomb were noted as well.

182

Figure 5.14. Tomb plan of Grave V, Umm an-Nar Island (Frifelt 1991:261).

In 1961, a third expedition to Umm an-Nar Island focused primarily on the massive Grave II after preliminary excavations in 1960 had delineated the ringwall of this tomb (Thorvildsen 1963; Frifelt 1991). Positioned just 80 m to the southeast of Tomb I,

Tomb II boasted the largest diameter of any funerary monument on the island at 12 m, was constructed in typical Umm an-Nar fashion with a circular double ringwall and multiple interior cross-walls forming eight chambers as well as a central passage, and was flanked on both its north and south sides by two entry points (Figure 5.15) (Frifelt 1991).

However, the most fascinating aspect of this tomb lies in five limestone blocks carved in

183

Figure 5.15. Tomb plan of Grave II, Umm an-Nar Island (Frifelt 1991:259).

bas-relief (Frifelt 1991). These carvings depict a bull, a camel and oryx, a human-like figure, a camel, and two snakes, respectively, and likely embellished the dual entrances of the tomb (Frifelt 1991). At least 55 ceramic vessels, 6 copper objects, grinding slabs, netsinkers, hammerstones, and over 5000 beads had been placed in Tomb II during its use between ca. 2500-2300 BC (Frifelt 1991; Benton 2006). Poorly preserved and commingled human skeletal material accompanied these grave goods, with an estimated

MNI of 34 (Kunter 1991) to 38 (Højgaard 1980), including at least 20 males, 6 females, 4

184 individuals of indeterminate sex, and 4 subadults (Kunter 1991). 18 teeth from 15 individuals were analyzed from Tomb II for this study.

In addition to the excavation of seven Umm an-Nar tombs over the course of three field seasons (1959-1961), excavations of a contemporary settlement (approximately 200-

300 m2) on the northeastern shores of the island took place as well (Hoch 1979). While a test trench was started in 1959 during the first archaeological campaign on Umm an-Nar

Island by the Danes, this work was abandoned so that Tomb I could be properly addressed (Frifelt 1991, 1995). It was only during the fourth expedition to the island in

1962-1963 that excavations resumed in the settlement area, and continued in subsequent field seasons (1964, 1965) (Frifelt 1995). During these campaigns, numerous limestone foundations were unearthed, and based on their construction, it is likely that these buildings were covered not with stone or mudbrick but by palm leaves or other plants in barasti-like fashion (Frifelt 1995). Interestingly, the rooms examined contained impressive quantities of imported pottery sherds, shell, and copper objects and fragments, suggesting that the more permanent structures of the settlement may have actually served not as private residences but as workshops (Frifelt 1995). One particularly impressive complex that has been dubbed a ‘warehouse,’ comprised of seven rooms and measuring

16 m2, contained a casting mold, copper refuse, and ingots indicative of copper working, as well as faunal remains implying the curing of meats and fish (Frifelt 1995). Large numbers of Mesopotamian ceramic vessels within the structure also point to the building’s use as one of storage, and perhaps redistribution, of liquids such as oil

(Mynors 1983; Frifelt 1995; Potts 2001). Domestic dwellings were almost certainly constructed of organic material like palm fronds (Frifelt 1995). Unlike some sites also

185 dating to this period (e.g., Tell Abraq, Hili), no evidence of a fortified tower exists on this island settlement; however, this may simply be a product of areas chosen for excavation, as only a very small percentage of the site has been exposed.

The inhabitants of Umm an-Nar Island were clearly involved in interregional trade (Frifelt 1995). Umm an-Nar Island’s strategic position in the Gulf made it a major trading port on the Oman Peninsula during the early and middle Umm an-Nar period until its abandonment ca. 2200 BC, around the time that other coastal sites like Tell Abraq began to flourish in the late third millennium BC (Frifelt 1995). Significant quantities of ceramic sherds from Mesopotamia, the Indus Valley, and southeastern Iran illustrate the development of these exchange networks during the first part of the Umm an-Nar period

(Frifelt 1995). Additional finds such as etched carnelian beads of probable Harappan origin, imported incised grey and black-on-grey Iranian wares, and an unusual seal impression reminiscent of Syrian forms further speak to interregional contacts and relationships (Amiet 1975, 1985; Frifelt 1991, 1995; Potts 2005).

Subsistence on Umm an-Nar Island revolved primarily around maritime resources, with large quantities of bone and shell originating from fish, molluscs, dugong, turtles, sharks, stingrays, and whales recovered from settlement middens (Hoch 1979,

1995). This dependence on the sea is further illustrated by the considerable number of netsinkers, fish-hooks, and other marine artifacts unearthed both at the settlement and within the tombs (Frifelt 1991, 1995; Benton 2006). Coastal birds, especially the cormorant, seem to have been frequently trapped as well (Hoch 1979, 1995; Potts 1990).

These remains vastly outnumber both domestic and wild terrestrial mammals, which included sheep, goat, cattle, gazelle, oryx, and camel (Hoch 1979, 1995). While no plant

186 remains were recovered directly from the settlement area, plant impressions on materials such as ceramic sherds, mudbricks, and bitumen show the presence of wheat, barley, straw, and date stones (Willcox 1995). The presence of a few quernstones, again both deposited in graves and within the workshops of the settlement, hint towards at least some cereal processing on site (Frifelt 1995). Interestingly, the island contained no freshwater supply, which would have made it necessary for its Bronze Age inhabitants to frequently travel to the mainland (Frifelt 1991). From the settlement, 15 fauna were sampled, including one oryx, two cattle, and 12 sheep/goat.

Various analyses of human teeth from Graves I, II, and V have been conducted

(Højgaard 1980, 1981; Kunter 1991) to discern dietary patterns on Umm an-Nar Island.

In conjunction with zooarchaeological evidence, severe wear, the relative absence of caries, the presence of calculus, and a low prevalence of antemortem tooth loss has been interpreted as indicative of a diet dominated by coarse marine resources and not agricultural products, which likely played only a supplementary role (Højgaard 1980,

1981; Kunter 1991). No raw data are given for these studies. Based on both dental metric and nonmetric traits, including agenesis of the third molar, Højgaard (1980, 1981),

Kunter (1991), and later Alt et al. (1995) suggest that these individuals represent a homogeneous, endogamous kin-based group.

Unar 1, Emirate of Ras al-Khaimah

Located on the Shimal Plain, along the foothills of the Hajjar Mountains in the

Emirate of Ras al-Khaimah, the tomb of Unar 1 is one of only two Umm an-Nar funerary structures at the vast Shimal necropolis, dominated almost entirely by Wadi Suq-period

187 graves. Unfortunately, Unar 1 is often overshadowed by its larger and more renowned

Umm an-Nar counterpart, Unar 2, positioned just 200 m south of Unar 1 (Sahm 1988).

No publications on the site exist to describe the archaeological excavations and recovered material culture; therefore, the following descriptions are based on piecemeal compilations of unpublished reports, dissertations, and other brief mentions in the literature.

Excavations at Unar 1 commenced in 1987 by the German Mission to Ras al-

Khaimah after the accidental discovery of the tomb during a construction project (Kästner

1988; Schutkowski 1988; Blau 1998). A second campaign in 1988 saw the completion of formal excavations, although a team returned for a third time in 1989 to evaluate the skeletal material (Schutkowski 1988, 1989). As was typical of Umm an-Nar funerary architecture, the Unar 1 grave was a large, circular, collective burial monument with a diameter of 11.5 m and a median crosswall dividing the interior of the tomb (Figure 5.16)

(Blau 1998). Three smaller walls, oriented east-west and thus perpendicular to the central crosswall, further divided the internal space of the tomb into eight chambers, although looters dismantled most of these in antiquity, possibly as early as the second millennium BC (Sahm 1988; Blau 1998; Weeks 2003a).

Grave goods associated with human remains, including local black-on-red ceramic sherds, numerous copper awls, pins, and rings, and a variety of beads, date the tomb to the early to mid-Umm an-Nar, namely from ca. 2400-2200 BC (Sahm 1988; Blau

1998; Weeks 2003a, 2003b). In addition to local wares, non-local products were also found in the tomb, such as an etched carnelian bead and incised gray as well as

188

Figure 5.16. Unar 1, a circular Umm an-Nar tomb on the Shimal Plain (from Weeks 2003a:62, photo by Christian Velde).

black-on-gray Iranian pottery, suggestive of Unar 1’s potential involvement in larger interregional exchange networks (Sahm 1988; Weeks 2003a).

At least 438 individuals were interred at Unar 1 during its use over 200 years

(Blau 2001b). Evidence for both inhumation and cremation exists, and cremation does not appear to have been selectively applied for males, females, or subadults (Schutkowski

1989, 1989). Remains were cremated to varying degrees (charred to calcined), and clusters of burned bones were found throughout the tomb (Schutkowski 1988). The ratio of males to females is balanced, and infants comprise approximately 22-24% of the assemblage (Schutkowski 1988, 1989). While the vast majority of remains were fragmentary and commingled, the presence of a few articulated skeletal segments suggest that individuals were initially buried in a tightly flexed position (Figure 5.17) 189

Figure 5.17. Partially articulated skeletal segments from the Umm an-Nar tomb of Unar 1, Unit L 11 G (from Schutkowski 1988: Figure 1).

(Schutkowski 1988). Of the 279 teeth examined by Blau (2001b, 2007), low rates of caries (3.9%), linear enamel hypoplasias (1.8%), dental abscesses (0.7%), and calculus

(1.1%) all suggest mixed subsistence practices and a broad diet.

United Arab Emirates: Gulf of Oman (East) Coast

The Emirate of Fujairah is the only emirate of seven to entirely lie alongside the east coast of the Gulf of Oman (Figure 5.18). Its landscape is in stark contrast to the deserts of the western Emirates along the Persian Gulf coast, with the fertile, rich soils of the gravel plains coupled with a relatively high annual rainfall providing a coastal oasis environment (King and Maren-Griesbach 1999). Surrounding these narrow plains, the

Hajar Mountains extend across Fujairah and into Oman. Very little has been published

190

Figure 5.18. Map of the northern United Arab Emirates. The Emirate of Fujairah is highlighted in gray on the right. Note that Fujairah has two exclaves to the north and south of its main territory, where the capital city is located.

on the Bronze and Iron Age archaeology of the Emirate of Fujairah, and few human remains recovered. Nevertheless, a brief summary of each site where samples were taken is included below.

Bidya, Emirate of Fujairah

Located in northern Emirate of Fujairah, approximately 38 km north of the capital city of Fujairah, the coastal site of Bidya sits within a palm oasis bordering the nearby

Hajar Mountains (Figure 5.19). Lying at the mouth of a wadi, the ancient inhabitants of

Bidya would have had ready access to fresh water as well as marine resources from the

Batinah Coast. Excavations at the site were undertaken in 1989 by the Al Ain

Department of Antiquities and Tourism of Abu Dhabi, led by Dr. W.Y. Al Tikriti (Al

Tikriti 1989b; King and Maren-Griesbach 1999; Brass and Britton 2004). While Bidya is

191

Figure 5.19. Major archaeological sites in the northern Emirates, including the coastal sites of Bidya, Dibba, and Qidfa in the Emirate of Fujairah, as well as Kalba in the Emirate of Sharjah (From Vogt 1998:274).

192 well known for its historic archaeology, including the 17th century AD Bidya mosque and

Portuguese fort (King and Maren-Griesbach 1999; Ziolkowski 2008), its oldest occupation dates to the third millennium BC, as evidenced by the remains of a substantial round tower dating to the Umm an-Nar period (Al Tikriti 1989b). Similar to the circular fortress towers found at Tell Abraq and Hili, and the earliest tower unearthed from the

Emirate of Fujairah to date, the Bidya tower was likely the hub of a larger surrounding oasis settlement (Hellyer 1994; Potts 2009). Unfortunately, due to heavy bulldozing and the recent construction of a village on the site, little of the Umm an-Nar settlement remains (Velde and Moellering 2010).

Nearby, a series of Wadi Suq- and Iron Age-period collective tombs have been surveyed, although most remain unexcavated. All appear to have been plundered in antiquity, and it is likely that most were re-used with some regularity, a practice not uncommon throughout this region, particularly during the Iron Age (Hellyer 1994).

Beginning with a 1987 expedition led by Al Tikriti, a large, semi-subterranean tomb denoted as Bidya 1 was excavated (Al Tikriti 1989b). This communal tomb, approximately 30.7 m x 2.0 m, lay mostly underground but was marked on the surface by two parallel stone lines extending 25 m and oriented north-south (Figure 5.20) (Al Tikriti

1989b). Interestingly, while these external stones were taken from the mountains, the inner tomb was paved with marine stones acquired from the beach (Al Tikriti 1989b).

Such pavement within the tomb is atypical for Wadi Suq mortuary structures (Riley and

Petrie 1999). Large stone slabs, rectangular in shape, provided a protective covering over the tomb (Brass and Britton 2004). The tomb shares its rectangular shape with other

193

Figure 5.20. Tomb plan of Bidya 1 in the Emirate of Fujairah (from Al Tikriti 1989a: Plate 61).

Wadi Suq tombs in the region, including those at Shimal in the Emirate of Ras al-

Khaimah (Hellyer 1994), but is also reminiscent of the late third millennium subterranean rectangular tombs found at Ajman and Hili and may thus reflect an evolution of mortuary architecture (Al Tikriti 1989a).

The Bidya 1 tomb contained a substantial amount of commingled human skeletal material, with no complete skeletons recovered. Al Tikriti (1989b) estimated that the

194 tomb initially contained over 100 individuals; however, as with other tombs in the area,

Bidya 1 had been looted, contributing to the severe fragmentation of the skeletons interred there. Twelve skulls in poor condition were recovered from the western portion of the tomb, along with most other postcranial remains. The three human molars sampled from Bidya 1 in this study came from this portion of the tomb. Grave goods from the tomb indicate that it was in use during the early second millennium BC, at approximately

1800 BC (Al Tikriti 1989b; Barker 2002). However, because tomb reuse may have occurred, it cannot be said with certainty whether these individuals date to the Wadi Suq period or to the subsequent Iron Age (Hellyer 1994), so caution must be taken in interpreting isotopic results without definitive dating to the Middle Bronze Age. An additional, secondary Hellenistic deposition of bone in the top-most and central portion of the tomb contained the remains of five individuals from the 1st century AD (Al Tikriti

1989b).

Dadna, Emirate of Fujairah

Located just seven kilometers southeast of Dibba on the eastern coast of the

United Arab Emirates, the tomb at Dadna sits amidst a modern village and has thus sustained damage resulting from recent construction projects (Figure 5.21) (Benoist

2002; Benoist and Hassan 2010). Dadna was the site of a rescue excavation after workers installing a cable unearthed a collective burial (Benoist and Hassan 2010). Salah Ali

Hassan of the Department of Antiquities in the Emirate of Fujairah oversaw excavations of the tomb in December 1995, although the site was not published until 2010 (Benoist

195

Figure 5.21. Location of the sites of Dibba and Dadna, Emirate of Fujairah, on the coast of the Gulf of Oman (from Brass and Britton 2004:150).

and Hassan 2010). This rescue expedition uncovered two subterranean, rectilinear chambers belonging to a single grave and divided by a central interior crosswall (Figure

5.22) (Benoist 2002; Benoist and Hassan 2010). Damage to the original exterior wall, particularly on its north/northeastern side, makes it unclear if the structure was rectangular or U-shaped; however, U-shaped structures from the second millennium are known from the nearby sites of Qidfa and Mereshid (Benoist 2002; Benoist and Hassan

2010). Walls were constructed of multiple rows of unworked stone with a fill comprised of sand, gravel and small rocks (Benoist 2002; Benoist and Hassan 2010).

Despite disturbances both in antiquity and in the present day, a considerable number of grave goods and large quantities of fragmentary, commingled skeletal material were left in the Dadna tomb (Benoist and Hassan 2010). While an analysis of these

196

Figure 5.22. Tomb plan from Dadna (from Benoist and Hassan 2010:85).

human remains has yet to take place, the many artifacts from this grave – including bronze arrowheads, a bronze bowl, beads manufactured from multiple stone and shell materials, pottery, and softstone containers – illustrate a long history of occupation from the second millennium Wadi Suq to ca. 600 BC, corresponding to the Iron Age II-III in this region (Benoist 2002; Benoist and Hassan 2010). Due to poor preservation, a single tooth was all that remained for stable isotope analysis, and because reuse of Wadi Suq funerary structures during the Iron Age was common practice, a second millennium date could not be firmly established.

197 Dibba, Emirate of Fujairah

Dibba sits in a large harbor on the eastern coastline of the Emirate of Fujairah, with the Hajar Mountains situated just southeast of the site (Figures 5.19, 5.21) (de Cardi and Doe 1971). Located at the base of two wadis, the Dibba oasis was formed by the buildup of rich soils into a sizeable gravel fan (Velde 2010). Sedimentary deposits have obscured evidence of much of its prehistoric settlement, although a fertile terrestrial and marine environment likely gave rise to an important trading settlement here as early as the Wadi Suq period (Velde 2010).

Dibba has one of the longest excavation histories in Fujairah. In 1962, a series of pre-Islamic artifacts and materials, including steatite vessel fragments, pottery sherds, a bronze arrowhead, and bones, was unearthed during a training exercise as the Trucial

Oman Scouts were practicing digging trenches (Bibby 1966). Two years later, as part of the Danish Archaeological Expedition in 1964, archaeologists T.G. Bibby and P.V. Glob made a brief stop in Dibba to view this trench and evaluate these materials (Bibby 1966).

Other materials were later collected by Captain W.F. Stockdale and H.G. Balfour-Paul

(de Cardi and Doe 1971). Subsequent surveys of the Batinah Coast were undertaken in

1968 by a British archaeological team led by Beatrice de Cardi (de Cardi and Doe 1971;

Brass and Britton 2004). Twenty-five years later, during the 1993-1994 field season, the

Fujairah Museum excavated Dibba 76, from which three human teeth and one faunal tooth for this study were analyzed (Brass and Britton 2004). A 1995 archaeological survey of Fujairah by an Australian expedition examined numerous sites along the eastern coast, including multiple graves and a tower at Dibba, but did not return to the

Dibba 76 tomb (Brass and Britton 2004).

198 The Dibba 76 tombs consist of two long graves, oriented parallel to each other

(Ziolkowski and Al-Sharqi 2006). Originally constructed during the second millennium

BC, these subterranean mortuary structures each contained artifacts spanning thousands of years, with finds ranging in age from Wadi Suq through the Hellenistic period.

However, most finds date to the earliest periods of tomb use, namely the Wadi Suq and

Iron Age (Barker and Hassan 2004). As with Bidya 1, then, the reuse of these graves, particularly during the Iron Age, precludes any definitive dating of its skeletal remains

(Vogt 1998). Three individuals from Dibba 76 were sampled as part of this study.

Mereshid, Emirate of Fujairah

No publications on the Mereshid excavations currently exist, and the site is often simply mentioned in passing (e.g., Barker 2004; Benoist 2007; Velde 2010); however, an unpublished excavation report (Benoist 2002) does provide a brief description of the tomb and its findings.

Located in an extensive palm garden just a few kilometers north of Kalba, the site of Mereshid was home to a second millennium tomb discovered in the southwest quarter of the town of Fujairah during construction work (Figure 5.23) (Benoist 2002; Brass and

Britton 2004; Velde 2010). Excavations of the tomb were undertaken in 1997 by the

Fujairah Museum under Salah Ali Hassan before its eventual destruction to make room for a modern residential structure (Benoist 2002). This collective, U-shaped grave measured 8 m in length and was subterranean in its construction (Benoist 2002). While the entrance to the tomb was not found, stones formed the walls, which were oriented east-west (Benoist 2002). As with other tombs scattered along the eastern coast of the

199

Figure 5.23. Map of Middle Bronze Age (Wadi Suq) sites along the eastern coast of the United Arab Emirates, including the tomb at Mereshid (from Velde 2010:12).

Emirates, hundreds of ceramic sherds, three softstone vessels, three bronze spearheads, one bronze bracelet, one bronze ring, 14 carnelian and shell beads, and two shell rings characteristic of the Wadi Suq period were present (Benoist 2002, 2007; Barker 2004).

200 Additionally, five ceramic bowls and a rim fragment are comparable to Iron Age I and II finds elsewhere (Benoist 2002).

Human skeletal remains from Mereshid are not discussed in this literature, although Benoist (2002) does mention that bones were collected before the tomb’s demolition to be stored at the Fujairah Museum. Upon inspection of these remains by the author at the Fujairah Museum, the Mereshid skeletal material appeared extremely fragmentary, commingled, and in a state of poor preservation. One individual from the

Mereshid tomb was sampled as part of this study.

Qidfa, Emirate of Fujairah

Qidfa is one of a number of sites along the eastern coast of the Emirate of

Fujairah containing numerous collective Wadi Suq and Iron Age graves, speaking to the importance of the area during the Middle Bronze and Iron Ages (Figures 5.18, 5.21)

(Hellyer 1994; Vogt 1998). A resident farmer discovered the site, now located near the small, modern village of Qidfa, with formal excavations carried out in 1986 by an Emirati team from the Al Ain Department of Antiquities and Tourism, led by W.Y. Al Tikriti (Al

Tikriti 1989b; Hellyer 1994; Brass and Britton 2004). This expedition excavated part of a second millennium settlement as well as the tomb of Qidfa 4, from which two human molars and one faunal tooth were sampled in this study (Brass and Britton 2004). Qidfa

4 was a large, subterranean, collective grave constructed in the shape of a horseshoe (Al

Tikriti 1989b). While Al Tikriti (1989b) tentatively dated Qidfa 4 to the latter half of the second millennium, subsequent analyses of the artifacts unearthed from this uniquely undisturbed tomb place it in the early Iron Age, at around 1000 BC (Hellyer 1994). An

201 impressive collection of grave goods accompanied the human remains, including a rich array of bronze-made weapons (axes, daggers, swords), bracelets, and containers, as well as pottery, stone vessels, and stone beads (Al Tikriti 1989b; Hellyer 1994).

Comparative Sites

Due to the absence of comparative isotopic data in this region, Bronze Age human and fauna samples from six sites outside of the United Arab Emirates were selected for additional analyses (Figure 5.24, Table 5.2). These sites were chosen from areas known to be actively engaged in trade with the Oman Peninsula during the Bronze Age. These include the Dilmun sites of the A’ali Mound Field (ca. 2200-1800 BC; e.g., Frohlich

1983; Frifelt 1986; Højlund 2007) and Barbar Temple (ca. 2000 BC; e.g., Mortensen

Figure 5.24. Map of the Arabian Peninsula and South Asia, illustrating the location of comparative Bronze Age sites across the Persian Gulf used in this study (Image from Google Earth).

202 Table 5.2. Comparative Bronze Age human and faunal teeth sampled in this study.

Site Country Human n Faunal n A'ali Mound Field Bahrain 6 5 Barbar Temple Bahrain 0 15 Failaka Island Kuwait 0 15 Tepe Yahya Iran 0 10 Al-Khubayb Oman 3 0 Allahdino Pakistan 0 10 Balakot Pakistan 0 10

1971, 1986; Andersen 1986; Andersen and Højlund 2000, 2003; Højlund et al. 2005) in

Bahrain, the inland site of Tepe Yahya (ca. 5500 BC to AD 300, with samples dating to

Period IVA and B, ca. 2400-1700 BC; e.g., Lamberg-Karlovsky 1970, 1971, 1972, 1977;

Lamberg-Karlovsky and Beale 1986; Lamberg-Karlovsky and Potts 2001; Beale 2004;

Potts 2004b; Magee 2005) in modern-day Iran, the Dilmun colony of Failaka Island

(early second millennium BC; e.g., Salles 1984; Calvet and Salles 1986; Højlund 1987;

Calvet and Gachet 1990) off the coast of Kuwait, and the two coastal Harappan sites of

Allahdino (ca. 2500-1700 BC; e.g., Fairservis 1973, 1976, 1993; Hoffman and Shaffer

1976; Hoffman and Cleland 1977; Belcher 2005) and Balakot (ca. 2500-2100 BC; e.g.,

Dales 1974, 1979a, 1979b, 1981; Dales and Kenoyer 1977; Meadow 1979; Franke-Vogt

1997; Belcher 2000, 2005) from Pakistan.

Methods

Enamel Extraction

250 human and faunal teeth were collected from the sites described above (Tables

5.1, 5.2). Due to the disarticulated nature of the remains, almost all teeth were not found 203 in situ but were isolated and sometimes fragmentary. Subsequently, in order to prevent repetitive sampling from the same individual, the most common molar of a particular dental quadrant (e.g., LM1) present was selected from each site. Whenever possible, first molars were preferred. Because the majority of teeth sampled were loose, molars were differentiated based on guidelines found in Hillson (1996) and White and Folkens (2005).

Traits used to identify tooth type included:

1. Cusp pattern: For mandibular molars, first molars were distinguished based on

the presence of five cusps, including a well-developed hypoconulid, with a large

crown that typically displays a Y-5 pattern. Second molars possessed slightly

smaller crowns and were usually more rectangular in shape with a four- instead of

five-cusp arrangement. Third molars normally had the smallest crowns and

exhibited the most variability regarding overall shape and cusp arrangement. For

maxillary molars, first molars exhibited the largest crown size, the most

pronounced hypocone, and a rhombiodal shape that becomes more square in the

second molar. The hypocone and overall crown size are also reduced in the

second molar. Furthermore, as has been noted at a variety of other sites in the

Near East, upper second molars can frequently be characterized by the presence

of a reduced metacone (Ullinger 2010; Gauld et al. In press). The variable third

molar often displays irregular cusp patterns and is typically more triangular in

shape. For both lower and upper third molars, crowns showed more crenulation

on the occlusal surface.

2. Interproximal contact facets (ICPFs): Two wear facets were present on the

mesial and distal surfaces of both upper and lower first and second molars. The

204 distal facet was absent in third molars.

3. Root divergence: For mandibular molars, first molars possess two long and

distinct roots that diverge close to the cervix of the tooth. This divergence

successively decreases in second and third molars, and in many cases, third molar

roots are fused. Similarly, for maxillary molars, first molars display the most

divergent roots, with increasingly diminished divergence in second and third

molars, respectively. As with mandibular molars, roots of the third maxillary

molars are commonly fused.

In addition, deciduous molars were distinguished from permanent molars by their more bulbous crowns, as well as their shorter, thinner, and more divergent roots, which were

(in some cases) resorbed (White and Folkens 2005). Moreover, the unusual morphology of deciduous first molar crowns sets these teeth apart from their permanent counterparts,

1 with prominent protoconid (dm1) and paracone (dm ) surfaces, respectively (White and

Folkens 2005). For faunal dentitions, another strategy had to be employed due to the scarcity of well-preserved, intact animal teeth from many of the tombs and settlements.

In this case, every attempt was made to sample teeth from different areas of excavation to prevent the analysis of multiple teeth from the same animal.

Dental molds of intact human teeth were made using Coltene/Whaledent Inc.

President regular body with a MicroSystem Jet Blue Dispenser, as described in Fiorenza et al. (2009). Dental casts were made using Struer’s EpoFix. Molars were gently cleaned by hand with a toothbrush to remove visible surface contaminants. Tooth surfaces were then cleaned mechanically by drilling, removing the most superficial layer of enamel.

Because cementum coats the enamel crown of some mammals, including the ungulates

205 sampled here, this outer cementum layer was removed before surface abrasion of enamel could take place (Figure 5.25) (O’Connor 2000). Following cleaning procedures, 5-10 mg of inner enamel from each tooth was isolated with a carbide drill bit attached to a Dremel

300 Series Rotary Tool.

Strontium

Sample preparation of enamel for strontium isotope analysis was drawn from

Perry and Coleman et al. (2008, 2009). Samples were prepared and measured at the

Isotope Geochemistry Laboratory in the Department of Geological Sciences at the

University of North Carolina at Chapel Hill. 3-5 mg of enamel was weighed into a

Teflon beaker and dissolved overnight in 500 L 3.5 M HNO3 (nitric acid). 50-100 m

Figure 5.25. Ovicaprine RM3 from Balakot, Pakistan after removal of outer coating of cementum and enamel abrasion (Sample ID: Bal 84; Curation ID: 4Blk 292 #4). Photo courtesy of Dr. Richard Meadow, Peabody Museum, Harvard University.

206 EiChrom (SR-B100-S) Sr-Spec resin was utilized in column extraction of strontium from the dissolved sample. 30 L of 0.1 M H3PO4 (phosphoric acid) was added to extracted samples before drying down the solution on a hot plate at 145C. Samples were later re- dissolved using 2 L TaCl5 (tantalum chloride) before being loaded onto rhenium (Re) filaments and dried down with an electrical current. Stable isotope ratios were analyzed on VG Micromass Sector 54 thermal ionization mass spectrometer (TIMS) in quintuple- collector dynamic mode. To correct for mass fractionation, an internal ratio of 86Sr/88Sr =

0.1194 was used. Ratios are reported relative to a value of 0.710270  0.000014 (2) for the NBS-987 standard. Internal precision for strontium runs is typically 0.000012 to

0.000018% (2) standard error based on 100 dynamic cycles of data collection.

Oxygen and Carbon

Sample preparation of enamel carbonate for oxygen and carbon isotope analysis was drawn from Krigbaum (2003) and Garvie-Lok et al. (2004). Collected enamel powder was placed into sterilized, lidded 1.5-1.7 mL microcentrifuge tubes. 1.5 mL of

2% sodium hypochlorite (NaOCl) solution was added to each tube, and samples were allowed to soak for 13 hours. This process removed any organic contaminants present in the sample, including humic acids. During this treatment, samples were stirred every three hours to ensure exposure of the entire powdered sample. After 13 hours, sample tubes were centrifuged to consolidate the samples so that the NaOCl supernatant could be removed; samples were then rinsed with distilled, deionized water before being stirred.

This cycle was typically repeated three to five times until samples achieved neutrality.

Samples were dried overnight in a desiccator cabinet. In order to remove soluble 207 diagenetic contaminants as well as adsorbed carbonate, samples were next treated with

1.5 mL 0.1 M acetic acid solution (CH3COOH) for 4 hours. During this soak, sample tubes were stirred every hour. Samples were rinsed to neutrality, frozen overnight, and then lyophilized for a minimum of 48 hours.

A 75-95 µg carbonate subsample was analyzed for δ18O (relative to Vienna

Peedee Belemnite Limestone standard (VPDB)) and δ13C (relative to VPDB) using an automated Carbonate Kiel device coupled to a Finnigan Delta IV Plus stable isotope ratio mass spectrometer at the Stable Isotope Biogeochemistry Laboratory at The Ohio State

University. Samples were acidified under vacuum with 100% ortho-phosphoric acid, the resulting CO2 cryogenically purified, and delivered to the mass spectrometer.

Approximately 10% of all samples were run in duplicate. The standard deviation of repeated measurements of an internal standard was ± 0.03‰ for δ13C and ± 0.06‰ for

δ18O.

In following with previous oxygen isotope studies (e.g., Knudson 2009),

18 18 conversion equations were utilized in order to adapt δ Oc (VPDB) values to δ Odw

(VSMOW) ratios (dw = drinking water) so that locality may be generally assessed by

18 comparisons with meteoric water δ Omw maps. These include the following three conversion steps:

18 18 δ Oc(VSMOW) = (1.03091 x δ Oc(VPDB)) + 30.91 Coplen et al. 1983

18 18 δ Op(VSMOW) = (0.98 x δ Oc(VSMOW)) – 8.5 Iacumin et al. 1996

18 18 δ Odw(VSMOW) = (δ Op(VSMOW) – 22.70)  0.78 Luz et al. 1984

208 These conversions, along with the potential effects of errors associated with regression calibration (Pollard et al. 2011), will be discussed further in Chapter 7.

Statistical Analyses

All statistical analyses were performed using Statistical Analysis Software (SAS) version 9.2. The D’Agostino-Pearson K2 test was utilized in order to determine whether each dataset was normally distributed. The null hypothesis holds that values will be normally distributed, while the alternate hypothesis anticipates that samples do not possess a normal distribution. Because samples from Mowaihat, Shimal, and Fujairah number less than 20 individuals, the probability of failing to reject the null hypothesis increases; as such, these datasets will be assumed to have non-normal distributions.

The null hypothesis was rejected – meaning that the data could not be treated as normal – for the majority of value distributions at each site and for each isotope. The following three sample distributions failed to reject the null hypothesis: Umm an-Nar

Island (18O; KSQ=1.98); Tell Abraq (13C; KSQ=0.19); and Unar 1 (13C; KSQ=3.98).

Because only three of 18 tests failed to reject the null hypothesis, for comparative statistical purposes, all data were assumed to be non-normally distributed. Additionally, data for both oxygen and carbon isotopes is ordinal because it represents a type of percentage (‰, or ‘per mil’) that does not show quantity, but instead, illustrates rank, precluding the use of parametric statistics. As such, only statistics that do not require normality (e.g., non-parametric) were used in this study.

The non-parametric Mann-Whitney U (Wilcoxon rank-sum) test was selected because it does not assume equality of variance or normal distribution between two sets

209 of samples and because it does not require sample sets to possess an equal number of observations. This test assesses if two independent groups of samples are drawn from similar distributions (or, in biostatistics, populations), and is essentially equivalent to a parametric independent-sample t-test. The null hypothesis assumes that the two independent sample sets were drawn from the same distribution (x¯ 1= x¯ 2), while the alternate hypothesis anticipates that these samples originated from different distributions

(x¯ 1 ≠ x¯ 2).

Monte Carlo analyses randomly reorganize data in order to test whether sample observations differ when arbitrarily allocated to different groups. As with the Mann-

Whitney, this test is useful because it does not assume a normal distribution. The null hypothesis states that sample means will not differ (x¯ 1 = x¯ 2), while the alternate hypothesis expects a difference in means after randomized sampling has taken place

(x¯ 1 ≠ x¯ 2). A simulation of 9999 at a 0.05 significance level was run for each Monte

Carlo test.

Levene’s test was used to test the equality of variances between the Umm an-Nar and Wadi Suq periods for all three isotopes. The null hypothesis assumes a homogeneity of variance (v1=v2), while the alternate hypothesis expects variances between samples to differ (v1≠v2).

***

In summary, human and faunal dental enamel was sampled from six Umm an-Nar and seven Wadi Suq Bronze Age tombs in the United Arab Emirates. Comparative samples from across the Gulf were taken from various sites in Bahrain, Kuwait, Iran,

Oman, and Pakistan. Strontium isotope sample preparation involves the drilling and

210 isolation of enamel samples, with the resultant enamel powder cleaned, treated with and dissolved in nitric acid, and purified by extraction chromatography before being loaded onto filaments for analysis with a Finnigan-MAT thermal ionization mass spectrometer

(TIMS). Oxygen and carbon sample preparation involves the drilling and isolation of enamel samples, with the resultant enamel powder cleaned with sodium hypochlorite, treated with acetic acid, and analyzed with a Finnigan Delta Plus IV mass spectrometer coupled with a Kiel III carbonate device. Statistical analyses include the non-parametric

Mann-Whitney U test, Monte Carlo analyses, and Levene’s test for equality of variances.

211

CHAPTER 6

RESULTS

Geological Setting of the United Arab Emirates

The United Arab Emirates may be partitioned into a variety of geologic zones: sabkhas, sand dunes, alluvial (wadi) fans, and the Hajar Mountains. Its western coastline is characterized by sabkhas, an Arabic transliteration for ‘salt flats,’ representing extremely flat Quaternary supratidal zones often located along arid coastal regions and comprised primarily of relatively pure carbonate sediment (Glennie 1998; Goudie et al.

2000). In the Emirates, these sabkhas consist of calcareous and gypsiferous sand and silt

(Government of the UAE Ministry of Petroleum and Mineral Resources 1979). While few modern studies have examined regional geomorphology in the United Arab Emirates

(Al Farraj and Harvey 2004), strontium isotope analysis of both sediment and water samples across the Emirate of Abu Dhabi by Muller and colleagues (1990) revealed the proportional contributions of strontium from both continental and coastal water origins to sabkha sediments. Muller et al. (1990) identified three sabkha sectors based on such Sr values, finding (1) Sr values largely of marine influence from the coast to as far inland as two kilometers, suggestive of periodic coastal recharge via flooding, (2) a combination of continental and marine influence on Sr between around 2.0-4.4 km inland, and (3) a

212 predominantly continental origin of Sr values beyond 4.4 km inland (Table 6.1). Marine zone Sr values fit generally with the expected modern seawater value of 0.70923 (Sealy et al. 1991), although a coastal water sample taken off the coast of Abu Dhabi possessed a slightly lower ratio at 0.709145 (Muller et al. 1990). Local modern carbonate values

Table 6.1. Strontium isotope values from modern water samples across Abu Dhabi, United Arab Emirates (adapted from data in Muller et al. 1990: 619).

Distance Sample Sabkha Sample Location from Coast 87Sr/86Sr Type Zone (km) Evans' Line AD 103 coast water +20.0 0.709145

AD 107 lagoon water +3.5 0.709126 Marine AD 166 inland water 0.3 0.709096 AD 165 inland water 0.3 0.709063 AD 180 inland water 0.7 0.709142 AD 170 inland water 0.7 0.709095 AD 171 inland water 0.7 0.709075 AD 174 inland water 1.4 0.709134

AD 178 inland water 2.5 0.709082 Mixed AD 183 inland water 3.3 0.709015 AD 159 inland water 4.0 0.709038

AD 163 inland water 4.4 0.708921 AD 131 inland water 6.4 0.708847 AD 126 inland water 7.8 0.708801

Continental AD 125 inland water 8.7 0.708853 AD 115 well water 20.0 0.708631 AD 112 well water 20.0 0.708780 AD 113 well water 20.0 0.708633 AD 122 oasis water 50.0 0.708744 Kinsman’s area

AD 114 LP 1 water 6.0 0.708844 AD 185 LP 1 water 6.0 0.708866

213 from coastal and lagoon sediments display similar 87Sr values ranging from 0.709093 to

0.709130 (Table 6.2) (Muller et al. 1990). The eastern shoreline is similarly characterized by sabkha and other Quarternary fluviatile and beach deposits, but is much narrower than its western counterpart, and more abruptly bordered by the Hajar

Mountains to the west.

Table 6.2. Modern carbonate sediments sampled off the coast of the Emirate of Abu Dhabi (adapted from Muller et al. 1990:620).

Sample Sediment Distance from Sample Location 87Sr/86Sr Type Type Coast (km) AD 13 lagoon sediment carbonate +3.5 0.709112 AD 14 lagoon sediment carbonate +3.5 0.709120 AD 16 coast sediment carbonate +20.0 0.709113 AD 18 lagoon sediment carbonate +3.5 0.709130 AD 20 lagoon sediment carbonate +3.5 0.709093

A northeastern extension of sand dunes from the Rub’ al-Khali, or Empty Quarter, of the Arabian Peninsula dominates much of the southern land mass of the Emirates

(Glennie 1998). Formed primarily during the Quaternary period and shaped by aeolian

(wind) processes, the sand dunes are interspersed with strips of desert plain deposits, small outcrops of calcareous sandstone, and minor exposed Tertiary period evaporite sequences, including sandy limestones, sandstones, and marl (Government of the UAE

Ministry of Petroleum and Mineral Resources 1979). These dunes also stretch into the northern Emirates, wedged between the coastal sabkhas to the west and the gravel fans of the Hajar Mountains to the east.

214 While presently, the Emirates lack active water channels, their former presence is evident based on extensive gravel deposits in the form of alluvial fans along the western front of the Hajar Mountains (Figure 6.1) (Al Farraj 1995; Parker and Goudie 2008).

Created though erosion by the drainage of multiple wadis from this mountain range,

Figure 6.1. The geomorphology of the northern Emirates (from Parker and Goudie 2008:460). Note the extensive alluvial fan belt extending from north to south along the western front of the Hajjar (Oman) Mountains. 215 gravel sediments comprised mainly of limestone, dolomite, and some chert characterize these Quaternary alluvial deposits, now home to modern rural villages and groves of planted date palm (Kennet 1997; Al Farraj and Harvey 2004; Parker et al. 2006).

Stretching some 200 km from north to south, the ages of those fans dated thus far range from 30,000 to over 400,000 years old, reflecting deposition as early as the Plio-

Pleistocene and into the Holocene (Sanlaville 1992; Glennie 1998; Juyal et al. 1998).

The Hajar Mountains along the eastern border of the Emirates and larger

Musandam Peninsula reflect a complex geologic history and represent the oldest rock formations in the country (Mery 1991). While generally dominated by limestones, this mountain chain may be roughly partitioned into northern and southern divisions (Figure

6.2) (Parker et al. 2006). In the north, grey limestone and dolomitic limestone form the

Musandam Group and date to the Cretaceous-Jurassic, although significant eastern permeations of Triassic (Elphinstone Group) and Triassic-Permian (Ruus Al Jibal Group) limestone, dolomitic limestone, dolomite, marl, shale, and chert exist (Government of the

UAE Ministry of Petroleum and Mineral Resources 1979; Mery 1991; Parker and Goudie

2008). The southern portion of these mountains is more variable, comprised largely of

Silutrian gabbros and ultrabasic rocks, but also with small projections of metamorphics

(including quartz and quartzite), limestone, dolomite, sandstone, chert, and granite

(Government of the UAE Ministry of Petroleum and Mineral Resources 1979; Parker and

Goudie 2008). While no settlements or funerary monuments have been unearthed from the Hajar Mountains, wadi drainage from these mountains into the Shimal Plain likely

216

Figure 6.2. Geologic map of the eastern United Arab Emirates, dominated by the Hajjar Mountains, which may be divided geologically into northern and southern portions (from Government of the UAE Ministry of Petroleum and Mineral Resources 1979).

217 influenced strontium isotope values by those living on the alluvial fans of the northern

Emirates, including those interred at Unar 1, Shimal 95, and Shimal 103.

Strontium Isotope Ratios of the United Arab Emirates

Although the geology and geomorphology of the United Arab Emirates are generally known, 87Sr/86Sr values cannot be determined without direct testing. Only one, geologically-limited study in Abu Dhabi (Muller et al. 1990) has been undertaken.

However, as discussed previously in Chapter 3, because Sr content can be extremely variable in a small geographic region, this Abu Dhabi study alone cannot be depended upon for interpreting human values and defining what it means to be ‘local’ in the United

Arab Emirates (Price et al. 2002). In lieu of relying on geologic 87Sr/86Sr values, biologically available strontium measured from fauna broadly reflects local isotope ratios because animals (particularly herbivores) typically consume a variety of plants over a wide area (Sealy et al. 1991; Sillen et al. 1998). Due to isotopic averaging effects during enamel formation, faunal 87Sr/86Sr ratio means consistently exhibit low standard deviations, making those animals living in the same environment as their human counterparts the ideal candidates for determining the bioavailability of local strontium in a given locale (Koch et al. 1992; Sillen et al. 1998; Price et al. 2002; Bentley et al. 2004;

Turner et al. 2009). After determining the mean 87Sr/86Sr ratio from faunal remains, a recommended ±2 s.d. from this mean is applied to define ‘local’ human values, while anything outside of these limits is considered nonlocal or immigrant (Grupe et al. 1997).

All fauna selected date to the Bronze Age in order to avoid complications associated with the utilization of strontium ratios from modern animals, including the

218 influence of fertilizers or imported diets (Böhlke and Horan 2000; Price et al. 2002;

Bentley and Knipper 2005). Faunal 87Sr/86Sr ratios from the United Arab Emirates are shown in Figures 6.3 and 6.4. Overlap of these ratios is present at all Emirati sites, although slight variations between sites speak to variable bioavailability within modern- day borders (Figure 6.3). Similar values are also seen when values are broken down by species (Figure 6.4). Ovicaprines (n=31), the largest faunal group sampled, display the greatest range of 87Sr/86Sr values from 0.708535 to 0.708973 and represent the sites of

Umm an-Nar Island, Dibba and Qidfa from the Emirate of Fujairah, Tell Abraq, and

Shimal. Cattle (n=6) samples from Umm an-Nar Island and Tell Abraq possess more

0.7100

0.7095

86 0.7090

Sr/ 87

0.7085

0.7080 Umm an-Nar Fujairah Tell Abraq Shimal 0 1 2 3 4 5 Island

Figure 6.3. Strontium isotope ratio ranges for Bronze Age fauna in the United Arab Emirates, used to define local ranges at each site.

219 0.7100

0.7095

86 0.7090

Sr/ 87

0.7085

0.7080 Cow Sheep/Goat Pig Oryx 0 1 2 3 4 5

Figure 6.4. Strontium isotope ratio ranges by species for Bronze Age fauna in the United Arab Emirates.

tightly constrained 87Sr/86Sr ratios ranging from 0.708752 to 0.708900, likely a result of similar geology at both western coastal sites, whereas sheep and goat originated from more diverse sites in the foothills as well as across the western and eastern coasts. A single pig tooth from Tell Abraq (87Sr/86Sr=0.708862) was also analyzed and sits in the upper portion of both ovicaprine and cattle 87Sr/86Sr value ranges. Finally, an oryx

(87Sr/86Sr =0.708775) tooth recovered from Umm an-Nar Island represents the only wild ungulate sampled in this study and falls neatly within mid-ranges shared by sheep/goat and cattle.

220 Mowaihat

At the Umm an-Nar site of Mowaihat in the Emirate of Ajman, no faunal remains were recovered in the rectangular Tomb B. However, because of their similar geologic setting and the close proximity of this tomb to that of Tell Abraq in the Emirate of

Sharjah (Ajman and Sharjah sit less than eight kilometers apart by modern coastal road today), fauna from Tell Abraq have been applied in defining local strontium isotope signatures for Mowaihat. At Tell Abraq, the average 87Sr/86Sr ratio of fauna (n=12: goat, sheep, cattle, and pig) is 0.708776  0.000057 (1), with a local range from 0.708661 to

0.708890 (Figure 6.5). These values are the most tightly clustered of all sites (UAE or otherwise) in this study, which may indicate that these domesticates (a) were kept locally and did not travel seasonally over great distances for grazing purposes, or (b) did travel over some unknown distances, but little geologic variability was present across these pasturelands.

87 86 The Sr/ Sr values from the Mowaihat human LM1 molars (n=13) range from

0.708582 to 0.708879 and have an average of 0.708838  0.000078 (1), displaying little variability as evident by this extremely low standard deviation (Figure 6.5). With one exception, all individuals fell into local ranges defined by the Tell Abraq fauna.

Interestingly, these 12 individuals closely approach or reach maximum local values.

Their high 87Sr/86Sr ratios relative to the local average may be indicative of a greater maritime influence during human enamel formation as a result of diet, as seawater possesses a worldwide average (0.70923) higher than the average expressed by terrestrial mammals analyzed here.

221 0.7100 M1 M2

0.7096 M3

0.7092

86 Sr/ 87 0.7088

0.7084

0.7080 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Figure 6.5. Strontium isotope ratios of individuals interred at Mowaihat, Emirate of Ajman.

While the vast majority of teeth examined in this study were isolated and not found in context with other, associated molars, six pairs of teeth were recovered from six individuals interred at Mowaihat, either found in situ in the mandible or in association with a fragmented skull. Of the individuals with multiple molars sampled, none display remarkable changes in 87Sr/86Sr ratios between first and second/third mandibular molars

(Figure 6.6). This suggests that these individuals were likely born in this area (as mandibular first molar crowns begin forming in utero, with completion at around three

222

tooth sampling at Mowaihat tooth Mowaihat six at for sampling individuals. - Inter

. Figure 6.6 Figure

223 years of age) and continued to live there at least into their mid-teens (when third molar crown completion typically occurs) (Hillson 1996). Only one individual (MW 197) out of 13 (8%) from the Mowaihat LM1 sample falls outside this locally defined range, with a

87Sr/86Sr value of 0.708582. While this immigrant does not fall far below the boundary of the minimum local range, deviating from the local average by -0.000194, its obvious difference with all other human ratios hugging the maximum values of the local range suggest a non-local origin. Subsequently, this individual does not appear to have resided in this area during childhood when enamel formation of the first mandibular molar took place.

Tell Abraq

At the Umm an-Nar site of Tell Abraq in the Emirate of Sharjah, dental enamel samples from 12 fauna were analyzed and produced a mean 87Sr/86Sr ratio of 0.708776 

0.000057 (1), with a local range from 0.708661 to 0.708890 (Figure 6.7).

87 86 The Sr/ Sr values of human LM1 enamel from the 29 individuals sampled at

Tell Abraq span from 0.708179 to 0.710661 and have an average of 0.708911  0.000361

(1), displaying more variability than at Mowaihat because of the presence of two individuals that are well outside the local range (Figure 6.7). Excluding these presumed non-locals, this variability is considerably reduced, with a mean 87Sr/86Sr ratio of

0.708873  0.000020 (1) and a greatly constricted range from 0.708820 to 0.708906.

Local adults (n=16) show a mean of 0.708866  0.000019 (1) with a range of 0.708820 to 0.708888, and all of these individuals possessed 87Sr/86Sr ratios consistent with local ranges set by fauna on site. In one adult individual with an intact mandibular fragment 224 0.7110 Adults Subadults 0.7105

0.7100

86 0.7095

Sr/ 87

225

0.7090

0.7085

0.7080 0 5 10 15 20 25 30

Figure 6.7. Strontium isotope ratios of subadult and adult individuals interred at Tell Abraq, Emirate of Sharjah. containing multiple teeth in situ, both LM1 and LM2 were sampled and displayed almost identical values of 0.708866 and 0.708869, respectively. Such similarity suggests that this individual lived in the area throughout early childhood (up to around seven years of age, when the second molar crown completion takes place).

As at Mowaihat, all local individuals (adult and subadult) from Tell Abraq display elevated 87Sr/86Sr ratios relative to the local average. This likely indicates the inclusion of more marine-based resources into human diet, resources with higher 87Sr/86Sr values that terrestrial domesticates like sheep and cattle did not consume with the same regularity (if at all) as the human inhabitants at the site.

Subadults (n=11) as well as adults were included in this analysis. With a mean of

0.708883  0.000015 (1) and a range from 0.708862 to 0.708906, these immature individuals possess 87Sr/86Sr values very similar to those of local adults; nevertheless, subadult values are generally elevated relative to adult ratios and are significantly different (U=203.50; z=2.05; p=0.04) (Figure 6.8). All subadult values adhere closely to the maximum local value of 0.708890, and while seven individuals fit within local ranges, four slightly exceed this delineation with differences up to +0.000015. As the portion of a population most likely uninvolved in long-distance movement and trade, these children are not interpreted as immigrants to Tell Abraq but as individuals with slightly more access to marine resources during enamel formation. It should be noted that while the trace element strontium does vary with breastfeeding and weaning in infant bone and teeth (relative to calcium), stable strontium isotope ratios do not undergo fractionation between trophic levels during the breastfeeding and weaning process and should thus not be considered as an influencing factor in this disparity.

226 0.7092

Adults

Subadults

0.7090

86 0.7088

Sr/ 87

227

0.7086

0.7084 0 5 10 15 20 25 30

Figure 6.8. Detailed scale showing local subadults and adults from Tell Abraq (non-locals excluded). Note the subtle differences in value scatters between age groups. Two out of a total of 18 adults (11%) at Tell Abraq exhibited 87Sr/86Sr ratios far beyond the local range. The first non-local (TA 161) has a 87Sr/86Sr value of 0.710661, deviating radically from local ranges and falling well above the local average

(+0.001885). With 87Sr/86Sr = 0.708179, the second immigrant individual (TA 165) diverges from the local average (-0.000597) in the opposite direction, indicating that these two foreigners spent at least part of their childhoods in regions geologically dissimilar to one another.

Umm an-Nar Island

At the Umm an-Nar settlement on Umm an-Nar Island off the coast of the

Emirate of Abu Dhabi, dental enamel samples from 15 fauna were analyzed and produced a mean 87Sr/86Sr ratio of 0.708788  0.000092 (1), with a local range from

0.708605 to 0.708971 (Figure 6.9). This standard deviation is somewhat larger than those previously seen at other sites along the west coast, including Mowaihat and Tell

Abraq, because of the increased variability in 87Sr/86Sr ratios among these ungulates.

Individuals interred in three separate tombs on Umm an-Nar Island were included in this

87 86 study. At Tomb I, the Sr/ Sr ratios of human RM1 enamel from four individuals produced a mean of 0.708893  0.000085 (1) and a range from 0.708795 to 0.709001

(Figure 6.9). An individual (UaN 122) with the highest 87Sr/86Sr ratio in Tomb I

(87Sr/86Sr= 0.709001) fell just above (+0.00003) the maximum value for the local range.

Inter-tooth sampling performed on the RM1 and RM3 from a single individual (UaN

123/124) displayed similar values at 0.708905 and 0.708926, respectively (Figure 6.10).

228

0.7100 M1 Tomb I M1 Tomb II M1 Tomb V M2 0.7095

Sr Sr

86 0.7090

Sr/ 87

229

0.7085

0.7080

Figure 6.9. Strontium isotope ratios of individuals interred in Tombs I, II, and V at Umm an-Nar Island, Emirate of Abu Dhabi. 87 86 At Tomb II, Sr/ Sr ratios for 15 individuals (LM1) range from 0.708649 to

0.708998 with a mean of 0.708870  0.000090 (1) (Figure 6.9). 87Sr/86Sr values from this tomb are by far the most variable on Umm an-Nar Island, with individuals scattered on either side of the local average. While differences between Tomb II and Tomb I

(U=66.00; z=0.41; p=0.65) are not statistically significant, Tombs II and V exhibit significant differences (U=319.50; z=2.31; p=0.02) in 87Sr/86Sr ratios as a result of this variability. This suggests that, while these individuals were still local from a regional perspective, they spent their childhoods in different localities within this region, each of which possessed minor differences in strontium bioavailability, because of slight disparities in either geologic makeup or possibly dietary influences.

Three of the 15 individuals in Tomb II were represented by mandibular fragments with in situ first and third molars that were subsequently tested to examine inter-tooth changes in 87Sr/86Sr ratios potentially indicative of migration later in life (Figure 6.10).

87Sr/86Sr values generally reflect local signatures for all three individuals and underwent little change during subsequent molar crown formation. Nonetheless, one of these individuals (UaN 128/129) experienced an increase in 87Sr/86Sr ratios over time that

87 86 resulted in a M3 Sr/ Sr value (0.709003) extending just over the local range maximum value (+0.000032).

Tomb II on Umm an-Nar Island contained two individuals (UaN 129 and 139) with 87Sr/86Sr values slightly above the maximum value defined by the local range at

0.709003 and 0.708998, respectively. As mentioned previously, one of these values

(UaN 129: 0.709003) is actually representative of a third molar, whereas the 87Sr/86Sr ratio of the first molar from that same individual fell within the local range. A second

230

129;

1 3 9 8 0 7 . 0

8 1 9 8 0 7 . 0 UaN 144/145 144/145 UaN

(UaN 125/126; 128/

6 2 7 8 0 7 . 0

9 4 6 8 0 7 . 0 Tomb UaN 130/131 130/131 UaN ,

3 0 0 9 0 7 . 0

(UaN 123/1 6 0 9 8 0 7 . 0 I M3 M3

UaN 128/129 128/129 UaN

M1 M1

9 8 0 7 . 0 8 0

8 8 0 7 . 0 6 9 . UaN 125/126 125/126 UaN

Nar Island Tomb from Nar Island

6 2 9 8 0 7 . 0 (UaN 144/145)

5 0 9 8 0 7 . 0 Umm an at ng UaN 123/124 123/124 UaN Tomb and and

, tooth sampli

0.7080 0.7090 0.7085 0.7100 0.7095

r S / r S

7 8 6 8 . Inter 130/131) Figure 6.10 Figure 231

individual (UaN 139) present in the tomb also exhibited a 87Sr/86Sr value (0.708998) just outside of values (+0.000027) for those expected to be local residents.

87 86 At Tomb V, 14 individuals (LM1) display a mean Sr/ Sr ratio of 0.708940 

0.000049 (1) and range in value from 0.708871 to 0.709035 (Figure 6.9). Values from this monument are not as variable as those individuals deposited in Tomb II, but instead cluster more closely around the upper reaches of the local range. Unlike Tombs V and II

(discussed previously), Tombs V and I show no statistically significant differences

(U=39.00; z=-1.14; p=0.24), with the majority of 87Sr/86Sr values from both tombs inclined towards the local maximum. One individual (UaN 144/145) was subjected to inter-tooth sampling of LM1 and LM3 and showed negligible change over time (Figure

6.10). Two individuals (UaN 143 and 150) had 87Sr/86Sr ratios of 0.709035 and

0.709029, just over the maximum local borderline by +0.000064 and +0.000058, respectively.

In total, five individuals from Tombs I, II, and V fall just above the value assigned as the maximum local range. However, their close proximity to other human values on the island warrants further consideration as to whether such individuals can be considered non-locals. Residence on the island undoubtedly influenced dietary practices, and it is likely that marine resources were consumed regularly and often. Because the strontium value of seawater (0.70923) is higher than local values determined by terrestrial domesticates, a diet dominated by fish and shellfish in humans would have produced elevated signals relative to terrestrially bioavailable strontium. Subsequently, these individuals may have simply ingested more sea-based foods than others living on the island. 232 Small shifts from a more terrestrial to a more marine diet with age are evident based on inter-tooth sampling, with 87Sr/86Sr ratios increasing in each individual from all three tombs (Figure 6.10). This trend, the opposite of which is seen at Tell Abraq, may reflect more limited dietary choices on the island where marine resources would have been especially prevalent. This same trend may also be a product of shifting geographic residence within the region, perhaps from the mainland to the island from youth to adulthood.

The isotopic variability seen on Umm an-Nar Island, particularly at Tomb II, suggests that either (a) individuals from a variety of local sites with slightly different geologic makeup may have come together to form the community at Umm an-Nar, or that (b) there was a great deal of dietary variability during crown formation of the first molars between individuals on the island (and see Chapter 7).

Shimal Plain: Unar 1, Shimal 95, and Shimal 103

At the Wadi Suq settlement at Shimal and the nearby Umm an-Nar tomb of Unar

1 in the Emirate of Ras al-Khaimah, 10 ovicaprine teeth were sampled and range in

87Sr/86Sr value from 0.708646 to 0.708973 with a mean ratio of 0.708788  0.000113

(1) (Figure 6.11).

The LM1 enamel of those interred within Umm an-Nar monumental tomb of Unar

1 (n=25) at Shimal has a mean 87Sr/86Sr ratio of 0.708805  0.000065 (1) with a range of 0.708737 to 0.709012 (Figure 6.11). All human values fall within the local range for

Shimal, with the majority of these hugging the 87Sr/86Sr local average. This distribution differs from the previous sites discussed here in that the 87Sr/86Sr ratios of the Unar 1 233 0.7100 M1-Shimal 95 M1-Shimal 103 M1-Unar 1

0.7095 M2

Sr Sr

86 0.7090

Sr/ 87

234

0.7085

0.7080 0 5 10 15 20 25 30 35

Figure 6.11. Strontium isotope ratios of individuals interred at the Umm an-Nar tomb of Unar 1 and the Wadi Suq tombs of Shimal 95 and 103, Emirate of Ras al-Khaimah. individuals congregate around the local average, and not the local maximum value, as is seen with other sites along the western United Arab Emirates. Such a disparity is likely not the result of geologic differences between these sites, as local averages and ranges of strontium bioavailability as defined by fauna are quite similar. Instead, this indicates that human 87Sr/86Sr values are very similar to those of terrestrial mammals, and therefore, that these humans probably consumed a more terrestrial-based diet that did not include as many marine resources as their counterparts on the coast and on Umm an-Nar Island.

Likely a product of its interior location at the foothills of the Hajjar Mountains, the rich alluvial soils and active wadis of the Shimal Plain would have provided ideal pasturelands for herds and allowed for the successful cultivation of palm gardens.

Only two individuals, RAK 221 (87Sr/86Sr=0.709012) and RAK 238

(87Sr/86Sr=0.708968), begin to approach 87Sr/86Sr ratios associated with seawater, and may thus represent a more mixed terrestrial-marine diet. A third individual (RAK 240) with a 87Sr/86Sr ratio of 0.708900 also draws closer to the local maximum value, but comes from a second molar, possibly suggesting an increase in the contribution of maritime resources to the diet with age for this particular person.

Both adults (n=19) and subadults (n=6) from the Unar 1 tomb were included in this study. Adult 87Sr/86Sr ratios range from 0.708750 to 0.708968 with a mean of

0.708805  0.000051 (1), while subadults have an identical mean 87Sr/86Sr value of

0.708805  0.000104 (1) and but a slightly larger range of 0.708783 to 0.709012

(Figure 6.12). Subadult values are not significantly different than adults (U=57.00; z=-1.30; p=0.18), indicating a similar geographic origin for individuals of both age categories. 235 0.7100 Adult Subadult

0.7095

86 0.7090

Sr/ 87

236

0.7085

0.7080

Figure 6.12. Strontium isotope ratios of subadult and adult individuals interred at Unar 1, Emirate of Ras al-Khaimah. Of these 25 individuals, three possessed in situ first and second left mandibular molars and were analyzed to discern any temporal shifts in 87Sr/86Sr value (Figure 6.13).

No major changes representing immigration from one geologic zone to another are evident between these molars, particularly with individuals RAK 226/277 and 233/234, each of whom display very little difference in 87Sr/86Sr ratios over time.

At the Wadi Suq tomb of Shimal 95, an analysis of the molars (RM1) of two individuals produces an average 87Sr/86Sr value of 0.708814  0.000024 (1) with a range of 0.708797 to 0.708831 (Figure 6.11). While two individuals are obviously not representative of the entire tomb population (MNI=15), both 87Sr/86Sr ratios fall within the local range and close to the local average, as at Unar 1.

0.7100

0.7095

86 0.7090

Sr/

0.7085

0.708750 0.708763 0.708820 0.708900

0.708814 0.708820 0.7080 RAK 226/227 RAK 333/334 RAK 239/240

LM1 LM2

Figure 6.13. Inter-tooth sampling for three individuals from Unar 1.

237 At Shimal 103, another tomb dating to the Wadi Suq period in Ras al-Khaimah, a

87 86 mean Sr/ Sr of 0.708828  0.000039 (1) was found for seven individuals (LM1), with an associated range of 87Sr/86Sr values from 0.708764 to 0.708880 (Figure 6.11). As with the other Shimal Plain tombs from this study, the 87Sr/86Sr ratios of those interred within

Shimal 103 are all contained within the locally defined range, and once again, these values tend to cluster around the local average.

Fujairah: Bidya 1, Dadna, Mereshid, Qidfa 4, and Dibba 76

At the Wadi Suq mortuary sites of Qidfa 4 (n=1) and Dibba 76 (n=1), sites along the eastern coast of the United Arab Emirates in the Emirate of Fujairah, two faunal teeth were recovered and exhibit a 87Sr/86Sr range from 0.708535 to 0.708686 with a mean ratio of 0.708611  0.000107 (1) (Figure 6.14). With only two individuals to set the local range for this area, caution must be taken when interpreting human 87Sr/86Sr values, as these two ovicaprines may not be representative of the entire spectra of strontium bioavailability in this environment.

Because of poor preservation and low MNI for each of these tombs, very few teeth (faunal or human) were recovered. However, due to the close proximity and similar geology of all eastern Emirati sites in this study, 87Sr/86Sr ratios from all Fujairah individuals will be considered together. Furthermore, a single individual from the Iron

Age site of Qidfa 4 is included as supplemental isotopic data indicative of local 87Sr/86Sr values. Finally, unlike previous scatter plots, inter-tooth sampling data for three individuals was plotted in Figure 6.14 in addition to the column chart of Figure 6.15 in order to better illustrate the placement of two of these outside local ranges. 238 0.7100 Bidya 1 Dadna 0.7095 Mereshid

Qidfa 4

Dibba 76 Sr Sr

86 0.7090

Sr/ 87

0.7085

0.7080 0 2 4 6 8 10 12

Figure 6.14. Strontium isotope ratios of individuals interred across the Emirate of Fujairah.

The individuals of Fujairah (n=8) display an average 87Sr/86Sr ratio of 0.708710 

0.000198 (1) and range in value from 0.708227 to 0.709055. Of these, two individuals

(25%) fall outside the local range, albeit at different times in life. Excluding these possible non-locals, the 87Sr/86Sr range is greatly reduced from 0.708640 to 0.708831 and exhibit a mean of 0.708725  0.000070 (1). Three individuals possessed two in situ molars within either maxillary or mandibular bone fragments, each of which was analyzed to examine changes in 87Sr/86Sr ratios over time (Figure 6.15). Interestingly, one of these three individuals possesses disparate inter-tooth 87Sr/86Sr values that suggest migration may have taken place. From Bidya 1, individual Bid 245/246 depicts an

239 0.7100

0.7095

0.7090 LM2 LM3 Sr 1

86 RM RM1 RM2 Sr/

87 0.7085

RM2

0.7080

0.708640 0.708674 0.708696 0.708798 0.708831 0.708227 0.7075 Bid 245/246 Qid 250/251 Dib 252/253

Figure 6.15. Inter-tooth sampling from Wadi Suq tombs across the Emirate of Fujairah. Note individual Bid 245/246, whose 87Sr/86Sr value shifts considerably with age.

intriguing pattern in which the first molar (0.708640) is contained with the local Fujairah range, but the second molar (0.708227) falls well below the local minimum value. This pattern is unusual in that the individual appears to have spent the first three years of life in the area, but then moved to a geologically different region shortly thereafter.

Nevertheless, this individual later returned to Fujairah and was buried there. A second non-local (Dib 254) from Dibba 76 was also identified, originating from an unknown region as evidenced by the elevated ratio of this first molar (0.709055) which not only falls outside of the local range of Fujairah but also above every local maximum value throughout the United Arab Emirates.

240 A summary of mean 87Sr/86Sr human values for all sites in the United Arab

Emirates is presented in Table 6.3. When comparing 87Sr/86Sr ratios between the Umm an-Nar and subsequent Wadi Suq periods, striking similarities among the bulk of samples is evident (Figure 6.16). The majority of individuals cluster together in value and display little variation, although a few outliers are present during both periods. The mean Wadi

Suq 87Sr/86Sr value (0.708761) is slightly lower relative to Umm an-Nar (0.708872).

Table 6.3. Mean 87Sr/86Sr ratios in human tooth enamel from Bronze Age sites in the United Arab Emirates. Parentheses refer to sample number after outliers are removed.

Site n 87Sr/86Sr 87Sr/86Sr (excluding outliers) Bidya 2 (1) 0.708666 0.708693 Dadna 1 0.708790 0.708790 Dibba 3 (2) 0.708869 0.708777 Mereshid 1 0.708646 0.708646 Mowaihat 13(12) 0.708838 0.708859 Qidfa 1 0.708674 0.708674 Shimal 95 2 0.708814 0.708814 Shimal 103 7 0.708828 0.708828 Tell Abraq 29(27) 0.708911 0.708873 Umm an-Nar Island 33 0.708902 0.708902 Unar 1 25 0.708805 0.708805

This difference of 0.000111 is more easily visualized in Figure 6.17. A statistically significant difference in means between the Umm an-Nar and Wadi Suq periods was also noted (p=0.04) using a Monte Carlo test (simulation=9999) at a 0.05 significance level when both locals and non-locals were included, and likewise, p<0.0001 when comparing only locals. Significant differences in variance were also recognized using Levene’s test

241 0.7110

0.7105

0.7100

86 0.7095

Sr/ 87 0.7090

0.7085

0.7080 0 Umm an1 -Nar Wadi2 Suq 3

Figure 6.16. Variance in strontium isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods.

between these periods for all individuals (p=0.02) as well as between only those falling within local ranges (p<0.0001).

Strontium Isotope Ratios of Comparative Regions Across the Persian Gulf

87Sr/86Sr ratios were also assessed for a variety of sites throughout the Persian

Gulf as a means of generating a regional isotopic map, useful for potentially determining where non-locals interred in the United Arab Emirates may have initially originated.

Sites were chosen based on known trade routes and interregional relationships during the

Bronze Age as evidenced by artifact distribution.

242 0.7092

0.7091

0.7090

0.7089

Sr 0.7088 86

Sr/ 0.7087 87 0.7086

0.7085

0.7084

0.7083 4-Jan 5-Jan Umm an-Nar Wadi Suq

Figure 6.17. Mean  1 s.d. in strontium isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods.

Bahrain

Two sites in Bahrain were examined: the A’ali Mound Field and Barbar Temple.

First, at the A’ali Mound Field, dental enamel samples from five fauna produce an average 87Sr/86Sr value of 0.708279  0.000141 (1) and possess a range of ratios from

0.708168 to 0.708523 and produce a local range of 0.707998 to 0.708560 (Figure 6.18).

Five human individuals were also analyzed from these burial mounds and show a mean

87Sr/86Sr ratio of 0.708252  0.000067 (1) and a range from 0.708175 to 0.708351

(Figure 6.18). All human values cluster around the local (faunal) average and fall neatly within local ranges.

243 0.7100

0.7095

0.7090

Sr Sr

86 Sr/

87 0.7085

0.7080

0.7075

Figure 6.18. Strontium isotope ratios of individuals interred at the A’ali Mound Field, Bahrain.

At the second Bahraini site of Barbar Temple, cattle (n=8) and sheep/goat (n=7) together generate a mean 87Sr/86Sr value of 0.708265  0.000157 (1), with ratios ranging from 0.708082 to 0.708572 and a local range of 0.707951 to 0.708578. No associated human remains were available for analysis. Because 87Sr/86Sr ratios from both sites were so similar, these data were combined to create local ranges for Bahrain.

Together, these sites possess an average 87Sr/86Sr value of 0.708268  0.000149 (1) and range in total from 0.708082 to 0.708572.

When compared with the local ranges from both the A’ali Mound Field and

Barbar Temple, individuals from all sites utilized in the United Arab Emirates show a

244 distinct 87Sr/86Sr isotopic separation (Figure 6.19). Overlap occurs in only two cases –

Tell Abraq and Bidya 1 – and interestingly, both of these individuals are classified as potential immigrants to the UAE, with 87Sr/86Sr ratios well below the local minimum value at each site. Each of these non-local individuals from Tell Abraq (TA 165) and

Bidya 1 (Bid 246) lands near the Bahraini local average. A third individual from

Mowaihat (MW 197) categorized as a non-local falls just above (+0.000015) the local maximum value for Bahrain.

Iran

At the inland site of Tepe Yahya in Iran, enamel from cattle (n=3), sheep/goat

(n=5), and pig (n=2) range in 87Sr/86Sr value from 0.708094 to 0.708652 and have a mean

87Sr/86Sr ratio of 0.708257  0.000164 (1) with a local range of 0.707928 to 0.708585.

No associated human remains were available for analysis.

Local individuals from all sites in the United Arab Emirates exhibit 87Sr/86Sr ratios distinct from those of Tepe Yahya (Figure 6.20). The local range from Iran is almost indistinguishable from the A’ali Mound Field and Barbar Temple sites of Bahrain, and as with the Bahrain range, the immigrants interred at Tell Abraq (TA 165) and Bidya

1 (Bid 246) are each contained within the local values of this Irani site and land near the

Tepe Yahya local average. This indicates that the geology and resulting bioavailable strontium of both Bahrain and Tepe Yahya are very much alike. Additionally, a third non-local from Mowaihat (MW 197) that fell outside of the Bahrain local range falls just within the maximum value set for Tepe Yahya.

245

Sh 95 Sh 103 Bidya 1 Dadna Mereshid 76 Dibba Qidfa 4 UaN Island I UaN Island II UaN Island V UaN Island Unar 1 Mowaihat Abraq Tell , organized chronologically to left from chronologically right, , organized . Bahrain tios of all individuals from the United Arab United the tios individuals of all from Emirates

anges from anges from

0.7085 0.7080 0.7075 0.7100 0.7095 0.7090 0.7110 0.7105

Sr Sr Sr/

86 87 . Strontium isotope ra

246 Figure 6.19 Figure with local r compared

Shimal 95 Shimal 103 Shimal Bidya 1 Dadna Mereshid 76 Dibba Qidfa 4 UaN Island I UaN Island II UaN Island V UaN Island Unar 1 Mowaihat Abraq Tell lly to left lly from right,

0.7080 0.7075 0.7100 0.7095 0.7090 0.7085 0.7110 0.7105

Sr Sr/

86 87 . Strontium isotope ratios of all individuals from the United Arab Emirates, organized chronologica United Arab the organized individuals. Strontium from of all Emirates, isotope ratios

247 Figure 6.20 Figure Tepe with local Yahya. ranges from compared

Kuwait

Dental enamel samples from cattle (n=7) and ovicaprines (n=8) display a mean

87Sr/86Sr ratio of 0.708645  0.000238 (1) with a range of values between 0.708392 to

0.709024 and a local ±2 s.d. range of 0.708169 to 0.709121. No associated human remains were available for analysis. Unlike previous comparative sites from Bahrain and

Iran, the fauna on Failaka Island demonstrate a sizeable standard deviation, generating a local 87Sr/86Sr range that encompasses every individual sampled from the United Arab

Emirates, with the exception of one immigrant interred at Tell Abraq (Figure 6.21). Such a range is indicative of considerable geologic diversity during the enamel formation of these fauna, which may point to a pattern of seasonal pastoralism in which these animals were exposed to rocks of varying age as they moved from one place to another.

Alternatively, these ungulates may have simply originated from different areas on the mainland of the Arabian Peninsula and were only later brought to the island settlement.

Pakistan

Faunal enamel from the Harappan coastal sites of Allahdino (n=10) and Balakot

(n=10) were analyzed and compared with those individuals interred in the United Arab

Emirates. Domestic ungulates from Allahdino produced a 87Sr/86Sr range from 0.708446 to 0.710417 and an average 87Sr/86Sr value of 0.708914  0.000554 (1), and as a result of this isotopic variability, generated a substantial local range (0.707806 to 0.710022) which contained all Emirati human samples except for a foreigner from Tell Abraq (TA

161) (Figure 6.22). In this region, rocks of considerable age (≥0.710), resulting from their originally high Rb/Sr ratios, were present and contributed to the elevated 87Sr/86Sr

248

Shimal 95 Shimal 103 Shimal Bidya 1 Dadna Mereshid 76 Dibba Qidfa 4 UaN Island I UaN Island II UaN Island V Island UaN Unar 1 Mowaihat Abraq Tell

. , Kuwait

0.7085 0.7080 0.7100 0.7095 0.7090 0.7110 0.7105

Sr Sr Sr/

86 87 . Strontium isotope ratios of all individuals from the United Arab Emirates, organized chronologically to left from chronologically right, United Arab the organized individuals. Strontium from of all Emirates, isotope ratios

249 Figure 6.21 Figure Failaka with local ranges from Island compared values of some of these fauna, particularly that of goat ALL 95 (87Sr/86Sr =0.710417).

Nevertheless, not all fauna from Pakistan possess such high ratios; like on Failaka Island, then, these fauna appear to have undergone enamel formation in somewhat diverse geologic locations, likely indicative of some type of seasonal herding, perhaps involving movement between the coast and the interior of the Indus Valley.

At Balakot, fauna have a mean 87Sr/86Sr ratio of 0.708980  0.000259 (1) with values ranging from 0.708745 to 0.709668 and a local range of 0.708463 to 0.709497

(Figure 6.22). Although considerably more limited in scope than Allahdino, this range is still extensive when compared to those of the Emirates, and encompasses all local individuals in the UAE, as well as the non-local from Mowaihat. These local values also exclude the two immigrants from Tell Abraq (TA 161 and 165) and the foreigner interred at Bidya 1 (Bid 246). No associated human remains were available for analysis at either site.

250

Shimal 95 Shimal 103 Shimal Bidya 1 Dadna Mereshid 76 Dibba Qidfa 4 UaN Island I UaN Island II UaN Island V UaN Island Unar 1 Mowaihat Abraq Tell ahdino (blue) and Balakot (red) Balakot (blue) and ahdino s of all individuals from the United Arab Emirates, organized chronologically to left from chronologically right, Arab United the organized individualss of all from Emirates,

0.7080 0.7075 0.7100 0.7095 0.7090 0.7085 0.7110 0.7105

Sr Sr Sr/

86 87 . Strontium isotope ratio

251

Figure 6.22 Figure All with local ranges from compared

Oxygen Isotope Ratios of the United Arab Emirates

As discussed in Chapter 3, ascertaining local ranges of 18O ratios for a given geographic region is particularly challenging given the myriad variables that influence fractionation and eventually, human 18O values, both environmental (e.g., evaporation, precipitation) and cultural (cooking, water storage, beverage preparation). Nevertheless, modern 18O precipitation maps can be useful as a broad indicator of expected 18O values available to past humans (Figure 6.23). Generally, high 18O ratios exemplify arid

18 regions like the Arabian Peninsula with little annual rainfall, and local  Odw(VSMOW) values ranging from approximately -4.0‰ to 0.0‰ were anticipated based on these

18 modern estimators. Average  Odw ratios for bottled water from a source in Abu Dhabi

(-1.9‰) confirm the elevated nature of local water values (Bowen et al. 2005).

Additionally, while faunal 18O ratios can provide some information about the bioavailability of 18O values in a specific environment, these cannot be used to define local or non-local ranges for human 18O values because animals (a) may be deriving

18O from different water and food sources than their human counterparts, (b) cannot store water in containers or process/cook food, which initiates additional fractionation processes that further alter 18O values in humans, and most importantly, (c) display significant variability in 18O due to body size, physiology and metabolism, diet, and drinking behavior that will overwhelm any local environmental 18O signatures useful for comparison with humans. Consequently, faunal 18O ratios are only briefly discussed below.

252

Figure 6.23. Stable oxygen isotope ratio (VSMOW) values of modern annual precipitation in Asia (from WaterIsotopes.org, accessed 2/7/2011). For a detailed map of the Arabian Peninsula, see Figure 6.27.

All fauna selected date to the Bronze Age in order to avoid complications associated with the consumption of modern water and other resources. Faunal teeth evaluated for oxygen isotopes were similarly analyzed for stable strontium and carbon isotopes. Faunal 18O ratios from the United Arab Emirates are shown in Figure 6.24 and

6.25. Overlap of these ratios is present between all Emirati sites, although considerably 253 12.0

10.0

8.0

6.0 O (‰) O

18 4.0 δ

2.0

0.0

-2.0 Umm an-Nar FujairahSite Name Tell Abraq Shimal Island

Figure 6.24. Oxygen isotope ratio ranges for Bronze Age fauna in the United Arab Emirates.

broad ranges of 18O values characterize each locale. At the settlement on Umm an-Nar

Island, fauna (n=15) possessed 18O values ranging from 1.6 to 10.2‰ with an average ratio of 4.8  2.2‰ (1). At the sites of Dibba and Qidfa from the Emirate of Fujairah, animal teeth (n=2) generated a mean 18O value of 1.9  1.8‰ (1), with ratios ranging from 0.6 to 3.1‰. 18O ratios from fauna (n=12) at Tell Abraq span from -1.2 to +4.9‰ and exhibit an average of 2.6  1.7‰ (1). Finally, at Shimal, which comprised both the

Wadi Suq Shimal settlement (n=9) and the Umm an-Nar tomb Unar 1 (n=1), a mean 18O value of 1.9  2.3‰ (1) with a range of -0.6 to +6.4‰ characterize the Bronze Age fauna. 254 12.0

10.0

8.0

) 6.0

‰ O ( O

18 4.0 δ

2.0

0.0

-2.0 Cow Sheep/GoatSpecies Pig Oryx

Figure 6.25. Oxygen isotope ratio ranges by species for Bronze Age fauna in the United Arab Emirates.

When values are broken down by species, different patterns emerge; however, caution must be taken when drawing conclusions based on individual fauna (e.g., pig, oryx) (Figure 6.25). The majority of faunal samples come from ovicaprines (n=31) representing the sites of Umm an-Nar Island, Dibba and Qidfa from the Emirate of

Fujairah, Tell Abraq, and Shimal, which show 18O ratios spanning 7.0‰, from -0.6 to

+6.4‰. Cattle (n=6) from Tell Abraq and Umm an-Nar Island possessed a range of 18O values from 1.9 to 8.3‰, representing a span similar to that of sheep and goat at 6.4‰ and suggesting a consumption of water similar sources. A single pig tooth from Tell

Abraq (18O = -1.2‰) was also analyzed, and although this ratio is comparable to a few

255 negative sheep/goat 18O values seen predominantly from Shimal, the pig displays the lowest value of all fauna measured and most approaches Bronze Age human values in this region. Conversely, from Umm an-Nar Island, enamel from a wild oryx displayed an

18O ratio of 10.2‰, higher above any other fauna sampled here.

Local human 18O values (n=131) across all Emirati sites range from -4.7 to

-0.3‰, with the majority (79%) falling between -3.0 to -1.0‰ (Figure 6.26). Using the

18 conversion equations outlined in Chapter 5, the  OVPDB ratios reported here were

18 18 converted to  Odw(VSMOW) values in order to (a) estimate  O ratios of drinking water available to Bronze Age populations in the Oman Peninsula, and (b) compare Bronze

18 Age drinking water values to those of modern-day precipitation ( Omw) using data

18 accessible by Waterisotopes.org (Figures 6.23 and 6.27).  OVPDB and conversion data

18 can be found in Appendix D. As discussed previously, modern  Omw ratios in this area

18 fall between -4.0 to 0.0‰; correspondingly, converted Bronze Age human  Odw values range from -7.2 to -1.5‰, suggesting that while overlap in oxygen isotope values exists between modern precipitation and Bronze Age drinking water, drinking water during the third and second millennia BC was somewhat low in 18O relative to the present,

18 particularly further inland at the site of Unar 1 on the Shimal Plain where  Odw ratios are lowest.

256 locals - across the United Arab Emirates.across Non the United Arab

human enamel Bronze Age Bronze

ratios from

in line with local range. in line local with

Ranges isotope of stable oxygen .

257 6.26 Figure squares isolated pictured as are 258

Figure 6.27. Stable oxygen isotope ratio (VSMOW) values of modern annual precipitation in Asia (from WaterIsotopes.org, accessed 2/7/2011). 19 Mowaihat

The 18O values from the Mowaihat human first mandibular left molars (n=13) range from -3.1 to -0.7‰ and have an average of -2.1  0.6‰ (1) (Figure 6.28). While

12 out of 13 individuals cluster between -3.1 and -1.8‰, a span of only 1.3‰, one individual (MW 197) possessed a value of -0.7‰, elevated relative to others interred at

Tomb B and suggestive of the utilization of a different, possibly more distant water source. While alone, using oxygen isotope data to identify MW 197 as a non-local is tentative, this 18O ratio, in conjunction with a deviant 87Sr/86Sr value belonging to the same person, indicates that this individual likely did not reside in the immediate area

0.0 M1 M2 -1.0

) -2.0

‰

O ( O 18

δ -3.0

-4.0

-5.0

Figure 6.28. Oxygen isotope ratios of individuals interred at Mowaihat, Emirate of Ajman. 259 during childhood when enamel formation of the first mandibular molar took place (Figure

6.29). The remaining 12 individuals closely match one another in both 87Sr/86Sr and 18O values, suggesting a local origin for the majority of those interred at Mowaihat.

6.0 5.0 Human Sheep 4.0 Goat 3.0

Cattle

2.0 Pig

1.0

O (‰) O 18

δ 0.0 -1.0 -2.0 -3.0 -4.0 0.7085 0.7086 0.7087 0.7088 0.7089 0.7090 87Sr/86Sr Figure 6.29. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from Mowaihat. Note the individual on the far left with non-local oxygen and strontium values.

Six individuals from Mowaihat were sampled to assess inter-tooth variation when multiple in situ teeth in the mandible were available (Table 6.4). Typically, 18O ratios in first molars are elevated by approximately 0.5 to 0.7‰ when compared to third molars, a product of breastfeeding during enamel formation (Wright and Schwarcz 1998). In all four M1/M3 pairs (MW 190/191; MW 192/193; MW 202/203; MW 204/205), M1s

260

Table 6.4. Inter-tooth 18O sampling at Mowaihat for six individuals.

Individual Site Tooth δ18O (‰) ∆δ18O

MW LM -2.0 Mowaihat 1 -0.7 190/191 LM3 -2.7 MW LM -2.0 Mowaihat 1 -0.3 192/193 LM3 -2.3 MW LM -2.0 Mowaihat 1 -0.4 195/196 LM2 -2.4 MW LM -3.1 Mowaihat 1 +0.2 200/201 LM2 -2.9 MW LM -2.6 Mowaihat 1 -0.4 202/203 LM3 -3.0 MW LM -1.8 Mowaihat 1 -0.8 204/205 LM3 -2.6

possessed 18O values elevated by 0.3 to 0.8‰ relative to M3s, with a mean increase of

+0.6‰. For individual MW 195/196, a difference of 0.4‰ was noted between M1 and

M2, suggestive of a declining 18O ratio as environmental sources of food and water were gradually introduced into the diet during weaning. This slightly smaller difference is expected, given that M1 and M2 values, and not M1 and M3, are being measured. Only one individual (MW 200/201) displays the opposite pattern, with its 18O ratio actually increasing from M1 to M2. This atypical configuration is possibly a product of the early introduction of environmental water sources to this individual during enamel formation of the first molar, or of some other kind of dietary influence overriding characteristics

18O weaning trends. No 18O differences between M1s and M2/M3s are large enough to

261 suggest substantial changes in water sources associated with movement from one area to another during enamel formation.

Tell Abraq

18 The  O values of human molar (LM1) enamel from the 29 individuals sampled at Tell Abraq span from -6.0 to -1.2‰ and have an average of -2.2  0.9‰ (1) (Figure

6.30). This range displays considerably more variability than at Mowaihat because of a single outlier (TA 165); after removing this potential non-local, such variability is reduced, with a mean 18O ratio of -2.1  0.5‰ (1) and a more concentrated range of -

3.2 to -1.2‰. Local adults (n= 17) exhibit an average of -2.2  0.5‰ (1) and range in value from -3.2 to -1.2‰. In one adult (TA 186/187) with an intact mandibular fragment containing multiple teeth in situ, both LM1 and LM2 enamel samples displayed similar values of -2.5‰ and -2.4‰, respectively (Table 6.5). This suggests that the individual lived in the area until at least seven years of age with the completion of M2 crown, although the relatively lower 18O ratios associated with weaning are not present.

In addition to adults, subadults (n=11) were included in this analysis. Subadults range in 18O value from -2.8 to -1.3‰ and exhibit a mean of -1.9  0.5‰ (1), similar to but elevated relative to the local adult mean of -2.2‰, and are not significantly different (U=193.50; z=1.58; p=0.11) from adult ratios (Figures 6.30 and 6.32). One adult out of a total of 19 (5%) at Tell Abraq exhibited a 18O ratio well beyond the cluster generated by the remaining adult individuals. This outlier (TA 165) has a 18O value of -

6.0‰, deviating from the local adult mean by 3.8‰. The assignment of this individual to

262 0.0

-1.0

-2.0

-3.0

-4.0

O (‰) O

18 δ

263 -5.0

-6.0

-7.0 Adult Subadult -8.0

Figure 6.30. Oxygen isotope ratios of adult and subadult individuals interred at Tell Abraq. Table 6.5. Inter-tooth 18O sampling at Tell Abraq for one individual.

Individual Site Tooth δ18O (‰) ∆δ18O

LM -2.5 TA 186/187 Tell Abraq 1 +0.1 LM2 -2.4

non-local status is further confirmed when 18O values are plotted against associated

87Sr/86Sr ratios from the site (Figures 6.30 and 6.31). However, a second individual (TA

161) identified as an immigrant by strontium does not deviate from local 18O values.

This does not exclude the possibility of a non-local origin for this individual, particularly

6.0

4.0

2.0

0.0

Human Adult O (‰) O

18 -2.0

δ Human Subadult Sheep -4.0 Goat Cattle -6.0 Pig -8.0 0.7080 0.7085 0.7090 0.7095 0.7100 0.7105 0.7110 87 86 Sr/ Sr

Figure 6.31. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from Tell Abraq.

264 0.0

Human Adult Human Subadult -1.0

Pig

-2.0

O (‰) O

18 δ

-3.0

-4.0 0.7085 0.7086 0.7087 0.7088 0.7089 0.7090 87 86 Sr/ Sr

Figure 6.32. Detailed scale of strontium-oxygen bivariate scatter plot showing local subadults and adults from Tell Abraq (non-locals excluded). Note the subtle but statistically significant differences in value scatters between age groups in strontium, but not oxygen.

as the 87Sr/86Sr ratio diverges radically from the local range; instead, it simply suggests that water sources in the region where this individual underwent enamel formation possessed similar 18O values to those of the United Arab Emirates.

Umm an-Nar Island

Individuals from three different tombs (I, II, and V) on Umm an-Nar Island were also included for stable oxygen isotope analysis. In Tomb I, the 18O ratios of human enamel (RM1) from four individuals produced a mean of -1.9  0.1‰ (1) and a range

265 18 from -2.0 to -1.8‰ (Figure 6.33).  O ratios for 15 individuals (LM1) from Tombs II and V are more variable (likely a result of larger samples sizes), with Tomb II values spanning from -2.7 to -1.1‰ with an average of -2.3  0.4‰ (1) and Tomb V ratios ranging from -3.1 to -1.1‰ with a mean of -2.3  0.6‰ (1) (Figure 6.33). While differences between Tombs II and V (U=253.00; z=-0.05; p=0.94) are statistically significant, Tombs I and II (U=86.00; z=1.90; p=0.05) do not exhibit significant differences in 18O, contrary to strontium results (p=0.02 and p=0.65, respectively).

Despite this reversed pattern, 18O ranges from all tombs on this island are rather tightly constricted; the isotopic variability present is thus likely a result of local 18O variability present in water sources. It is still possible that individuals from different localities within this region came together on the island, as suggested by the variation in 87Sr/86Sr values, but based on both strontium and oxygen isotopic ratios, no non-locals appear in any of these island funerary structures.

Across all three tombs, five individuals (UaN 122, 129, 139, 143, 150) possessed

87Sr/86Sr values that fell slightly above the maximum value defined by the local range.

Based on strontium ratios alone, these individuals were not considered non-locals because (a) of the close proximity of their 87Sr/86Sr values to other human values on the island, and (b) the regular consumption of marine resources during enamel formation would have produced elevated 87Sr/86Sr ratios relative to terrestrially bioavailable strontium. This conclusion is further supported by 18O data from the same individuals, all of whom exhibit similar values (U=81.00; z=0.48; p=0.61), and does not suggest the presence of immigrants to Umm an-Nar Island (Figure 6.34).

266 0.0 Tomb I (M1) Tomb II (M1) Tomb V (M1) -1.0 M2 M3

-2.0

O (‰) O

18 δ -3.0

267

-4.0

-5.0

Figure 6.33. Oxygen isotope ratios of individuals interred at Umm an-Nar Island.

12.0 Human (Tomb I) 10.0 Human (Tomb II) Human (Tomb V) 8.0 Cattle

6.0 Sheep/Goat Oryx

4.0

O (‰) O 18 δ 2.0

0.0

-2.0

-4.0 0.7085 0.7086 0.7087 0.7088 0.7089 0.7090 0.7091 87 86 Sr/ Sr

Figure 6.34. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from Umm an-Nar Island. Note the variability in 87Sr/86Sr but not 18O values.

Inter-tooth sampling was performed on five individuals from the three Umm an-

Nar Island tombs (Table 6.6). In four of these five adults, a comparison of temporal changes in 18O between M1 and M3 was made. Three of these individuals (UaN

123/124, 128/129, 130/131) show elevated 18O ratios in M1s relative to M3s ranging from +0.4 to 0.9‰, with a mean change of +0.7‰, falling within the 0.5 to 0.7‰ elevation expected as a result of breastfeeding. Overall, M1s display an average 18O of -

1.6  0.4‰ (1) and M3s a mean 18O of -2.3  0.3‰ (1). On the other hand, the

268 Table 6.6. Inter-tooth 18O sampling at Umm an-Nar Island for five individuals.

Individual Tomb Tooth δ18O (‰) ∆δ18O

UaN Umm an-Nar RM1 -1.9 -0.4 123/124 Island I RM3 -2.3 UaN Umm an-Nar LM -2.2 1 -1.0 125/126 Island II LM2 -3.2 UaN Umm an-Nar LM -1.8 1 -0.7 128/129 Island II LM3 -2.5 UaN Umm an-Nar LM -1.1 1 -0.9 130/131 Island II LM3 -2.0 UaN Umm an-Nar LM -2.3 1 +0.9 144/145 Island V LM3 -1.4

fourth individual (UaN 144/145) demonstrates a conflicting pattern, initially displaying a low 18O ratio that increases over time. The variability present in this individual cannot be accounted for by breastfeeding and weaning, and must be attributed to some other overriding dietary influence. In all four individuals, then, a mean M1 18O value of -

1.8‰  0.5‰ (1) and of -2.1  0.5‰ (1) for M3s produce an average increase of only

0.3‰. Finally, the M1 and M2 of a fifth individual were also sampled, and while the anticipated pattern of 18O enrichment in the M1 is present, a difference of 1.0‰ between these two teeth suggests a relatively rapid change in sources of imbibed water.

In summary, the strontium isotopic variability seen on Umm an-Nar Island, particularly for individuals interred in Tomb II, is not as evident in 18O ratios of the same individuals, suggesting that these individuals originated from a similar geographic area and obtained water from sources comparable in 18O value.

269 Shimal: Unar 1, Shimal 95, and Shimal 103

On the Shimal Plain, human dental enamel from the third millennium BC tomb of

Unar 1 and the second millennium tombs of Shimal 95 and 103 was examined

18 isotopically. At Unar 1, 25 individuals (LM1) have an average  O ratio of -4.1  2.3‰

(1) with a range of -12.2 to -2.5‰ (Figure 6.35). The considerable variability present within this tomb is primarily a consequence of two outliers (RAK 220 and 224) extremely depleted in 18O relative to the majority of individuals sampled. However, these two values cannot be considered because both teeth had undergone burning to various degrees – RAK 220 was in a calcined state, whereas RAK 224 was charred.

While burning will not affect radiogenic strontium (Knudson and Tung 2007) or stable carbon (Schurr 2008; Schurr and Hayes 2008) isotopes, stable oxygen isotopes are susceptible to changes associated with temperatures over 300C (Munro et al. 2007;

Knudson, personal communication 6/9/2011).

Excluding these individuals, a more tightly constricted local range of -4.7 to

-2.5‰ with a mean of -3.5  0.6‰ (1) is generated. Overall, 18O values at Unar 1 are more negative than from individuals interred along both the western and eastern coasts of the Oman Peninsula, providing some isotopic differentiation between the coast and the foothills of Shimal during the Umm an-Nar period. Statistically significant differences are present between local values from Unar 1 and the Umm an-Nar tombs at Mowaihat

(U=345.50; z=4.49 ; p<0.0001 ), Tell Abraq (U=292.00; z=-5.79; p<0.0001), and Umm an-Nar Island Tomb I (U=102.00; z=3.11; p=0.002), Tomb II (U=456.50; z=4.89; p<0.0001), and Tomb V (U=406.50; z=4.39; p<0.0001). Even when compared with

270 0.0

-2.0

-4.0

) -6.0

‰

O ( O 18

δ -8.0

271

-10.0 M1-Unar 1 M1-Shimal 95 -12.0 M1-Shimal 103 M2 -14.0

Figure 6.35. Oxygen isotope ratios of individuals interred at the Umm an-Nar tomb of Unar 1 and the Wadi Suq tombs of Shimal 95 and 103, Emirate of Ras al-Khaimah. Note the two outliers from Unar 1 in the lower left corner, a result of diagenetically altered enamel and not from immigration. locals from the Wadi Suq tombs sharing the Shimal Plain, a similar pattern emerges. As at other Umm an-Nar tombs in the United Arab Emirates, 18O values from the Umm an-

Nar tomb at Unar 1 differ significantly from those of the later tomb of Shimal 103

(U=142.50; z=2.80; p=0.005). Interestingly, the Wadi Suq tomb of Shimal 95 approaches but does not statistically differ from Unar 1 (U=43.50; z=1.71; p=0.08), although this may largely be due to its small sample size (n=2). Finally, the Wadi Suq tombs along the eastern coast of Fujairah also exhibit results statistically distinct

(U=300.50; z=3.96; p<0.0001) from that of Unar 1. Consequently, while all individuals from Unar 1 appear to form a local 18O cluster, these values are nevertheless distinctly depleted in 18O from the remainder of sites in the United Arab Emirates from both the

Umm an-Nar and Wadi Suq periods and suggest a unique source of water for its Umm an-Nar inhabitants.

Slightly elevated 87Sr/86Sr ratios in three individuals (RAK 221, 238, 240) from

Unar 1 were previously noted. These were not interpreted as non-locals but as possibly indicative of the consumption of a higher proportion of marine foods during enamel formation. Corresponding 18O values show no indication of non-local water sources and support the assertion that these individuals’ 18O ratios represent a product of local intra- site variability (Figure 6.36).

LM1 enamel from both adult (n=19) and subadult (n=4) individuals interred at

Unar 1 was examined (Figure 6.37). Six subadults were initially analyzed in total; however, two of these teeth (RAK 220 and 224) were burned, and because stable oxygen isotope ratios are altered by high temperatures, these samples were excluded from this

272 8.0 Unar 1 6.0 Shimal 95 Shimal 103 4.0 Sheep/Goat

2.0

O 18 δ 0.0

-2.0

-4.0

-6.0 0.7086 0.7087 0.7088 0.7089 0.7090 0.7091 87Sr/86Sr

Figure 6.36. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from the Shimal Plain, Ras al-Khaimah. Note a single, 18O-enriched outlier from Shimal 103.

summary. Adults exhibit a range of -4.7 to -2.5‰ with a mean 18O value of -3.5 

0.6‰ (1), while subadults display a range from -3.8 to -2.7‰ and a similar mean to that of adults at -3.3  0.4‰ (1) (Figure 6.37). These age groups are not statistically different from one another (U=57.00; z=0.69; p=0.47), indicative of a similar, local origin.

Of these 25 individuals, three possessed in situ first and second left mandibular molars and were analyzed to discern any temporal shifts in 18O value (Table 6.7). Two of these (RAK 233/234 and 239/240) display elevated M1 18O values by 0.8 and

273 0.0 Adult -1.0 Subadult -2.0

-3.0

-4.0

-5.0 O (‰) O

274

18 

-6.0

-7.0

-8.0

-9.0

-10.0

Figure 6.37. Oxygen isotope ratios for subadults and adults from the tomb of Unar 1, Emirate of Ras al-Khaimah. Table 6.7. Inter-tooth 18O sampling at Unar 1 for three individuals.

Individual Site Tooth δ18O (‰) ∆δ18O

RAK LM -3.5 Unar 1 1 +0.1 226/227 LM2 -3.4 RAK LM -2.8 Unar 1 1 -0.8 233/234 LM2 -3.6 RAK LM -3.0 Unar 1 1 -1.1 239/240 LM2 -4.1

1.1‰, respectively, a slightly larger divergence than predicted by the 0.5 to 0.7‰ between M1s and M3s. A third individual (RAK 226/227) shows 18O-enrichment of the second molar by +0.1‰ instead of the first. Overall, average 18O ratios for M1s fall at

-3.1  0.4‰ (1), while M2 18O values exhibit a mean of -3.7  0.4‰ (1).

Subsequently, 18O ratios are elevated by +0.6‰ and fit well with documented changes in oxygen isotope values for M1/M3s; however, these measures were conducted for

M1/M2s. This, along with the pattern seen in individual RAK 226/227, suggest some influence by environmental water resources superseding expected 18O breastfeeding and weaning trends.

At the Wadi Suq tomb of Shimal 95, an analysis of right mandibular first molars

(n=2) produces an average 18O value of -2.6  0.5‰ (1) with a range of -3.0 to -2.2‰

(Figure 6.35). While only two individuals were available for sampling out of an estimated MNI of 15 for the tomb, both 18O ratios fall well within the ranges shown at other tombs across the western Oman Peninsula and, as with strontium, appear to be of local origin.

275 Also dating to the Wadi Suq, the nearby tomb of Shimal 103 produced a mean of

18 18  O of -2.3  1.0‰ (1) for seven individuals (LM1), with an associated range of  O values from -3.2 to -0.3‰ (Figure 6.35). However, a single outlier (SH 214) possesses a value of -0.3‰, considerably elevated relative to the local range produced by all three

Shimal tombs (n=31; -4.7 to -2.2‰). While the associated 87Sr/86Sr ratio (0.708828) for this individual appears local, the 18O value suggests that during enamel formation, this individual imbibed water sources with a considerably different 18O signature than the remainder of those buried at Shimal 103. After removing this non-local, the 18O mean for the tomb was -2.7  0.4‰ (1) with a more contracted range of -3.2 to -2.2‰.

Despite its proximity to Unar 1, the Shimal 103 tomb exhibits 18O values significantly enriched in 18O (U=142.50; z=2.80; p=0.005) relative to its Umm an-Nar counterpart.

This difference may be the result of either (a) the gradual desiccation of the area, leading to more positive 18O values in this later period despite exploiting similar water sources, or (b) the utilization of isotopically dissimilar water sources between the Early and

Middle Bronze Age on the Shimal Plain.

In summary, the 18O ratios of those interred within Shimal 103 as well as Shimal

95 and Unar 1 generally cluster within a range that appears local, with the exception of a single individual possessing an elevated 18O value suggestive of the possible presence of a non-local immigrant for this time and area. Associated strontium ratios all display values falling into a local range, corroborating the likelihood that the majority of these individuals originated from this area.

276

Fujairah: Bidya 1, Dadna, Mereshid, Qidfa 4, and Dibba 76

As with strontium, all samples (n=8) from the eastern Emirati sites of Fujairah will be considered together, primarily due to low sample sizes but also because of similar environmental conditions amongst these five sites. The individuals interred in Fujairah tombs display an average 18O value of -2.4  0.5‰ (1) and range in value from -3.5 to

-1.8‰ (Figure 6.38). 87Sr/86Sr ratios from two individuals (Bid 246 and Dib 254) fell outside locally defined strontium ranges for the area, indicative of the presence of non- locals in the Bidya 1 and Dibba 76 tombs, but associated 18O values for the same individuals do not suggest a non-local origin. Still, this may simply point to similar

0.0

-1.0

-2.0 Bidya 1

Dadna O (‰) (‰) O

18 Mereshid δ -3.0 Qidfa 4 Dibba 76 -4.0

-5.0

Figure 6.38. Oxygen isotope ratios of individuals interred in the Emirate of Fujairah. 277 hydrological conditions producing comparable 18O values across different regions. The relative homogeneity of 18O ratios across the Wadi Suq tombs of Fujairah is evident when plotted against corresponding strontium values, which exhibit considerably more variability (Figure 6.39). This bivariate Sr/O plot continues to highlight the two apparently non-local individuals interred in Fujairah during the Middle Bronze Age.

Three individuals possessed two in situ molars within either maxillary or mandibular bone fragments, each of which was analyzed to examine temporal changes in

18O ratios (Table 6.8). Interestingly, two of these three individuals possess an M1 (Bid

245/246) or an M2 (Dib 252/253) with low 18O values relative to enamel forming

4 Human (Bidya 1) 3 Human (Dadna) Human (Mereshid) 2 Human (Dibba 76)

1 Human (Qidfa 4) Sheep/Goat

O (‰) O 18 δ -1

-2

-3

-4 0.7080 0.7082 0.7084 0.7086 0.7088 0.7090 0.7092 87Sr/86Sr

Figure 6.39. Bivariate plot of strontium and oxygen isotope ratios for human and faunal enamel from the Emirate of Fujairah.

278 Table 6.8. Inter-tooth 18O sampling at Bidya, Qidfa and Dibba in the Emirate of Fujairah.

Individual Site Tooth δ18O (‰) ∆δ18O

RM1 -1.9 Bid 245/246 Bidya 1 +0.1 RM2 -1.8 RM -2.3 Qid 250/251 Qidfa 4 1 -0.1 RM2 -2.4 LM -2.7 Dib 252/253 Dibba 76 2 +0.2 LM3 -2.5

later in life, a pattern contrary to that expected as a result of breastfeeding. While one individual (Qid 250/251) does possess an M1 with a slightly elevated 18O ratio when compared to its M2 counterpart, this 18O-enrichment (+0.1‰) is minimal. Overall, a mean 18O ratio of -2.1  0.3‰ (1) for M1s, -2.3  0.5‰ (1) for M2s, and -2.5 for a single M3 shows little change in 18O over time.

A lack of oxygen isotope fractionation associated with breastfeeding may in part be masked by residential mobility evident in two of these individuals based on strontium ratios. For instance, the individual from Bidya 1 displays a somewhat atypical strontium pattern in which the first molar is included within the locally defined range, but by the time enamel formation of the second molar has taken place, this individual displays non- local values signifying some geographic change in residence undertaken during early childhood. Nevertheless, 18O values between these two teeth differ by only 0.1‰. It is possible that the 0.5-0.7‰ change expected from trophic level effects of breastfeeding has been subsequently masked by different 18O ratios present in this new geographic area where this individual moved.

279 A summary of mean 18O human values for all sites in the United Arab Emirates is presented in Table 6.9. When comparing 18O ratios between the Umm an-Nar and subsequent Wadi Suq periods, the overlap between the Early and Middle Bronze Age is apparent (Figure 6.40). The majority of individual values during both periods clusters together and displays little variation, although as with strontium, a few deviant ratios are present. The mean Wadi Suq 18O value of -2.4  0.7‰ (1) is only slightly elevated relative to the Umm an-Nar average of -2.5  0.8‰ (1). This mean difference of only

0.1‰, as well as overall variance, is more clearly illustrated in Figure 6.41. When analyzed using the Monte Carlo method (simulation=9999), these means are not statistically significant (p=0.36 for all individuals; p=0.92 for locals only) between these two time periods. In addition, no significant differences in variance were recognized using Levene’s test between these periods for all individuals (p=0.36) as well as between only locals (p=0.91).

Table 6.9. Mean 18O ratios in human tooth enamel from Bronze Age sites in the United Arab Emirates. Parentheses refer to sample number after outliers are removed.

Site n δ18O (‰) δ18O (excluding outliers) Bidya 2 -2.3 -2.3 Dadna 1 -2.2 -2.2 Dibba 3 -2.4 -2.4 Mereshid 1 -3.5 -3.5 Mowaihat 13 (12) -2.1 -2.2 Qidfa 1 -2.3 -2.3 Shimal 95 2 -2.6 -2.6 Shimal 103 7 (6) -2.3 -2.7 Tell Abraq 29 (28) -2.2 -2.1 Umm an-Nar Island 33 -2.3 -2.3 Unar 1 25 (23) -4.1 -3.5 280

0.0

-1.0

-2.0

-3.0 O (‰) O

18 -4.0 

-5.0

-6.0

-7.0 0 Umm1 an-Nar Wadi2 Suq 3 Figure 6.40. Variance in oxygen isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods.

Oxygen Isotope Ratios of Comparative Regions Across the Persian Gulf

A preliminary analysis of human 18O ratios was also conducted for two sites outside the United Arab Emirates as a means of generating a regional isotopic map, useful for potentially determining where non-locals interred in the United Arab Emirates may have initially originated. Unfortunately, for the majority of comparative sites utilized in this study, samples consisted of faunal, and not human, dental enamel, as no associated human remains were available for analysis. While such comparative data is extremely useful for defining local strontium ranges and for reconstructions of trophic

281 0.0

-1.0

-2.0

O (‰) O 18

 -3.0

-4.0

-5.0 4-Jan 5-Jan Umm an-Nar Wadi Suq

Figure 6.41. Mean  1 s.d. in oxygen isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods.

level systems using stable carbon isotopes, its applicability to oxygen isotope analysis is limited because of major metabolic differences between humans and animals, in which oxygen isotope fractionation is not well understood.

A’ali Mound Field, Bahrain

Two sites in Bahrain were examined as part of this study: the A’ali Mound Field and Barbar Temple. Of these, only the A’ali Mound Field yielded 18O ratios from six human molars (LM1) as well as four ovicaprines. Fauna produce a mean 18O value of

4.7  5.8‰ (1) and possess an expansive range from -1.2 to +11.8‰. Human

282 individuals from these burial mounds yielded a much more constricted 18O range of -3.4 to -2.2‰ and an average of -2.9  0.5‰ (1). Bahraini 18O ratios fall well within the range defined by individuals interred in the United Arab Emirates (Figure 6.42). These values are not statistically different (U=278.00; z=-1.47; p=0.14) from those found in the

UAE and are thus not useful in distinguishing migrants from this island.

2.0

1.0

0.0

-1.0

-2.0 O (‰) O

18 -3.0 δ -4.0

-5.0

-6.0

-7.0 0 1 2 3 UAE Bahrain

Figure 6.42. Oxygen isotope ratios of all human individuals from the United Arab Emirates compared with those from the A’ali Mound Field in Bahrain.

Al- Khubayb, Oman

Preliminary excavations of Hafit and transitional tombs from the al-Khubayb necropolis of Oman made available three human left mandibular molars for analysis.

283 Results from this dental enamel include a mean 18O value of 4.2  2.3‰ (1) and a range of 2.8 to 6.8‰. Unlike human values from the A’ali Mound Field of Bahrain,

Omani 18O ratios are considerably higher relative to values from the UAE, and no overlap is present (Figure 6.43). Unsurprisingly, these values differ significantly

(U=405.00; z=2.95; p=0.003) from those found in the UAE, and are henceforth considered useful in distinguishing individuals from the eastern and western portions of the Oman Peninsula. Based on this information, no other individuals interred in the UAE but identified as non-locals appear to have originated from this region of Oman.

8.0

6.0

4.0

2.0

0.0

O (‰) O 18 δ -2.0

-4.0

-6.0

-8.0 0 UAE1 Oman2 3

Figure 6.43. Oxygen isotope ratios of all human individuals from the United Arab Emirates compared with those from al-Khubayb in Oman.

284 Carbon Isotope Ratios of the United Arab Emirates

Unlike stable oxygen isotopes, in which the 18O values of animals cannot be utilized as a proxy from which human ratios can be compared, plant and faunal 13C values can provide crucial information regarding trophic level and human diet within a given environment. Lake sediment sequences from the Emirate of Ras al-Khaimah in the

United Arab Emirates reveal a history of the environmental transition from an area dominated by C3 grasses and woodlands to mixed C3-C4 vegetation at the beginning of the sixth millennium BC (Parker et al. 2004). By 4100 BP, increasingly arid conditions encouraged C4 vegetation to become the dominant photosynthetic pathway in the Oman

Peninsula (Parker et al. 2004). An analysis of herbivore enamel from the Late Miocene

Baynunah Formation in the Emirate of Abu Dhabi correspondingly shows that while both

C3 and C4 plants comprised the vegetation of the area, fauna primarily consumed C4 grasses (Kingston 1999). The presence of such grazers – fauna relying on C4 plants – has later been recorded in northern Saudi Arabia during the Pleistocene (Thomas et al. 1998).

A modern floral inventory of grasses present in the Rub’ al-Khali of the UAE illustrates the continued dominance of C4 grasses (63 species), although smaller factions of C3 plants (27 species) continue to coexist in this environment (Western 1989; Mandaville

1985; Parker et al. 2004).

This is not to suggest that C3 plants were not available or played no role in human diet. In a study by Potts (1993c), radiocarbon dating from the site of Tell Abraq yielded

13C values for charcoal as well as dates, a food that would have been consumed regularly at this and other sites during the Bronze Age (Table 6.10). As a known C3 plant, the carbonized dates sampled by Potts (n=7) exhibit low 13C values expected as 285 part of this photosynthetic pathway. Similar 13C values are also expressed by the charcoal (n=3) samples analyzed by Potts, some of which may have been from the trunk of a date palm. Because dates played such an important role in Bronze Age agricultural systems, some contribution by these C3 plants is expected in the human dental enamel sampled as part of this study.

Table 6.10. Bronze Age 13C ratios from Tell Abraq (from Potts 1993d).

Lab No. Sample Context cal. C14 Date δ13C (‰) K-5580 carbonized date I. locus 15 1860-1770 BC -23.3 K-5583 carbonized date I. locus 3 1690 BC -23.1 K-5578 carbonized date III. locus 27 2170-2140 BC -22.6 K-5579 carbonized date I. 4.94-4.74 1520 BC -22.3 K-5581 carbonized date II. 3.96-3.76 2300 BC -22.0 K-5576 carbonized date II. locus 22 2140 BC -21.8 K-5577 carbonized date II. locus 4 2190-2140 BC -21.7 K-5575 charcoal I. locus 23, 7.97-7.77 2190 BC -24.7 K-5574 charcoal I. 7.90 2130 BC -23.9 K-5582 charcoal III. 2.76-2.50 2570-2510 BC -23.5

Bronze Age faunal 13C ratios from the United Arab Emirates are shown in

Figures 6.44 and 6.45. Considerable overlap in 13C values is present across both sites and species. On Umm an-Nar Island, 15 animals, including cattle (n=2), ovicaprines

(n=12), and a wild oryx (n=1), were analyzed and possess a mean of -5.4  3.6‰ (1) and an expansive range of -9.0 to +5.9‰ (Figure 6.44). This sizeable range is due primarily to the oryx, the only wild fauna utilized for this study and the only animal from the island to have a positive 13C ratio. Excluding this value, the mean decreases to -6.2

 1.7‰ (1). As discussed previously, in the bioapatite of large modern mammals, 286

Figure 6.44. Carbon isotope ratio ranges for Bronze Age fauna in the United Arab Emirates. Diet 13C ranges from Cerling et al. (1997).

elevated values of approximately +14‰ relative to diet has been observed so that an

13 apatite  C ratio of -13‰ denotes a 100% C3 diet, while a value of +1‰ signifies a total, exclusive contribution by C4 plants (Cerling et al. 1997). The fauna of Umm an-Nar

Island consumed plants with 13C values ranging from -23.0 to -8.1‰, indicative of a predominantly mixed C3-C4 diet. The exception to this mixed diet was the wild oryx, which seems to have differed from its domestic counterparts and consumed a diet

13 dominated by C4 plants. However, because marine foods can mimic  C signatures associated with a mixed C3-C4 diet, it is also possible that domestic animals were being fed some marine products while living on the island. Unfortunately, without stable

287 nitrogen isotope ratios, it is difficult to distinguish between terrestrial and marine contributions to diet.

In the Emirate of Fujairah, two ovicaprines from Dibba (n=1) and Qidfa (n=1) range in 13C value from -7.0 to -1.3‰ and produce a mean of -4.1  4.0‰ (1) (Figure

6.44). While the small number of fauna available for analysis from Fujairah precludes any definitive conclusions regarding diet, these two animals consumed plants with 13C ratios approximating -21.0 to -15.3‰, producing similar values to those seen on Umm an-Nar Island and suggestive of a mixed C3-C4 diet.

At Tell Abraq, dental enamel from four species of domestic animals (n=12) was utilized for isotopic analysis, including goat (n=2), sheep (n=5), cattle (n=4), and pig

(n=1). These fauna display an average 13C value of -0.7  3.0‰ (1) and range from

-6.5 to +4.6‰. Tell Abraq fauna exhibit the most elevated 13C ratios relative to other

Bronze Age sites in the United Arab Emirates, although as with Umm an-Nar Island and

Fujairah, the majority of these individuals exploited a mixed diet of both C3 and C4 plants. Again, the coastal location of Tell Abraq could hypothetically have provided domesticates with some marine aspect to their diets, which may in part explain the 13C values seen here. Nevertheless, wild C4 grasses would have been readily available in this environment for grazing, and it is evident that C4 plants played an important role in the diet of these fauna; even where a mixed C3-C4 diet is observed, proportionally more C4 plants were consumed relative to C3 plants for the majority of the animals at Tell Abraq.

Finally, the Shimal Plain is represented by nine ovicaprines from the Wadi Suq

Shimal settlement as well as a single ovicaprine from the Umm an-Nar tomb of Unar 1.

Together, these generate a 13C range of -8.9 to -2.3‰ and a mean of -6.2  1.9‰ (1). 288 This equates to plants with 13C ratios extending from -22.9 to -16.3‰, comparable to other Emirati sites where a mixture of C3 and C4 plants in diet took place.

When 13C results are considered by species instead of by site, a few tentative trends emerge. Cattle (n=6) possess an average 13C value of -0.7  3.1‰ (1) with an associated range from -4.5 to +4.6‰ (Figure 6.45). All but two of these animals fall into

13 the  C gamut expected for a mixed C3-C4 diet, and each of these values tends towards the C4 end of the spectrum, indicating a greater availability and/or preference for C4 plants (or perhaps marine foods) by domesticated cattle. Further, the 13C ratios of two of these cattle are contained within the C4-dominated diet category, again reiterating the

Figure 6.45. Carbon isotope ratio ranges by species for Bronze Age fauna in the United Arab Emirates. Diet 13C ranges from Cerling et al. (1997).

289

importance of the C4 photosynthetic pathway in southeastern Arabia during the Bronze

Age.

By far, the most numerous animals sampled in this study were ovicaprines

(n=31), producing 13C values ranging from -9.0 to +0.6‰ and thus spanning almost the

13 entirety of the mixed C3-C4 diet category (Figure 6.45). The average ovicaprine  C value of -5.2  2.9‰ (1) is considerably low relative to cattle, suggesting that sheep and goat were exposed to more plant varieties, and specifically, more C3 plants. Inter-site differences in faunal consumption patterns are also evident. Ovicaprines from Tell Abraq possess the highest 13C ratios (as did cattle from Tell Abraq), while the lowest ratios come from Umm an-Nar Island and the Shimal Plain.

A single pig (1.3‰) and oryx (5.9‰) were also included in stable carbon isotope analysis. The oryx displays the most elevated 13C value of any animal (domestic or wild) in this study, consistent with a diet dominated by C4 plants. As the only wild animal included, the oryx is also only one of only four fauna to have a diet almost exclusively influenced by C4 plants. Without an analysis of more non-domesticated animals, it is difficult to conclude why such a difference exists between wild and domesticated fauna. As one potential scenario, domesticated C3 crops including wheat, barley, and dates were introduced as fodder into the diets of these mammals by humans, thus increasing the availability of C3 plants that would not have been available in the wild and decreasing the 13C ratios present in the enamel of these fauna. The 13C ratio for the pig also falls just above the threshold for a C4-dominated diet and represents the third-

290 13 most C-enriched diet of the entire faunal pool, again suggesting that C4 plants were readily available for consumption in this environment.

Mowaihat

At the Umm an-Nar site of Mowaihat in the Emirate of Ajman, no faunal remains were recovered in the rectangular Tomb B. However, because of the close proximity of this tomb to that of Tell Abraq, fauna from Tell Abraq have been utilized to compare against human isotope signatures for Mowaihat.

Thirteen individuals interred within Tomb B of Mowaihat have an average 13C of -9.8  2.0‰ (1) and range in value from -13.0 to -6.2‰ (Figure 6.46). These (and

Figure 6.46. LM1 carbon isotope ratios of individuals interred at Mowaihat, Emirate of Ajman. 291 subsequent 13C values presented here) were plotted in ranked order to better illustrate

13 their position relative to predominantly C3 and C4 diets. The  C values from Mowaihat

13 are consistent with a mixed C3-C4 diet, but unlike their faunal counterparts, human  C ratios fall closer to the C3 end of the photosynthetic spectrum, indicating that humans had considerably more access to C3 sources (Figure 6.46; see also Figures 6.47 and 6.48).

This assessment fits with Bronze Age archaeobotanical data from the Oman Peninsula, which has revealed the presence of domesticated C3 crops as part of local palm garden cultivation, including dates and cereals.

Interestingly, the lowest of these human values (-13.0‰), separated from the remainder of the group by at least 1.7‰ from the next lowest ratio, belongs to MW 197, the same individual identified as a non-local by both strontium and oxygen isotope analysis (Figures 6.47 and 6.48). This low ratio indicates that MW 197 consumed more

C3 plants and/or C3 plant-consuming fauna relative to the other tomb members; in fact, this individual falls just below the estimated 12.5‰ value expected for those completely reliant on C3 plants and may suggest a lack of marine foods in the diet. While not notably different in value, the 13C ratio of this individual adds further credence to the assignment of this person as a non-local whose enamel formed in an area geologically and environmentally distinct from the majority of those laid to rest in this Umm an-Nar tomb. Excluding this individual, the mean of the Mowaihat group remains similar at -9.6

 1.9‰ (1) and ranges in value from -11.3 to -6.2‰. Of these locals, 13C ratios span

5.1‰, showing some degree of dietary variability in the proportion of C3/C4 and terrestrial/marine foods consumed by different individuals.

292 6.0

4.0

2.0

18 δ 0.0 Human Sheep Goat -2.0 Cattle Pig -4.0 -15.0 -10.0 -5.0 0.0 5.0 10.0 δ13C

Figure 6.47. Bivariate plot of carbon and oxygen isotope ratios for human enamel from Mowaihat and comparative faunal enamel from Tell Abraq.

Of the 13 individuals sampled here, six possessed multiple in situ mandibular molars suitable for evaluating inter-tooth variability associated with changes in diet, including temporal changes in 13C as a result of breastfeeding and weaning (Table 6.11).

13  Cco ratios elevated by approximately +1‰ in bone collagen have been recorded in breastfeeding infants (e.g., Dupras et al. 2001; Richards et al. 2002; Fuller et al. 2006;

Gregoricka and Sheridan In press) and has been explained as a trophic level effect.

13 Fewer studies have focused on changes in bioapatite ( Cap). Wright and Schwarcz

13 (1998) found that enamel  Cap values are elevated in M1s relative to M3s by approximately +0.6, while Dupras and Tocheri (2007) found a similar difference of

293 0.7090

0.7089

0.7088

Sr 86

Sr/ Human

87 0.7087 Sheep Goat 0.7086 Cattle Pig 0.7085 -15.0 -10.0 -5.0 0.0 5.0 10.0 δ13C

Figure 6.48. Bivariate plot of carbon and strontium isotope ratios for human enamel from Mowaihat and comparative faunal enamel from Tell Abraq.

Table 6.11. Inter-tooth 13C sampling at Mowaihat for six individuals.

Individual Site Tooth δ13C (‰) ∆δ13C

MW LM -10.5 Mowaihat 1 +1.2 190/191 LM3 -9.7 MW LM -10.7 Mowaihat 1 +1.2 192/193 LM3 -9.5 MW LM -8.2 Mowaihat 1 -0.5 195/196 LM2 -8.7 MW LM -6.2 Mowaihat 1 +1.1 200/201 LM2 -5.1 MW LM -6.9 Mowaihat 1 +0.9 202/203 LM3 -6.0 MW LM -10.9 Mowaihat 1 +0.7 204/205 LM3 -10.2 294 +0.7‰ between M1s and M2s (Wright and Schwarcz 1998). Such enrichment in 13C has been explained as a result of an increasing reliance on C4 foods in the diet.

Out of the six Mowaihat individuals, the four M1/M3 pairs all possessed first molars with values depleted in 13C relative to third molars. The mean 13C ratio for these

M1s (-9.8  1.7‰, 1) differs by +0.9 from M3s (-8.9  1.7‰, 1) and thus follows a

13 comparable pattern of temporal enrichment of C as discussed above, likely signifying a greater dependence on marine and/or C4-based sources of food with age. As the Bronze

Age inhabitants of this region employed not only agricultural but also pastoral systems of subsistence, it is likely that animal products (primarily secondary products including milk) played an important role both in weanling and adult diets. As has already been established, fauna from this era utilized both C3 and C4 plants in their diets, but

13 preference was given to C4 plants. Such C-enriched values would subsequently be passed onto those humans consuming animals and their products, and would explain the

13 increasingly positive  Cap seen over time.

Two other individuals with an M1/M2 pair were also sampled. Individual MW

200/201 likewise shows an elevation in value from M1 to M2 by +1.1‰ and again may derive from a diet containing more C4-based sources of food during second molar enamel formation. Conversely, individual MW 195/196 displays the opposite pattern, in which

13C values actually decrease over time. This depletion in 13C points to an increased

13 dependence on C3 foods with lower  C values than breast milk given by the mother and follows a pattern similar to that seen in bone collagen. None of these individuals exhibit

13C differences between M1s and M2/M3s large enough to suggest substantial changes

295 in diet associated with movement from one area to another during enamel formation, but it is clear that marine and/or C4 foods played a primary role in subsistence.

Tell Abraq

13 The  C values of human molar (LM1) enamel from the 29 individuals sampled at Tell Abraq span from -13.6 to -5.4‰ and have an average of -8.9  1.9‰ (1) (Figure

6.49). These ratios suggest that the majority of the individuals interred in this Umm an-

Nar tomb consumed a mixed C3-C4 diet, with the spectrum generally weighted towards

13 C3-based sources. The range of  C values at Tell Abraq is significantly extended by an outlier (TA 165) 2.0‰ more negative than the 13C ratios of other tomb members,

Figure 6.49. Carbon isotope ratios of adult and subadult individuals interred at Tell Abraq. 296 indicative of a C3-dominated diet. This individual was similarly identified as a non-local by both strontium and oxygen isotopes (see Figures 6.50 and 6.51); a second individual

(TA 161) was also distinguished as an immigrant by strontium, although oxygen and carbon values both fall within local ranges and are thus included in the recalculated mean. Removing individual TA 165 produces an average of -8.7  1.7‰ (1).

Local adults (n=18) display a mean 13C of -8.4  1.7‰ (1) and fall into a range between -10.6 to -5.7‰ (Figure 6.49). Subadult (n=11) 13C ratios are generally lower than adults but also possess a larger range of –11.6 to -5.4‰ and average -9.1 1.6‰

(1). Both age groups show sizeable 13C ranges that, as at Mowaihat, indicate a good deal of dietary variability within sites. In addition, subadult 13C ratios are not

0.7110 Human Adult Human Subadult 0.7105 Sheep Goat 0.7100

Cattle Sr

86 0.7095 Pig

Sr/ 87 0.7090

0.7085

0.7080 -15.0 -10.0 -5.0 0.0 5.0 10.0 δ13C

Figure 6.50. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from Tell Abraq. 297 6.0

4.0

2.0

0.0

O 18 δ -2.0 Human Adult Human Subadult -4.0 Sheep Goat -6.0 Cattle -8.0 Pig -15.0 -10.0 -5.0 0.0 5.0 10.0 δ13C

Figure 6.51. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from Tell Abraq.

significantly different (U=141.00; z=-0.85; p=0.38) from those of local adults, implying a continuity of dietary practices between multiple generations. Local inhabitants of Tell

Abraq also do not statistically differ from those at nearby Mowaihat (U=184.50; z=-1.80; p=0.07).

Enamel from a single adult from Tell Abraq with an in situ LM1 and LM2 was analyzed to examine inter-tooth variability (Table 6.12). A change in 13C of only

+0.2‰ suggests that very little change in diet took place between the enamel formation of these two molars. While this individual appears to have consumed a mixed C3-C4 diet,

13 the  C ratios in both teeth suggest that C4 and/or marine resources played a more

13 important role than C3 plants and the animals consuming them. The elevated  C values 298 Table 6.12. Inter-tooth 13C sampling at Tell Abraq for one individual.

Individual Site Tooth δ13C (‰) ∆δ13C

LM -6.6 TA 186/187 Tell Abraq 1 +0.2 LM2 -6.4

associated with C4 dietary sources may have ‘hidden’ the expected trophic level effect associated with breastfeeding and thus caused a lack of disparity between the molars.

Umm an-Nar Island

Stable carbon isotopes for 33 individuals (RM1) from three tombs (I, II, and V) on

Umm an-Nar Island were assessed. In Tomb I, the 13C ratios of human enamel from four individuals produced a mean of -5.6  4.3‰ (1) and a range from -8.4 to +0.9‰

(Figure 6.52). This mean becomes significantly more depleted (-7.7  0.7‰, 1) and its range more constricted (-8.4 to -7.1‰) when outlier UaN 122 is eliminated. Strontium and oxygen isotope signatures did not identify this individual as a non-local, but UaN 122 did possess the highest 87Sr/86Sr ratio of Tomb I. This elevated strontium value reflects the habitual consumption of marine foods during enamel formation; subsequently, the associated 13C value of +0.9‰ can likewise be interpreted as the result of a predominantly marine-based (and/or C4-based) diet.

The remainder of those interred in Tomb I exhibit ratios consisted with a mixed

13 C3-C4 diet. 15 individuals (LM1) from Tomb II on Umm an-Nar Island show  C values spanning from -11.3 to -1.7‰ with an average of -6.0  2.5‰ (1) (Figure 6.52). Tomb

299

Figure 6.52. Carbon isotope ratios of individuals interred at Umm an-Nar Island.

V (n=14, LM1) shows a smaller spectrum of ratios from -8.5 to -2.8‰ and a comparable mean 13C value of -5.9  1.6‰ (1). The 13C variability present in these two tombs indicates a wide array of dietary options available to these individuals during enamel formation of the first molar. Although all of these tomb members fall within ranges expected for a mixed C3-C4 diet, it is evident that different individuals consumed differing proportions of these C3/C4 and terrestrial/marine sources. Combined with the strontium variability present among these same individuals, the 13C data lends credence to the idea that individuals from different localities along the western Oman Peninsula came together to live and die on the island, although none of these are strictly defined as non-locals (Figures 6.53 and 6.54). 300 0.7091

0.7090

0.7089

86 0.7088 Human (Tomb I)

Sr/ Human (Tomb II) 87 0.7087 Human (Tomb V) Cattle 0.7086 Sheep/Goat Oryx 0.7085 -12.0 -10.0 -8.0 -6.0 -4.0 -2.0 0.0 2.0 4.0 6.0 8.0 δ13C

Figure 6.53. Bivariate plot of strontium and carbon isotope ratios for human and faunal enamel from Umm an-Nar Island.

There are no statistically significant differences between any of the tombs (I-II:

U=15.00; z=-1.54; p=0.11; I-V: U=13.00; z=-1.70; p=0.08; II-V: U=212.00; z=0.07; p=0.93). As a single entity, the Umm an-Nar Island tombs are significantly different from both Mowaihat (U=125.50; z=-3.80; p=0.0001) and Tell Abraq (U=554.50; z=

-4.43; p<0.0001), indicating that different dietary strategies were employed on the island as opposed to the coastal mainland. This disparity stems from marine- and/or C4 resources playing a more important dietary role on Umm an-Nar Island than at either

Mowaihat or Tell Abraq.

Inter-tooth sampling was performed on five individuals from the three Umm an-

Nar Island tombs (Table 6.13). In four of these five adults, a comparison of temporal 301 12.0

10.0

8.0

6.0

O 4.0 Human (Tomb I) 18

δ Human (Tomb II) 2.0 Human (Tomb V) 0.0 Cattle Sheep/Goat -2.0 Oryx -4.0 -15.0 -10.0 -5.0 0.0 5.0 10.0 δ13C

Figure 6.54. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from Umm an-Nar Island.

Table 6.13. Inter-tooth 13C sampling at Umm an-Nar Island for five individuals.

Individual Site Tooth δ13C (‰) ∆δ13C

UaN RM1 -7.5 Umm an-Nar Island I +4.9 123/124 RM3 -2.6 UaN LM -5.4 Umm an-Nar Island II 1 +0.8 125/126 LM2 -4.6 UaN LM -6.0 Umm an-Nar Island II 1 +4.7 128/129 LM3 -1.3 UaN LM -11.3 Umm an-Nar Island II 1 +2.9 130/131 LM3 -8.4 UaN LM -3.9 Umm an-Nar Island V 1 -0.7 144/145 LM3 -4.6

302 changes in 13C between M1 and M3 was made. Three of these individuals (UaN

123/124, 128/129, 130/131) show elevated 13C ratios in M3s relative to M1s ranging from +2.9 to 4.9‰. Such a dramatic increase indicates some overriding dietary influence, likely related to a considerable increase in the consumption of marine and/or

C4 foods after three years of age when first molar enamel formation is complete. On the other hand, a fourth individual (UaN 144/145) demonstrates a conflicting pattern in which 13C values become more depleted in 13C over time. Interestingly, this individual presents some of the most elevated 13C values in the Umm an-Nar tombs for both molars, suggestive of a diet consistently high in marine and/or C4-based foods. Despite this 13C-enriched intake, however, foods with slightly diminished 13C ratios were preferred with age. This case further illustrates the dietary variability present among the people of the Oman Peninsula after breastfeeding took place and supports the notion that these Bronze Age inhabitants implemented numerous subsistence strategies. For all four individuals, a mean M1 13C value of -7.2‰  3.1‰ (1) and of -4.2  3.1‰ (1) for

M3s signifies the importance of contributions from 13C-enriched foods, including both marine resources and C4 plants and/or plant-consuming animals, post-infancy. The M1 and M2 of a fifth individual (UaN 125/126) were also sampled, and though 13C- enrichment was not as extreme as between M1/M3 pairs, follows previously observed patterns indicative of a marine/C4 focus.

Shimal: Unar 1, Shimal 95, and Shimal 103

Dental enamel from individuals interred within one Umm an-Nar tomb (Unar 1) and two Wadi Suq tombs (Shimal 95 and 103) on the Shimal Plain was analyzed for 303 13 stable carbon isotopes. Individuals from Unar 1 (n=25, LM1) possess a mean  C ratio of -11.2  1.0‰ (1) with a range of -13.9 to -8.9‰ (Figure 6.55). These ratios span only 5‰ and thus represent the tightest cluster of 13C values seen at any Bronze Age site in the UAE thus far. The majority of these individuals (both adult and subadult)

13 consumed a mixed C3-C4 diet but primarily derived  C from C3 sources, and as a group, are considerably more 13C-depleted relative to sites along the western coast. When segregated by age, adults (n=19) display an average 13C value of -11.1  0.8‰ (1) with a range of -12.3 to -8.9‰, while subadults (n=6) span -13.9 to -9.8‰ and possess a mean

Figure 6.55. Carbon isotope ratios of adults and subadults interred at the Umm an-Nar tomb of Unar 1, Emirate of Ras al-Khaimah.

304 of -11.5  1.4‰ (1). However, this division is not significant (U=71.00; z=-0.42; p=0.65) and speaks to a relatively uniform diet across age categories.

When compared to other Emirati sites, statistically significant differences are present between 13C ratios from Unar 1 and Mowaihat (U=322.00; z=3.05; p=0.002),

Tell Abraq (U=382.50; z=-5.21; p<0.0001), and Umm an-Nar Island (U=342.00; z=-6.12; p<0.0001). Such disparities indicate that while similar food resources and a generally mixed C3-C4 and marine-based diet may have been utilized across the Oman Peninsula, dietary differences did exist between island, coastal, and foothill environs. Although discussed further in the following section, Unar 1 does not significantly differ from either

Shimal 95 (U=40.50; z=1.11; p=0.24) or Shimal 103 (U=108.00; z=-0.32; p=0.73). As all three sites are located on the Shimal Plain in close proximity to one another, it follows that comparable foodstuffs were exploited throughout the Umm an-Nar and later Wadi

Suq period.

Elevated 87Sr/86Sr ratios identified three individuals (RAK 221, 238, 240) as locals who relied heavily on marine resources during the time of enamel formation

(Figure 6.56). 18O ratios show no indication that non-local sources of water were utilized (Figure 6.57), and corresponding local 13C values support this assertion. In addition, two of these three individuals exhibit the highest 13C ratios (-9.8 and

-8.6‰) at Unar 1, adding further support to the assertion that marine foods played some role in diet in the foothills of Shimal.

Of the 25 individuals sampled from Unar 1, three possessed in situ first and second left mandibular molars and were analyzed to discern any temporal shifts in 13C value (Table 6.14). Two of these (RAK 233/234 and 239/240) display 13C values that 305 0.7091 Unar 1 0.7090 Shimal 95 Shimal 103 Sheep/Goat

0.7089

86 Sr/

87 0.7088

0.7087

0.7086 -16.0 -14.0 -12.0 -10.0 -8.0 -6.0 -4.0 -2.0 0.0 δ13C

Figure 6.56. Bivariate plot of strontium and carbon isotope ratios for human and faunal enamel from the Shimal Plain, Ras al-Khaimah.

become increasingly elevated over time; while all three M1 values exhibit ratios indicative of a predominantly C3-based diet, an increase in these values with time suggests the gradual inclusion of more C4 plants and/or animals consuming them, and were perhaps first introduced as weaning foods (e.g., goat milk). However, a third individual (RAK 226/227) shows a pattern of 13C-depletion with time, following a more traditional weaning curve suggestive of a diet emphasizing C3 sources to a greater extent.Overall, average 13C ratios for M1s fall at -11.4  0.4‰ (1), while M2 13C values exhibit a mean of -10.2  1.5‰ (1). Subsequently, an increase of +1.2‰ is present and fits with previously described models of a post-breastfeeding introduction of

C4- and/or marine-based foods. 306

10.0

5.0

0.0

18 δ -5.0 Unar 1 Shimal 95 -10.0 Shimal 103 Sheep/Goat -15.0 -16.0 -14.0 -12.0 -10.0 -8.0 -6.0 -4.0 -2.0 0.0 δ13C

Figure 6.57. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from the Shimal Plain, Ras al-Khaimah. Note that two outliers identified by 18O ratios were subsequently discarded because of burning.

Table 6.14. Inter-tooth 13C sampling at Unar 1 for three individuals.

Individual Site Tooth δ13C (‰) ∆δ13C

RAK LM -11.0 Unar 1 1 -0.5 226/227 LM2 -11.5 RAK LM -11.6 Unar 1 1 +1.2 233/234 LM2 -10.4 RAK LM -11.7 Unar 1 1 +3.1 239/240 LM2 -8.6

307 Also on the Shimal Plain but dating to the subsequent Wadi Suq period, the tomb

13 of Shimal 95 (n= 2; RM1) shows a range from -11.0 to -10.7‰ with a mean  C ratio of

-10.8  0.2‰ (1) (Figure 6.58). These values suggest a mixed but predominantly C3- based diet, potentially signifying a heavier reliance on C3 agricultural (and not more C4- oriented pastoral) products. Further, the site’s inland location at the foothills of the

Hajjar Mountains may have made access to 13C-enriched marine foods more difficult.

While only two individuals were available for sampling out of an estimated MNI of 15 for the tomb, both 13C ratios are comparable with values seen elsewhere, both on the

Figure 6.58. Carbon isotope ratios of individuals interred at the Wadi Suq tombs of Shimal 95 and 103, Emirate of Ras al-Khaimah.

308 Shimal Plain and in the larger United Arab Emirates, and do not suggest a non-local origin for either of these individuals.

A second Wadi Suq tomb on the Shimal Plain, Shimal 103 (n=7; LM1) exhibits an average 13C value of -11.3  0.9‰ (1) with a range from -12.6 to -9.9‰ (Figure 6.58).

As at Shimal 95 and Unar 1, these ratios point to a diet high in C3 sources, possibly reflecting an increased dependence on agricultural products in this area of the UAE.

Additionally, these data points cluster tightly together, spanning less than 3‰; while this may in part be due to small sample size, it also tentatively suggests a local community with a relatively uniform diet during the Middle Bronze Age. Despite this uniformity, a non-local 18O signature for individual SH 214 raises the question of whether a foreign

13C value is present. While all 13C ratios at Shimal 103 are relatively similar, SH 214 possesses the lowest value (-12.6‰) of the group, and is the only individual whose diet

13 was likely completely dominated by C3 foods. While this  C ratio could not be used to identify SH 214 as an immigrant, its negative 13C value coupled with its apparently non- local 18O status does bring to light the possibility of a non-local childhood geographic residence.

13C values from Shimal 103 are not significantly different from those individuals interred at Shimal 95 (U=12.50; z=0.59; z=0.46) or the earlier tomb of Unar 1

(U=162.50; z=0.18; p=0.84), supporting the hypothesis that cultivars like dates and cereals may have played a more dominant role on the Shimal Plain than elsewhere in southeastern Arabian. Correspondingly, then, 13C values from both Shimal 95 and 103 differ significantly from the island site of Umm an-Nar (U=50.00; z=-4.36; p<0.0001) as

309 well as the coastal sites of Mowaihat (U=69.50; z=-2.07; p=0.04) and Tell Abraq

(U=63.00; z=-3.81; p=0.0001).

Fujairah: Bidya 1, Dadna, Mereshid, Qidfa 4, and Dibba 76

As with strontium and oxygen, all samples from the eastern Emirati sites of

Fujairah will be considered together, primarily due to low sample sizes but also because of similar environmental (and presumably dietary) conditions amongst these five sites.

The individuals interred in Fujairah tombs display an average 13C value of -11.1  0.7‰

(1) and range in value from -11.8 to -9.4‰ (Figure 6.59). Two individuals (Bid 246 and Dib 254) producing non-local 87Sr/86Sr (Figure 6.60) but local 18O (Figure 6.61)

Figure 6.59. Carbon isotope ratios of individuals interred in the Emirate of Fujairah. 310

0.7092

0.7090

0.7088

Bidya 1 Sr

86 Dadna 0.7086

Mereshid Sr/

87 Dibba 76 0.7084 Qidfa 4 Sheep/Goat 0.7082

0.7080 -15.0 -10.0 -5.0 0.0 δ13C

Figure 6.60. Bivariate plot of strontium and carbon isotope ratios for human and faunal enamel from the Emirate of Fujairah.

values also exhibit local 13C ratios that do not deviate from a tightly clustered span of

2.4‰. This is not to deny that these individuals were immigrants; instead, their differences are emphasized even further when 13C ratios are plotted against associated strontium isotopic values (Figure 6.60). This scatterplot highlights the possibility of multiple geographic regions possessing disparate sources of strontium but isotopically comparable diets. Interestingly, individual Dib 254 possesses the most elevated 13C ratio at -9.4‰, and while not an extreme deviant, offers additional evidence of at least some residential mobility at Dibba 76, albeit (likely) from nearby sites.

311 4.0

3.0

2.0

1.0 Bidya 1

O Dadna

0.0 18

δ Mereshid -1.0 Dibba 76 Qidfa 4 -2.0 Sheep/Goat -3.0

-4.0 -15.0 -10.0 -5.0 0.0 δ13C

Figure 6.61. Bivariate plot of oxygen and carbon isotope ratios for human and faunal enamel from the Emirate of Fujairah.

The considerable homogeneity present here suggests that individuals across the eastern coast of the United Arab Emirates consumed similar foodstuffs during the course of the Wadi Suq period. This diet, comprised of mixed C3-C4 sources but dependent

13 primarily on C-depleted C3 plants and the animals consuming them, appears analogous to the consumption patterns seen on the Shimal Plain (Unar 1: U=208.50; z=0.16; p=0.86; Shimal: U=93.50; z=-0.04; p=0.94) and may once again point to an increased reliance on agricultural products and a decreased reliance on secondary products from domestic herds in these areas. In contrast, statistically significant differences were found between Fujairah and the Umm an-Nar coastal sites of Mowaihat (U=88.50; z=-2.66;

312 p=0.007), Tell Abraq (U=86.50; z=-4.15; p<0.0001), and Umm an-Nar Island (U=72.50; z=-4.71; p<0.0001), and again speak to dissimilarities in diet both spatial and temporal.

Three individuals possessed two in situ molars within either maxillary or mandibular bone fragments, each of which was examined for temporal changes in 13C ratios (Table 6.15). Within the life history of each of these individuals (Bid 245/246, Qid

250/251, Dib 252/253), changes are minimal and variously show smaller increases or decreases in 13C with time, reflecting some minor variability in subsistence practices.

Table 6.15. Inter-tooth 13C sampling in the Emirate of Fujairah.

Individual Site Tooth δ13C (‰) ∆δ13C RM1 -11.1 Bid 245/246 Bidya 1 -0.6 RM2 -11.7 RM -11.8 Qid 250/251 Qidfa 4 1 +0.3 RM2 -11.5 LM -11.0 Dib 252/253 Dibba 76 2 +0.5 LM3 -10.5

A summary of mean 13C human values for all sites in the United Arab Emirates is presented in Table 6.16. When comparing 13C ratios between the Umm an-Nar and subsequent Wadi Suq, the overlap between the Early and Middle Bronze Age is apparent

(Figure 6.62), although major differences in variability and value clusters distinguish these periods (Figures 6.63 and 6.64). Individual values during the Umm an-Nar period vary widely and span much greater 13C ranges than their Wadi Suq counterparts.

313 Table 6.16. Mean 13C ratios in human tooth enamel from Bronze Age sites in the United Arab Emirates. Parentheses refer to sample number after outliers are removed.

Site n δ13C (‰) δ13C (excluding outliers) Bidya 2 -11.1 -11.1 Dadna 1 -11.4 -11.4 Dibba 3(2) -10.6 -11.2 Mereshid 1 -11.0 -11.0 Mowaihat 13 (12) -9.8 -9.6 Qidfa 1 -11.8 -11.8 Shimal 95 2 -10.8 -10.8 Shimal 103 7(6) -11.3 -11.0 Tell Abraq 29 (28) -8.9 -8.7 Umm an-Nar Island 33 (32) -5.6 -5.8 Unar 1 25 -11.2 -11.2

2.0 Mowaihat 0.0 Tell Abraq

-2.0 Umm an-Nar Island Unar 1 -4.0 Shimal 95 Shimal 103 -6.0

C Fujairah 13  -8.0

-10.0

-12.0

-14.0

-16.0

Figure 6.62. A comparison of carbon isotope ratios of individuals interred at all sites in the United Arab Emirates. 314 2.0

0.0 Umm an-Nar Wadi Suq -2.0

-4.0

-6.0 C (‰) C

13 -8.0  -10.0

-12.0

-14.0

-16.0

Figure 6.63. Variance in carbon isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods.

When analyzed using the Monte Carlo method (simulation=9999), sample means between the Umm an-Nar and Wadi Suq are statistically significant (p=0.0001 for all individuals; p<0.0001 for locals only). Levene’s test demonstrates heteroscedasticity, or heterogeneity of variances, between there periods, including both locals and nonlocals

(p<0.0001) as well as just locals (p<0.0001).

Carbon Isotope Ratios of Comparative Regions Across the Persian Gulf

A preliminary analysis of human 13C ratios was also conducted for two sites outside the United Arab Emirates as a means of generating a regional isotopic map of

315 0.0

-2.0

-4.0

-6.0 C (‰) C

13 -8.0 

-10.0

-12.0

-14.0 Umm4-Jan an-Nar Wadi5-Jan Suq

Figure 6.64. Mean  1 s.d. in carbon isotope ratios from human individuals dating to the Umm an-Nar and Wadi Suq periods.

dietary carbon bioavailability, useful for evaluating differences in diet between regions as a further means of determining the presence of non-locals. For the majority of comparative sites utilized in this study, samples consisted of faunal, and not human, dental enamel, as no associated human remains were available for analysis. Such comparative data is useful for reconstructions of trophic level systems using stable carbon isotopes, although without associated human values, its applicability for our purposes is limited. Human samples from two sites (A’ali Mound Field, Bahrain and al-Khubayb,

Oman) will be discussed first, followed by a brief assessment of Bronze Age faunal 13C signatures from other Persian Gulf countries.

316 A’ali Mound Field, Bahrain

The A’ali Mound Field on the island of Bahrain provided 13C ratios from six human molars (LM1) as well as four ovicaprines. Fauna produce a mean 13C ratio of

-7.4  3.7‰ (1) and range in value from -9.6 to -1.9‰. Human 13C values prove much more tightly clustered, ranging from only -12.5 to -10.9‰ with an average of -12.0 

0.7‰ (1) (Figure 6.65). This range, indicative of a mixed but primarily C3-based diet, places those individuals interred in the mound field well within the ranges demonstrated for the United Arab Emirates and are thus not useful in distinguishing either dietary disparities or migrants from this island who may have been laid to rest in the Bronze Age

2.0

0.0

-2.0

-4.0

-6.0 C (‰) C

13 -8.0 δ -10.0

-12.0

-14.0

-16.0 0 UAE1 Bahrain2 3 Figure 6.65. Carbon isotope ratios of all human individuals from the United Arab Emirates compared with a sample of individuals from the A’ali Mound Field, Bahrain.

317 tombs of southeastern Arabia. Nevertheless, these values are statistically different

(U=102.00; z=-3.31; p=0.0009) from those found in the UAE.

Al-Khubayb, Oman

Preliminary excavations of Hafit and transitional tombs from the al-Khubayb necropolis of Oman provided three human left mandibular molars for stable isotope analysis. Dental enamel samples exhibit a mean 13C value of -10.7  0.6‰ (1) and a range from -11.3 to -10.1‰ (Figure 6.66). Unlike human 18O values, which display a significant 18O-enrichment over those in the UAE, 13C Omani ratios are

2.0

0.0

-2.0

-4.0

-6.0 C (‰) C

13 -8.0 δ -10.0

-12.0

-14.0

-16.0 0 UAE1 Oman2 3 Figure 6.66. Carbon isotope ratios of all human individuals from the United Arab Emirates compared with a sample of individuals from the al-Khubayb necropolis, Oman.

318 indistinguishable from those found in the Emirates (U=137.00; z=-1.04; p=0.30), indicative of the consumption of isotopically similar, predominantly C3 plants and plant- consuming animals.

Dental enamel from various fauna across the Persian Gulf was also sampled to detect any major disparities in carbon availability that might assist with differentiating individuals from dissimilar regions (Figure 6.67). In total, Bronze Age fauna from the

UAE average -4.1  3.8‰ (1) with a 13C range of -9.0 to +5.9‰, displaying considerably more variability than their human counterparts. This variability is mirrored in fauna from comparative sites across the Gulf, in part a consequence of the inclusion of multiple species with differing dietary preferences. Mean 13C ratios and associated ranges are found in Table 6.17. As with animals from the UAE, the majority of fauna from each comparative site engaged in the consumption of a mixed C3-C4 diet, although the preference for and/or availability of C4 plants found in the UAE is not always present elsewhere. This is particularly the case for the sites of A’ali Mound Field and Allahdino,

13 whose average faunal  C ratios suggest that C3 plants played a slightly more dominant role than C4-based sources. With the exception of Allahdino, cattle from each site consistently display the most elevated 13C values of all taxa sampled, although in some cases also exhibit the lowest ratios within the same site. Such range is again indicative of a substantial degree of dietary variability present amongst a single species.

Statistical inter-site comparisons of these fauna are found in Table 6.18. The majority of these fauna do not differ significantly in 13C value, suggesting an isotopically similar diet across a wide geographic area. Interestingly, despite the close

319 320

Figure 6.67. Carbon isotope ratios of fauna from the United Arab Emirates compared with fauna from sites across the Persian Gulf. 252 Table 6.17. Mean 13C and ranges for comparative fauna at sites across the Persian Gulf.

Country Site Mean Range

A'ali Mound Field -7.4  3.7‰ (1) -9.6 to -1.9‰ Bahrain Barbar -6.2  2.8‰ (1) -11.2 to -0.9‰ Allahdino -8.4  3.1‰ (1) -13.8 to -3.4‰ Pakistan Balakot -2.3  4.7‰ (1) -10.9 to +3.6‰ Iran Tepe Yahya -5.3  4.3‰ (1) -11.2 to +1.4‰ Kuwait Failaka Island -4.5  3.7‰ (1) -8.9 to +4.1‰

proximity and similar coastal settings of the Pakistani sites of Allahdino and Balakot, their Bronze Age fauna possessed isotopically distinct diets (U=72.00; z=-2.46; p=0.01) signifying considerable differences in grazing strategies employed by the herders of these domestic animals. Such a difference may reflect greater mobility and transhumance employed by one site over the other, increasing the potential for encountering foreign

13C values. In addition, the possibility exists that domesticates were fed marine foods at one site but not at another.

***

In summary, while the majority of 87Sr/86Sr and 18O ratios exhibit little variation, a few deviant ratios occur throughout both the Umm an-Nar and Wadi Suq of the Bronze

Age. Significant differences in mean and variance between periods are present for strontium but not oxygen. Additionally, major differences in 13C variability distinguish the Umm an-Nar and Wadi Suq periods. Values during the Umm an-Nar period vary widely and span much greater 13C ranges than their Wadi Suq counterparts, and expectedly, statistically significant differences in both mean and variance were evident between these time periods. 321

Table 6.18. Mann-Whitney (U) comparative statistics for carbon isotope ratios of fauna from the United Arab Emirates and across the Persian Gulf.

d l e i

F d

n d a a n l y s u o h I o n t i a r a o M d a Y k k i h b e a l a l E r a l a l p i ' l A a a e a

A B A B T F U A'ali Mound Field p=0.32 p=0.67 p=0.07 p=0.57 p=0.07 p=0.04 Barbar p=0.11 p=0.03 p=0.58 p=0.42 p=0.11 Allahdino p=0.01 p=0.13 p=0.01 p=0.002 Balakot p=0.15 p=0.18 p=0.15 Tepe Yahya p=0.74 p=0.55 Failaka Island p=0.74

UAE

322

CHAPTER 7

DISCUSSION

This chapter provides a general discussion of the results, evaluates each hypothesis (outlined in Chapter 2) based on the biogeochemical evidence provided by this study, and then further contextualizes these findings in light of corresponding archaeological evidence from the region. This chapter concludes by utilizing both isotopic and archaeological data to assess multiple theoretical approaches attempting to explicate political economy, social organization, and identity in the Bronze Age.

***

While the majority of Umm an-Nar and Wadi Suq individual 87Sr/86Sr ratios display little variation, a few deviant ratios likely point to the presence of a small number of non-locals throughout the Bronze Age. The mean Wadi Suq 87Sr/86Sr value

(0.708761) is slightly lower relative to Umm an-Nar (0.708872). Although a statistically significant difference in mean between these periods may indicate some difference in mobility, it more likely identifies how the geographic distribution of sites differed over the course of the Bronze Age. Significant differences in variance were also recognized and likewise suggest that these two groups did not come from populations with the same

323 variance, and that consequently, mobility during the Umm an-Nar period differed from the subsequent Wadi Suq.

As with strontium, a few outlying 18O ratios likely point to the presence of a small number of non-locals during both the Umm an-Nar and Wadi Suq periods of the

Bronze Age. Nonetheless, the majority of individuals possessed values that clustered together and displayed little variability, indicative of a largely local population that acquired water from isotopically similar sources. Only slightly elevated (+0.1‰) relative to its Umm an-Nar counterpart, the Wadi Suq 18O mean is not statistically significant from the preceding period, suggesting that local inhabitants of the region imbibed isotopically similar water throughout the course of the Bronze Age (though not necessarily from the same sources). No significant differences in variance imply that, despite the larger span of 18O values from individuals dating to the Umm an-Nar period, mobility did not change significantly from the Early to the Middle Bronze Age.

While overlap in 13C ratios throughout the Bronze Age is apparent, major differences in variability distinguish the Umm an-Nar and Wadi Suq periods, as Umm an-

Nar 13C values vary dramatically and span a much greater range than those interred in

Wadi Suq graves. While this may in part be due to differences in sample size, this trend speaks to the greater dietary variability present in the Umm an-Nar period, perhaps related to a multitude of subsistence strategies that gradually tapered by the Wadi Suq period, thereby generating much more limited 13C availability. Umm an-Nar values suggest an exploitation of resources utilizing both C3 and C4 photosynthetic pathways with a notable C3 preference. It is likely that the majority of C3 values stem from cultivars grown in shaded palm gardens, while C4 influences originate from the primary 324 and secondary animal product of C4-consuming fauna. The exception to this is on Umm an-Nar Island, where extremely elevated 13C ratios are a probable consequence of the regular consumption of 13C-enriched marine foods.

However, by the Wadi Suq, very few contributions from C4 sources are apparent, and variability much more restricted. This change could signify an increasing reliance on

C3-based agricultural products over time, coupled with a diminishing dependence on both marine foods as well as a pastoral way of life. This trend is substantiated by changes in the overall means of the periods themselves: the Umm an-Nar average 13C of -8.4 

2.9‰ (1) witnesses a decline to -11.1  0.7‰ (1) by the Wadi Suq. Non-local 13C signatures are also more frequent during the Umm an-Nar period, suggesting both that immigrants may have been present in these tombs and also that Umm an-Nar diet varied more substantially than in the subsequent Wadi Suq.

Statistically significant differences in mean and variance are indicative of substantial changes in subsistence practices between these two time periods. As mobility and diet are intimately related when dealing with populations engaging in herding and

(agro-) pastoralism, such a dramatic shift in 13C suggests that subsistence strategies were correspondingly modified in the Wadi Suq. Again, however, these differences in variance may in part be a result of decreased tomb membership in the Wadi Suq, offering fewer samples available for analysis.

***

325

Hypothesis I: The Umm an-Nar (2500-2000 BC) population of the Oman Peninsula would have been highly mobile as a result of increasingly complex and widespread interregional exchange networks.

Considerable isotopic variability was expected among locals as regional exchange networks across the peninsula expanded to meet growing demands for copper and as a standardization of mortuary practices as well as the local manufacturing of wares reflects increased communication between sites throughout the Emirates. However, mean

87Sr/86Sr ratios from local individuals interred at Mowaihat (0.7088630.000014, 1; n=12), Tell Abraq (0.7088730.000020, 1; n=27), Umm an-Nar Island (0.708902

0.000079, 1; n=33), and Unar 1 (0.7088050.000065, 1; n=25) all display very little isotopic variability, indicative of a population that was not highly mobile. This is not to suggest that Umm an-Nar peoples were uninvolved in interregional trade, as long- distance travel may have taken place after enamel formation was complete and would therefore not be recorded in the biogeochemical signatures of the teeth. With a hypothesized rise in mobility, increased isotopic variability was expected as more outsiders (both regional locals and interregional immigrants) interacted with and were potentially buried in local tombs. Instead, the data show a largely homogeneous group displaying ratios consistent with local ranges, indicating that (a) during enamel mineralization, inhabitants of the area were not regularly engaged in mobile behavior related to pastoralism or trade, and that (b) people born in this area were also interred there. Correspondingly, intra-site oxygen isotope ratios do not vary by more than 2.2‰, signifying that these individuals originated from a similar geographic area and obtained water from sources comparable in 18O value. Inter-tooth variability among Umm an-

326 Nar individuals for strontium and oxygen isotope ratios was generally low, adding further credence to the conclusion that group mobility was not an important part of everyday life for the local population of southeastern Arabia. 13C values are more variable and are discussed further in Hypothesis Ic. While the hypothesis that growing interregional exchange networks instigated a significant rise in mobility among local groups during the

Umm an-Nar period must therefore be rejected, these data do fit with other forms of archaeological evidence suggesting that groups became increasingly sedentary with the appearance of settlements, fortification towers, monumental tombs, and oasis agriculture.

Despite a lack of biogeochemical support for this hypothesis among the local population, the clear presence of non-locals (3% of Umm an-Nar individuals sampled) does indicate that the Oman Peninsula was actively engaged in interregional interaction to a certain extent, and that immigrants to this region were permitted to be buried in local tombs.

Hypothesis Ia: Non-locals will be present within Umm an-Nar communal tombs.

Three non-locals were identified out of a total of 100 Umm an-Nar individuals sampled in this study: two from Tell Abraq (TA 161 and 165) and one from Mowaihat

(MW 197). These individuals each possessed 87Sr/86Sr ratios that fell outside locally defined ranges and deviated considerably from local human value clusters. 18O and

13C values similarly classified two (MW 197 and TA 165) of these three individuals as immigrants, although a third (TA 161) from Tell Abraq exhibited 18O and 13C ratios consistent with other, local values. This does not mean that individual TA 161 was in fact local, but simply that he or she possessed an isotopically similar diet and had access 327 to isotopically similar water sources. Because of the extreme divergence of this individual’s 87Sr/86Sr ratio, TA 161 is still designated as a non-local. These three individuals were not designated as foreign in any way by burial position, grave good possession, or placement within the tomb, but were instead undifferentiated from other, local commingled remains. Consequently, the hypothesis that non-locals are present within Umm an-Nar collective graves is supported by the biogeochemical data and suggests that this region was actively engaged in interregional interaction. It should be noted that geologic and geographic similarities in both strontium and oxygen isotopes across large portions of the coastal Persian Gulf (based on geologic and precipitation maps) may mask the presence of additional immigrants that exhibit isotopically-similar enamel values.

The presence of foreigners at Tell Abraq as identified by stable isotope analysis fits with a previous assessment of biological affinity using craniometric data, which concludes that the tomb was used by a mixed population (Potts 1993d). Similarly, the absence of non-locals from Tombs I, II, and V on Umm an-Nar Island is consistent with dental metric and nonmetric biodistance analysis (Højgaard 1980, 1981; Kunter 1991; Alt et al. 1995), which suggest a homogeneous, endogamous population. No biodistance analyses have been performed on human remains from either Mowaihat or Unar 1.

Hypothesis Ib: Umm an-Nar tombs associated with large settlements will contain greater numbers of non-locals than tombs with no associated permanent settlement.

Two Umm an-Nar monumental tombs analyzed here were associated with large, contemporaneous settlements: Tell Abraq (n=29) and Umm an-Nar Island (n=33). Both

328 of these settlements and associated tombs contained large amounts of exotic wares, leading to speculation that these settlements acted as major trading posts along the western coast of the Oman Peninsula. The grandest of these settlements, that of Tell

Abraq, is represented by a large fortification tower less than 10 m from the third millennium tomb, both of which possessed a rich array of foreign pottery and ornaments.

Unsurprisingly, this funerary structure contains the most non-local individuals (n=2; TA

161 and 165) of any tomb sampled in this study and supports the hypothesis that large, trade-oriented settlements would have attracted foreign emissaries, traders, exogamous marriage partners, and/or other non-locals. On the other hand, no immigrants were identified from the three tombs sampled from Umm an-Nar Island despite its role as a seemingly major player in early Umm an-Nar coastal trade. Interestingly, Tombs I, II, and V are all positioned over one kilometer away from the settlement on the southern plateau of the island, while the majority of the other, unexcavated tombs lie much closer on the northern plateau, which may in part explain an absence of immigrants in these structures. These results do not support the above hypothesis.

A lack of permanent settlement and exotic goods associated with Umm an-Nar tombs at Mowaihat (n=13) and Unar 1 (n=25) was expected to reflect the more minor role of these sites in interregional trade, and correspondingly, it was anticipated that non- local individuals would not have been interred here. Surprisingly, despite an absence of foreign vessels and the recovery of only a few beads of non-local origin, one immigrant

(MW 197) was identified at Mowaihat. This may in part be related to Mowaihat’s close proximity to the larger, more prominent site of Tell Abraq. If speculation that the unusual, subterranean and rectilinear Tomb B at Mowaihat represented a secondary

329 repository after the more traditional, above-ground and circular Umm an-Nar Tomb A had been filled, this may mean that individual MW 197 was first interred in Tomb A before later being transferred – along with local tomb members – to Tomb B. These results do not support the above hypothesis.

Conversely, while a single carnelian bead and multiple sherds of Iranian pottery speak to at least some involvement with interregional exchange networks at Unar 1, no immigrants appear to have been interred within Unar 1 on the Shimal Plain. While these results fit with the hypothesis that non-locals are not expected at sites without an associated, permanent settlement, they are nevertheless interesting given that Unar 1 possessed one of the largest tomb populations (n=438) of any Umm an-Nar grave in the

UAE (and, in fact, was the largest tomb sampled in this study), likely reflecting a sizeable living population in the area.

Based on the distribution of immigrants in these six Umm an-Nar tombs, no relationship appears to exist between sizeable, permanent settlements and the presence of non-locals in local tombs, and this hypothesis must therefore be rejected.

Hypothesis Ic: Diet was extremely variable during the Umm an-Nar period.

The Umm an-Nar population of the Oman Peninsula utilized a variety of subsistence strategies, including agriculture, herding, hunting, and exploitation of coastal and marine environs. Because of this expansive repertoire, it was anticipated that intra- and inter-site 13C values would be correspondingly variable.

Considerable variability was indeed present among locals at all Umm an-Nar sites. Mowaihat (-11.3 to -6.2‰), Tell Abraq (-13.6 to -5.4‰), Umm an-Nar Island

330 Tomb I (-8.4 to -7.1‰), Umm an-Nar Island II (-11.3 to -1.7‰), Umm an-Nar Island V (-

8.5 to -2.8‰), and Unar 1 (-13.9 to -8.9‰) each exhibit intra-site variability by as much as 10‰, indicative of derived carbon from multiple photosynthetic pathways, trophic levels, and (likely) terrestrial and marine settings. Generally, these ranges are consistent with a mixed C3-C4 diet with a preference for C3-based plant and/or plant-consuming animal sources.

13C ranges are greatest at Umm an-Nar Island; unlike other third millennium BC sites, whose inhabitants relied more heavily on terrestrial-based C3 resources, individuals from Umm an-Nar consumed proportionally more marine-based foods, generating a wider 13C range with more positive values and distinguishing the site as statistically significant from Mowaihat, Tell Abraq, and Unar 1. This assessment fits with archaeological evidence from the settlement midden of the island’s reliance on fish, molluscs, dugong, turtles, sharks, stingrays, and whales, further corroborated by an abundance of artifacts associated with maritime exploitation, including net-sinkers and fish hooks (Hoch 1979, 1995; Frifelt 1991, 1995). Likewise, bioarchaeological analyses of dental wear, caries, calculus deposits, and antemortem tooth loss illustrate a diet rich in coarse marine resources, with plant domesticates playing only a minor dietary role

(Højgaard 1980, 1981; Kunter 1991). Nevertheless, individuals with low 13C values also resided on the island, suggesting that some of its inhabitants took particular advantage of agricultural products, evidenced by plant impressions of dates, wheat, and barley, and by quernstones recovered from both the settlement and within tombs. Almost all inter-tooth samples from Umm an-Nar Island show a substantial increase in 13C over time, indicative of a change in diet most likely related to a considerable increase in the 331 consumption of marine foods after three years of age when first molar enamel formation is complete.

Enamel from individuals interred in the Umm an-Nar monument of Unar 1, an inland site at the foothills of the Hajjar Mountains on the Shimal Plain, possess the least variable 13C ratios, spanning only 5‰. These values are significantly lower than in any other Umm an-Nar tomb, and suggest that while plain inhabitants did consume a mixed

C3-C4 diet, a primary reliance on C3 sources separates them statistically from their coastal counterparts who likely consumed more marine- and/or C4, pastoral-based resources.

Such disparities highlight the existence of dietary differences between island, coastal, and foothill locations. An evaluation of dental pathology at Unar 1 by Blau (2001b, 2007), including dental caries, abscesses, and calculus, was interpreted as representing a broad diet with mixed subsistence practices. However, the biochemical evidence presented here depicts a more restricted diet relative to neighboring Umm an-Nar sites that likely emphasized oasis agriculture, fitting with the interior location of Unar 1 that would not have allowed as much consistent access to coastal resources. This terrestrial-based diet is confirmed by strontium isotope ratios, which cluster around the local mean as defined by terrestrial fauna and not around the local maximum as at Mowaihat, Tell Abraq, and

Umm an-Nar Island, signifying the addition of elevated 87Sr/86Sr signatures from a marine environment.

13C ratios from Mowaihat and Tell Abraq fall between the two extremes of Umm an-Nar Island and Unar 1 but also show considerable 13C variability reflective of a mixed C3-C4 diet, with the value spectrum generally weighted towards C3-based sources.

Such variability may also point to differences in the availability of certain foods based on 332 social status, although inclusive mortuary practices make assessing societal stratification difficult. This assessment fits with the utilization of multiple subsistence strategies that characterize the Early Bronze Age in southeastern Arabia, including archaeobotanical data revealing the presence of domesticated C3 crops as part of local palm garden cultivation (Cleuziou 1996; Potts 1993a; Tengberg 2003a, 2003b). Bioarchaeological analyses of dental pathology from Mowaihat Tomb B and Tell Abraq support this appraisal, showing a high degree of dental wear indicative of the regular consumption of stone-ground cereal grains, a high frequency of antemortem tooth loss consistent with a agricultural diet high in carbohydrates and sugars, and calculus deposits suggestive of an increased dependence on plant domesticates (Martin 1996; Blau 2007). C4-consuming domesticated animals and marine sources supplemented human diet as well.

Domesticated sheep, goat, and cattle dominate archaeological assemblages at Tell Abraq, while wild terrestrial and marine animals played only a small role (Potts 1993d; Stephan

1995).

Finally, adult and subadult permanent dentition were sampled from Tell Abraq and Unar 1. No significant dietary differences in 13C were detected at either site, indicating that isotopically-similar foods were consumed, likely over multiple generations.

Hypothesis II: Interregional mobility decreased during the Wadi Suq (2000-1300 BC) in the Oman Peninsula as a result of a “collapse” of interregional exchange networks, while regional mobility increased as populations became more mobile.

A dramatic decline in isotopic variability was expected among Wadi Suq tomb members as interregional economic exchange networks across the Persian Gulf collapsed,

333 rendering obsolete (a) the presence of non-locals, and (b) regional trade and communication complexes related to copper and other local natural resources. Unlike the previous Umm an-Nar period, the Wadi Suq shows no evidence of copper working or large-scale copper production, and considerable variation in tomb construction speaks to a decrease in communication and an overall lack of cultural homogeneity (Potts 1990;

Weeks 1997; Carter 2003c; Jasim 2006). Despite the archaeological evidence, the presence of immigrants to and from the region (n=3, or 19% of Wadi Suq individuals sampled) is quite high, particularly when taking into account the small number of individuals (n=17) available for sampling from these tombs.

At the same time, however, mobility within the region itself was expected to increase, as archaeological evidence informs of a transition from a more sedentary to a more nomadic way of life for most areas of the Oman Peninsula. Multiple indicators of this shift – including less refined tomb construction, a decrease in the size, permanence, and number of settlements, and increasing regional aridification – all point to the adoption of an increasingly mobile lifestyle during the second millennium BC (Crawford

1998; Vogt 1998; Potts 2001; Carter 2003c). Moreover, while the highest frequencies of infectious disease, traumatic injury, DJD, and dental wear appear during the Umm an-Nar in conjunction with the emergence of large fortified settlements, monumental building projects, and a mixed subsistence economy, these trends decrease in the Wadi Suq, potentially indicative of population dispersal and a decline in physical labor associated with a loss of large-scale construction projects (Blau 2001a, 2007).

Biogeochemical evidence does not support the hypothesis that local populations became more mobile during this time. Mean 87Sr/86Sr ratios from local individuals

334 interred at Shimal 95 (0.708814  0.000024, 1; n=2), Shimal 103 (0.708828 

0.000039, 1; n=6), and Fujairah (0.708640  0.000070, 1; n=8) all display very little isotopic variability and hence do not suggest a nomadic existence. 18O values are similarly limited in scope, with intra-site values spanning less than 2‰, indicating that locals exploited isotopically-similar water sources. Does this mean that under no circumstances were these populations mobile? Not necessarily. The isotopic similarities seen here may in part be the result of comparable regional geology and precipitation, particularly along the coast, which may obscure some aspects of mobility from biogeochemical techniques. Finally, inter-tooth variability among local individuals for strontium and oxygen isotopes was low, adding further support to the assessment that inhtabitants of southeastern Arabia during the Wadi Suq were not particularly mobile.

While the hypothesis that regional mobility increased during the second millennium as local populations became more mobile must thus be rejected based on biogeochemical evidence alone, it should be recognized that the sample size presented here is small, and that abundant archaeological evidence for this increasingly nomadic lifestyle cannot be disregarded. Nevertheless, it also calls into question just how dramatic the transition from the Umm an-Nar to the Wadi Suq period actually was, and suggests that the often-depicted collapse of Umm an-Nar culture as ‘sudden’ and ‘drastic’ may have in fact been more gradual and not as extreme as depicted in the archaeological record, with a certain continuity of lifestyle that may have included at least some semi- sedentary practices.

335 Hypothesis IIa: Non-locals will be absent from Wadi Suq communal tombs.

Three possible non-locals were identified out of a total of 17 Wadi Suq individuals sampled in this study: one from Shimal 103 (SH 214), one from Bidya 1 (Bid

245/246), and one from Dibba 76 (Dib 254).

From Shimal, individual SH 214 did not possess a foreign 87Sr/86Sr ratio but did exhibit a 18O value considerably higher relative to the other individuals interred not only in Shimal 103 but also in the nearby tombs of Shimal 95 and Unar 1. While all 13C ratios at Shimal 103 were relatively similar, this individual also produced the lowest 13C value of any in Shimal 103 and is the only individual whose diet was likely completely

13 reliant on C3 foods. While alone, this  C ratio could not be used to identify SH 214 as an immigrant, its 13C value coupled with its apparently non-local 18O value does bring to light the possibility of a non-local geographic residence in childhood. Further mortuary evidence, particularly the occurrence of an unusually high concentration of four

Dilmun-type ceramic vessels within the tomb itself, has led some to speculate that these objects were intentionally placed with the burial of a non-local trader from the island of

Bahrain (Velde nd-b). While it is unlikely that the foreigner identified here is from

Bahrain based on comparative human and faunal 87Sr/86Sr and 18O ratios from the A’ali

Mound Field and Barbar Temple, this individual nevertheless appears to have consumed an isotopically different water source and was more heavily reliant on C3-based resources.

The immigrant from Bidya is an especially interesting case, in part because both

M1 and M2 are present, allowing a more thorough examination of this individual’s life history. While a local strontium isotopic value was assigned to enamel from the first 336 molar, the 87Sr/86Sr ratio from the second molar deviates substantially from the Fujairah local range, indicating that between approximately 3 and 7-8 years of age, this individual migrated to a region geologically dissimilar from that of Fujairah. A similarly foreign

87Sr/86Sr value is also noted from Tell Abraq (TA 165). Nevertheless, at some point in this individual’s life (or death), he or she returned to Bidya and was interred in a local tomb. This case indicates that mobility and migration were not necessarily activities limited to adult traders, but may have involved children and families as well.

Conversely, the immigrant to Dibba 76 displays an elevated M1 87Sr/86Sr ratio.

This 87Sr/86Sr value not only falls outside the local range for all of Fujairah, but also represents one of the highest ratios of this study, second only to individual TA 161 from

Tell Abraq. However, as this ratio closely approaches other, Umm an-Nar signatures from the western coast of the Emirates, it is possible that Dib 254 represents a regional and not interregional migrant. Interestingly, this individual also possesses highest 13C ratio of any tomb member from across Fujairah; while not an extreme deviant, this offers additional evidence of at least some residential mobility in the Emirates during the Wadi

Suq, albeit (likely) from sites within the peninsula itself.

As with non-locals identified from Umm an-Nar tombs, these three Wadi Suq individuals were not designated as foreign in any way by burial position, grave good possession, or placement within the tomb, and were instead undifferentiated from other commingled remains belonging to individuals from the local community. Consequently, the hypothesis that non-locals are absent within Wadi Suq collective graves is not supported by the biogeochemical data and suggests that this region continued to be actively engaged in interregional interaction despite drastic changes in the archaeological

337 record. It is interesting to note that while the majority of archaeological evidence seems to indicate that the Wadi Suq represents a period of cultural isolation in the Oman

Peninsula, contact was briefly maintained with the Indus Valley just after the collapse of the Mature Harappan culture, as evidenced by the presence of a few Post-Harappan sherds at Tell Abraq as well as a Harappan jar and Indus-style weight in Tomb 6 of the

Shimal Plain (de Cardi 1989; Potts 2001). Subsequently, it appears that trade was not completely disrupted and would have allowed for the continued presence of non-locals in local funerary structures of southeastern Arabia.

Hypothesis IIb: Dietary variability decreased during the Wadi Suq period.

According to archaeological sources, subsistence practices employed by the Wadi

Suq inhabitants of the Oman Peninsula were not as variable as in the preceding Umm an-

Nar period, not only because of shift towards a more mobile way of life, but also because of increasing aridity in the region, leading to increased dependence on marine resources and (with the exception of rare sites like Tell Abraq whose wells were not adversely affected by dropping water tables) a decreased dependence on agro-pastoralism. It was anticipated that intra- and inter-site 13C values would correspondingly experience a decline in variability in response to these changes.

Variability was considerably diminished relative to that seen during the Umm an-

Nar period. Shimal 95 (-11.0 to -10.7‰), Shimal 103 (-12.6 to -9.9‰), and Fujairah

(-11.8 to -9.4‰) each exhibit limited 13C ranges spanning less than 3‰. These values suggest a mixed but predominantly C3-based diet, signifying an increasingly heavy reliance on C3 agricultural products. A considerable decrease in the consumption of 338 marine and/or C4-based plants as well as primary and secondary products from domestic animals is evident. While again, small sample sizes may skew actual ranges present at this time, these data tentatively suggest a local community with a relatively uniform, non- varied diet during the Middle Bronze Age that consisted primarily of C3 cultivars like dates, wheat, and barley, despite the coastal provenance of many of these sites. No statistically significant differences were found between Wadi Suq western (Shimal 95 and 103) and eastern (Fujairah) sites, but with the exception of Unar 1 (on the Shimal

Plain in close proximity to Shimal 95 and 103), significant disparities in diet were found between Wadi Suq sites and all Umm an-Nar sites, likely reflecting both spatial and temporal dissimilarities.

This biochemical assessment of diet directly contradicts previous archaeological and bioarchaeological reports. A zooarchaeological analysis of the shell mounds at the

Shimal settlement shows that maritime resources, including fish and shellfish, dominated faunal assemblages (approximately 90%), with terrestrial mammals representing only a small portion of these deposits (Vogt and Franke-Vogt 1987; Grupe and Schutkowski

1989; Potts 1990; von den Driesh 1994; Glover 1998; Vogt 1998). No evidence of oasis agriculture has been unearthed on the Shimal Plain, again indicative of a dietary shift from agro-pastoralism towards marine resources (Potts 1990; Hellyer 1998). Finally, relative to the Umm an-Nar occupation at Tell Abraq, Wadi Suq levels show an increasing dependence on maritime resources (Potts 1995, 1997a).

Also at Shimal, trace element analysis of skeletal material from three tombs dating to the early, middle, and late Wadi Suq display a gradual shift from a more varied diet with contributions from both terrestrial and marine to a diet dominated by marine

339 foods, consistent with the massive shell middens in the area (Grupe and Schutkowski

1989). However, as was previously noted, bone is extremely susceptible to diagenesis, and as many trace elements have been shown to be unreliable indicators of diet (e.g.,

Ezzo 1994), these results must be approached with caution.

The Shimal skeletons were additionally evaluated for dental wear (Wells 1984;

Schutkowski and Herrmann 1987; Littleton and Frohlich 1993) as well as dental caries and antemortem tooth loss (Schutkowski and Herrmann 1987; Littleton and Frohlich

1993). Minimal wear, coupled with abundant fish remains on site, led the authors to conclude that Wadi Suq peoples consumed a diet heavily reliant on shellfish and fish and that cereals played only a minor role, while low caries and calculus frequencies suggest a decreasing dependence on carbohydrate-rich foods like dates (Wells 1984; Schutkowski and Herrmann 1987; Littleton and Frohlich 1993). (Interestingly, however, relatively high frequencies of AMTL at both Shimal 95 (15.8%) and 103 (21.9%) may at least partially explain the lack of caries in this group.)

Blau (2001a, 2007) similarly reports on a decline in dental attrition from the Umm an-Nar to Wadi Suq period. This trend, together with an increase in the frequencies of linear enamel hypoplasias (LEH), antemortem tooth loss (AMTL), and cribra orbitalia, characterized the dental and skeletal health of Wadi Suq skeletons. Both LEH and cribra orbitalia reflect stress during early childhood, and Blau (2001a, 2007) proposed that with a shift from a broad Umm an-Nar diet (with contributions from a variety of terrestrial and maritime sources) to a more narrow Wadi Suq diet dominated by seafood, such skeletal indicators may signify poor nutrition. Moreover, AMTL occurred with increasing frequency while dental caries and wear simultaneously decreased in number, leading

340 Blau (2001a) to conclude that abscesses may have facilitated AMTL, and that all together, this biological evidence signifies a general change in subsistence strategy and diet from terrestrial to marine resources.

Despite these interpretations, some archaeological evidence does hint at contiuity in diet between the Early and Middle Bronze Age. Although not typically seen as representative of most Wadi Suq sites, at Tell Abraq, abundant faunal remains from domesticated animals illustrate that pastoralism was not completely abandoned during the second millennium (Potts 1990). Large numbers of dates and grinding stones suggest some stability in agricultural practices as well; in fact, a small number of grinding stones were also recovered at Shimal (Potts 1990).

Interestingly, this dietary evaluation using stable carbon isotope ratios fits with associated evidence of mobility using strontium and oxygen isotope signatures, which illustrate that local individuals from Wadi Suq tombs appear to have (a) not been highly mobile and (a) had consistent access to isotopically similar sources of food and water that did not vary as one would expect in a more mobile population. This interpretation corresponds with the ranges of carbon ratios seen in the same Wadi Suq individuals, whose values are most closely associated with terrestrial C3 plants available via oasis agriculture and that do not possess a variability consistent with higher degrees of mobility. This appraisal is further supported by re-examining the 13C values from carbonized dates recovered at Tell Abraq (Table 6.10). These date stones, which range in value from -24.7 to -21.7‰, correspond to dietary carbonate values of approximately

-13.2 to -10.2‰ in humans; when compared with the total Wadi Suq range of -12.6 to

-9.9‰, these ranges overlap considerably. Conversely, the elevated 13C ratios produced

341 by a diet dominated by the consumption of marine foods and grazing herd animals are not present, suggesting that these strategies did not play a major role. Together, then, the biogeochemical evidence suggests a relatively sedentary community who continued to rely on agriculture as its main subsistence strategy.

Modeling the Past: Using Stable Isotopes to Evaluate Theory

World-System’s Theory

Biogeochemical, archaeological, and mortuary data may be used to test the applicability and explanatory power of world-systems theory in interpreting past social structure. Frank (1993) has made a case for a single Bronze Age world-system that has persisted into the present day, drawing extensively on demographic reconstructions and artifactual evidence from across the Near East, South Asia, and northern Africa and identifying relatively simultaneous cyclical economic expansions and downswings experienced by both the core and the periphery (see also Edens and Kohl 1993 for similar observations on these cycles in the Near East), thus illustrating a high degree of economic interconnectivity and the presence of a world-system. However, if such a core hegemonic structure were truly in place, why then do we see such little material evidence of dependence on Mesopotamia or the Indus Valley in the Oman Peninsula, and instead find distinctive local ceramic traditions, architecture, metallurgic practices, and mortuary customs signifying a relatively autonomous society? These seemingly contradictory datasets suggest that we must use caution in the application of modern world-systems concepts to pre-capitalist societies. Chase-Dunn and Hall (1993) argue this very point, emphasizing that typical world-systems patterns, particularly with regards to core

342 domination and peripheral subordination, must be re-evaluated in any investigation of past interregional exchange. Potts (1990) contends that such terminology be thrown out altogether and discourages thinking about the economic landscape of the Bronze Age in terms of core ‘megacenters’ and peripheral outposts, arguing that the geographical and temporal intricacies of these exchange relations were far too complex for a formulaic core/periphery model to explain with any accuracy.

Furthermore, the presence of non-locals in the highly standardized Umm an-Nar tombs of southeastern Arabia would not suggest subordination by foreigners from the core as world-systems theory would predict, since no political or ideological influence appears to have visibly altered these mortuary customs. Instead, the subtle incorporation of immigrants seen in both the Umm an-Nar and Wadi Suq periods would imply an adoption of local burial practices by non-locals, perhaps in an attempt to promote cooperative trade relations. Such a mortuary ‘compromise’ may also suggest that definitions of kinship became more flexible in an attempt to cope with an influx of foreign immigrants, and more generally, the increasing economic demands of their involvement in complex interregional systems at the local level. As such, traditional definitions of biological kinship may have been broadened, and a sort of ‘fictive kinship’ developed. A term derived from ethnographic literature, fictive kinship refers to a relationship between one or more individuals not biologically related but who nevertheless possess a quasi-kinship bond that usually serves some emotional, ritual, or economic function (Norbeck and Befu 1958). The practice of fictive kinship has been recorded around the globe, including Asia, Europe, Latin America, and North America

(Native Americans), and is often incorporated into a formal cultural system in which

343 these ficticious relationships are given recognized names (Freed 1963). Such relationships have been noted ethnographically amongst Arab tribes across the Middle

East (e.g., Salim 1962). Archaeologically, the existence of fictive, or “practical”

(Bourdieu 1977) kin has also been posited amongst the Neolithic inhabitants of

Çatalhöyük in south-central Turkey, where phenotypic patterns of metric and nonmetric dental traits show that biological affinity did not play an important role in determining the co-interment of individuals within the same house (Pilloud and Larsen 2011). Here, the implementation of practical kin relationships may have been prompted by economic activities such as planting and harvesting that required the cooperation of large groups, or by social motives related to inheritance or religion (Pilloud and Larsen 2011).

Fictive kinship represents a valuable cultural tool in the flexibility that it brings to social relationships, so that formal, kin-like bonds can be invented as opportunities present themselves. In Bronze Age Arabia, fictive relationships may have arisen out of a necessity for a more formalized and official union to exist between locals and non-locals.

Such a bond may have fulfilled an economic role as non-locals became increasingly involved in everyday commerce, but it also served a ritual role in permitting the entrance of non-locals into mortuary contexts, thereby loosening the restraints of tomb membership, which presumably were previously restricted to local community members.

Distance-Parity Model

The economic and political cycles as described by Frank (1993) may also be utilized in the evaluation of the distance-parity model; those peripheries farther from the core centers of Mesopotamia and Harappa should not necessarily show complete

344 synchronization with major economic upswings and downswings if distance does indeed impact economic and even hegemonic control. However, asynchronous series are not apparent in the archaeological record, with ‘peripheries’ of all distances in the Persian

Gulf seemingly intertwined into these interregional exchanges based on archaeological evidence (Frank 1993). Nonetheless, the seemingly autonomous nature of the Oman

Peninsula based on local, internal developments in material and mortuary culture may speak to the flexibility permitted by its distance from core centers despite economic ties.

Further archaeological investigation into the copper trade with Mesopotamia may reveal that the Oman Peninsula used this highly sought-after resource to its advantage in manipulating social interaction and organization during the Umm an-Nar period.

Along these lines, if the Oman Peninsula represented a region too distant for core centers to maintain any definitive control over, core individuals buried here may not have had as much power over their choice of interment. Biogeochemical evidence confirms that non-locals were indeed buried in southeastern Arabia with no distinguishing mortuary features – or, if exotic grave goods were once associated with these immigrants, these associations were soon lost as tomb contents were mixed to make room for additional tomb members. The control local communities possessed over their own mortuary behavior seemingly confirms a degree of autonomy that speaks to an overall lack of hegemonic dominance by centers who simultaneously exerted at least some control over economic facets of the same region.

345 Trade Diaspora Model

While the influence of foreign culture in the Oman Peninsula is evident based on the inclusion of non-local pottery, personal ornaments, and other objects deposited with the dead in local funerary monuments, speculation of the existence of foreign merchants and/or enclaves in local contexts has existed for decades. As early as 1971, evidence of the Jemdet Nasr culture from Mesopotamia in Oman led During Caspers (1971) to suppose the possibility of a merchant colony or trading post controlled by Mesopotamian settlers in the early centuries of the Hafit period to manage the growing copper trade.

After her excavation of the Wadi Suq Tomb 6 in Shimal, de Cardi (1989) argued that the presence of both a Harappan weight and a Harappan jar indicated ownership by an individual originating from the Indus Valley, likely a merchant, and that this person’s residence at Shimal had ultimately given him/her access to burial rights with the local community. At the Umm an-Nar cemetery of Wadi ‘Asimah in the Emirate of Ras al-

Khaimah, unusual stone alignments, platforms, and associated tombs are without parallel anywhere in the Oman Peninsula, and in conjunction with a significant quantity of metal objects and a pedestal chalice similar to those unearthed in Iran as well as Harappan vessels comprising 20% of all containers on site, suggested to Vogt (1994) a significant foreign presence, if not actual foreign settlers, residing in the area. Edens (1992) rightfully pointed out that such foreign contact was not necessarily restricted to economic dealings, and that political relationships, diplomatic envoys, exogamous marriage practices, and hegemonic strategies may have all contributed to the presence of non- locals in the Oman Peninsula.

346 Parallels between manufacturing and stylistic techniques in ceramic sherds found in southeastern Arabia have also instigated hypotheses of a foreign presence in or near local communities. De Cardi and Doe (1971) suggested that Bampur (Iranian

Baluchistan) ceramics at both Hili and on Umm an-Nar Island were of such similar design and form that their presence might be explained by migrants from this region to southeastern Arabia. Additionally, two new trends in local ceramic production during the second millennium BC directly mirror that of a developing Harappan tradition. The first modification consisted of “string-cut bases”; while rare in the Umm an-Nar period, this technique became representative of the Wadi Suq, particularly at Tell Abraq (Potts 1990,

1993d). Secondly, impressions of string, occurring in one or multiple rows, can be seen on large jars requiring roped support during the firing process (Potts 1993c). While this is not attested to at Tell Abraq, other sites, including Maysar 1, Hili 8, and Ras al-Junayz, exhibit this feature (Edens 1993). Subsequently, Potts (1993c) has interpreted these second millennium innovations as the product of Indus Valley colonists contributing to local knowledge of ceramic production and a resultant adoption of Harappan techniques via acculturation.

The sudden emergence of a local Umm an-Nar ceramic tradition in the mid-third millennium BC has also caused archaeologists to inquire after the actual origins of these innovations. Southeastern Arabia represents a fascinating case study in that, despite multiple millennia of contact with civilizations actively engaged in ceramic production – vessels of which made their way to the Oman Peninsula as early as the ‘Ubaid period – there appears to have been no local interest in the manufacture of such vessels until the

Umm an-Nar period (Potts 2005). The evolution of production techniques, vessel shape,

347 and motifs characteristic of a particular culture can normally be traced over time, reflecting a growing knowledge base and increased expertise in the manufacture of pottery; however, in southeastern Arabia, these skills already appear fully formed with the advent of ceramics (Potts 2005). This abrupt sophistication, along with the uncanny resemblance of this local pottery (both technically and stylistically) to that produced in southeastern Iran, prompted Potts (2005) to conclude that Iranian potters migrating to the

Oman Peninsula instigated this local ceramic industry.

In addition to speculation of non-local settlement within southeastern Arabia, the

Umm an-Nar inhabitants of the Oman Peninsula may have established merchant colonies themselves. A huge number of Umm an-Nar ceramic vessels manufactured in the Oman

Peninsula were recovered from Qala’at al-Bahrain, the Barbar Temple, and to a lesser extent, at Saar, and were a common inclusion in Dilmun graves (Bibby 1986; Vogt 1996;

Carter 2003b). A similarly widespread occurrence of Umm an-Nar soft-stone vessels throughout Bahrain paralleled the distribution of Umm an-Nar pottery in Dilmun (David

1996), while isolated assemblages consisting of only Umm an-Nar sherds have been reported at Hamad Town (Lowe 1983; Vogt 1996). In addition, a minority of Dilmun funerary monuments strongly resemble the collective tombs of the Umm an-Nar period in southeastern Arabia, as illustrated by their circular construction and internal dividing walls (Crawford 1998; Carter 2003c). Furthermore, an architectural technique known as ashlar masonry represents a development unique to the Oman Peninsula and was not found outside it until the construction of the Barbar Temple II in Bahrain in the early second millennium BC, implying the deliberate adoption of Umm an-Nar construction methods or possibly the actual presence of builders from southeastern Arabia (Crawford

348 1998; Carter 2003c). Consequently, the frequent occurrence of Umm an-Nar artifacts in conjunction with the emulation of Umm an-Nar architectural techniques and mortuary traditions as well as sizeable cemeteries that appear too extensive for the native Dilmun population all suggest the possibility of a physical presence of Umm an-Nar merchants or colonists in Dilmun, perhaps as a means of facilitating the trade of copper with

Mesopotamia in the north (Vogt 1996; Carter 2003c).

Speculation towards the existence of foreign enclaves is not limited to the southern reaches of the Persian Gulf but appears well documented archaeologically by

Mesopotamian colonists. In an effort to obtain raw materials lacking in its alluvial plains, the Uruk expansion out of southern Mesopotamia not only witnessed the establishment of numerous, isolated trading outposts in northern Syro-Mesopotamia and eastern Iran but also small, intrusive colonies situated within larger indigenous settlements as far away as

Anatolia in the west and Iran in the east (Algaze 1989, 2001, 2002a; Stein 1999b). These merchant colonies typify the trade diaspora model outlined in Chapter 4, displaying a distinctive cultural identity within a larger, foreign setting as illustrated by architecture, mortuary practices, and stylistic elements characteristic of that homeland’s material culture, and not simply isolated imports (Stein 1999a). For example, at the fourth millennium BC site of Hacınebi Tepe in eastern Anatolia, an Uruk residential enclave seemingly housing a small number of traders from Mesopotamia maintained a distinct and autonomous identity – expressed with both domestic and public forms of Sumerian architecture, administrative tools, ceramics, and other iconographies – despite its position within the northeastern portion of an Anatolian community serving as its ‘host’

(Stein 1999b, 2001, 2002a; Algaze 2001). Similarly, in central-western Iran, the pre-

349 existing indigenous settlement of Godin Tepe (Godin IV) surrounded a small

Mesopotamian fortification (Godin V) with a cultural assemblage dominated by Uruk wares consistent with the presence of a trading enclave (Weiss and Young 1975; Young

1986; Algaze 1989; Badley 1998; Stein 2002a). While these colonies were likely established as a means of exerting control over distant exchange networks, they by no means appeared to have subjugated the indigenous communities in which they were placed; instead, the continued existence of these vulnerable Mesopotamian enclaves would have required goodwill between the two ethnic groups placed in such close proximity to one another (Stein 2002a).

Few studies of biological affinity among the individuals interred in the Oman

Peninsula have been performed that might shed light on the presence of foreigners at these sites and in local tombs, largely due to the poor preservation and fragmentary nature of human skeletal material from the Bronze Age. Despite these challenges, a few preliminary examinations show conflicting results. Højgaard’s (1980, 1981) evaluation of isolated teeth from Tombs I, II, and V on Umm an-Nar Island using dental nonmetric traits, including agenesis of the third molar, led her to conclude the presence of

“Caucasoid but no Mongoloid traits” (1980:355) among an extremely homogeneous group of people. A subsequent appraisal of reconstructed crania from these same tombs by Kunter (1991) similarly revealed uniform morphological features. Both of these assessments generally agree with the isotopic data derived here, which display strontium, oxygen and carbon ratios consistent with local inhabitants in all three Early Bronze Age tombs.

350 At Tell Abraq and Unar 1, craniometric and nonmetric analyses performed by both Wright (unpublished; see Potts 1993d) and Blau (1998) suggest that while the majority of individuals exhibit traits of European descent, 25-38% possessed non-

European morphological features possibly indicative of either (a) admixture between different populations or (b) the actual occurrence of foreigners in these tombs.

Interestingly, both strontium and oxygen isotope ratios exposed the presence of two non- locals interred at Tell Abraq, corroborating the conclusions reached by Wright and Blau that different populations were interacting (and possibly exchanging genes) with one another. Though limited in scope due to severe dental wear on the majority of teeth recovered from these tombs, further evaluations of dental nonmetric traits (e.g.,

Carabelli’s cusp, molar cusp number, and molar groove pattern) of individuals from numerous Umm an-Nar burial monuments by Blau (1998) likewise revealed trait complexes associated with western Eurasian populations. Consequently, evidence from studies in both biodistance and stable isotope analysis shows a largely homogeneous population during the Umm an-Nar period, but one that appears to be interacting with other populations to at least some degree.

During the subsequent Wadi Suq period, nonmetric traits from the Site 1 tomb on the Shimal Plain were reported by Wells (1984) as reflective of a homogeneous population structure, and El-Najjar (1985:39) described the dentition of individuals from

Tomb A at Hili North as belonging to the “Caucasian racial stock.” Blau (1998) also described crania from Shimal 602 and Sharm as suggestive of “Caucasian” origins, although 20% of those from Sharm displayed “ambiguous” features potentially indicative of some population heterogeneity. These conclusions correspond somewhat with the

351 isotopic data presented here, as the Shimal Plain Wadi Suq tombs of Shimal 95 and 103

(with one exception: SH 214) displayed local strontium and oxygen values, in contrast to the Wadi Suq tombs of the east coast (where Sharm is located), which produced two non- locals out of the seven total individuals sampled, signifying a foreign presence of almost

30%. Although it must be noted that in all Wadi Suq studies of biodistance, extremely small sample sizes preclude definitive conclusions from being reached, it appears that – as in the preceding Umm an-Nar period – while the majority of individuals interred in the

Emirates during the Wadi Suq were local, some degree of population heterogeneity and a non-local presence is evident.

However, archaeological, and particularly mortuary, evidence from the Oman

Peninsula does not meet Stein’s three criteria for the existence of a trade diaspora (see

Chapter 4). At the height of these interregional exchange networks with Mesopotamia, the Indus Valley, Dilmun, Elam, Central Asia, and the Indo-Iranian borderlands during the Umm an-Nar period, tower tombs unique to the region were in use across southeastern Arabia. Despite this impressive geographic distribution, these tombs retained a high degree of standardization found at all sites, including collective burial, a circular structure, and the unique ring wall construction of outer dressed stone and an inner layer of unworked limestone (Potts 1990; Blau 2001b). While some foreign artifacts were deposited in these burial chambers, local assemblages dominated the grave furniture uncovered. Although the possibility exists that individuals were initially interred with grave goods particular to them (e.g., de Cardi 1989), commingling within the tombs precludes their associated with any particular individual. Subsequently, an expression of foreign identity as emphasized through distinct mortuary ritual is not

352 present in the Umm an-Nar during this time, precluding the presence of a true trade diaspora.

The one exception to this standardization lies in the two unusual rectangular/ ovoid graves found at Mowaihat Tomb B and Hili N, and it may be argued that this distinct style of burial speaks to the potential existence of a foreign enclave at both sites.

However, like their circular counterparts, these graves held hundreds of disarticulated individuals, displayed no evidence of selective burial treatment, and largely contained locally manufactured goods (Al Tikriti and Mery 2000). Moreover, the presence of Wadi

Suq-like objects in the most recently used layer of the grave at Hili N, in addition to the shape of the tomb itself, suggests that this tomb architecture was not a radical departure from local norms but instead represented a transitory form between the circular Umm an-

Nar and generally ovoid Wadi Suq tombs (Crawford 1998). While significant mortuary variability is introduced with the advent of the Wadi Suq period, there is no evidence that any of these funerary monumental forms reflect the deliberate maintenance of foreign identity. Thus, in its strictest sense, the trade diasporas model cannot be applied in either the Umm an-Nar or Wadi Suq periods in the Oman Peninsula.

This does not mean that the trade diasporas model is useless in conceptualizing exchange relationships in Bronze Age Arabia. Instead, we may view it as potentially working in a much less formal way, where Arabian communities may have dealt with individual merchants or ambassadors who readily adopted the practices of the host community, even in death, in lieu of visibly maintaining their own social identities. Such a hypothesis is supported based on an apparent lack of foreign enclaves and overall foreign influence on settlement construction and mortuary customs as well as by isotopic

353 evidence for the internment of non-locals at Mowaihat, Tell Abraq, Shimal 103, Bidya 1, and Dibba 76 without any effort to distinguish or isolate these individuals in any way. As

Algaze (1989) has argued, it would have been immensely difficult for ‘core’ areas like

Mesopotamia to sustain control of the market, defend an outpost, and efficiently obtain raw materials so far from its center. Instead, cooperating with indigenous communities who have already established local means of resource exploitation and transportation represents a more cost-effective way of acquiring these valued goods. This strategy permits the continued autonomy of these local groups and does not require the kind of colonialism, domination, and subjugation brought to mind by European models of imperialism (Stein 2002a).

Cleuziou’s Social Dynamics Model

Cleuziou (2007) viewed the so-called breakdown of societal organization in southeastern Arabia as the result of a fundamental conflict between traditional kin/tribal systems and developing social hierarchies that came to a head by the end of the third millennium BC. According to Cleuziou, traditional tribal organization and kinship-based systems rooted in cooperation, redistribution, and egalitarianism would have been increasingly challenged as these elites gained power and wealth through their involvement in increasingly complex interregional exchange networks. This conflict, creating considerable stress and opposing ideological forces, rapidly propelled the Oman

Peninsula away from its path towards a state-level hierarchical system and into collapse.

Such a collapse, marking the commencement of the Wadi Suq period in the early second

354 millennium BC, exposes an Arabian economy that could not be sustained as a result of a lack of resolution between these two societal factions (Cleuziou 2007).

The biogeochemical evidence of this study can be applied to Cleuziou’s local agency model. If, as Cleuziou speculates, an increasingly unhappy and tradition-based faction emphasizing kinship eventually rose up to prevent further societal development, it is perhaps possible that this discontent could have in part stemmed from the inclusion of immigrants into local monumental tombs during the Umm an-Nar period, particularly since these inclusions do not appear in the early Umm an-Nar (e.g., Umm an-Nar Island) but only towards the end of the third millennium. By including foreigners in traditional rites and placing them side-by-side with local ancestors, this creates a message that biological kinship is no longer required for tomb membership, and although a formal relationship of fictive kinship may have been established, this was not accepted by all members of Umm an-Nar society.

However, if this anti-elite, tribal faction was so concerned with a return to traditional ways of life rooted in an ethos of biological kinship and egalitarianism, it is interesting to note that foreigners continued to be included as local tomb members well into the Wadi Suq period, after this supposed ideological uprising had challenged and overcome the developing social hierarchies of the Umm an-Nar. Shouldn’t this collapse, so centered around the consolidation of elite power through the control of exchange networks, show that foreigners – the physical embodiment of these outside economic influences that created a local social elite – are absent from Wadi Suq tombs as the traditional, kinship faction wins out? One possibility may lie with defining what constitutes traditional kinship in the first place. If practices of fictive kinship were so

355 much a part of daily life that the inclusion of non-blood relatives into local tombs would in fact represent a long-held and traditional approach that would not conflict with local customs, then perhaps this conflict acutely targeted economic developments, leaving ideas of kinship (whether biological or fictive) and tomb membership intact and thus allowing inclusive, egalitarian-based mortuary practices to continue. This resolution would account for the continued presence of non-locals in Wadi Suq tombs.

356

CHAPTER 8

CONCLUSIONS

Mobility in southeastern Arabia during the Bronze Age involved a complex arrangement of social and economic factors, all of which impacted local inhabitants of the peninsula on both a regional and interregional level. The presence of non-locals in

Umm an-Nar and Wadi Suq tombs not only confirms the intimate involvement of this so- called periphery in a pan-Gulf interaction sphere but also sheds new light on the nature of economic exchanges from this less recognized perspective. Far from a subordinate position in a hegemonic network established by larger, more complex centers like

Mesopotamia or the Indus Valley, the people of the Oman Peninsula appear to have lived relatively autonomous lives. Monumental tombs from both the Umm an-Nar and Wadi

Suq periods embody the continued maintenance of a local social identity by the inclusion of immigrants in tombs wherein the majority of tomb members represent multiple generations of local ancestors. The absence of any specific expression of foreign identity coupled with particular individuals in these graves reinforces the idea that these non- locals may have readily adopted the practices of their local host community, even in death. Such a relationship may indicate a form of fictive kinship in place as a means of more formally cementing economic ties. Subsequently, it appears that the relationship

357 between ‘core’ centers and indigenous tribal communities was more likely to involve cooperation in lieu of subjugation. Nevertheless, local communities were highly invested in these interregional systems of trade, so much so that a breakdown of economic relationships, first with Mesopotamia and then the Indus Valley, triggered a purported

“Dark Age” of cultural isolation. In addition to these external economic influences, this collapse may have also been partially influenced by internal, local agency, although biogeochemical evidence disputes the presumption of an ideological conflict between a traditional faction and a growing elite.

In summary, this dissertation suggests that despite the burgeoning exchange networks developing in the Early Bronze Age Umm an-Nar period, only 3% of tomb members were identified as non-local immigrants to the region. Mobility was relatively restricted for local inhabitants of the Oman Peninsula, likely a reflection of an increasingly sedentary lifestyle associated with a reliance on cultivars from oasis gardens and coastal gathering/fishing, although surprising given the increased involvement of the peninsula in regional and interregional trade networks. Also, despite archaeological evidence to the contrary, Wadi Suq tomb members did not appear more mobile than their

Umm an-Nar counterparts, although a statistically significant difference in mean strontium values between these groups suggests that these individuals were exploiting slightly different geographic areas. The broad, mixed C3-C4 diet of the Umm an-Nar population as gleaned from stable carbon isotope ratios fits with evidence of the employment of a variety of subsistence strategies, although preference was given to C3- based sources of food. Carbon values between this period and the Wadi Suq differed significantly as well, indicative of a considerable modification to subsistence practices

35 8 involving a greater reliance on isotopically heavier, C3-based foodstuffs and a narrowed dietary focus – not on marine resources, as predicted by the archaeological record, but on what appears to be oasis agriculture.

This study contributes substantially to the bioarchaeology of the Persian Gulf. As the first biogeochemical study in the region, this research provides critical insight into temporal patterns of mobility and diet as well as changing Bronze Age interregional relationships during a formative archaeological transition in the Oman Peninsula. As a result, these findings create an important framework against which other biological, archaeological, and theoretical questions may be posited. This work is also significant in that it takes on a broad geographic area extending across both the Middle East and South

Asia. Finally, it is important to recognize the potential of bioarchaeological data to more critically evaluate theoretical models, and from this union, get at more elusive aspects of social identity in the past, including status, power, and group membership.

A variety of limitations restrict the interpretations that can be made in this study.

Because of the commingled nature of the remains and the isolated (i.e., not in situ) condition of the majority of teeth, individual skeletons cannot be fully evaluated nor a complete biological profile developed. Additionally, because of a combination of environmental factors and mortuary behaviors, bone is poorly preserved and no longer contains collagen, precluding DNA extraction as well as an analysis of stable carbon and nitrogen isotopes to give further insight into relatedness, sex, diet, trophic level position, and weaning practices. Furthermore, comparative samples were largely limited to fauna, and while faunal ratios are useful for defining local strontium ranges and for

359 reconstructing carbon trophic level systems, they cannot be directly compared with human oxygen values due to metabolic differences between humans and animals.

Another major issue relates to sampling. The poor preservation, frequent tomb reuse, and smaller numbers of interred individuals in Wadi Suq burial monuments all contribute to the low sample size included in this study. The inclusion of supplementary skeletons from these and other Middle Bronze Age sites in the UAE would significantly improve the tentative conclusions drawn from the small data pool presented here, although more excavations are needed to increase the number of human remains available for analysis. For future research, additional human samples should be obtained from those involved in trade with the Oman Peninsula, including Mesopotamia, the Indus

Valley, Elam, Central Asia, and Dilmun. With these samples, a more thorough knowledge of the isoscapes that make up the Persian Gulf can be achieved, and the direction of interregional exchange perhaps better understaood as the geographic origins of non-locals become more easily identifiable.

Isotopic sampling of enamel should also be expanded to include the earlier

Neolithic and Hafit periods. Neolithic isotopic signatures of paleomobility should reflect a mobile, pastoral lifestyle with consistent seasonal migrations between the mountains and the coast. Such a sample would make a useful comparison with biogeochemical data from the supposedly semi-nomadic Wadi Suq peoples. Samples from Hafit and later transitional beehive tombs also represent a fascinating liminal period between the

Neolithic and Umm an-Nar, during which time major cultural changes had taken place, although to date, few such tombs have been excavated, and human remains are thus exceedingly rare. Other sampling avenues might include a collection of modern soil,

360 water, and faunal samples as a supplement to archaeological values and as a means of refining our understanding of the regional bioavailability of these isotopes.

This study would benefit from other areas of future research as well. Biodistance analysis using cranial nonmetric as well as dental metric and nonmetric traits would offer intriguing evidence as to who was buried in these tombs. Were these biological kin groups, or does fictive kinship play a larger role than the isotope data can tell us? Did tomb membership and biological affinity differ between the Umm an-Nar and Wadi Suq?

Such data would contribute to a greater understanding of genetic relatedness both at the site and regional level and would complement the identification of non-locals by stable isotope analysis. Moreover, temporal trends in femoral diaphyseal cross-sectional geometry between Umm an-Nar and Wadi Suq tomb members would shed additional light on possible changes in mobility between the Early and Middle Bronze Age (Ruff et al. 1984; Larsen 1997; but see also Pearson and Buikstra 2006).

This research has drawn on biogeochemical techniques to elucidate patterns of mobility, interregional exchange, and the construction of social identity in mortuary settings. It is hoped that this bioarchaeological analysis will contribute new perspectives to the interpretation of the archaeological record and the socioeconomic changes that so drastically changed the cultural landscape of the Bronze Age in southeastern Arabia.

361

REFERENCES CITED

Abu Dhabi Authority for Culture & Heritage. 2010. Jebel Hafit tombs. . Accessed 5/23/2011.

Adams R. 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. 1993. The Uruk world system: the dynamics of expansion of early Mesopotamian civilization. Chicago: University of Chicago Press.

Al Farraj A. 1995. Geomorphology of the Wadi Bih, Ras al-Khaimah, UAE. Ph.D. Dissertation. Liverpool: University of Liverpool.

Al Farraj A, Harvey A. 2004. Late Quaternary interactions between aeolian and fluvial processes: a case study in the northern UAE. J Arid Environ 56(2):235-248. al-Jahwari N. 2009. The agricultural basis of Umm an-Nar society in the northern Oman Peninsula (2500-2000 BC). Arab Archaeol Epigr 20(2):122-133.

Al Saad A, Yasin W. 2005. Hafit cairns. Al Ain National Museum. Department of Antiquities and Tourism. . Accessed 5/17/2010.

Alt K, Vach W, Frifelt K, Kunter M. 1995. Familienanalyse in kupferzeitlichen Kollektivgrabern aus Umm an-Nar; Abu Dhabi. Arab Archaeol Epigr 6:65-80.

Al Tikriti W. 1982. Reconsideration of the late fourth and third millennium B.C. in the Arabian Gulf with special reference to the United Arab Emirates. Ph.D. Dissertation, Department of Archaeology. Cambridge: University of Cambridge.

362 Al Tikriti W. 1985. The archaeological investigations on Ghanadha Island 1982-1984: further evidence for the coastal Umm an-Nar culture. Archaeol United Arab Emirates 4:9-19.

Al Tikriti W. 1989a. Umm an-Nar culture in the northern Emirates: third millennium BC tombs at Ajman. Archaeol United Arab Emirates 9:89-98.

Al Tikriti W. 1989b. The excavations at Bidya, Fujairah: the 3rd and 2nd millennia B.C. culture. Archaeol United Arab Emirates 5:101-114.

Al Tikriti W, Mery S. 2000. Tomb N at Hili and the question of the subterranean graves during the Umm an-Nar period. Proc Sem Arab Stud 30:205-219.

Ambrose S. 1987. Chemical and isotopic techniques of diet reconstruction in eastern North America. In: Keegan W, ed. Emergent horticultural economies of the eastern woodlands. Carbondale: Southern Illinois University of Carbondale: Center for Archaeological Investigations. p 87-107.

Ambrose S. 1990. Preparation and characterization of bone and tooth collagen for isotopic analysis. J Archaeol Sci 17:431-451.

Ambrose S, Norr L. 1993. Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In: Lambert J, and Grupe G, eds. Prehistoric human bone: archaeology at the molecular level. New York: Springer-Verlag. p 1-37.

Amiet P. 1975. A cylinder seal impression found at Umm an-Nar. East and West 25:425- 426.

Amiet P. 1985. Quelques temoins des contacts de Suse avec les pays du Levant au IIIe et IIe millenaires. In: Durand J-M, and Kupper J-R, editors. Miscellanea Babylonica: melanges offerts a Maurice Birot. Paris: Editions Recherche sur les Civilisations. p 9-15.

Amiet P. 1986. Susa and the Dilmun culture. In: Al Khalifa H, and Rice M, eds. Bahrain through the ages: the archaeology. London: Kegan Paul International. p 262-268.

Amirkhanov H. 1996. Bilinear cultural parallelism in the Arabian Early Neolithic. In: Afanas'ev G, Cleuziou S, Lukacs J, and Tosi M, eds. The prehistory of Asia and Oceania: XIII International congress of prehistoric and protohistoric sciences. Forli: ABACO. p 135-139.

Andersen H. 1986. The Barbar Temple: stratigraphy, architecture and interpretation. In: Al-Khalifa H, Rice M, eds. Bahrain through the ages: the archaeology. London: Kegan Paul International. p 166-177.

363 Andersen H, Højlund F. 2000. The Barbar temples. Traces of paradise: the archaeology of Bahrain 2500 BC - 300 AD. London: The Bahrain National Museum. p 89-92.

Andersen H, Højlund F. 2003. The Barbar temples, vol. I. Aarhus: Jutland Archaeological Society.

Anonymous. 1994. At Tell Abraq - the earliest recorded find of polio. Sydney, Australia: The University of Syndey - Research, External Relations Division. p 20-21.

Aspinall S. 1998. Harvesting the sea. In: Hellyer P, ed. Waves of time: the maritime history of the United Arab Emirates. London: Trident Press. p 150-165.

Aufderheide A. 1989. Chemical analysis of skeletal remains. In: İşcan M, Kennedy K, eds. Reconstruction of life from the skeleton. New York: Wiley-Liss. p 237-260.

Baadsgaard A. 2008. Trends, traditions, and transformations: fashions in dress in Early Dynastic Mesopotamia. Ph.D. Dissertation, Department of Anthropology. Philadelphia: University of Pennsylvania.

Bajjali W, Abu-Jaber N. 2001. Climatological signals of the paleogroundwater in Jordan. J Hydrology 243(1-2):133-147.

Barker D. 2002. Wadi Suq and Iron Age period ceramics from Sharm, Fujairah (U.A.E.). Arab Archaeol Epigr 13(1):1-94.

Barker D. 2004. Miniature shell and bone 'hilts' or 'pulleys' from Sharm. Arabian Archaeology and Epigraphy 15(1):115-119.

Barker D, Hassan S. 2004. Aspects of east-coast Hellenism and beyond: late pre-Islamic ceramics from Dibba 76 and Dibba Moraba'a, Fujairah, United Arab Emirates. Seminar for Arabian Studies. British Museum, London, UK.

Bar-Matthews M, Ayalon A, and Kaufman A. 1997. Late Quaternary paleoclimate in the eastern Mediterranean region from stable isotope analysis of speleotherms at Soreq Cave, Israel. Quaternary Res 47(2):155-168.

Bar-Matthews M, Ayalon A, Kaufman A, and Wasserburg G. 1999. The eastern Mediterranean paleoclimate as a reflection of regional events: Soreq cave, Israel. Earth Planet Sci Lett 166:85-95.

Barralet J, Best S, Bonfield W. 1998. Carbonate substitution in precipitated hydroxyapatite: an investigation into the effects of reaction temperature and bicarboante ion concentration. J Biomedical Material Res 41(1):79-86.

364 Baustian K. 2010. Health status of infants and children from the Bronze Age tomb at Tell Abraq, United Arab Emirates. MA Thesis. Las Vegas: University of Nevada Las Vegas.

Baustian K, Martin D. 2010. Patterns of mortality in a Bronze Age tomb from Tell Abraq. In: Weeks L, ed. Death and burial in Arabia and beyond: multidisciplinary perspectives. Oxford: Archaeopress. p 55-59.

Beale T. 2004. Excavations at Tepe Yahya, Iran, 1967-1975, vol. I: the early periods. Lamberg-Karlovsky C, ed. Cambridge: Peabody Museum Press.

Beard B, Johnson C. 2000. Strontium isotope composition of skeletal material can determine the birth place and geographic mobility of humans and animals. J Forensic Sci 45(5):1049-1061.

Beech M. 2000. Preliminary report on the faunal remains from an 'Ubaid settlement on Dalma Island, United Arab Emirates. In: Mashkour M, Choyke A, Buitenhuis H, and Poplin F, eds. Archaeozoology of the Near East IV, vol B: proceedings of the fourth international symposium on the archaeozoology of southwestern Asia and adjacent areas. Groningen: ARC Publicatie 32. p 68-78.

Beech M. 2002. Fishing in the 'Ubaid: a review of fish-bone assemblages from early prehistoric coastal settlements in the Arabian Gulf. J Oman Stud 12:25-40.

Beech M. 2003a. Archaeobotanical evidence for early date consumption in the Arabian Gulf. The date palm - from traditional resource to green wealth. Abu Dhabi: The Emirates Centre for Strategic Studies and Research. p 11-31.

Beech M. 2003b. The development of fishing in the U.A.E.: a zooarchaeological perspective. In: Potts D, Al Naboodah H, and Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident Press. p 290-308.

Beech M, al-Husaini M. 2005. Preliminary report on the vertebrate fauna from site H3, Sabiyah: an Arabian Neolithic/'Ubaid site in Kuwait. In: Buitenhuis H, Choyke A, Martin L, Bartosiewicz L, and Mashkour M, eds. Archaeozoology of the Near East VI: proceedings of the sixth international symposium on the archaeozoology of southwestern Asia and adjacent areas. Groningen: ARC-Publicaties 123. p 124-138.

Beech M, Al Shaiba N. 2004. Intertidal archaeology on Marawah Island: new evidence for ancient boat mooring sites. Tribulus 14(2):8-11.

Beech M, Cuttler R, Moscrop D, Kallweit H, Martin J. 2005. New evidence for the Neolithic settlement of Marawah Island, Abu Dhabi, United Arab Emirates. Proc Sem Arab Stud 35:37-56.

365 Beech M, Elders J. 1999. An 'Ubaid-related settlement on Dalma Island, Abu Dhabi Emirate, United Arab Emirates. Bull Soc Arab Stud 4:17-21.

Beech M, Elders J, Shepherd E. 2000. Reconsidering the 'Ubaid of the southern Gulf: new results from excavations on Dalma Island, U.A.E. Proc Sem Arab Stud 30:41-47.

Beech M, Glover E. 2005. The environment and economy of an Ubaid-related settlement on Dalma Island, United Arab Emirates. Paleorient 31(1):97-107.

Beech M, Kallweit H, Cuttler R, Al-Tikriti W. 2006. Neolithic sites in Umm az-Zamul, SE desert of Abu Dhabi, UAE. Bull Soc Arab Stud 11:17-26.

Beech M, Shepherd E. 2001. Archaeobotanical evidence for early date consumption on Dalma Island, United Arab Emirates. Antiquity 75(287):83-89.

Begemann F, Hauptmann A, Schmitt-Strecker S, and Weisgerber G. 2010. Lead isotope and chemical signature of copper from Oman and its occurrence in Mesopotamia and sites on the Arabian Gulf coast. Arab Archaeol Epigr 21(2):135-169.

Begemann F, Schmitt-Strecker S. 2009. Uber das fruhe Kuffer Mesopotamiens. Iranica Antiqua 44:1-46.

Belcher W. 2000. Marine subsistence of the Indus Valley tradition: fish remains from Balakot (Pakistan). In: Taddei M, ed. South Asian archaeology 1997. Rome: Istituto Italiano per l'Africa e l'Orientale. p 3-16.

Belcher W. 2005. Marine exploitation in the third millennium BC - the eastern coast of Pakistan. Paleorient 31(1):79-85.

Benoist A. 2002. Archaeological investigations in Fujairah 2000: a preliminary report. Lyon: Unpublished report from the French Archaeological Mission in the U.A.E. to the Department of Antiquities and National Heritage of Fujairah. p 11-36.

Benoist A. 2007. An Iron Age II snake cult in the Oman Peninsula: evidence from Bithnah (Emirate of Fujairah). Arab Archaeol Epigr 18:34-54.

Benoist A, Hassan S. 2010. An inventory of the objects in a collective burial at Dadna (Emirate of Fujairah). In: Weeks L, ed. Death and burial in Arabia and beyond: multidisciplinary perspectives. Oxford: Archaeopress. p 85-100.

Bentley R. 2001. Human migration in early Neolithic Europe: a study by strontium and lead isotope analysis of archaeological skeletons. Ph.D. Dissertation. Madison: University of Wisconsin-Madison.

366 Bentley R. 2004. Appendix: Characterising human mobility by strontium analysis of the skeletons. In: Higham C, Thosarat R, eds. The excavation of Khok Phanom Di. Oxford: Oxbow Books. p 159-166.

Bentley R. 2006. Strontium isotopes from the earth to the archaeological skeleton: a review. J Archaeol Method Theory 13(3):135-187.

Bentley A. 2007. Mobility, specialisation and community diversity in the Linearbandkeramik: isotopic evidence from the skeletons. In: Whittle A, Cummings V, eds. Going over: the Mesolithic-Neolithic transition in north-west Europe. Oxford: Oxford University Press. p 117-140.

Bentley R, Buckley H, Spriggs M, Bedford S, Ottley C, Nowell G, Macpherson C, Pearson D. 2007. Lapita migrants in the Pacific's oldest cemetery: isotopic analysis at Teouma, Vanuatu. Am Antiquity 72(4):645-656.

Bentley A, Knipper C. 2005. Geographical patterns in biologically available strontium, carbon and oxygen isotope signatures in prehistoric SW Germany. Archaeometry 47(3):629-644.

Bentley R, Krause R, Price T, Kaufman B. 2003. Human mobility at the early Neolithic settlement of Vaihingen, Germany: evidence from strontium isotope analysis. Archaeometry 45(3):471-486.

Bentley R, Pietrusewsky M, Douglas M, Atkinson T. 2005. Matrilocality during the prehistoric transition to agriculture in Thailand? Antiquity 79(306):865-881.

Bentley R, Price T, Luning J, Gronenborn D, Wahl J, Fullagar P. 2002. Prehistoric migration in Europe: strontium isotope analysis of early Neolithic skeletons. Curr Anthropol 43(5):799-804.

Bentley R, Price T, Stephan E. 2004. Determining the local 87Sr/86Sr range for archaeological skeletons: a case study from Neolithic Europe. J Archaeol Sci 31:365-375.

Benton J. 1996. Excavations at al-Sufouh: a third millennium site in the emirate of Dubai, United Arab Emirates. Turnhout: Abiel 1.

Benton J. 2006. Burial practices of the third millennium BC in the Oman Peninsula: a reconsideration. Ph.D. Dissertation, School of Archaeology. Sydney: University of Sydney.

Benton J, Potts D. 1994. Jebel al-Emalah 1993-1994: report compiled for the Department of Culture and Information. Sharjah, UAE: Unpublished report submitted to the Directorate of Antiquities, Government of Sharjah, UAE.

367 Berthoud T, Cleuziou S. 1983. Farming communities of the Oman Peninsula and the copper of Makkan. J Oman Stud 6(2):239-247.

Betts F, Blumenthal N, Posner A. 1981. Bone mineralization. J Crystal Growth 53:63-73.

Biagi P, Torke W, Tosi M, Uerpmann H-P. 1984. Qurum: a case study of coastal archaeology in northern Oman. World Archaeol 16(1):43-61.

Biagi P, Maggi R, Nisbet R. 1989. Excavations at the aceramic coastal settlement of RH5 (Muscat, Sultanate of Oman) 1983-5. In: Frifelt K, and Sorensen P, eds. South Asian archaeology 1985. London: Curzon Press. p 1-8.

Biagi P, Maggi R. 1990. Archaeological surveys along the Oman coast: preliminary results of five years of research (1983-1987). In: Taddei M, editor. South Asian archaeology 1987. Rome: Serie Orientale Roma. p 543-553.

Biagi P, Nisbet R. 1992. Environmental history and plant exploitation at the aceramic sites of RH5 and RH6 near the mangrove swamp of Qurm (Muscat - Oman). In: Vernet J- L, ed. Les charbons de bois, les anciens ecosystems et le role de l'homme: Bulletin de la Societe Botanique de France 139. p 571-578.

Biagi P, Nisbet R. 2006. The prehistoric fisher-gatherers of the western coast of the Arabian Sea: a case of seasonal sedentarization? World Archaeol 38(2):220-238.

Bickle P, Hofmann D. 2007. Moving on: the contribution of isotope studies to the early Neolithic of central Europe. Antiquity 81(314):1029-1041.

Bibby T. 1964. A forgotten civilization of Abu Dhabi. BP Magazine 13:28-33.

Bibby G. 1965. Arabian Gulf archaeology. Kuml: Årbog for Jysk Arkæologisk Selskab 1964:144-152.

Bibby T. 1966. Arabian Gulf archaeology. Kuml: Årbog for Jysk Arkæologisk Selskab 1965:133-152.

Bibby G. 1970. Looking for Dilmun. London: Collins.

Blackman M, Mery S, Wright R. 1989. Production and exchange of ceramics on the Oman Peninsula from the perspective of Hili. J Field Archaeol 16(1):61-77.

Blau S. 1996. Attempting to identify activities in the past: preliminary investigations of the third millennium BC population at Tell Abraq. Arab Archaeol Epigr 7:143-176.

368 Blau S. 1998. Finally the skeleton: an analysis of archaeological human skeletal remains from the United Arab Emirates. Ph.D. Dissertation. Sydney, Australia: University of Sydney.

Blau S. 1999a. Of water and oil: exploitation of natural resources and social change in eastern Arabia. In: Gosden C, and Hather J, editors. The prehistory of food: appetites for change. London: Routledge. p 83-98.

Blau S. 1999b. The people at Sharm: an analysis of the archaeological human skeletal remains. Arab Archaeol Epigr 10:190-204.

Blau S. 1999c. Studies of human skeletal remains in the United Arab Emirates: where are we now? Proc Sem Arab Stud 28:7-13.

Blau S. 2001a. Limited yet informative: pathological alterations observed on human skeletal remains from third and second millennia BC collective burials in the United Arab Emirates. Int J Osteoarchaeol 11:173-205.

Blau S. 2001b. Fragmentary endings: a discussion of 3rd-millennium BC burial practices in the Oman peninsula. Antiquity 75:557-570.

Blau S. 2001c. Young and alone: discussion of an articulated third millennium BC burial at Tell Abraq, United Arab Emirates. Adumatu 3:7-14.

Blau S. 2004. Out of anonymity - a central location for 'peripheral' places through people: the contributions made by Karen Frifelt and Beatrice de Cardi to an understanding of the archaeology of the United Arab Emirates. Arab Archaeol Epigr 15(1):11-19.

Blau S. 2007. Skeletal and dental health and subsistence change in the United Arab Emirates. In: Cohen M, and Crane-Kramer G, eds. Ancient health: skeletal indicators of agricultural and economic intensification. Gainesville: University Press of Florida. p 190- 206.

Blau S, Beech M. 1999. One woman and her dog: an Umm an-Nar example from the United Arab Emirates. Arab Archaeol Epigr 10:34-42.

Blin O. 2007. Al-Ayn: a small settlement and palm tree garden in eastern Oman. In: Cleuziou S, and Tosi M, editors. In the shadow of the ancestors: the prehistoric foundations of the early Arabian civilization in Oman. Oman: Ministry of Heritage and Culture. p 248-250.

Bocquet-Appel J-P, Naji S. 2006. Testing the hypothesis of a worldwide Neolithic demographic transition. Curr Anthropol 47(2):341-365.

369 Böhlke J, Horan M. 2000. Strontium isotope geochemistry of groundwaters and streams affected by agriculture, Locust Grove, MD. Appl Geochem 15:599-609.

Boivin N, Blench R, Fuller D. 2009. Archaeological, linguistic and historical sources on ancient seafaring: a multidisciplinary approach to the study of early maritime contact and exchange in the Arabian Peninsula. In: Petraglia M, and Rose J, eds. The evolution of human populations in Arabia: palaeoenvironments, prehistory and genetics. Dordrecht: Springer. p 251-278.

Bondioli L, Coppa A, Macchiarelli R. 1998. From the coast to the oasis in prehistoric Arabia: what the human osteodental remains tell us about the transition from a foraging to the exchange economy? Evidence from Ra's al-Hamra (Oman) and Hili North (U.A.E.). Proceedings of the XIII congress Forli, Italy. p 229-234.

Bortolini E, Tosi M. 2008. Prehistoric burial cairns and control of passage in Early Bronze Age Oman: the Oasis of Zukayt (Ad Dakhiliyah, northern Oman). Preliminary results and possible developments; 2008; The British Museum, London. Society for Arabian Studies Bienniel Conference, November 27-29. p 6.

Boucharlat R, Dalongeville R, Hesse A, Millet M. 1991a. Occupation humaine et environnement au 5e et au 4e millenaire sur la cote Sharjah-Umm al-Qaiwain (U.A.E.). Arab Archaeol Epigr 2:93-106.

Boucharlat R, Haerinck E, Phillips C, Potts D. 1991b. Note on an Ubaid-pottery site in the Emirate of Umm al-Qaiwain. Arab Archaeol Epigr 2:65-71.

Bourdieu P. 1977. Outline of a theory of practice. Cambridge: Cambridge University Press.

Boutton T, Klein P, Lynott M, Price J, Tieszen L. 1984. Stable carbon isotope ratios as indicators of prehistoric human diet. In: Turnlund J, Johnson P, eds. Stable isotopes in nutrition. Washington, D.C.: American Chemical Society. p 191-204.

Bowen G, Winter D, Spero H, Zierenberg R, Reeder M, Cerling T, Ehleringer J. 2005. Stable hydrogen and oxygen isotope ratios of bottled waters of the world. Rapid Comm Mass Spectrometry 19:3442-3450.

Boyde A. 1989. Handbook of microscopic anatomy: teeth. Berlin: Springer-Verlag.

Brass L, Britton G. 2004. An archaeological survey of northern Fujairah, United Arab Emirates. Arab Archaeol Epigr 15(2):149-196.

Brothwell D, ed. 1963. Dental anthropology. London: Pergamon Press.

370 Brothwell D. 1981. Digging up bones: the excavation, treatment and study of human skeletal remains. Ithaca: Cornell University Press.

Bruckner M. 2009. A primer on stable isotopes and some common uses in hydrology. . Accessed 3/11/2011.

Bryant J, Koch P, Froelich P, Showers W, Genna B. 1996. Oxygen isotope partitioning between phosphate and carbonate in mammalian apatite. Geochim Cosmochim Acta 60(24):5145-5148.

Budd P, Montgomery J, Barreiro B, Thomas R. 2000. Differential diagenesis of strontium in archaeological human dental tissues. Appl Geochem 15(5):687-694.

Budd P, Millard A, Chenery C, Lucy S, Roberts C. 2004. Investigating population movement by stable isotope analysis: a report from Britain. Antiquity 78(299):127-141.

Buikstra JE, Konigsberg LW, Bullington J. 1986. Fertility and the development of agriculture in the prehistoric Midwest. Am Antiq 51:528-546.

Buikstra J, Price T, Wright L, Burton J. 2003. Tombs from the Copan acropolis: a life- history approach. In: Bell E, Canuto M, Sharer R, eds. Understanding early Classic Copan. Philadelphia: University of Pennsylvania Museum Publications. p 191-212.

Burton J, Price T. 1999. Evaluation of bone strontium as a measure of seafood consumption. Int J Osteoarchaeol 9:233-236.

Buzon M, Simonetti A, Creaser R. 2006. Migration in the Nile Valley during the New Kingdom period: a preliminary strontium isotope study. J Archaeol Sci 34:1391-1401.

Calvet Y, Gachet J. 1990. Failaka fouilles francaises 1986-1988. Lyon: Travaux de la Maison de l'Orient 18.

Calvet Y, Salles J-F. 1986. Failaka fouilles francaises 1984-1985. Lyon: Travaux de la Maison de l'Orient 12.

Carter R. 1997. Defining the Late Bronze Age in southeast Arabia: ceramic evolution and settlement during the second millennium BC. Ph.D. Dissertation. London: University College London.

Carter R. 2002. Ubaid-period boat remains from As-Sabiyah: excavations by the British Archaeological Expedition to Kuwait. Proc Sem Arab Stud 32:13-30.

Carter R. 2003a. The Neolithic origins of seafaring in the Arabian Gulf. Archaeol Int 2002/2003:44-47.

371

Carter R. 2003b. Tracing Bronze Age trade in the Arabian Gulf: evidence for way- stations of the merchants of Dilmun between Bahrain and the northern Emirates. In: Potts D, Al Naboodah H, and Hellyer P, editors. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident Press. p 123-131.

Carter R. 2003c. Restructuring Bronze Age trade: Bahrain, southeast Arabia and the copper question. In: Crawford H, editor. Proceedings of a seminar held on Monday 24th July 2000 to mark the exhibition 'Traces of Paradise' at the Brunei Gallery, SOAS, London. Oxford: BAR International Series 1189. p 31-42.

Carter R. 2006. Boat remains and maritime trade in the Persian Gulf during the sixth and fifth millennia BC. Antiquity 80(307):52-63.

Carter R. 2008. Excavations and Ubaid-period boat remains at H3, As-Sabiyah (Kuwait). In: Olijdam E, and Spoor R, editors. Intercultural relations between south and southwest Asia: studies in commemoration of ECL During Caspers (1934-1996). Oxford: BAR International Series 1826. p 92-102.

Carter R, Crawford H, Mellalieu S, and Barrett D. 1999. The Kuwait-British Archaeological Expedition to As-Sabiyah: report on the first season's work. Iraq 61:43- 58.

Carter R, Crawford H. 2003. The Kuwait-British Archaeological Expedition to As- Sabiyah: report on the fourth season's work. Iraq 65:77-90.

Cerling T, Harris J, MacFadden B, Leakey M, Quade J, Eisemann V, Ehleringer J. 1997. Global vegetation change through the Miocene-Pliocene boundary. Nature 389:153-158.

Charles D. 1992. Woodland demographic and social dynamics in the American Midwest: analysis of a burial mound survey. World Archaeol 24:175-197.

Charles D. 1995. Diachronic regional social dynamics: mortuary sites in the Illinois Valley/American bottom region. In: Beck L, editor. Regional approaches to mortuary analysis. New York: Plenum Press. p 77-99.

Charpentier V, Blin O, Tosi M. 1998. Un village de pecheurs neolithiques de la peninsule d'Oman: Suwayh 2 (SWY-2), premiere campagne de fouille. Proc Sem Arab Stud 28:21- 37.

Charpentier V, Angelucci D, Mery S, Saliege J-F. 2000. Autour de la mangrove morte de Suwayh, l'habitat VIe-Ve millenaires de Suwayh SWY-11, Sultanat d'Oman. Proc Sem Arab Stud 30:69-85.

372 Charpentier V, Marquis P, Pelle E. 2003. La necropole et les derniers horizons Ve millenaire du site du Gorbat al-Mahar (Suwayh, SWY-1, Sultanat d'Oman): premiers resultats. Proc Sem Arab Stud 33:11-19.

Chase-Dunn C, Grimes P. 1995. World-systems analysis. Ann Rev Sociol 21:387-417.

Chase-Dunn C, Hall T. 1993. Comparing world-systems: concepts and working hypotheses. Social Forces 71(4):851-886.

Cheddadi R, Lamb H, Guiot J, van der Kaars S. 1998. Holocene climate change in Morocco: a quantitative reconstruction from pollen data. Climate Dynamics 14(12):883- 890.

Chisholm B. 1989. Variation in diet reconstructions based on stable carbon isotopic evidence. In: Price T, ed. The chemistry of prehistoric human bone. Cambridge: Cambridge University Press. p 10-37.

Cleuziou S. 1978-79. The 2nd and 3rd seasons of excavation at Hili 8. Archaeol United Arab Emirates 2/3:30-69.

Cleuziou S. 1979. Les deuxieme et troisième campagnes de fouilles a Hili 8. Archaeol United Arab Emirates 2/3:19-29.

Cleuziou S. 1980. Three seasons at Hili: toward a chronology and cultural history of the Oman Peninsula in the 3rd millennium B.C. Proc Sem Arab Stud 10:19-32.

Cleuziou S. 1981. Oman Peninsula in the early second millennium B.C. . In: Hartel H, ed. South Asian archaeology 1979. Berlin: Riemer. p 279-293.

Cleuziou S. 1982. Hili and the beginnings of oasis life in eastern Arabia. Proc Sem Arab Stud 12:15-22.

Cleuziou S. 1989. Excavations at Hili 8: a preliminary report on the 4th to 7th campaigns. Archaeol United Arab Emirates 5:61-87.

Cleuziou S. 1992. The Oman Peninsula and the Indus civilization: a reassessment. Man and Environment 17(2):93-103.

Cleuziou S. 1996. The emergence of oases and towns in eastern and southern Arabia. In: Afanas'ev G, Cleuziou S, Lukacs J, and Tosi M, eds. The prehistory of Asia and Oceania. Forli: ABACO Edizioni. p 159-165.

373

Cleuziou S. 1997. Construire et proteger son terroir: les oasis d'Oman a l'Age du Bronze. La dynamique des paysages protohistoriques, antiques, medievaux et modernes: XVIIe rencontres internationales d'archeologie et d'histoire d'Antibes, Editions APDCA, Sophia Antipolis. p 389-412.

Cleuziou S. 2000. Dilmun: origins and early development. Traces of paradise: the archaeology of Bahrain 2500 BC - 300 AD. London: The Bahrain National Museum. p 24-27.

Cleuziou S. 2002. The Early Bronze Age of the Oman Peninsula from chronology to the dialectics of tribe and state formation. In: Cleuziou S, Tosi M, Zarins J, eds. Essays on the late prehistory of the Arabian Peninsula. Rome: Serie Orientale 93. p 191-236.

Cleuziou S. 2003. Early Bronze Age trade in the Gulf and the Arabian Sea: the society behind the boats. In: Potts D, Al Naboodah H, Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident Press. p 134-149.

Cleuziou S. 2007. Evolution toward complexity in a coastal desert environment: the early Bronze Age in the Ja'alan, Sultanate of Oman. In: Kohler T, van der Leeuw S, eds. The model-based archaeology of socionatural systems. Santa Fe, NM: School for Advanced Research. p 209-227.

Cleuziou S, Costantini L. 1980. Premiers elements sur l'agriculture protohistorique de l'Arabie orientale. Paleorient 6(1):245-251.

Cleuziou S, and Constantini L. 1982. A l'origine des oasis. La Recherche 137:1180-1182.

Cleuziou S, Gnoli G, Robin C, Tosi M. 1994. Cachets inscrits de la fin du IIIe millenaire av. notre ere a Ra's al-Junays, sultanat d'Oman (note d'information). Comptes-rendus des seances de l'Academie des inscriptions et belles-lettres 138(2):453-468.

Cleuziou S, Tosi M. 1989. The south-eastern frontier of the ancient Near East. In: Frifelt K, and Sorensen P, eds. South Asian archaeology 1985. London: Curzon Press. p 15-48.

Cleuziou S, Tosi M. 1997. Hommes, climats et environnements de la Peninsule arabique a l'Holocene. Paleorient 23(2):121-135.

Cleuziou S, Tosi M. 2000. Ra's al-Jinz and the prehistoric coastal cultures of the Ja'alan. J Oman Stud 11:19-74.

Cleuziou S, Tosi M. 2007. In the shadow of the ancestors: the prehistoric foundations of the early Arabian civilization in Oman. Oman: Ministry of Heritage & Culture.

374 Cleuziou S, Vogt B. 1983. Umm an-Nar burial customs: new evidence from Tomb A at Hili North. Proc Sem Arab Stud 13:37-45.

Cleuziou S, Tosi M, Zarins J. 2002. Introduction. In: Cleuziou S, Tosi M, and Zarins J, editors. Essays on the late prehistory of the Arabian Peninsula. Rome: Serie Orientale Roma 93: Instituto Italiano per l'Africa e l'Oriente. p 9-27.

Cohen A. 1971. Cultural strategies in the organization of trading diasporas. In: Baldwin R, ed. The development of indigenous trade and markets in west commercial policy. Chicago: University of Chicago Press. p 266-281.

Connan J, Carter R, Crawford H, Tobey M, Charrie-Duhaut A, Jarvie D, Albrecht P, Norman K. 2005. A comparative geochemical study of bituminous boat remains from H3, As-Sabiyah (Kuwait), and RJ-2, Ra's al-Jinz (Oman). Arab Archaeol Epigr 15(1):21- 66.

Cooper A, Ullinger J, Coleman D, Sheridan S. 2007. Assessing mobility patterns for the Early Bronze Age community of Bab edh-Dhra' using radiogenic isotope analysis of human dental enamel. American Journal of Physical Anthropology (Abstract). 132(S44):90-91.

Cope J. 2010. Discerning health, disease and activity patterns in a Bronze Age population from Tell Abraq, United Arab Emirates. In: Weeks L, ed. Death and burial in Arabia and beyond: multidisciplinary perspectives. Oxford: Archaeopress. p 61-70.

Cope J, Berryman A, Martin D, Potts D. 2005. Robusticity and osteoarthritis at the trapeziometacarpal joint in a Bronze Age population from Tell Abraq, United Arab Emirates. Am J Phys Anthropol 126:391-400.

Coplen T, Kendall C, Hopple J. 1983. Comparison of stable isotope reference samples. Nature 302:236-238.

Coppa A, Cucina A, Mack M. 1993. Frequenza e distribuzione cronologica dell'ipoplasia dello smalto in un campione scheletrico di RH5. Antropologia Contemporanea 16(1- 4):75-80.

Coppa A, Damadio S, Armelagos G, Mancinelli D, Vargiu R. 1990. Paleobiology and paleopathology: a preliminary study of the prehistoric fishing population of Ra's al- Hamra 5 (Qurum, Sultanate of Oman, 3,700-3,200 BC). Antropologia Contemporanea 13(4):329-336.

Coppa A, Macchiarelli R, Salvatori S, Santini G. 1985. The prehistoric graveyard of Ra's al-Hamra (RH5): (a short preliminary report on the 1981-83 excavations). J Oman Stud 8:97-102.

375 Corboud P, Castella A-C, Hapka R, Im-Obersteg P, Garczynski P. 1990. Archaeological survey of Fujairah, 2 (1988-1989): preliminary report second and third campaigns of the archaeological survey of Fujairah (United Arab Emirates). Geneva: Swiss-Liechtenstein Foundation for Archaeological Research Abroad.

Corboud P, Castella A-C, Hapka R, im-Obersteg P. 1996. Les tombes protohistoriques de Bithnah, Fujairah, Emirats Arabes Unis: terra archaeologica, vol. I. Mainz: Verlag Philipp von Zabern.

Costa P, Wilkinson T. 1987. Settlement and copper exploitation in the 'Arja area. J Oman Stud 9:93-131.

Costantini L. 1979. Palaeoethnobotany: identification of two rows barley and early domesticated sorghum. Archaeol United Arab Emirates 2/3:70-71.

Costantini L. 1984a. The beginning of agriculture in the Kachi Plain: the evidence of Mehgarh. In: Allchin B, ed. South Asian archaeology 1981. Cambridge: Cambridge University Press. p 29-33.

Costantini L. 1984b. Plant impressions in Bronze Age pottery from Yemen Arab Republic. East and West 34:107-115.

Costantini L. 1985. Laboratory of bioarchaeology: paleoethnobotanical investigations in the Middle East and Arabian Peninsula (1985). East and West 35:331-336, 373-374.

Costantini L, Audisio P. 2000. Plant and insect remains from the Bronze Age site of Ra's al-Jinz (RJ-2), Sultanate of Oman. Paleorient 26(1):143-156.

Costin C. 1991. Craft specialization: issues in defining, documenting, and explaining the organization of production. Archaeol Method The 3:1-56.

Cox G, Sealy J. 1997. Investigating identity and life histories: isotopic analysis and historical documentation of slave skeletons found on the Cape Town Foreshore, South Africa. Int J Historical Archaeol 1(3):207-224.

Crawford H. 1973. Mesopotamia's invisible exports in the third millennium B.C. World Archaeol 5(2):232-241.

Crawford H. 1998. Dilmun and its Gulf neighbours. Cambridge: Cambridge University Press.

Crombe P, De Dapper M, Haerinck E. 2001. An archaeological survey of Hawar Island (Bahrain). Arab Archaeol Epigr 12(2):143-155.

376 Crumley C. 1995. Heterarchy and the analysis of complex societies. In: Ehrenreich R, Crumley C, and Levy J, eds. Heterarchy and the analysis of complex societies. Arlington, VA: Archaeological Papers of the American Anthropological Association No. 6. p 1-5.

Cullen H, deMenocal P, Hemming S, Hemming G, Brown F, Guilderson T, Sirocko F. 2000. Climate change and the collapse of the Akkadian Empire: evidence from the deep sea. Geology 28(4):379-382.

Cuttler R, Beech M, Kallweit H, Zander A, and al-Tikriti W. 2007. Pastoral nomadic communities of the Holocene climatic optimum: excavation and research at Kharimat Khor al-Manahil and Khor Al Manahil in the Rub'al-Khali, Abu Dhabi. Proc Sem Arab Stud 37:61-78.

Dales G. 1964. The mythical massacre at Mohenjodaro. Expedition 6(3):36-43.

Dales G. 1974. Excavations at Balakot, Pakistan, 1973. J Field Archaeol 1(1-2):3-22.

Dales G. 1979a. The Balakot project: summary of four years excavation in Pakistan. In: Taddei M, ed. South Asian archaeology 1977. Naples: Istituto Universitario Orientale. p 241-274.

Dales G. 1979b. The Balakot project: summary of four years of excavations in Pakistan. Man Environ 2:45-53.

Dales G. 1981. Reflections on four years of excavations at Balakot. In: Dani A, editor. Indus civilisation: new perspectives. Islamabad: Quaid-i-Azam University. p 25-36.

Dales G, Kenoyer J. 1977. Shell working at ancient Balakot, Pakistan. Expedition 19(2):13-19.

Dansgaard W. 1961. The isotopic composition of natural waters. Medd om Gronland 165(2):1-120.

Dansgaard W. 1964. Stable isotopes in precipitation. Tellus 16:436-468.

Dasch E. 1969. Strontium isotopes in weathering profiles, deep-sea sediments, and sedimentary rocks. Geochim Cosmochim Acta 33:1521-1552.

David H. 1996. Styles and evolution: soft stone vessels during the Bronze Age in the Oman Peninsula. Proc Sem Arab Stud 26:31-46.

Daux V, Lecuyer C, Heran M-A, Amiot R, Simon L, Fourel F, Martineau F, Lynnerup N, Reychler H, Escarguel G. 2008. Oxygen isotope fractionation between human phosphate and water revisited. J Hum Evol 55(6):1138-1147.

377 de Beauclair R, Jasim S, Uerpmann H-P. 2006. New results on the Neolithic jewellery from al-Buhais 18, UAE. Proc Sem Arab Stud 36:175-187. de Cardi B, ed. 1978. Qatar archaeological report: excavations 1973. Oxford: Oxford University Press. de Cardi B. 1986. Some aspects of Neolithic settlement in Bahrain and adjacent regions. In: Al-Khalifa S, and Rice M, eds. Bahrain through the ages: the archaeology. New York: KPI. p 87-93. de Cardi B. 1988. The grave-goods from Shimal Tomb 6 in Ras al-Khaimah, U.A.E. In: Potts D, editor. Araby the Blest: studies in Arabian archaeology. Copenhagen: Museum Tusculanum Press. p 45-72. de Cardi B. 1989. Harappan finds from Tomb 6 at Shimal, Ras al-Khaimah, United Arab Emirates. In: Frifelt K, and Sorensen P, editors. South Asian archaeology 1985. London: Curzon Press. p 9-14. de Cardi B, Bell R, Starling N. 1979. Excavations at Tawi Silaim and Tawi Sa'id in the Sharqiya, 1978. J Oman Stud 5:61-94. de Cardi B, Doe B. 1971. Archaeological survey in the northern Trucial states. East and West 21(3-4):225-289. de Cardi B, Kennet D, Stocks R. 1994. Five thousand years of settlement at Khatt, U.A.E. Proc Sem Arab Stud 24:35-95.

DeNiro M. 1985. Postmortem preservation and alteration of in vivo bone collagen isotope ratios in relation to palaeodietary reconstruction. Nature 317:806-809.

DeNiro M. 1987. Stable isotopy and archaeology. Am Scientist 75:182-191.

DeNiro M, Epstein S. 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochim Cosmochim Acta 42:495-506.

DeNiro M, Schoeninger M. 1983. Stable carbon and nitrogen isotope ratios of bone collagen: variations within individuals, between sexes, and within populations raised on monotonous diets. J Archaeol Sci 10(3):199-203.

Dickin A. 2005. Radiogenic isotope geology. New York: Cambridge University Press.

Donaldson P. 1984. Prehistoric tombs of Ras al-Khaimah. Oriens Antiquus 23:191-312.

Donaldson P. 1985. Prehistoric tombs of Ras al-Khaimah. Oriens Antiquus 24:85-94.

378 Drechsler P. 2007. The Neolithic dispersal into Arabia. Proc Sem Arab Stud 37:93-109.

Dupras T, Schwarcz H. 2001. Strangers in a strange land: stable isotope evidence for human migration in the Dakhleh Oasis, Egypt. J Archaeol Sci 28:1199-1208.

Dupras T, Schwarcz H, Fairgrieve S. 2001. Infant feeding and weaning practices in Roman Egypt. Am J Phys Anthropol 115:204-212.

Dupras T, Tocheri M. 2007. Reconstructing infant weaning histories at Roman period Kellis, Egypt using stable isotope analysis of dentition. Am J Phys Anthropol 134:63-74.

During Caspers E. 1970. Trucial Oman in the third millennium B.C.: new evidence for contacts with Sumer, Baluchistan and the Indus Valley. Origini 4:205-276.

During Caspers E. 1972. Harappan trade in the Arabian Gulf in the third millennium B.C. Mesopotamia 7:167-191.

During Caspers E. 1992. Intercultural/mercantile contacts between the Arabian Gulf and South Asia at the close of the third millennium B.C. Proc Sem Arab Stud 22:3-28.

During Caspers E. 1994. Further evidence for 'Central Asian' materials from the Arabian Gulf. J Econ Soc Hist Orient 37(1):33-53.

Edens C. 1982. Towards a definition of the western ar-Rub' al-Khali "Neolithic". ATLAL: J Saudi Arab Archaeol 6:109-129.

Edens C. 1992. Dynamics of trade in the ancient Mesopotamia "world system". Am Anthropol 94(1):118-139.

Edens C. 1993. Indus-Arabian interaction during the Bronze Age: a review of evidence. In: Possehl G, ed. Harappan civilization: a recent perspective. 2 ed. New Delhi: Oxford & IBH Publishing Co. p 335-363.

Edens C, Kohl P. 1993. Trade and world systems in Early Bronze Age western Asia. In: Scarre C, Healy F, eds. Trade and exchange in prehistoric Europe. Oxford: Oxbow Monographs No. 33. p 17-34.

Elias R, Hirao Y, Patterson C. 1982. The circumvention of the natural biopurification of calcium along nutrient pathways by atmospheric inputs of industrial lead. Geochim Cosmochim Acta 46:2561-2580.

Englund R. 1983. Dilmun in the archaic Uruk corpus. In: Potts D, ed. Dilmun: new studies in the archaeology and early history of Bahrain. Berlin: Reimer. p 34-55.

379 Ericson J. 1985. Strontium isotope characterization in the study of prehistoric human ecology. J Hum Evol 14:503-514.

Evans J. 2005. Redefining the sculpture of Early Dynastic Mesopotamia: New York University.

Evans J. 2007. The Square Temple at Tell Asmar and the construction of Early Dynastic Mesopotamia, ca. 2900-2350 B.C.E. Am J Archaeol 111(4):599-632.

Evans J, Chenery C, Fitzpatrick A. 2006a. Bronze Age childhood migration of individuals near Stonehenge, revealed by strontium and oxygen isotope tooth enamel analysis. Archaeometry 48(2):309-321.

Evans G, Schmidt V, Bush P, Nelson H. 1969. Stratigraphy and geologic history of Sabkha, Abu Dhabi, Persian Gulf. Sedimentology 12:145-169.

Evans J, Stoodley N, Cherney C. 2006b. A strontium and oxygen isotope assessment of a possible fourth century immigrant population in a Hampshire cemetery. J Archaeol Sci 33(2):265-272.

Ezzo J. 1994. Putting the "chemistry" back into archaeological bone chemistry analysis: modeling potential paleodietary indicators. J Anthropol Archaeol 13:1-34.

Ezzo J, Price T. 2002. Migration, regional reorganization, and spatial group composition at Grasshopper Pueblo, Arizona. J Archaeol Sci 29:499-520.

Fairservis W. 1973. Preliminary report on excavations at Allahdino (first season, 1973). Pakistan Archaeol 9:95-102.

Fairservis W. 1976. Excavations at Allahdino I: seals and inscribed material. New York: American Museum of Natural History.

Fairservis W. 1993. Allahdino: an excavation of a small Harappan site. In: Possehl G, ed. Harappan civilization: a recent perspective. New Delhi: Oxford & IBH Publishing Co. p 107-112.

Faure G. 1986. Principles of isotope geology. New York: John Wiley & Sons.

Fincham A, Moradian-Oldak J, Simmer J. 1999. The structural biology of the developing dental enamel matrix. J Structural Biology 126(3):270-299.

Fincham A, Simmer J. 1997. Amelogenin proteins of developing dental enamel. In: Chadwick D, Cardew G, eds. Dental enamel: proceedings of the Ciba Foundation Symposium 205. Chichester: John Wiley & Sons. p 118-134.

380 Fiorenza L, Benazzi S, Kullmer O. 2009. Do it yourself: morphology, wear and 3D digital surface models: materials and techniques to create high-resolution replicas of teeth. J Anthropol Sci 87:211-218.

Flavin K, Shepherd E. 1994. Fishing in the Gulf: preliminary investigations at an Ubaid site, Dalma (UAE). Proc Sem Arab Stud 24:115-134.

Frank A. 1993. Bronze Age world system cycles. Curr Anthropol 34(4):383-405.

Franke-Vogt U. 1997. Reopening research on Balakot: a summary of perspectives and first results. In: Allchin R, Allchin B, eds. South Asian archaeology 1995. New Delhi: Oxford and IBH Publishing Co.

Frankfort H. 1934. Iraq excavations of the Oriental Institute 1932/33: third preliminary report of the Iraq excavation. Chicago: University of Chicago Press.

Frankfort H. 1954. The art and architecture of the ancient Orient. Baltimore: Penguin Books.

Freed S. 1963. Fictive kinship in a north Indian village. Ethnology 2(1):86-103.

Fricke H, O'Neil J. 1996. Inter- and intra-tooth variation in the oxygen isotope composition of mammalian tooth enamel phosphate: implications for palaeoclimatological and palaeobiological research. Palaeogeogr Palaeocl 126:91-99.

Frifelt K. 1971. Jamdat Nasr Fund fra Oman [Jamdat Nasr graves in the Oman]. Kuml: Årbog for Jysk Arkæologisk Selskab 1970:355-383.

Frifelt K. 1975a. On prehistoric settlement and chronology of the Oman Peninsula. East and West 25(3-4):359-424.

Frifelt K. 1975b. A possible link between the Jemdet Nasr and the Umm an-Nar graves of Oman. J Oman Stud 1:57-80.

Frifelt K. 1979. The Umm an-Nar and Jemdet Nasr of Oman and their relations abroad. In: van Lohuizen-de Leeuw J, ed. South Asian archaeology 1975. Leiden: Brill. p 43-57.

Frifelt K. 1980. "Jemdet Nasr graves" on the Oman peninsula. In: Alster B, ed. Death in Mesopotamia. Copenhagen: Akademisk Forlag. p 273-279.

Frifelt K. 1986. Burial mounds near Ali excavated by the Danish Expedition. In: Al- Khalifa H, Rice M, eds. Bahrain through the ages: the archaeology. London: Kegan Paul International. p 125-134.

381 Frifelt K. 1989. 'Ubaid in the Gulf area. In: Henrickson E, and Thuesen I, eds. Upon this foundation: the 'Ubaid reconsidered. Copenhagen: Carsten Niebuhr Institute Publication 10. p 405-415.

Frifelt K. 1991. The island of Umm an-Nar, vol I: third millennium graves. Aarhus: Jutland Archaeological Society Publications XXVI:I.

Frifelt K. 1995. The island of Umm an-Nar, vol II: the third millennium settlement. Aarhus: Jutland Archaeological Society Publications XXVI:2.

Frifelt K. 2002. Bat, a centre in third millennium Oman. In: Cleuziou S, Tosi M, and Zarins J, eds. Essays on the late prehistory of the Arabian Peninsula. Rome: Istituto Italiano per l'Africa e l'Oriente. p 101-110.

Frohlich B. 1983. The Bahrain burial mounds. Dilmun: J Bahrain Hist Archaeol Soc:4-9.

Frohlich B. 1986. The human biological history of the Early Bronze Age population in Bahrain. In: al-Khalifa S, and Rice M, eds. Bahrain through the ages: the archaeology. London: KPI. p 47-63.

Frumkin A, Magaritz M, Carmi I, Zak I. 1991. The Holocene climatic record of the salt caves of Mount Sedom, Israel. The Holocene 1(3):191-200.

Fuller B, Richards M, Mays S. 2003. Stable carbon and nitrogen isotope variations in tooth dentine serial sections from Wharram Percy. J Archaeol Sci 30:1673-1684.

Fuller B, Molleson T, Harris D, Gilmour L, Hedges R. 2006. Isotopic evidence for breastfeeding and possible adult dietary differences from Late/Sub-Roman Britain. Am J Phys Anthropol 129:45-54.

Garvie-Lok S, Varney T, Katzenberg M. 2004. Preparation of bone carbonate for stable isotope analysis: the effects of treatment time and acid concentration. J Archaeol Sci 31(6):763-776.

Gasse F. 2000. Hydrological changes in the African tropics since the Last Glacial Maximum. Quaternary Sci Rev 19(1-5):189-211.

Gasse F, Van Campo E. 1994. Abrupt post-glacial climate events in west Asia and north Africa monsoon domains. Earth Planet Sci Lett 126:435-456.

Gat J. 1996. Oxygen and hydrogen isotopes in the hydrologic cycle. Ann Rev Earth Planet Sci 24:225-262.

Gat J, Dansgaard W. 1972. Stable isotope survey of the fresh water occurrences in Israel and the northern Jordan rift valley. J Hydrology 16:177-211.

382

Gauld SC, Oliver J, Kansa SW. In press. Mortality patterns in the Pottery Neolithic: evidence of catastrophic death and cannibalism in the Domuztepe death pit. In: Perry MA, ed. Bioarchaeology of the Near East and Eastern Mediterranean.

Gaultier M, Guy H, Munoz O, Tosi M, Usai D. 2005. Settlement structures and cemetery at Wadi Shab-GAS1, Sultanate of Oman: report on the 2002 and 2003 field seasons. Arab Archaeol Epigr 16(1):1-20.

Gentelle P, Frifelt K. 1989. About the distribution of third millennium graves and settlement in the Ibri area of Oman. In: Costa P, and Tosi M, eds. Oman studies. Rome: Serie Orientale Roma LXIII. p 119-126.

Giblin J. 2009. Strontium isotope analysis of Neolithic and Copper Age populations on the Great Hungarian Plain. J Archaeol Sci 36(2):491-497.

Glennie K. 1998. The desert of southeast Arabia: a product of Quaternary climatic change. In: Alsharhan A, Glennie K, Whittle G, Kendall G, eds. Quaternary deserts and climatic change. Rotterdam: Balkema. p 279-291.

Glennie K, Pugh J, Goodall T. 1994. Late Quaternary Arabian desert models of Permian Rotliegend reservoirs. Exploration Bulletin 274:1-19.

Glob P. 1959. Rekognoscering i Abu Dhabi (Reconnaissance in Abu Dhabi). Kuml: Årbog for Jysk Arkæologisk Selskab 1959:164-165.

Glover E. 1991. The molluscan fauna from Shimal, Ras al-Khaimah, United Arab Emirates. In: Schippmann K, Herling A, Salles J-F, eds. Golf archaologie Mesopotamien, Iran, Kuwait, Bahrain, Vereinigte Arabische Emiraten und Oman. Buch am Erlach: Internationale Archaologie, Bd. 6. p 205-220.

Glover E. 1998. Mangroves, molluscs and man: archaeological evidence for biogeographical changes in mangrove around the Arabian Peninsula. In: Phillips C, Potts D, and Searight S, eds. Arabia and her neighbours: essays on prehistorical and historical developments presented in honour of Beatrice de Cardi. Turnhout: Abiel II. p 63-78.

Goudie A, Parker A, Al-Farraj A. 2000. Coastal change in Ras Al Khaimah (United Arab Emirates): a cartographic analysis. Geographic J 166(1):14-25.

Government of the UAE Ministry of Petroleum and Mineral Resources. 1979. Geological map of the United Arab Emirates.

Graustein W. 1989. 87Sr/86Sr ratios measure the sources and flow of strontium in terrestrial ecosystems. In: Rundel P, Ehleringer R, Nagy K, eds. Stable isotopes in ecological research. New York: Springer-Verlag. p 491-512.

383

Gregoricka LA, Sheridan SG. Submitted for publication. Stable isotope analysis of Byzantine diet from an urban monastic community. J Archaeol Sci.

Grupe G. 1995. Reconstructing migration in the bell beaker period by 87Sr/86Sr isotope ratios in teeth and bones. In: Radlanski R, Renz H, eds. Proceedings of the 10th international symposium on dental morphology. Berlin: "M" Marketing Services. p 339- 342.

Grupe G, Piepenbrink H, Schoeninger M. 1989. Note on microbial influence on stable carbon and nitrogen isotopes in bone. Appl Geochem 4:299.

Grupe G, Price T, Schroter P, Sollner F, Johnson C, Beard B. 1997. Mobility of the Bell Beaker people revealed by strontium isotope ratios of tooth and bone: a study of southern Bavarian skeletal remains. Appl Geochem 12:517-525.

Grupe G, Price T, Sollner F. 1999. Mobility of Bell Beaker people revealed by strontium isotope ratios of tooth and bone: a study of southern Bavarian skeletal remains. A reply to the comment by Peter Horn and Dieter Muller-Sohnius. Appl Geochem 14:271-275.

Grupe G, Schutkowski H. 1989. Dietary shift during the 2nd millennium BC in prehistoric Shimal, Oman Peninsula. Paleorient 15(2):77-84.

Haddu M. 1989. A preliminary report on the excavation of Tomb N at Hili. Archaeology in the United Arab Emirates 5:55-71.

Haerinck E. 1991. The rectangular Umm an-Nar-period grave at Mowaihat (Emirate of Ajman, United Arab Emirates). Gentse Bijdragen, tot de Kunstgeschiedenis en Oudheidkunde 29:1-30.

Halfman J, Johnson T. 1988. High-resolution record of cyclic climatic change during the past 4 ka from Lake Turkana, Kenya. Geology 16(6):496-500.

Hansen D. 1971. Some Early Dynastic I sealings from Nippur. In: Mitten D, Pedley J, and Scott J, eds. Studies presented to George M A Hanfmann. Mainz am Rhein: Monographs in Art and Archaeology, no. Fogg Art Museum, Harvard University. Verlag Philipp von Zabern. p 47-54.

Haser J. 1990. Soft-stone vessels of the 2nd millennium B.C. in the Gulf region. Proc Sem Arab Stud 20:43-54.

Haser J. 1991. Soft-stone vessels (1) from Shimal and Dhayah/Ras al-Khaimah, U.A.E. In: Schippmann K, Herling A, and Salles J-F, eds. Golf-Archaologie, Mesopotamien, Iran, Kuwait, Bahrain, Vereinigte Arabische Emirate und Oman. Buch am Erlach: Internationale Archaologie, Bd. 6. p 221-232.

384

Hastings A, Humphries J, Meadow R. 1975. Oman in the third millennium BCE. J Oman Stud 1:9-55.

Hauptmann A. 1985. 5000 Jahre Kupfer in Oman. Band 1: Die Entwicklung der Kupfermetallurgie vom. 3 Jahrtausend bis zur Neuzeit. Bochum.: Beiheft 4. Deutsches Bergbau Museum.

Hauptmann A, Weisgerber G, Bachmann H. 1988. Early copper metallurgy in Oman. In: Maddin R, ed. The beginning of the use of metals and alloys. Cambridge: MIT Press. p 34-51.

Haverkort C, Weber A, Katzenberg M, Goriunova O, Simonetti A, Creaser R. 2008. Hunter-gatherer mobility strategies and resource use based on strontium isotope (87Sr/86Sr) analysis: a case study from Middle Holocene Lake Baikal, Siberia. J Archaeol Sci 35:1265-1280.

Hedges R, van Klinken G. 2000. "Consider a spherical cow..." - on modeling and diet. In: Ambrose S, Katzenberg M, eds. Biogeochemical approaches to paleodietary analysis. New York: Kluwer Academic/Plenum Publishers. p 211-241.

Hellyer P. 1993. Kalba archaeology. Tribulus: Journal of the Emirates Natural History Group 3(1):25.

Hellyer P. 1994. Fujairah: an Arabian jewel. Dubai: Motivate Publishing.

Hellyer P. 1998. Hidden riches: an archaeological introduction to the United Arab Emirates. Abu Dhabi: Union National Bank.

Hill P. 1998. Bone remodelling. British J Orthodont 25(2):101-107.

Hillson S. 1996. Dental anthropology. Cambridge: Cambridge University Press.

Hillson S. 2005. Teeth. Cambridge: Cambridge University Press.

Hoch E. 1979. Reflections on prehistoric life at Umm an-Nar (Trucial Oman) based on faunal remains from the third millennium BC. In: Taddei M, ed. South Asian archaeology 1977. Naples: Istututo Universitario Orientale. p 589-638.

Hoch E. 1995. Animal bones from the Umm an-Nar settlement. In: Frifelt K, ed. The island of Umm an-Nar, vol II: the third millennium settlement. Munskgaard: JASP XXVI:2. p 249-256.

385 Hodell D, Quinn R, Brenner M, Kamenov G. 2004. Spatial variation of strontium isotopes (87Sr/86Sr) in the Maya region: a tool for tracking ancient human migration. J Archaeol Sci 31:585-601.

Hoefs J. 2004. Stable isotope geochemistry. Berlin: Springer.

Hoffman M, Shaffer J. 1976. The Harappan settlement at Allahdino: analyzing the sociology of an archaeological site. In: Kennedy K, and Possehl G, editors. Ecological backgrounds of South Asian prehistory. Ithaca: Cornell University. p 94-117.

Hoffman M, Cleland J. 1977. Excavations at Allahdino II: the lithic industry at Allahdino: a metric and quantitative analysis of an Harappan activity system. New York.

Hoogewerff J, Papesch W, Kralik M, Berner M, Vroon P, Miesbauer H, Gaber O, Kunzel K-H, Kleinjans J. 2001. The last domicile of the iceman from Hauslabjoch: a geochemical approach using Sr, C and O isotopes and trace element signatures. J Archaeol Sci 28:983-989.

Hoppe K, Koch P, Furutani T. 2003. Assessing the preservation of biogenic strontium in fossil bones and tooth enamel. Int J Osteoarchaeol 13:20-28.

Horn P, Müller-Sohnius D. 1999. Comment on "Mobility of Bell Beaker people revealed by strontium isotope ratios of tooth and bone: a study of southern Bavarian skeletal remains" by Gisela Grupe, T. Douglas Price, Peter Schroter, Frank Sollner, Clark M. Johnson and Brian L. Beard. Appl Geochem 14(2):263-269.

Højgaard K. 1980. Dentition on Umm an-Nar (Trucial Oman), 2500 B.C. Scandinavian J Dental Res 88:355-364.

Højgaard K. 1981. Dentition on Umm an-Nar, c. 2500 B.C. Proc Sem Arab Stud 11:31- 36.

Højgaard K. 1985. SEM (scanning electron microscopic) examination of teeth from the third millennium BC excavated in Wadi Jizzi and Hafit. In: Schotsmans J, and Taddei M, eds. South Asian archaeology 1983. Naples: Instituto Universitano Orientale. p 151-156.

Højlund F. 1987. Failaka/Dilmun, the second millennium settlements, vol. II: The Bronze Age pottery. Aarhus: Jutland Archaeological Society.

Højlund F. 2007. The burial mounds of Bahrain: social complexity in early Dilmun. Moesgard: Jutland Archaeological Society.

Højlund F, Andersen H. 1994. Qala'at al-Bahrain, vol I: the northern city wall and the Islamic fortress. Aarhus: Jutland Archaeological Society.

386 Højlund F, Bangsgaard P, Hansen J, Haue N, Kjærum P, and Lund D. 2005. New excavations at the Barbar Temple, Bahrain. Arab Archaeol Epigr 16:105-128.

Hublin J-J, Tillier A-M, and Vandermeersch B. 1988. Etude du squelette humain decouvert a Khor (Qatar). In: Inizan M-L, ed. Mission archeologique francaise a Qatar tome 2. Paris: Editions Recherche sur les Civilisations. p 185-194.

Hummel S. 1988. Report on the anthropological activities during the 1988 campaign at the site of Shimal, Settlement SX, Tomb N, Ras al-Khaimah, U.A.E. Ras al-Khaimah, U.A.E. p 1-6.

Hurst R, Davis T. 1981. Strontium isotopes as tracers of airborne fly ash from coal-fired power plants. Environ Geology 3(6):363-367.

Iacumin P, Bocherens H, Mariotti A, Longinelli A. 1996. Oxygen isotope analyses of co- existing carbonate and phosphate in biogenic apatite: a way to monitor diagenetic alteration of bone phosphate? Earth Planet Sci Lett 142(1-2):1-6.

Inizan M-L. 1988. Prehistoire a Qatar: mission archeologique francaise a Qatar, tome 2. Paris: Editions Recherche sur les Civilisations.

Jasim S. 1996. An 'Ubaid site in the Emirate of Sharjah (U.A.E.). Arab Archaeol Epigr 7:1-12.

Jasim S. 2006. The archaeological sites of Jebel al-Buhais. In: Uerpmann H-P, Uerpmann M, and Jasim S, eds. Funeral monuments and human remains from Jebel al-Buhais: Department of Culture and Information, Government of Sharjah, United Arab Emirates. p 13-63.

Jasim S, Uerpmann H-P, Uerpmann M. 2005. Neolithic life and death in the desert - 8 seasons of excavations at Jebel al-Buhais. In: Hellyer P, and Ziolkowski M, eds. Emirates heritage, vol I: proceedings of the 1st annual symposium on recent palaeontological & "archaeological discoveries in the Emirates", Al Ain 2003. Al Ain: Zayed Center for Heritage and History. p 29-35.

Jowsey J. 1961. Age changes in human bone. Clinical Orthopaedics 17:210-218.

Jowsey J, Gordan G. 1971. Bone turnover and osteoporosis. In: Bourne G, ed. The biochemistry and physiology of bone III: development and growth. New York: Academic Press. p 202-238.

Jørgensen N, Morthorst J, Holm P. 1999. Strontium-isotope studies of "brown water" (organic-rich groundwater) from Denmark. Hydrogeology J 7:533-539.

387 Juyal N, Singhvi A, Glennie K. 1998. Chronology and paleoenvironmental significance of the Quaternary desert sediments in SE Arabia. In: Alsharhan A, Glennie K, Whittle G, Kendall G, eds. Quaternary deserts and climatic change. Rotterdam: Balkema. p 315-325.

Kallweit H. 2003. Remarks on the late Stone Age in the U.A.E. In: Potts D, Al Naboodah H, and Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident. p 56-64.

Kallweit H. 2006. A new type of tool from Umm az-Zamul Neolithic sites, United Arab Emirates. Tribulus: J Emirates Nat Hist Group 16(1):14-15.

Kallweit H, Beech M, al-Tikriti W. 2005. Kharimat Khor al-Manahil and Khor Al Manahil - new Neolithic sites in the south-eastern desert of the UAE. Proc Sem Arab Stud 35:97-113.

Kapel H. 1967. Atlas of the Stone-Age cultures of Qatar. Aarhus: Jutland Archaeological Society Publications 6.

Kästner J-M. 1988. Preliminary report concerning the investigation of tomb structures from the 2nd millennium B.C. in Dhayah/Ras al-Khaimah. Ras al-Khaimah, U.A.E.

Kästner J-M. 1991. Some preliminary remarks concerning two recently excavated tombs in Dhayah/Ras al-Khaimah. In: Schippmann K, Herling A, and Salles J-F, eds. Golf- Archaologie. Leidorf: Buch am Erlbach. p 233-244.

Katzenberg M. 2000. Stable isotope analysis: a tool for studying past diet, demography, and life history. In: Katzenberg M, Saunders S, eds. Biological anthropology of the human skeleton. New York: John Wiley and Sons. p 305-327.

Katzenberg M, Saunders S, Fitzgerald W. 1993. Age differences in stable carbon and nitrogen isotope ratios in a population of prehistoric maize horticulturalists. Am J Phys Anthropol 90:267-281.

Katzenberg M, Herring D, Saunders S. 1996. Weaning and infant mortality: evaluating the skeletal evidence. Yearb Phys Anthropol 39:177-199.

Kennedy K. 1994. Identification of sacrificial and massacre victims in archaeological sites: the skeletal evidence. Man Environ 19(1-2):247-251.

Kennedy K, Hemphill B, Lukacs J. 2000. Bring back the bones: hard evidence. Man Environ 25(1):105-109.

Kennet D. 1997. Kush: a Sasanian and Islamic-period archaeological tell in Ras al Khaimah (U.A.E.). Arab Archaeol Epigr 8:284-302.

388 Kennet D, Velde C. 1995. Third and early second-millennium occupation at Nud Zhiba, Khatt (U.A.E.). Arab Archaeol Epigr 6:81-99.

Kenoyer J. 1991. The Indus Valley Tradition of Pakistan and western India. J World Prehist 5(4):331-385.

Kenoyer J. 2000. Wealth and socio-economic hierarchies of the Indus Valley civilization. In: Richards J, and Van Buren M, eds. Order, legitimacy and wealth in early states. Cambridge: Cambridge University Press. p 90-112.

Kiesewetter H. 2003. The Neolithic population at Jebel Buhais 18: remarks on funerary practices, palaeodemography and palaeopathology. In: Potts D, Al Naboodah H, and Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident Press. p 36-43.

Kiesewetter H. 2006. Analyses of the human remains from the Neolithic cemetery at al- Buhais 18 (excavations 1996-2000). In: Uerpmann H-P, Uerpmann M, and Jasim S, eds. Funeral monuments and human remains from Jebel al-Buhais. Sharjah: Department of Culture and Information. p 103-380.

Kiesewetter H, Uerpmann H-P, Jasim S. 2000. Neolithic jewellery from Jebel al-Buhais 18. Proc Sem Arab Stud 30:137-146.

Kimber C. 2000. Origins of domesticated sorghum and its early diffusion to India and China. In: Smith C, and Frederiksen R, eds. Sorghum: origin, history, technology, and production. New York: John Wiley & Sons, Inc. p 3-98.

King G, Maren-Griesbach H. 1999. A preliminary survey of the archaeology of the Wadi Safad, Fujairah, UAE. Tribulus: J Emirates Nat Hist Group 9(2):10-18.

Kingston J. 1999. Isotopes and environments of the Baynunah Formation, Emirate of Abu Dhabi, United Arab Emirates. In: Whybrow P, Hill A, eds. Fossil vertebrates of Arabia: with emphasis on the late Miocene faunas, geology, and palaeoenvironments of the Emirate of Abu Dhabi. New Haven: Yale University Press. p 354-372.

Kluge M, and Ting I. 1978. Crassulacean acid metabolism: analysis of an ecological adaptation. New York: Springer-Verlag.

Knudson K. 2004. Tiwanaku residential mobility in the south central Andes: identifying archaeological human migration through strontium isotope analysis. Ph.D. DissertationMadison: University of Wisconsin at Madison.

Knudson K. 2009. Oxygen isotope analysis in a land of environmental extremes: the complexities of isotopic work in the Andes. Int J Osteoarchaeol 19(2):171-191.

389 Knudson K, Blom D. 2009. The complex relationship between Tiwanaku mortuary identity and geographic origin in the south central Andes. In: Knudson K, Stojanowski C, eds. Bioarchaeology and identity in the Americas. Gainesville: University Press of Florida. p 194-211.

Knudson K, Buikstra J. 2007. Residential mobility and resource use in the Chiribaya Polity of Southern Peru: strontium isotope analysis of archaeological tooth enamel and bone. Int J Osteoarchaeol 17:563-580.

Knudson K, Price T. 2007. Utility of multiple chemical techniques in archaeological residential mobility studies: case studies from Tiwanaku- and Chiribaya-affiliated sites in the Andes. Am J Phys Anthropol 132(1):25-39.

Knudson K, Price T, Buikstra J, Blom D. 2004. The use of strontium isotope analysis to investigate Tiwanaku migration and mortuary ritual in Bolivia and Peru. Archaeometry 46:5-18.

Knudson K, Price T, Buikstra J, Blom D. 2005. Strontium-isotope analysis and migration in the south central Andes: Tiwanaku colonization of the Moquegua Valley. In: Kars H, and Burke E, eds. Proceedings of the 33rd International Symposium on Archaeometry, 2002. Amsterdam: Vrije Universitat. p 477-482.

Koch P, Behrensmeyer A, Tuross N, Fogel M. 1990. Isotopic fidelity during bone weathering and burial. Carnegie Institution, annual report of the director of the geophysical laboratory, 1989-1990. Washington, D.C.: Smithsonian Institution. p 105- 110.

Koch P, Halliday A, Walter L, Stearley R, Huston T, Smith G. 1992. Sr isotopic composition of hydroxyapatite from recent and fossil salmon: the record of lifetime migration and diagenesis. Earth Planet Sci Lett 108:277-287.

Koch P, Tuross N, Fogel M. 1997. The effects of sample treatment and diagenesis on the isotopic integrity of carbonate in biogenic hydroxylapatite. J Archaeol Sci 24:417-429.

Kohn M, Schoeninger M, Valley J. 1998. Variability in herbivore tooth oxygen isotope compositions: reflections of seasonality or developmental physiology? Chem Geol 152:97-112.

Kohn M, Schoeninger M, Barker W. 1999. Altered states: effects of diagenesis on fossil tooth chemistry. Geochim Cosmochim Acta 63(18):2737-2747.

Kolodny Y, Luz B, Sander M, Clemens W. 1996. Dinosaur bones: fossils or pseudomorphs? The pitfalls of physiology reconstruction from apatitic fossils. Palaeogeography, Palaeoclimatology, Palaeoecology 126:161-171.

390 Krigbaum J. 2003. Neolithic subsistence patterns in northern Borneo reconstructed with stable carbon isotopes of enamel. J Anthropol Archaeol 22:292-304.

Krueger H. 1991. Exchange of carbon with biological apatite. J Archaeol Sci 18:355-361.

Krueger H, Sullivan C. 1984. Models for carbon isotope fractionation between diet and bone. In: Turnlund J, Johnson P, eds. Stable isotopes in nutrition. Washington, D.C.: American Chemical Society. p 205-220.

Kunter M. 1983. Chronological and regional differences found in pathological tooth remains from the Arabian Peninsula. Archaologisches Korrespondenzblatt 13:339-343.

Kunter M. 1991. The human skeletal remains from the graves of Umm an-Nar, Abu Dhabi, U.A.E. (third millennium BC). In: Frifelt K, ed. The island of Umm an-Nar, vol I: third millennium graves. Aarhus: Jutland Archaeological Society Publications. p 163- 179.

Kunter M. 1996. Human skeletal remains from the Oman Peninsula dating from the 4th to the 2nd millennium BC. A survey. Homo 47(1-3):43-60.

Kutterer A, de Beauclair R. 2008. FAY-NE15 - another Neolithic graveyard in the central region of the Sharjah Emirate? Arab Archaeol Epigr 19(2):134-143.

Lamberg-Karlovsky C. 1970. Excavations at Tepe Yahya, Iran 1967-1969: progress report I. Cambridge: Harvary University Press.

Lamberg-Karlovsky C. 1971. Excavations at Tepe Yahya. Artibus Asiae 33(4):302-306.

Lamberg-Karlovsky C. 1972. Tepe Yahya 1971: Mesopotamia and the Indo-Iranian borderlands. Iran 10:89-100.

Lamberg-Karlovsky C. 1977. Foreign relations in the third millennium at Tepe Yahya. Le Plateau Iranien et L'asie Centrale des origines a la conquete Islamique: Colloques Internationaux du C.N.R.S. p 33-43.

Lamberg-Karlovsky C. 2009. Structure, agency and commerce in the ancient Near East. Iranica Antiqua 44:47-88.

Lamberg-Karlovsky C, Beale T, eds. 1986. Excavations at Tepe Yahya, Iran 1967-1975: the early periods. Cambridge: Harvard University Press.

Lamberg-Karlovsky C, Potts D. 2001. Excavations at Tepe Yahya, Iran: the third millennium. Cambridge: Harvard Museum Peabody Museum of Archaeology and Ethnology.

391 Langham W, Anderson E. 1957. Strontium-90 and skeletal formation. Science 126(3266):205-206.

Larsen C. 1983a. The early environment and hydrology of ancient Bahrain. In: Potts D, ed. Dilmun: new studies in the archaeology and early history of Bahrain. Berlin: Reimer. p 3-37.

Larsen C. 1983b. Life and land use on the Bahrain Islands: the geoarchaeological of an ancient society. Chicago: University of Chicago Press.

Larsen CS. 1997. Bioarchaeology: interpreting behavior from the human skeleton. Cambridge: Cambridge University Press.

Larsen CS, Ruff C, Kelly R. 1995. Structural analysis of the Stillwater postcranial human remains: behavioral implications of articular joint pathology and long bone diaphyseal morphology. In: Bioarchaeology of Stillwater Marsh: prehistoric human adaptation in the western Great Basin. New York: Anthropological Papers of the American Museum of Natural History 77. p 107-133.

Laursen S. 2008. Early Dilmun and its rulers: new evidence of the burial mounds of the elite and the development of social complexity, c. 2200-1750 BC. Arab Archaeol Epigr 19(2):156-167.

Laursen S. 2009. The decline of Magan and the rise of Dilmun: Umm an-Nar ceramics from the burial mounds of Bahrain, c.2250-2000 BC. Arab Archaeol Epigr 20(2):134- 155.

Lawler A. 2010. The shipping news, circa 2200 B.C.E. Science 328:1094-1095.

LeGeros R. 1981. Apatites in biological systems. Prog Crystal Growth Characteristics 4:1-45.

LeGeros R. 1991. Calcium phosphates in oral biology and medicine. Basel: Karger.

LeGeros R, LeGeros J. 1984. Phosphate minerals in human tissues. In: Nriago J, and Moore P, eds. Phosphate minerals. New York: Springer-Verlag. p 351-385.

Leemans W. 1960. Foreign trade in the Old Babylonian period. Leiden: Brill.

Lee-Thorp J, Sealy J, van der Merwe N. 1989. Stable carbon isotope ratio differences between bone collagen and bone apatite, and their relationship to diet. J Archaeol Sci 16(6):585-599.

Lee-Thorp J, van der Merwe N. 1987. Carbon isotope analysis of fossil bone apatite. S African J Sci 83:712-715.

392

Lee-Thorp J, van der Merwe N. 1991. Aspects of the chemistry of modern and fossil biological apatites. J Archaeol Sci 18:343-354.

Lemcke G, Sturm M. 1997. 18O and trace element measurements as proxy for the reconstruction of climate changes at Lake Van (Turkey): preliminary results. In: Dalfes H, Kukla G, and Weiss H, eds. Third millennium BC climate change and Old World collapse. Berlin: Springer, NATO ASI 49. p 653-678.

Levy J. 1995. Heterarchy in Bronze Age Denmark: settlement pattern, gender, and ritual. In: Ehrenreich R, Crumley C, and Levy J, eds. Heterarchy and the analysis of complex societies. Arlington, VA: Archaeological Papers of the American Anthropological Association No. 6. p 41-52.

Lieverse A. 1999. Diet and the aetiology of dental calculus. Int J Osteoarchaeol 9:219- 232.

Littleton J. 1987. A delicious torment: an analysis of dental pathology. Ph.D. Dissertation. Canberra: Australian National University.

Littleton J. 1990. A delicious torment: date palms and toothache in old Bahrain. In: Freedman L, ed. Is our future limited by our past? Proceedings of the third conference of the Australasian Society for Human Biology. Nedlands, Western Australia: Centre of Human Biology, The University of Western Australia. p 241-248.

Littleton J. 1997. The human remains from excavation 519 at Qala'at al-Bahrain. In: Højlund F, and Andersen H, eds. Qala'at al-Bahrain, vol II: the central monumental buildings. Munskgaard: JASP XXX:2. p 217-233.

Littleton J. 1998a. A Middle Eastern paradox: rickets in skeletons from Bahrain. J Paleopathol 10(1):13-30.

Littleton J. 1998b. Skeletons and social composition: Bahrain 300 BC - AD 250. Oxford: BAR International Series 703.

Littleton J. 1999. Paleopathology of skeletal fluorosis. Am J Phys Anthropol 109(4):465- 483.

Littleton J. 2007. The political ecology of health in Bahrain. In: Cohen M, and Crane- Kramer G, editors. Ancient health: skeletal indicators of agricultural and economic intensification. Gainesville: University Press of Florida. p 176-189.

Littleton J, Frifelt K. 2006. Trepanations from Oman: a case of diffusion? Arab Archaeol Epigr 17(2):139-151.

393 Littleton J, Frolich B. 1989. An analysis of dental pathology and diet on historic Bahrain. Paleorient 15(2):59-75.

Littleton J, Frolich B. 1993. Fish-eaters and farmers: dental pathology in the Arabian Gulf. Am J Phys Anthropol 92(4):427-447.

Longinelli A. 1984. Oxygen isotopes in mammal bone phosphate: a new tool for paleohydrological and paleoclimatogical research? Geochim Cosmochim Acta 48:385- 390.

Lowenstam H, Weiner S. 1989. On biomineralization. Oxford: Oxford University Press.

Luz B, Kolodny Y. 1989. Oxygen isotope variation in bone phosphate. Appl Geochem 4:317-323.

Luz B, Kolodny Y. 1985. Oxygen isotope variations in bone phosphate of biogenic apatites. IV. Mammal teeth and bones. Earth Planet Sci Lett 75:29-36.

Luz B, Kolodny Y, Horowitz M. 1984. Fractionation of oxygen isotopes between mammalian bone-phosphate and environmental drinking water. Geochim Cosmochim Acta 48:1689-1693.

Macchiarelli R. 1985. Antropologia dentaria, contesti culturali ed ambiente: iI parallelismo nelle popolazioni pre-protostoriche dal Vicino Oriente all'Asia media. In: Vogel G, and Gambacorta G, eds. Storia dell'odontoiatria. Milano: Ars Medica Antiqua. p 35-46.

Macchiarelli R. 1989. Prehistoric "fish-eaters" along the eastern Arabian coasts: dental variation, morphology, and oral health in the Ra's al-Hamra community (Qurum, Sulanate of Oman, 5th-4th millennia BC). Am J Phys Anthropol 78:575-594.

Mack M, Coppa A. 1992. Frequency and chronological distribution of enamel hypoplasias from the Ra's al-Hamra-5 (RH5) skeletal collection (Oman). J Paleopathol 2:131-141.

Mackay E. 1931. Report on excavations at Jemdet Nasr, Iraq. Chicago: Field Museum of Natural History, Anthropological Memoirs Vol.I/3.

Magee P. 2005. Excavations at Tepe Yahya, Iran, 1967-1975, vol. IV: the Iron Age settlement. Cambridge: Peabody Museum Press.

Magee P, Uerpmann H-P, Uerpmann M, Jasim S, Handel M, Barber D, Fritz C, and Hammer E. 2009. Multi-disciplinary research on the past human ecology of the east Arabian coast: excavations at Hamriya and Tell Abraq (Emirate of Sharjah, United Arab Emirates). Arab Archaeol Epigr 20(1):18-29.

394

Mandaville J. 1985. A botanical reconnaissance of the Musandam region of Oman. J Oman Stud 7:9-28.

Marino B, Elroy M. 1991. Isotopic composition of atmospheric CO2 inferred from carbon in C4 plant cellulose. Nature 349:127-131.

Martin D. 1996. Paleopathology of the Mowaihat human remains from Ajman, United Arab Emirates (2300-2100 BC): 1995-1996 field season. Amherst: Hampshire College. p 1-5.

Martin D. 2007. Bioarchaeology in the United Arab Emirates. Arab Archaeol Epigr 18(2):124-131.

Masry A. 1974. Prehistory in northeastern Arabia: the problem of interregional interaction. Coconut Grove, Miami: Field Research Projects.

Matthews R. 2002. Secrets of the dark mound: Jemdet Nasr 1926-1928. London: British School of Archaeology in Iraq.

Mayewski P, Rohling E, Stager J, Karlen W, Maasch K, Meeker L, Meyerson E, Gasse F, van Kreveld S, Holmgren K and others. 2004. Holocene climate variability. Quaternary Res 62(3):243-255.

McIntosh S. 1995. Paleobotanical and human osteological remains. In: McIntosh S, ed. Excavations at Jenne-Jeno, Hambarketolo, and Kaniana Inland Niger Delta, Mali, the 1981 season Anthropology vol 20. Berkeley, CA: University of California Press. p 348- 359.

McSweeney K, Mery S, Macchiarelli R. 2008. Rewriting the end of the Early Bronze Age in the United Arab Emirates through the anthropological and artefactual evaluation of two collective Umm an-Nar graves at Hili (eastern region of Abu Dhabi). Arab Archaeol Epigr 19:1-14.

McSweeney K, Mery S, al Tikriti W. 2010. Life and death in an Early Bronze Age community from Hili, Al Ain, UAE. In: Weeks L, ed. Death and burial in Arabia and beyond: multidisciplinary perspectives. Oxford: Archaeopress. p 45-54.

Meadow R. 1979. Prehistoric subsistence at Balakot. In: Taddei M, ed. South Asian archaeology 1977. Naples: Istituto Universitario Orientale. p 275-315.

Meadow R, Humphries J, Hastings A. 1976. Explorations in Oman, 1973 and 1975: prehistoric settlements and ancient copper smelting with its comparative aspects in Iran. In: Bagherzadeh F, ed. Proceedings of the IVth annual symposium on archaeological research in Iran. Teheran: Iranian Centre for Archaeological Research. p 110-129.

395

Mery S. 1988. Ceramics from RJ-2. In: Cleuziou S, and Tosi M, eds. The Joint Hadd Project: summary report on the second season, November 1986-January 1987. Naples: Instituto Universitario Orientale. p 41-47.

Mery S. 1991. Wadi Suq fine wares from Shimal and HIli sites (United Arab Emirates): a technological and provenience analysis. In: Schippmann K, Herling A, Salles J-F, eds. Golf-Archaologie. Buch am Erlbach: Leidorf. p 245-255.

Mery S. 1997. A funerary assemblage from the Umm an-Nar period: the ceramics from tomb A at Hili North, UAE. Proc Sem Arab Stud 27:171-191.

Mery S. 2000. Les ceramiques d'Oman et l'Asie moyenne: une archeologie des echanges a l'Age du Bronze. Paris: CNRS Editions.

Mery S, Charpentier V. 2002. Around Suwayh, Ja'alan: a summary of some recent discoveries from coastal shell-middens of eastern Arabia. J Oman Stud 12:181-195.

Mery S, Charpentier V, Auxiette G, Pelle E. 2009. A dugong bone mound: the Neolithic ritual site on Akab in Umm al-Quwain, United Arab Emirates. Antiquity 83:696-708.

Mery S, McSweeney K, Rouquet J, Basset G, Al Tikriti W. 2008. New evidence of funerary practices at the end of the Early Bronze Age at Hili, United Arab Emirates. In:

Olijdam E, and Spoor R, eds. Intercultural relations between south and southwest Asia: studies in commemoration of ECL During Caspers (1934-1996). Oxford: Archaeopress BAR International Series 1826. p 200-208.

Mery S, McSweeney K, Van der Leeuw S, Al Tikriti W. 2004. New approaches to a collective grave from the Umm an-Nar period at Hili (UAE). Paleorient 30(1):163-178.

Mery S, Phillips C, Calvet Y. 1998. Dilmun potteries in Mesopotamia and Magan from the end of the 3rd and beginning of the 2nd millennium B.C. In: Phillips C, Potts D, and Searight S, eds. Arabia and her neighbours: essays on prehistorical and historical developments presented in honour of Beatrice de Cardi. Leyden: Brepols. p 165-180.

Mery S, Tengberg M. 2009. Food for eternity? The analysis of a date offering from a 3rd millennium BC grave at Hili N, Abu Dhabi (United Arab Emirates). J Archaeol Sci 36:2012-2017.

Midant-Reynes B. 1985. Un ensemble de sepultures en fosses sous cairn a Khor (Qatar): etude des rites funeraires. Paleorient 11(1):129-144.

Molnar S. 1971. Human tooth wear, tooth function and cultural variability. Am J Phys Anthropol 34(2):175-190.

396

Montgomery J, Budd P, Cox A, Krause P, Thomas R. 1999. LA-ICP-MS evidence for the distribution of Pb and Sr in Romano-British medieval and modern human teeth: implications for life history and exposure reconstruction. Young S, Pollard A, Budd P, Ixer R, eds. Oxford: Archaeopress. p 258-261.

Montgomery J, Evans J. 2006. Immigrants on the Isle of Lewis - combining traditional funerary and modern isotope evidence to investigate social differentiation, migration and dietary change in the Outer Hebrides of Scotland. In: Gowland R, Knüsel C, eds. Social archaeology of funerary remains. Oxford: Oxbow Books. p 122-142.

Montgomery J, Evans J, Neighbour T. 2003. Sr isotope evidence for population movement within the Hebridean Norse community of NW Scotland. J Geological Soc 160:649-653.

Montgomery J, Evans J, Powlesland D, Roberts C. 2005. Continuity or colonization in Anglo-Saxon England? Isotope evidence for mobility, subsistence practice, and status at West Heslerton. Am J Phys Anthropol 126(2):123-138.

Mortensen P. 1971. On the date of the temple at Barbar in Bahrain. Artibus Asiae 33(4):299-302.

Mortensen P. 1986. The Barbar temple: its chronology and foreign relations reconsidered. In: Al Khalifa H, Rice M, eds. Bahrain through the ages: the archaeology. London: Kegan Paul International. p 178-185.

Mulhern D, van Gerven D. 1997. Patterns of femoral bone remodeling dynamics in a medieval Nubian population. Am J Phys Anthropol 104(1):133-146.

Müller D, McKenzie J, Mueller P. 1990. Abu Dhabi sabkha, Persian Gulf, revisited: application of strontium isotopes to test an early dolomitization model. Geology (Boulder)18(7):618-621.

Mynors H. 1983. An examination of Mesopotamian ceramics using petrographic and neutron activation analysis. In: Aspinall A, Warren S, eds. The proceedings of the 22nd symposium on archaeometry. Bradford: Schools of Physics and Archaeological Sciences. p 377-387.

Nanci A. 2008. Ten Cate's oral histology: development, structure, and function. St. Louis: Mosby Elsevier.

Nayeem M. 1990. Saudi Arabia: prehistory and protohistory of the Arabian Peninsula. Hyderabad, India: Hyderabad Publishers.

397 Nayeem M. 1996. The Sultanate of Oman: prehistory and protohistory from the most ancient times (c. 1,000,000 B.C. to 100 B.C.). Hyderabad, India: Hyderabad Publishers.

Negrel P, Deschamps P. 1996. Natural and anthropogenic budgets of a small watershed in the Massif Central (France): chemical and strontium isotopic characterization of water and sediments. Aquatic Geochemistry 2:1-27.

Nelson B, DeNiro M, Schoeninger M, De Paolo D. 1986. Effects of diagenesis on strontium, carbon, nitrogen and oxygen concentration and isotopic composition of bone. Geochim Cosmochim Acta 50:1941-1949.

Nelson G, Lukacs J, Yule P. 1999. Dates, caries, and early tooth loss during the Iron Age of Oman. Am J Phys Anthropol 108(3):333-343.

Neumann K. 1999. Charcoal from west African savanna sites: questions of identification and interpretation. The exploitation of plant resources in ancient Africa. New York: Kluwer Academic/Plenum p 205-219.

Nisbet R. 1985. Evidence of sorghum at site RH5, Qurm, Muscat. East and West 35:415- 417.

Nissen H. 1988. The early history of the ancient Near East 9000-2000 BC. Chicago: University of Chicago Press.

Norbeck E, Befu H. 1958. Informal fictive kinship in Japan. Am Anthropol 60:102-117.

Oates J. 1976. Prehistory in northeastern Arabia. Antiquity 50:20-31.

Oates D. 1986. Dilmun and the Late Assyrian Empire. In: Al Khalifa H, and Rice M, eds. Bahrain through the ages: the archaeology. London. p 428-434.

Oates J. 1993. Trade and power in the fifth and fourth millennia BC: new evidence from northern Mesopotamia. World Archaeol 24(3):403-422.

Oates J. 2008. Babylon, revised edition. New York: Thames and Hudson.

Oates J, Davidson T, Kamilli D, McKerrell H. 1977. Seafaring merchants of Ur? Antiquity 51:221-234.

O'Connor T. 2000. The archaeology of animal bones. College Station, TX: Texas A&M University Press.

O'Leary M. 1981. Carbon isotope fractionation in plants. Phytochemistry 20(4):553-567.

398 Orchard J, Stanger G. 1994. Third millennium oasis towns and environmental constraints on settlement in the al-Hajar region, part I: the al-Hajar oasis towns. Iraq 56:63-88.

Palmieri A, Coppa A, Francalacci P. 1994. Problems of an aquatic diet in trace element analysis: the coastal site of Qurum RH5 (Sultanate of Oman, 3.700-3.200 b.C.). Hum Evol 9(2):155-164.

Parker A, Eckersley L, Smith M, Goudie A, Stokes S, Ward S, White K, Hodson M. 2004. Holocene vegetation dynamics in the northeastern Rub' al-Khali desert, Arabian Peninsula: a phytolith, pollen and carbon isotope study. J Quaternary Sci 19(7):665-676.

Parker A, Goudie A. 2007. Development of the Bronze Age landscape in the southeastern Arabian Gulf: new evidence from a buried shell midden in the eastern extremity of the Rub' al-Khali desert, Emirate of Ras al-Khaimah, U.A.E. Arab Archaeol Epigr 18(2):132- 138.

Parker A, Goudie A. 2008. Geomorphological and palaeoenvironmental investigations in the southeastern Arabian Gulf region and the implication for the archaeology of the region. Geomorphology 101:458-470.

Parker A, Goudie A, Stokes S, White K, Hodson M, Manning M, Kennet D. 2006. A record of Holocene climate change from lake geochemical analyses in southeastern Arabia. Quaternary Res 66:465-476.

Passey B, Robinson T, Ayliffe L, Cerling T, Sponheimer M, Dearing M, Roeder B, Ehleringer J. 2005. Carbon isotope fractionation between diet, breath CO2, and bioapatite in different mammals. J Archaeol Sci 32:1459-1470.

Peake H. 1928. The copper mountain of Magan. Antiquity 2:452-457.

Pearson O, and Buikstra J. 2006. Behavior and the bones. In: Buikstra J, Beck L, eds. Bioarchaeology: the contextual analysis of human remains. London: Academic Press. p 207-225.

Peng T-H, Broecker W, Freyer H, and Trumbore S. 1983. A deconvolution of the tree ring based d13C record. J Geophys Res 88(C6):3609-3620.

Perry MA, Coleman D, Delhopital N. 2008. Mobility and exile at 2nd century A.D. Khirbet edh-Dharih: strontium isotope analysis of human migration in western Jordan. Geoarchaeology 23(4):528-549.

Perry M, Coleman D, Dettman D, and Halim al-Shiyab A. 2009. An isotopic perspective on the transport of Byzantine mining camp laborers into southwestern Jordan. Am J Phys Anthropol 140(3):429-441.

399

Petraglia M. 2007. Mind the gap: factoring the Arabian Peninsula and the Indian subcontinent into Out of Africa models. In: Mellars P, Boyle K, Bar-Yosef O, Stringer C, editors. The human revolution revisited. Cambridge: McDonald Institute Archaeological Publications. p 383-394.

Phillips C. 1987. Wadi al Qawr, Fashga 1: the excavation of a prehistoric burial structure in Ras Al Khaimah, U.A.E., 1986: a preliminary report. Edinburgh: University of Edinburgh.

Phillips C. 2002. Prehistoric middens and a cemetery from the southern Arabian Gulf. In: Cleuziou S, Tosi M, Zarins J, eds. Essays on the late prehistory of the Arabian Peninsula. Rome: IsIAO (Istituto Italiano per l'Africa e l'Oriente). p 169-186.

Phillips C. 2007. The third-millennium tombs and settlement at Mowaihat in the Emirate of Ajman, U.A.E. Arabian Archaeology and Epigraphy 18(1):1-7.

Phillips C, Mosseri-Marlio C. 2002. Sustaining change: the emerging picture of the Neolithic to Iron Age subsistence economy at Kalba, Sharjah Emirate, UAE. In: Buitenhuis H, Choyke A, Mashkour M, al-Shiyab A-H, eds. Archaeozoology of the Near East V. p 195-210.

Piesinger C. 1983. Legacy of Dilmun: the roots of ancient maritime trade in eastern coastal Arabia in the 4th/3rd millennium B.C. Ph.D. Dissertation. Madison: University of Wisconsin Madison.

Pilloud M, Larsen CS. 2011. "Official" and "practical" kin: inferring social and community structure from dental phenotype at Neolithic Catalhoyuk, Turkey. Am J Phys Anthropol 145:519-530.

Pollard A, Pellegrini M, Lee-Thorp J. 2011. Technical note: some observations on the conversion of dental enamel d18Op values to d18Ow to determine human mobility. Am J Phys Anthropol 145(3):499-504.

Pollock S. 1992. Bureaucrats and managers, peasants and pastoralists, imperialists and traders: research on the Uruk and Jemdet Nasr periods in Mesopotamia. J World Prehist 6(3):297-336.

Pollock S. 1999. Ancient Mesopotamia: the eden that never was. New York: Cambridge University Press.

Pollock S. 2001. The Uruk period in southern Mesopotamia. In: Rothman M, ed. Uruk Mesopotamia and its neighbors: cross-cultural interactions in the era of state formation. Santa Fe: School of American Research Press. p 181-232.

400 Popescu E. 2003. The Neolithic settlement sites on the islands of Dalma and Marawah, U.A.E. In: Potts D, Al Naboodah H, Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident. p 45-54.

Porada E, Hansen D, Dunham S, Babcock S. 1992. The chronology of Mesopotamia, ca. 7000-1600 B.C. In: Ehrich R, ed. Chronologies in Old World archaeology. Chicago: University of Chicago Press. p 77-121.

Possehl G. 1996. Meluhha. In: Reade J, ed. The Indian Ocean in antiquity. London: Keagan Paul. p 133-208.

Potts D. 1986a. Eastern Arabia and the Oman Peninsula during the late fourth and early third millennium B.C. In: Finkbeiner U, and Rollig W, editors. Gamdat Nasr: period or regional style? Wiesbaden: Tubinger Atlas des Vorderen Orients Beiheft B 62. p 121- 170.

Potts D. 1986b. The booty of Magan. Oriens Antiquus 26:271-285.

Potts D. 1989. Excavations at Tell Abraq, 1989. Paleorient 15(1):269-271.

Potts D. 1990. The Arabian Gulf in antiquity, vol. I: from prehistory to the fall of the Achaemenid Empire. Oxford: Clarendon Press.

Potts D. 1992. The chronology of the archaeological assemblages from the head of the Arabian Gulf to the Arabian Sea, 8000-1750 B.C. In: Ehrich R, ed. Chronologies in Old World archaeology, vol I. 3rd ed. Chicago: University of Chicago Press. p 63-76.

Potts D. 1993a. The late Prehistoric, Protohistoric, and Early Historic periods in eastern Arabia (ca. 5000-1200 B.C.). J World Prehist 7(2):163-212.

Potts T. 1993b. Patterns of trade in third-millennium BC in Mesopotamia and Iran. World Archaeol 24(3):379-402.

Potts D. 1993c. Tell Abraq and the Harappan tradition in southeastern Arabia. In: Possehl G, ed. Harappan civilization: a recent perspective. 2nd ed. New Delhi: Oxford &IBH Publishers Co. p 323-333.

Potts D. 1993d. Four seasons of excavation at Tell Abraq (1989-1993). Proc Sem Arab Stud 23:117-126.

Potts D. 1993e. A new Bactrian find from southeastern Arabia. Antiquity 67(256):591- 596.

401 Potts D. 1994a. Contributions to the agrarian history of eastern Arabia II. The cultivars. Arab Archaeol Epigr 5:236-275.

Potts D. 1994b. South and central Asian elements at Tell Abraq (Emirate of Umm al- Qaiwain, United Arab Emirates), c. 2200 BC - AD 400. In: Parpola A, Koskikallio P, eds. South Asian archaeology 1993. Helsinki: Annales Academiae Scientarum Fennicae Ser B-271. p 615-628.

Potts D. 1995. Watercraft of the lower sea. In: Dittmann R, Finkbeiner U, and Hauptamann H, editors. Beitrage zur Kulturgeschichte Vorderasiens: Festschrift fur Rainer M Boehmer Mainz: Von Zabern. p 559-571.

Potts D. 1997a. Before the emirates: an archaeological and historical account of developments in the region c. 5000 BC to 676 AD. In: Ghareeb E, Al Abed I, eds. Perspectives on the United Arab Emirates. London: Trident Press. p 36-73.

Potts D. 1997b. Rewriting the late prehistory of south-eastern Arabia: a reply to Jocelyn Orchard. Iraq 59:63-71.

Potts D. 1998. Maritime beginnings. In: Hellyer P, ed. Waves of time: the maritime history of the United Arab Emirates. London: Trident Press. p 8-43.

Potts D. 2000. Ancient Magan: the secrets of Tell Abraq. London: Trident Press.

Potts D. 2001. Before the emirates: an archaeological and historical account of developments in the region c. 5000 BC to 676 AD. In: Al Abed I, Hellyer P, eds. United Arab Emirates: a new perspective. London: Trident Press Ltd. p 28-69.

Potts D. 2004a. The Uruk explosion: more heat than light? The Review of Archaeology 25(2):19-28.

Potts D. 2004b. Excavations at Tepe Yahya, Iran, 1967-1975, vol. III: the third millennium. Lamberg-Karlovsky C, editor. Cambridge: Peabody Museum of Archaeology and Ethnology.

Potts D. 2005. In the beginning: Marhashi and the origins of Magan's ceramic industry in the third millennium BC. Arab Archaeol Epigr 26:67-78.

Potts D. 2008. Arabian Peninsula. In: Pearsall D, ed. Encyclopedia of archaeology. New York: Elsevier. p 827-834.

Potts D. 2009. The archaeology and early history of the Persian Gulf. In: Potter L, ed. The Persian Gulf in history. New York: Palgrave Macmillan. p 27-53.

402 Potts D, Weeks L. 1999. An AMS radiocarbon chronology for the late Umm an-Nar type tomb at Tell Abraq. Tribulus 9(1):9-10.

Price T. 1989. Bones, chemistry, and the human past. In: Price T, ed. The chemistry of prehistoric human bone. Cambridge: Cambridge University Press. p 1-9.

Price T, Bentley R, Luning J, Gronenborn D, Wahl J. 2001. Prehistoric human migration in the Linearbandkeramik of central Europe. Antiquity 75:593-603.

Price T, Blitz J, Burton J, Ezzo J. 1992. Diagenesis in prehistoric bone: problems and solutions. J Archaeol Sci 19:513-529.

Price T, Burton J, Bentley R. 2002. The characterization of biologically available strontium isotope ratios for the study of prehistoric migration. Archaeometry 44(1):117- 135.

Price T, Burton J, Sharer R, Buikstra J, Wright L, Traxler L, Miller K. 2010. Kings and commoners at Copan: isotopic evidence for origins and movement in the Classic Maya period. J Anthropol Archaeol 29:15-32.

Price T, Gestsdottir H. 2006. The first settlers of Iceland: an isotopic approach to colonisation. Antiquity 80(307):130-144.

Price T, Grupe G, Schroter P. 1994a. Reconstruction of migration patterns in the Bell Beaker period by stable strontium isotope analysis. Appl Geochem 9:413-417.

Price T, Grupe G, Schroter P. 1998. Migration in the Bell Beaker period of central Europe. Antiquity 72:405-411.

Price T, Johnson C, Ezzo J, Ericson J, Burton J. 1994b. Residential mobility in the prehistoric Southwest United States: a preliminary study using strontium isotope analysis. J Archaeol Sci 21:315-330.

Price T, Knipper C, Grupe G, Smrcka V. 2004. Strontium isotopes and prehistoric human migration: the Bell Beaker people in central Europe. European J Archaeol 7(1):9-40.

Price T, Manzanilla L, Middleton W. 2000. Immigration and the ancient city of Teotihuacan in Mexico: a study using strontium isotope ratios in human bone and teeth. J Archaeol Sci 27(10):903-913.

Price T, Schoeninger M, Armelagos G. 1985. Bone chemistry and past behavior: an overview. J Hum Evol 14:419-447.

Price T, Tiesler V, Burton J. 2006. Early African diaspora in colonial Campeche, Mexico: strontium isotopic evidence. Am J Phys Anthropol 130(4):485-490.

403

Probst A, El Gh'mari A, Aubert D, Fritz P, McNutt R. 2000. Strontium as a tracer of weathering processes in a silicate catchment polluted by acid atmospheric inputs, Strengbach, France. Chem Geol 170(1-4):203-219.

Probst A, Viville D, Fritz P, Ambroise B, Dambrine E. 1992. Hydrochemical budgets of a small forested granitic catchment exposed to acid deposition: the Strengbach catchment case study (Vosges Massif, France) Water, Air, Soil Pollut 62:337-347.

Prowse T, Schwarcz H, Garnsey P, Knyf M, Macchiarelli R, and Bondioli L. 2007. Isotopic evidence for age-related immigration to imperial Rome. Am J Phys Anthropol 132(4):510-519.

Prowse T, Schwarcz H, Saunders S, Macchiarelli R, Bondioli L. 2004. Isotopic paleodiet studies of skeletons from the Imperial Roman-age cemetery of Isola Sacra, Rome, Italy. J Archaeol Sci 31(3):259-272.

Ratnagar S. 1981. Encounters: the westerly trade of the Harappa civilization. Delhi: Oxford University Press.

Ratnagar S. 2001. The Bronze Age: unique instance of a pre-industrial world system? Curr Anthropol 42(3):351-365.

Reade J. 1979. Early etched beads and the Indus-Mesopotamia trade. London: British Museum.

Reade J. 2008. The Indus-Mesopotamian relationship reconsidered. In: Olijdam E, and Spoor R, eds. Intercultural relations between south and southwest Asia: studies in commemoration of ECL During Caspers (1934-1996). Oxford: BAR International Series 1826. p 12-18.

Reitsema L, Crews D. 2011. Brief communication: oxygen isotopes as a biomarker for sickle-cell disease? Results from transgenic mice expressing human hemoglobin S genes. Am J Phys Anthropol 145(3):495-498.

Renfrew J. 1984. Cereals cultivated in ancient Iraq. Bulletin on Sumerian Agriculture 1:32-44.

Rey C, Collins B, Goehl T, Dickson I, Glimcher M. 1989. The carbonate environment in bone mineral: a resolution-enhanced Fourier transform infrared spectroscopy study. Calcified Tissue Int 45:157-164.

404 Rey C, Renugopalakrishnan V, Shimizu M, Collins B, Glimcher M. 1991. A resolution- enhanced Fourier transform infrared spectroscopic study of the environment of the CO32- ion in the mineral phase of enamel during its formation and maturation. Calcified Tissue Int 49:259-268.

Richards M, Mays S, Fuller B. 2002. Stable carbon and nitrogen isotope values of bone and teeth reflect weaning age at the medieval Wharram Percy site, Yorkshire, UK. Am J Phys Anthropol 119:205-210.

Riley M, Petrie C. 1999. An analysis of the architecture of the tomb at Sharm, Fujairah, U.A.E. Arab Archaeol Epigr 10(2):180-189.

Roaf M, Galbraith J. 1994. Pottery and p-values: 'Seafaring merchants or Ur?' re- examined. Antiquity 68:770-783.

Rowley-Conwy P, Deakin W, Shaw C. 1997. Ancient DNA from archaeological sorghum (Sorghum bicolor) from Qasr Ibrim, Nubia: implications for domestication and evolution and a review of the archaeological evidence. Sahara 9:23-34.

Rozanski K, Araguas-Araguas L, Gonfiantini R. 1993. Isotopic patterns in modern global precipitation. In: Swart P, Lohmann K, McKenzie J, Savin S, eds. Climate change in continental isotopic records. Washington, D.C.: American Geophysical Union. p 1-36.

Ruff CB, Larsen CS, Hayes WC. 1984. Structural changes in the femur with the transition to agriculture on the Georgia Coast. Am J Phys Anthropol 64:125-136.

Sahm N. 1988. Preliminary report of the excavation of the Umm an-Nar tomb in Shimal North. In: Kästner J-M, Sahm N, and Velde C, eds. Excavations of the German archaeological mission to Ras al-Khaimah: report of the 4th season 1988. Göttingen: Seminar für Vorderasiatische Archaölogie. p 1-4.

Salim S. 1962. Marsh dwellers of the Euphrates Delta. New York: Humanities Press.

Salles J-F. 1984. Failaka fouilles Francaises 1983. Lyon: Travaux de la Maison de l'Orient No. 9.

Salvatori S. 1996. Death and ritual in a population of coastal food foragers in Oman. In: Afanas'ev G, Cleuziou S, Lukacs J, Tosi M, eds. The prehistory of Asia and Oceania: XIII International Congress of Prehistoric and Protohistoric Sciences. Forli: Abaco. p 205-222.

Salvatori S. 2001. Excavations at the funerary structures HD 10-3.1, 3.2, 4.1, 4.2 and 2.1 at Ra's al-Hadd (Sultanate of Oman). Rivista di Archeologia 25:67-77.

405 Sanlaville P. 1992. Changements climatiques dans la Peninsule Arabique durant le Pleistocene superieur et l'Holocene. Paleorient 18(1):5-26.

Santini G. 1987. Site RH-10 at Qurum and a preliminary analysis of its cemetery: an essay in stratigraphic discontinuity. Proc Sem Arab Stud 17:179-198.

Sasaki T, Takagi M, Yanagisawa T. 1997. Structure and function of secretory ameloblasts in enamel formation. In: Chadwick D, Cardew G, eds. Dental enamel: proceedings of the Ciba Foundation Symposium 205. Chichester: John Wiley. p 32-50.

Schneider J. 1977. Was there a pre-capitalist world-system? Peasant Stud 6(1):20-29.

Schoeninger M. 1985. Trophic level effects on 15N/14N and 13C/12C ratios in bone collagen and strontium levels in bone mineral. J Hum Evol 14:515-525.

Schoeninger M, Moore K. 1992. Bone stable isotope studies in archaeology. J World Prehist 6(2):247-296.

Schurr M. 1997. Stable nitrogen isotopes as evidence for the age of weaning at the Angel site: a comparison of isotopic and demographic measures of weaning age. J Archaeol Sci 24:919-927.

Schurr M. 1998. Using stable nitrogen-isotopes to study weaning behavior in past populations. World Archaeol 30:327-342.

Schurr M, and Powell M. 2005. The role of changing childhood diets in the prehistoric evolution of food production: an isotopic assessment. Am J Phys Anthropol 126:278-294.

Schutkowski H. 1988. Report on the anthropological activities during the 1988 campaign at the sites of Shimal and Dhayah, Ras al-Khaimah, U.A.E. Ras al-Khaimah, U.A.E. p 7- 10.

Schutkowski H. 1989. Report on the anthropological activities during the 1989 campaign and a brief sketch of first results. Ras al-Khaimah, U.A.E.

Schutkowski H, Hansen B, Wormuth M, Herrmann B. 2001. Signatures of stable strontium (Sr) isotopes in human hard tissue: utilisation in osteological identification. Res Legal Med 26:31-40.

Schutkowski H, Herrmann B. 1987. Anthropological report on human remains from the cemetery at Shimal. In: Vogt B, Franke-Vogt U, eds. Shimal 1985/1986: excavations of the German archaeological mission in Ras al-Khaimah, UAE. Berlin: Dietrich Reimer. p 55-65.

406 Schwarcz H, Melbye J, Katzenberg M, Knyf M. 1985. Stable isotopes in human skeletons of southern Ontario: reconstructing paleodiet. J Archaeol Sci 12:187-206.

Schwarcz H, Schoeninger M. 1991. Stable isotope analyses in human nutritional ecology. Yearb Phys Anthropol 34:283-321.

Schweissing M, Grupe G. 2000. Local or nonlocal? A research of strontium isotope ratios of teeth and bones on skeletal remains with artificial deformed skulls Anthropologischer Anzeiger 58(1):99-103.

Schweissing M, Grupe G. 2003a. Stable strontium isotopes in human teeth and bone: a key to migration events of late Roman period in Bavaria. J Archaeol Sci 30(11):1373- 1383.

Schweissing M, Grupe G. 2003b. Tracing migration events in man and cattle by stable strontium isotope analysis of appositionally grown mineralized tissue. Int J Osteoarchaeol 13:96-103.

Scott E. 1979. Dental wear scoring technique. Am J Phys Anthropol 51(2):213-217.

Sealy J, Armstrong R, Schrire C. 1995. Beyond lifetime averages: tracing life histories through isotopic analysis of different calcified tissues from archaeological human skeletons. Antiquity 69:290-300.

Sealy J, van der Merwe N, Sillen A, Kruger F, Krueger H. 1991. 87Sr/86Sr as a dietary indicator in modern and archaeological bone. J Archaeol Sci 18:399-416.

Shipman P, Foster G, Schoeninger M. 1984. Burnt bones and teeth: an experimental study of color, morphology, crystal structure and shrinkage. J Archaeol Sci 11:307-325.

Sillen A, Hall G, Richardson S, Armstrong R. 1998. 87Sr/86Sr ratios in modern and fossil food-webs of the Sterkfontein Valley: implications for early hominid habitat preference. Geochim Cosmochim Acta 62(14):2463-2473.

Smith G. 1978. Two prehistoric sites on Ras Abaruk, Site 4. In: de Cardi B, editor. Qatar archaeological report: excavations 1973. Oxford: Oxford University Press. p 80-106.

Smith B. 1984. Patterns of molar wear in hunter-gatherers and agriculturalists. Am J Phys Anthropol 63:39-56.

Smith B, Epstein S. 1971. Two categories of 13C/12C ratios for higher plants. Plant Physiol 47:380-384.

Sponheimer M, Lee-Thorp J. 1999. Oxygen isotopes in enamel carbonate and their ecological significance. J Archaeol Sci 26:723-728.

407

Srivastava K. 1991. Madinat Hamad: burial mounds 1984-85. Bahrain: Ministry of Information, Bahrain National Museum.

Staubwasser M, Sirocko F, Grootes P, Segl M. 2003. Climate change at the 4.2 ka BP termination of the Indus Valley civilisation and Holocene south Asian monsoon variability. Geophys Res Lett 30(8):1425-1428.

Stein G. 1999a. Rethinking world-systems: diasporas, colonies, and interaction in Uruk Mesopotamia. Tucson: The University of Arizona Press.

Stein G. 1999b. Material culture and social identity: the evidence for a 4th millennium BC Mesopotamian Uruk colony at Hacinebi, Turkey. Paleorient 25(1):11-22.

Stein G. 2001. Indigenous social complexity at Hacinebi (Turkey) and the organization of Uruk colonial contact. In: Rothman M, ed. Uruk Mesopotamia and its neighbors: cross- cultural interactions in the era of state formation. Santa Fe: School of American Research Press. p 265-306.

Stein G. 2002a. From passive periphery to active agents: emerging perspectives in the archaeology of interregional interaction. Am Anthropol 104(3):903-916.

Stein G. 2002b. Colonies without colonialism: a trade diaspora model of fourth millennium B.C. Mesopotamian enclaves in Anatolia. In: Lyons C, and Papadopoulos J, eds. The archaeology of colonialism. Los Angeles: The Getty Research Institute. p 27-64.

Stephan E. 1995. Preliminary report on the faunal remains of the first two seasons at Tell Abraq/Umm al-Quwain/United Arab Emirates. In: Buitenhuis H, Uerpmann H-P, eds. Archaeozoology of the Near East II. Leiden: Bachuys. p 52-63.

Strongman S. 1993. Preliminary report on the human remains excavated from site 2, Umm al Qaiwain, United Arab Emirates, April 1993. p 1-4.

Sullivan C, Krueger H. 1981. Carbon isotope analysis of separate chemical phases in modern and fossil bone. Nature 292:333-335.

Tafuri M, Bentley R, Manzi G, Di Lernia S. 2006. Mobility and kinship in the prehistoric Saraha: strontium isotope analysis of Holocene human skeletons from the Acacus Mts. (southwestern Libya). J Anthropol Archaeol 25:390-402.

Taha M. 1982-1983. The archaeology of the Arabian Gulf during the first millennium B.C. Al-Rafidan 3-4:75-87.

408 Tengberg M. 1998. Paleoenvironnements et economie vegetale en milieu aride - recherches archeobotaniques dans la region du Golfe arabo-persique et dans le Makran pakistanais (4eme millenaire av. notre ere - 1er millenaire de notre ere. Montpellier: University of Montpellier II.

Tengberg M. 1999. Crop husbandry at Miri Qalat, Makran, SW Pakistan (4000-2000 B.C.). Vegetation Hist Archaeobot 8:3-12.

Tengberg M. 2003a. Research into origins of date palm domestication. The date palm: from traditional resource to green wealth. Abu Dhabi: The Emirates Center for Strategic Studies and Research. p 51-62.

Tengberg M. 2003b. Archaeobotany in the Oman Peninsula and the role of eastern Arabia in the spread of African crops. In: Neumann K, Butler A, Kahlheber S, eds. Food, fuel and fields: progress in African archaeobotany. Koln: Heinrich-Barth-Institut. p 229- 237.

Thomas H, Geraads M, Janjou D, Vaslet D, Memesh A, Billiou D, Bocherens H, Dobigny G, Eisenmann V, Gayet M and others. 1998. First Pleistocene faunas from the Arabian Peninsula: an Nafud Desert, Saudi Arabia. Comptes Rendus de l'Academie des Sciences Series IIA Earth Planet Sci 226(2):145-152.

Thorvildsen K. 1963. Gravrøser på Umm an-Nar (Burial cairns on Umm an-Nar). Kuml: Årbog for Jysk Arkæologisk Selskab 1962:191-219.

Tieszen L, Chapman M. 1992. Carbon and nitrogen isotopic status of the major marine and terrestrial resources in the Atacama Desert of northern Chile. Proceedings of the First World Congress on Mummy Studies. p 409-425.

Tieszen L, Fagre T. 1993. Effect of diet quality and composition on the isotopic composition of respiratory CO2, bone collagen, bioapatite, and soft tissues. In: Lambert J, Grupe G, eds. Prehistoric human bone: archaeology at the molecular level. New York: Springer-Verlag. p 121-155.

Tixier J. 1980. Mission archeologique francaise a Qatar, 1976-77, 1977-78: tome 1. Doha: Ministry of Information, Department of Tourism and Antiquities.

Tixier J. 1986. The prehistory of the Gulf: recent finds. In: Al-Khalifa S, Rice M, eds. Bahrain through the ages: the archaeology. New York: KPI. p 76-78.

Tosi M. 1975. Notes on the distribution and exploitation of natural resources in ancient Oman. J Oman Stud 1:187-206.

Tosi M, and Usai D. 2003. Preliminary report on the excavations at Wadi Shab, Area 1, Sultanate of Oman. Arab Archaeol Epigr 14(1):8-23.

409

Turner B, Kamenov G, Kingston J, Armelagos G. 2009. Insights into immigration and social class at Machu Picchu, Peru based on oxygen, strontium, and lead isotopic analysis. J Archaeol Sci 36(2):317-332.

Uerpmann M. 1989. Some remarks on Late Stone Age industries from the coastal area of northern Oman. In: Costa P, Tosi M, eds. Oman studies. Rome: Serie Orientale Roma LXIII. p 169-177.

Uerpmann M. 1992. Structuring the Late Stone Age of southeastern Arabia. Arab Archaeol Epigr 3:65-109.

Uerpmann M. 2001. Remarks on the animal economy of Tell Abraq (Emirates of Sharjah and Umm al-Qaywayn, UAE). Proc Sem Arab Stud 31:227-234.

Uerpmann M. 2002. The dark millennium - remarks on the final Stone Age in the Emirates and Oman. In: Potts D, Al Naboodah H, Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident Press. p 74-81.

Uerpmann M, Uerpmann H-P. 1994. Animal bone finds from excavation 520 at Qala'at al-Bahrain. In: Højlund F, Andersen H, eds. Qala'at al-Bahrain, vol I: the northern city wall and the Islamic fortress. Aarhus: Jutland Archaeological Society Publications XXX:1. p 417-454.

Uerpmann M, Uerpmann H-P. 1996. 'Ubaid pottery in the eastern Gulf - new evidence from Umm al-Qaiwain (U.A.E.). Arab Archaeol Epigr 7:125-139.

Uerpmann M, Uerpmann H-P. 1997. Animal bones from Excavation 519 at Qala'at al- Bahrain. In: Højlund F, Andersen H, eds. Qala'at al-Bahrain, vol II: the central monumental buildings. Munskgaard: JASP XXX:2. p 235-264.

Uerpmann H-P, Uerpmann M. 1999. The camel burial of al-Buhais 12 (Sharjah, U.A.E.). In: Becker C, Manhart H, Peters J, Schiber J, eds. Historia Animalium Ex Ossibus - Beitrage zur Palaoanatomie, Archaologie, Agyptologi, Ethnologie u Geschichte der Tiermedizin; Festschrift fur Angela von den Driesch zum 65. Rahden: Leidorf 8. p 455- 462.

Uerpmann M, Uerpmann H-P. 2000. Faunal remains of Al-Buhais 18: an aceramic Neolithic site in the Emirate of Sharjah (SE-Arabia) - excavations 1995-1998. In: Mashkour M, Choyke A, Buitenhuis H, Poplin F, eds. Archaeozoology of the Near East IV, vol B: proceedings of the fourth international symposium on the archaeozoology of southwestern Asia and adjacent areas. Groningen: ARC. p 40-49.

410 Uerpmann H-P, Uerpmann M. 2002. The appearance of the domestic camel in south-east Arabia. J Oman Stud 12:235-260.

Uerpmann H-P, Uerpmann M. 2003. Conclusions: the development of culture, subsistence and the environment in northern Oman during the Late Stone Age. Stone Age sites and their natural environment: the capital area of northern Oman, Part III. Weisbaden: Dr. Ludwig Reichert. p 255-261.

Uerpmann H-P, Uerpmann M, Jasim S. 2006. Funeral monuments and human remains from Jebel al-Buhais. Sharjah, UAE: Department of Culture and Information.

Uerpmann M, Uerpmann H-P, Jasim S. 2000. Stone Age nomadism in SE-Arabia - palaeo-economic considerations on the Neolithic site of Al-Buhais 18 in the Emirate of Sharjah, U.A.E. Proc Sem Arab Stud 30:229-234.

Uerpmann H-P, Uerpmann M, Jasim S, eds. 2008. The natural environment of Jebel al- Buhais: past and present. Sharjah, UAE: Department of Culture and Information.

Uerpmann H-P, Potts D, Uerpmann M. 2009. Holocene (re-)occupation of eastern Arabia. In: Petraglia M, Rose J, eds. The evolution of human populations in Arabia: palaeoenvironments, prehistory and genetics. Netherlands: Springer. p 205-214.

Usai D. 2006. A fourth-millennium BC Oman site and its context: Wadi Shab-GAS1. Proc Sem Arab Stud 36:275-288.

Valentine B, Kamenov G, Krigbaum J. 2008. Reconstructing Neolithic groups in Sarawak, Malaysia, through lead and strontium isotope analysis. J Archaeol Sci 35:1463- 1473. van der Merwe N, Vogel J. 1978. 13C content of human collagen as a measure of prehistoric diet in woodland North America. Nature 276:815-816.

Velde C. 1991. Preliminary remarks on the settlement pottery in Shimal. In: Schippmann K, Herling A, Salles J-F, eds. Golf-Archaologie. Buch am Erlbach: Leidorf. p 265-288.

Velde C. 2003. Wadi Suq and Late Bronze Age in the Oman Peninsula. In: Potts D, Al Naboodah H, Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident Press. p 101-114.

Velde C. 2010. The landscape of the Middle Bronze Age in the UAE - where did people live? New perspectives on recording UAE history.

411 Velde C. nd-a. The cemeteries of Shimal and Dhayah (Middle Bronze Age/Wadi Suq period), Ras al-Khaimah, United Arab Emirates. Ras al-Khaimah: Publications of the National Museum of Ras al-Khaimah.

Velde C. nd-b. 14.0 SH 103. Ras al-Khaimah: Unpublished report of the Department of Antiquities & Museums. p 304-309.

Velde C. nd-c. 11.0 SH 95. Ras al-Khaimah: Unpublished report of the Department of Antiquities & Museums. p 178-185.

Velde C, Moellering I. 2010. Ras al-Khaimah heritage: Umm an-Nar tomb. Government of Ras al-Khaimah: RAK Tourism Office.

Vitousek P, Kennedy M, Derry L, Chadwick O. 1999. Weathering versus atmospheric sources of strontium in ecosystems on young volcanic soils. Oecologia 121(2):255-259.

Vogt B. 1985. The Umm an-Nar Tomb A at Hili North: a preliminary report on three seasons of excavation, 1982-1984. Archaeol United Arab Emirates 4:20-37.

Vogt B. 1994a. In search of coastal sites in pre-historic Makkan: mid Holocene "shell- eaters" in the coastal desert of Ras al-Khaimah, U.A.E. In: Kenoyer J, ed. From Sumer to Meluhha: contributions to the archaeology of south and west Asia in memory of George F Dales, Jr. Madison: Department of Anthropology, University of Wisconsin. p 113-128.

Vogt B. 1994b. Asimah: an account of two months rescue excavation in the mountains of Ras al-Khaimah, United Arab Emirates. Dubai: Shell Markets Middle East.

Vogt B. 1998. State, problems and perspectives of second millennium B.C. funerary studies in the Emirate of Ras al-Khaimah (U.A.E.). In: Phillips C, Potts D, Searight S, eds. Arabia and her neighbours: essays on prehistorical and historical developments presented in honour of Beatrice de Cardi. Leyden: Brepols. p 273-290.

Vogt B, Franke-Vogt U. 1987. Shimal 1985/1986: excavations of the German Archaeological Mission in Ras al-Khaimah, U.A.E., a preliminary report. Berlin: Berliner Beitrage zum Vorderen Orient 8.

Vogt B, Haser J, Kästner J-M, Schutkowski H, Velde C. 1989. Preliminary remarks on two recently excavated tombs in Shimal, Ras al-Khaimah (United Arab Emirates). South Asian archaeology 1985. p 62-73.

Vogt B, Velde C. 1987. Ghalilah tomb SH 103. In: Vogt B, Franke-Vogt U, eds. Shimal 1985/1986: excavations of the German archaeological mission in Ras al-Khaimah, UAE. Berlin: Dietrich Reimer Verlag. p 37-43.

412 von den Driesch A. 1994. Viehhaltung, Jagd und Fischfang in der bronzezeitlichen Siedlung von Shimal bei Ras al-Khaimah/U.A.E. In: Calmeyer P, Hecker K, Jacob-Post L, Walker C, eds. Beitrage zur Alorientlischen Archaologie und Altertumskunde: Festschrift fur Barthel Hrouda zum 65. Weisbaden: Harrassowitz. p 73-85.

Wahida G, Al-Tikriti W, Beech M, Al Meqbali A. 2009. A Middle Paleolithic assemblage from Jebel Barakah, coastal Abu Dhabi Emirate. In: Petraglia M, Rose J, eds. The evolution of human populations in Arabia: paleoenvironments, prehistory and genetics. The Netherlands: Springer. p 117-124.

Walker P, Miller K, Richman R. 2008. Time, temperature, and oxygen availability: an experimental study of the effects of environmental conditions on the color and organic content of cremated bone. In: Schmidt C, Symes S, eds. The analysis of burned human remains. Amsterdam: Academic Press. p 129-135.

Wallerstein I. 1974. The modern world-system: capitalist agriculture and the origins of the European world economy in the sixteenth century. New York: Academic Press.

Wallerstein I. 2000. The essential Wallerstein. New York: The New Press.

Wang Y, Cerling T, MacFadden B. 1994. Fossil horses and carbon isotopes: new evidence for Cenozoic dietary, habitat, and ecosystem changes in North America. Palaeogeography, Palaeoclimatology, Palaeoecology 107:269-279.

Weeks L. 1997. Prehistoric metallurgy at Tell Abraq, U.A.E. Arab Archaeol Epigr 8:11- 85.

Weeks L. 1999. Lead isotope analyses from Tell Abraq, United Arab Emirates: new data regarding the 'tin problem' in western Asia. Antiquity 73:49-64.

Weeks L. 2000. Pre-Islamic metallurgy of the Gulf. Ph.D. Dissertation. Sydney: University of Sydney.

Weeks L. 2003a. Early metallurgy of the Persian Gulf: technology, trade, and the Bronze Age World. Boston: Brill Academic Publishers, Inc.

Weeks L. 2003b. Prehistoric metallurgy in the U.A.E.: Bronze Age - Iron Age transitions. In: Potts D, Al Naboodah H, Hellyer P, eds. Archaeology of the United Arab Emirates: proceedings of the first international conference on the archaeology of the UAE. London: Trident Press. p 115-122.

Weeks L, Collerson K. 2005. Archaeometallurgical studies. In: Killick R, Moon J, eds. The Early Dilmun settlement at Saar. Ludlow: Archaeology International. p 309-324.

413 Weidemann F, Bocherens H, Mariotti A, von den Driesch A, and Grupe G. 1999. Methodological and archaeological implications of intra-tooth isotopic variation (d13C, d18O) in herbivores from Ain Ghazal (Jordan, Neolithic). J Archaeol Sci 26:697-704.

Weisgerber G. 1980. "...Und Kupfer in Oman" - Das Oman-Projekt des Deutschen Berhbau-Museums. Der Anschnitt 2/3:62-110.

Weisgerber G. 1981. Evidence of ancient mining sites in Oman: a preliminary report. J Oman Stud 4:15-28.

Weisgerber G. 1983. Copper production during the third millennium BC in Oman and the question of Makan. J Oman Stud 6(2):269-276.

Weisgerber G. 1984. Makan and Meluhha - third millennium BC copper production in Oman and the evidence of contact with the Indus Valley. In: Allchin B, ed. South Asian archaeology 1981. Cambridge: Cambridge University Press. p 196-201.

Weiss H, Bradley R. 2001. What drives societal collapse? Science 291(5504):609-610.

Wells C. 1984. Appendix 6: human bone. Oriens Antiquus 23:213-220, 268-282.

Western A. 1989. The flora of the United Arab Emirates: an introduction. UAE: United Arab Emirates University.

Whelan T, Sackett W, Benedict C. 1973. Enzymatic fractionation of carbon isotopes by phosphoenolpyruvate carboxylase from C4 plants. Plant Physiol 51(6):1051-1054.

Whipkey C, Capo R, Chadwick O, Stewart B. 2000. The importance of sea spray to the cation budget of a coastal Hawaii soil: a strontium isotope approach. Chem Geol 168(1- 2):37-48.

White W. 2007. Geochemistry: To be published at Johns Hopkins University Press, currently online at: . Accessed 3/22/11.

White C, Longstaffe F, Law K. 2004. Exploring the effects of environment, physiology and diet on oxygen isotope ratios in ancient Nubian bones and teeth. J Archaeol Sci 31(2):233-250.

White C, Price T, Longstaffe F. 2007. Residential histories of the human sacrifices at the Moon Pyramid, Teotihuacan: evidence from oxygen and strontium isotopes. Ancient Mesoam 18:159-172.

White C, Spence M, Longstaffe F, Law K. 2000. Testing the nature of Teotihuacan imperialism at Kaminaljuyu using phosphate oxygen-isotope ratios. J Anthropol Res 56:535-558.

414

Whitehouse D. 1975. Carnelian in the Persian Gulf. Antiquity 49:129f.

Wilkinson J. 1983. The origins of the Aflaj of Oman. J Oman Stud 6(1):177-194.

Wilkinson T. 1997. Holocene environments of the high plateau, Yemen: recent geological investigations. Geoarchaeology 12(8):833-864.

Willcox G. 1992. Timber and trees: ancient exploitation in the Middle East: evidence from plant remains. Bull Sumer Agric 6:1-32.

Willcox G. 1994. Archaeobotanical finds. In: Højlund F, Andersen H, eds. Qala'at al- Bahrain, vol I: the northern city wall and the Islamic fortress. Aarhus: Jutland Archaeological Publications XXX: 1. p 459-462.

Willcox G. 1995. Some plant impressions from Umm an-Nar Island. In: Frifelt K, ed. The island of Umm an-Nar, vol II: the third millennium settlement: Jutland Archaeological Publications XXVI: 2. p 257-259.

Williams J, White C, Longstaffe F. 2005. Trophic level and macronutrient shift effects associated with the weaning process in the Postclassic Maya. Am J Phys Anthropol 128:781-790.

Wong W, Benedict C, Kohel R. 1979. Enzymatic fractionation of the stable carbon isotopes of carbon dioxide by ribulose-1,5-bisphosphate carboxylase. Plant Physiol 63(5):852-856.

Wood JW, Milner GR, Harpending HC, Weiss KM. 1992. The osteological paradox: problems of inferring prehistoric health from skeletal samples. Curr Anthropol 33:343- 370.

Wright L. 2005. Identifying immigrants to Tikal, Guatemala: defining local variability in strontium isotope ratios of human tooth enamel. J Archaeol Sci 32:555-566.

Wright H, Johnson G. 1975. Population, exchange, and early state formation in southwestern Iran. Am Anthropol 77:267-289.

Wright L, Schwarcz H. 1998. Stable carbon and oxygen isotopes in human tooth enamel: identifying breastfeeding and weaning in prehistory. Am J Phys Anthropol 106:1-18.

Wright L, Schwarcz H. 1999. Correspondence between stable carbon, oxygen and nitrogen isotopes in human tooth enamel and dentine: infant diets at Kaminaljuyu. J Archaeol Sci 26:1159-1170.

415 Wright LE, Yoder CJ. 2003. Recent progress in bioarchaeology: approaches to the osteological paradox. J Archaeol Res 11(1):43-70.

Yule P. 2001. Die Graberfeld in Samad Al Shan (Sultanat Oman): Materialien zu einer Kulturgeschichte. Rahden/Westf: Verlag Marie Leidorf.

Yurtsever Y, Gat J. 1981. Atmospheric waters. In: Gat J, Gonflantini R, eds. Stable isotope hydrology: deuterium and oxygen-18 in the water cycle. Vienna: International Atomic Energy Agency. p 103-142.

Zarins J. 1989. Eastern Saudi Arabia and external relations: selected ceramic, steatite and textual evidence: 3500-1900 B.C. In: Frifelt K, Sorensen P, eds. South Asian archaeology 1985. London: Scandinavian Institute of Asian Studies Occasional Paper 4. p 74-103.

Ziolkowski M. 2008. Bidyah Mosque, Fujairah, U.A.E. Adumatu 17:7-32.

Ziolkowski M, al-Sharqi A. 2006. Dot-in-circle: an ethnoarchaeological approach to soft- stone vessel decoration. Arab Archaeol Epigr 17(2):152-162.

416

APPENDIX A

LIST OF ENAMEL SAMPLES AND SAMPLE IDENTIFICATION

417

ID Site Country Species Tooth Specimen Information

BM 1 Burial Mounds Bahrain sheep/goat LM1 A-244-B BM 2 Burial Mounds Bahrain sheep/goat LM2 A-217-B BM 3 Burial Mounds Bahrain sheep/goat LM3 A-219-B BM 4 Burial Mounds Bahrain sheep/goat LM2 A-243-B BM 5 Burial Mounds Bahrain sheep/goat LM2 A-245-B

Bar 6 Barbar Bahrain cattle LM1 ZMK 53/1980 517 ADZ 5570-5576 Bar 7 Barbar Bahrain cattle LM2 ZMK 53/1980 517 ABZ 4289-4291

Bar 8 Barbar Bahrain cattle RM3 ZMK 53/1980 517 ACB 4299-4308 Bar 9 418 Barbar Bahrain cattle RM1 ZMK 53/1980 517 AEA 1556-1562, 1566 Barbar Bahrain cattle LM Bar 10 1 ZMK 53/1980 517 AEA 1556-1562, 1566 Bar 11 Barbar Bahrain cattle RM1 ZMK 53/1980 517 AEA 1556-1562, 1566 Bar 12 Barbar Bahrain cattle LM2 ZMK 53/1980 517 AEA 1556-1562, 1566 Bar 13 Barbar Bahrain cattle LM2 ZMK 53/1980 517 AEA 1556-1562, 1566 Bar 14 Barbar Bahrain sheep/goat LM ZMK 53/1980 517 AEA 1546-1552, 1568- 3 1569, 1582, 1609-1614, 1665 Bar 15 Barbar Bahrain sheep/goat LM ZMK 53/1980 517 AEA 1546-1552, 1568- 3 1569, 1582, 1609-1614, 1665 Bar 16 Barbar Bahrain sheep/goat LM ZMK 53/1980 517 AEA 1546-1552, 1568- 3 1569, 1582, 1609-1614, 1665 Bar 17 Barbar Bahrain sheep/goat RM ZMK 53/1980 517 AEA 1546-1552, 1568- 3 1569, 1582, 1609-1614, 1665 Bar 18 Barbar Bahrain sheep/goat RM1 ZMK 53/1980 517 AEA 1546-1552, 1568- 1569, 1582, 1609-1614, 1665

ID Site Country Species Tooth Specimen Information

Bar 18 Barbar Bahrain sheep/goat RM1 ZMK 53/1980 517 AEA 1546-1552, 1568- 1569, 1582, 1609-1614, 1665 Bar 19 Barbar Bahrain sheep/goat RM ZMK 53/1980 517 AEA 1546-1552, 1568- 1 1569, 1582, 1609-1614, 1665 ZMK 53/1980 517 AEA 1546-1552, 1568- Bar 20 Barbar Bahrain sheep/goat LM1 1569, 1582, 1609-1614, 1665 FAI 21 Failaka Kuwait cattle LM2 ZMK 145/1975 F6 168-ER FAI 22 Failaka Kuwait cattle RM2 ZMK 145/1975 F6 366-MN

FAI 23 Failaka Kuwait cattle RM3

419 ZMK 145/1975 F6 366-MN

Failaka Kuwait cattle LM3 FAI 24 ZMK 145/1975 F6 441-PN FAI 25 Failaka Kuwait cattle LM2 ZMK 145/1975 F6 467-QO FAI 26 Failaka Kuwait cattle LM2 ZMK 145/1975 F6 467-QO FAI 27 Failaka Kuwait cattle LM2 ZMK 145/1975 F6 561-UI FAI 28 Failaka Kuwait sheep/goat LM3 ZMK 145/1975 F3 881-QL FAI 29 Failaka Kuwait sheep/goat LM2 ZMK 145/1975 F3 881-QT FAI 30 Failaka Kuwait sheep/goat RM2 ZMK 145/1975 F3 881-QT FAI 31 Failaka Kuwait sheep/goat RM3 ZMK 145/1975 F3 881-QT FAI 32 Failaka Kuwait sheep/goat LM3 ZMK 145/1975 F3 881-BMT FAI 33 Failaka Kuwait sheep/goat LM3 ZMK 145/1975 F3 881-BNE FAI 34 Failaka Kuwait sheep/goat LM1/2 ZMK 145/1975 F6 1129 FAI 35 Failaka Kuwait sheep/goat RM2 ZMK 145/1975 F6 296-JT

ID Site Country Species Tooth Specimen Information

UaN 36 Umm an-Nar Island UAE cattle LM1 ZMK 86/1979 8 lower-18/3/1979

UaN 37 Umm an-Nar Island UAE cattle RM3 ZMK 115/1966 8 upper-18/3/1979

UaN 38 Umm an-Nar Island UAE sheep/goat RM1 ZMK 115/1966 1 fill-11/3/1979

UaN 39 Umm an-Nar Island UAE sheep/goat LM1 ZMK 115/1966 2 lower-29/3/1979

UaN 40 Umm an-Nar Island UAE sheep/goat LM2 ZMK 115/1966 2 lower-29/3/1979

UaN 41 Umm an-Nar Island UAE sheep/goat RM2 ZMK 115/1966 2 upper-12/2/1979

UaN 42 Umm an-Nar Island UAE oryx RM3 ZMK 115/1966 7 lower-12/3/1979 2 420 UaN 43 Umm an-Nar Island UAE sheep/goat LM ZMK 115/1966 9 upper-5/3/1979

UaN 44 Umm an-Nar Island UAE sheep/goat RM1 ZMK 115/1966 9 upper-5/3/1979

UaN 45 Umm an-Nar Island UAE sheep/goat LM1 ZMK 115/1966 10 lower-27/2/1979 UaN 46 Umm an-Nar Island UAE sheep/goat RM1 ZMK 115/1966 10 lower-27/2/1979 UaN 47 Umm an-Nar Island UAE sheep/goat LM2 ZMK 115/1966 10 lower-27/2/1979 UaN 48 Umm an-Nar Island UAE sheep/goat RM2 ZMK 115/1966 10 lower-27/2/1979 UaN 49 Umm an-Nar Island UAE sheep/goat RM2 ZMK 115/1966 10 lower-27/2/1979 UaN 50 Umm an-Nar Island UAE sheep/goat RM2 ZMK 115/1966 9+8 lower-21/3/1979

Dib 51 Dibba UAE sheep/goat LM1 76, A:2, L:5 Qid 52 Qidfa UAE sheep/goat RM1 4-995, Sq D, N:5 TY 68 Tepe Yahya Iran cattle LM3 903-167 TY 69 Tepe Yahya Iran cattle RP3 61-127 TY 70 Tepe Yahya Iran cattle RM1/2 90-3 TY 71 Tepe Yahya Iran sheep/goat RM3 53-5 TY 72 Tepe Yahya Iran sheep/goat LM3 46-18

ID Site Country Species Tooth Specimen Information

TY 73 Tepe Yahya Iran sheep/goat LM3 115-34 TY 74 Tepe Yahya Iran sheep/goat LM3 113-17 TY 75 Tepe Yahya Iran sheep/goat LM3 23-50 TY 76 Tepe Yahya Iran pig RM3 61-1-125 TY 77 Tepe Yahya Iran pig RM3 579-56 Bal 78 Balakot Pakistan gazelle LM2 4Blk 240 #4 Bal 79 Balakot Pakistan pig RI1 4Blk 240 #2 Bal 80 Balakot Pakistan cattle RM1 4Blk 240 #3 Bal 81 Balakot Pakistan cattle LM1/2 4Blk 292-1 Bal 82 Balakot Pakistan sheep/goat RM1/2 4Blk 292 #2 3 421 Bal 83 Balakot Pakistan sheep/goat RM 4Blk 292 #3 3

Bal 84 Balakot Pakistan cattle RM 4Blk 292 #4 Bal 85 Balakot Pakistan cattle RM3 4Blk 147-242 Bal 86 Balakot Pakistan cattle LM2 4Blk 147-441 Bal 87 Balakot Pakistan cattle RM3 4Blk 140-10

ALL 88 Allahdino Pakistan sheep/goat RM2 A404 ALL 89 Allahdino Pakistan sheep/goat LM2/3 A982 ALL 90 Allahdino Pakistan goat RM2 A480 ALL 91 Allahdino Pakistan goat LM2 A1074 ALL 92 Allahdino Pakistan sheep/goat LM2 A1051 ALL 93 Allahdino Pakistan goat LM3 A492 ALL 94 Allahdino Pakistan cattle RP4 A977

mation

Nar Nar III? Nar III?

- -

E196 E213 E194 E194 E194 E213 A383 A1073 A491A 9 Wadi Suq II? 1 Wadi Suq II 7 Wadi Suq IV 2 Wadi Suq IV 3 Wadi Suq IV 4 Wadi Suq IV 6 Wadi Suq II? 5 Wadi Suq II? 926 Wadi Suq IV 91 E85 N117 locus 1002 E85 N117 925 Umm an 937 Umm an Specimen Infor Specimen

3 3 3 1 1 1 2 1 1 1 2 3 3 2 3 1/2 1/2 2/3 1/2 1/2 1/2 M LM LM LM LM LM LM LM RM RM RM R RM RM RM RM LM RM RM RM RM RM Tooth

pig ep/goat goat goat goat goat goat cattle cattle cattle cattle sheep sheep sheep sheep sheep Species sheep/goat sheep/goat sheep/goat sheep/goat she sheep/goat

UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE Pakistan Pakistan Pakistan Country

Site Shimal Shimal Shimal Shimal Shimal Shimal Allahdino Allahdino Allahdino Tell Abraq Tell Abraq Tell Tell Abraq Tell Abraq Tell Abraq Tell Abraq Tell Abraq Tell Abraq Tell Abraq Tell Abraq Tell Abraq Tell Tell Abraq Tell

ID TA 99 TA 98 SH 115 SH 110 SH 111 SH 112 SH 113 SH 114 TA 105 TA 106 TA 107 TA 108 TA 109 TA 101 TA 102 TA 103 TA 104 TA 100 ALL 96 ALL 97 ALL ALL 95 ALL

422

1 2 - -

1, Bag 31 1, Bag 31 1, Bag 50 2, Bag 43 3, Bag 43 3, Bag 38 4, Bag 38 4, Bag 42 5, Bag 12 6, Bag 32 7, Bag 22 8, Bag 11 9, Bag 10, Bag 11 10, Bag ------E194 E213 E213 - L16A, B L16A, 1010 AA 1010 NV 1010 NV 1010 BB 1010 BB ox 8 Box 8 Box 8 Box 8 Box 8 Box 8 Box 8 Box 8 Box 8 Box 8 B Box 8 Box 8 Box 8 ecimen Information ecimen Sp

1 1 1 1 1 3 1 1 1 1 1 3 1 1 2 1 1 3 1 3 1 3/4 LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM RM RM RM RM RM RM LP Tooth

human human human human human human human human human human human human human human human human human human Species sheep/goat sheep/goat sheep/goat sheep/goat

UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE Country

d I

Nar Nar I Island Nar Islan Nar I Island Nar I Island Nar I Island Nar Nar II Island Nar II Island Nar II Island Nar II Island Nar II Island Nar II Island Nar II Island Nar II Island Nar II Island Nar Nar II Island Nar II Island Nar II Island Nar II Island - - - - - Site ------Unar 1 Unar Shimal Shimal Shimal an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm Umm an Umm an

ID SH 118 SH 119 SH 116 SH 117 UaN 136 UaN 137 UaN UaN 131 UaN 132 UaN 133 UaN 134 UaN 135 UaN UaN 127 UaN 128 UaN 129 UaN 130 UaN UaN 122 UaN 123 UaN 124 UaN 125 UaN 126 UaN UaN 120 UaN 121 UaN

423

1 2 A 1, Bag 19 2, Bag 19 3, Bag 19 4, Bag 19 5, Bag 19 ------1, Bag 24 1, Bag 24 2, Bag - - 1, Bag 20 1, Bag 20 2, Bag 20 3, Bag - - - 1, Bag 4 1, Bag 4 2, Bag - - 1089 DL Box 8 rest 509 9342 Box 8 rest Box 8 rest Box 15, Tube 1 Box 15, Tube Box 8A Box 8A 1089 DP, Box 17 Box 13(2), Bag 6 Bag Box 13(2), 1089 DO, Box 16(II) 1089 DO, Box 16(II) 1089 DO, Box 16(III) Box 16(III) Box 16(III) Box 16(III)A Box 16(III)A Specimen Information Specimen 1089 DX, Box 18 1089 DX, Box 18 1089 DX, Box 18 1089 DX, Box 18 1089 DX, Box 18 1089 DX,

1 1 1 1 1 1 1 1 1 1 1 3 1 1 1 1 1 1 1 1 1 LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM Tooth

man human human human human human human human human human human human human hu human human human human human human human human Species

E AE UAE UAE UAE UA UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE U UAE UAE UAE UAE Country

nd II

and V and

Nar Nar Isla Nar Nar II Island Nar II Island Nar II Island Nar II Island Nar V Nar Island V Nar Island V Nar Island V Nar Island Nar Isl V Nar Island V Nar Island V Nar Island Nar V Nar Island V Nar Island V Nar Island V Nar Island V Nar Island V Nar Island V Nar Island Site ------Tell Abraq Tell Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an Umm an

138

ID TA 158 UaN 157 UaN UaN 153 UaN 154 UaN 155 UaN 156 UaN UaN 148 UaN 149 UaN 150 UaN 151 UaN 152 UaN UaN 144 UaN 145 UaN 146 UaN 147 UaN UaN 139 UaN 140 UaN 141 UaN 142 UaN 143 UaN UaN

424

ID Site Country Species Tooth Specimen Information

TA 159 Tell Abraq UAE human LM1 512

TA 160 Tell Abraq UAE human LM1 513 12570-C

TA 161 Tell Abraq UAE human LM1 514 6134-D

TA 162 Tell Abraq UAE human LM1 515 6175-A

TA 163 Tell Abraq UAE human LM1 516 #111

TA 164 Tell Abraq UAE human LM1 517 232

TA 165 Tell Abraq UAE human LM1 518 72/112-D

TA 166 Tell Abraq UAE human LM1 520 1121 L

TA 167 Tell Abraq UAE human LM1 527

425 TA 168 Tell Abraq UAE human LM1 534 17618-K

TA 169 Tell Abraq UAE human LM1 540

TA 170 Tell Abraq UAE human LM1 542 75/114-C

TA 171 Tell Abraq UAE human LM1 547 75/115A 6

TA 172 Tell Abraq UAE human LM1 553 74/115-C

TA 173 Tell Abraq UAE human LM1 554

TA 174 Tell Abraq UAE human LM1 558 75/115-A

TA 175 Tell Abraq UAE human LM1 580 75/114D

TA 176 Tell Abraq UAE human LM1 584 74/114D

TA 177 Tell Abraq UAE human LM1 587 74/115A

TA 178 Tell Abraq UAE human LM1 588 74/115A

TA 179 Tell Abraq UAE human LM1 828 234

ID Site Country Species Tooth Specimen Information

TA 180 Tell Abraq UAE human LM1 829 1094

TA 181 Tell Abraq UAE human LM1 831 626A

TA 182 Tell Abraq UAE human LM1 849 16610-A

TA 183 Tell Abraq UAE human LM1 851 #112

TA 184 Tell Abraq UAE human LM1 852 233

TA 185 Tell Abraq UAE human LM1 854 626A

TA 186 Tell Abraq UAE human LM1 1059 (from Sharjah Museum)

TA 187 Tell Abraq UAE human LM2 1059 (from Sharjah Museum) MW 188

426 Mowaihat UAE human LM1 Bag A-1

MW 189 Mowaihat UAE human LM1 Bag A-2

MW 190 Mowaihat UAE human LM1 Bag D

MW 191 Mowaihat UAE human LM3 Bag D

MW 192 Mowaihat UAE human LM1 Bag E, Individual 1

MW 193 Mowaihat UAE human LM3 Bag E, Individual 1

MW 194 Mowaihat UAE human LM1 Bag E, Individual 2

MW 195 Mowaihat UAE human LM1 Bag F

MW 196 Mowaihat UAE human LM2 Bag F

MW 197 Mowaihat UAE human LM1 Bag M

MW 198 Mowaihat UAE human LM1 Bag O

MW 199 Mowaihat UAE human LM1 Bag P

MW 200 Mowaihat UAE human LM1 Bag S

1 2 1 - - - L1 L2 L5 L5 L8 B15f B15f Bag S Bag Skull 1 Skull 1 B15 38 B15 44 B15 12 B16 41 B17 Skull 20 Skull 20 Skull 32 X5 N7C XS N7C B3 10/11 N B3 Specimen Information Specimen

1 1 1 1 1 2 1 3 1 3 1 1 1 1 1 1 1 1 1 1 LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM LM RM RM Tooth

man human human human human human human human human human human human human human human human human hu human human human Species

UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE UAE Country

Site Unar 1 Unar 1 Unar 1 Unar 1 Unar 1 Unar Mowaihat Mowaihat Mowaihat Mowaihat Mowaihat Mowaihat Shimal 95 Shimal 95 Shimal 103 Shimal 103 Shimal 103 Shimal 103 Shimal 103 Shimal 103 Shimal 103

ID SH 215 SH 211 SH 212 SH 213 SH 214 SH 207 SH 208 SH 209 SH 210 MW 202 MW 203 MW 204 MW 205 MW 206 MW 201 RAK 220 RAK RAK 216 RAK 217 RAK 218 RAK 219 RAK

427

ID Site Country Species Tooth Specimen Information

RAK 221 Unar 1 UAE human LM1 L8-2

RAK 222 Unar 1 UAE human LM1 L8-3

RAK 223 Unar 1 UAE human LM1 L9.81

RAK 224 Unar 1 UAE human LM1 L9E

RAK 225 Unar 1 UAE human LM1 L10.31

RAK 226 Unar 1 UAE human LM1 L11.22

RAK 227 Unar 1 UAE human LM2 L11.22

RAK 228 Unar 1 UAE human LM1 L11B

RAK 229 Unar 1 UAE human LM1 L12A

428

RAK 230 Unar 1 UAE human LM1 L12E

RAK 231 Unar 1 UAE human LM1 L12G

RAK 232 Unar 1 UAE human LM1 L12-C

RAK 233 Unar 1 UAE human LM1 L12L-1

RAK 234 Unar 1 UAE human LM2 L12L-1

RAK 235 Unar 1 UAE human LM1 L12L-2

RAK 236 Unar 1 UAE human LM1 L12L-3

RAK 237 Unar 1 UAE human LM1 L12L-4

RAK 238 Unar 1 UAE human LM1 L12L-5

RAK 239 Unar 1 UAE human LM1 L13-1

RAK 240 Unar 1 UAE human LM2 L13-1

RAK 241 Unar 1 UAE human LM1 L13-2

ID Site Country Species Tooth Specimen Information

RAK 242 Unar 1 UAE human LM1 L13-3

RAK 243 Unar 1 UAE human LM1 L16D Bid 244 Bidya 1 UAE human RM1 1987 Unit 18 Bid 245 Bidya 1 UAE human RM1 1987 Unit 18 Bid 246 Bidya 1 UAE human RM2 1987 DH 247 Dadna UAE human RM2 DHDN-1-95 SQ-B 50-60 28-11-95

MR 248 Mereshed UAE human RM1 Tomb Unit 4 1997 100 cm

Qid 250 Qidfa 4 UAE human RM1 1994 Sq B 16.11.94

Qid 251 Qidfa 4 UAE human RM2 1994 Sq B 16.11.94

429 Dib 252 Dibba 76 UAE human LM2 Sq 6 125-135 Cm 11 Jan 93, Bag #26

Dib 253 Dibba 76 UAE human LM3 Sq 6 125-135 Cm 11 Jan 93, Bag #26

Dib 254 Dibba 76 UAE human LM1 Sq 7, 125-135 cm, 12 Jan 94, Bag #29

Dib 255 Dibba 76 UAE human LM2 61A Sq 12 L12G, 'skull' BM 257 Bahrain Burial Mounds Bahrain human LM1 A-218.C BM 258 Bahrain Burial Mounds Bahrain human LM1 A-227.A BM 259 Bahrain Burial Mounds Bahrain human LM1 A-236.A BM 260 Bahrain Burial Mounds Bahrain human LM1 A-238.A BM 261 Bahrain Burial Mounds Bahrain human LM1 A-244.B BM 262 Bahrain Burial Mounds Bahrain human LM1 A-246.A OM 263 al-Khubayb Oman human LM 18-01-2010 Survey Unit#0001 Site 005 Locus 1 003 Lot 1 Bag #30

Unit#0001 Site 005 Locus Unit#0001 Site 005 Locus 003 Lot 1 Bag #30 1 Bag 003 Lot #30 1 Bag 003 Lot Specimen Information Specimen 2010 Survey 2010 Survey Unit#00012010 Survey Site 005 Locus - - 01 01 - - 18 18

2 3 LM LM Tooth

human human Species

Oman Oman Country

Site Khubayb Khubayb - - al al

ID OM 265 OM 264

430

APPENDIX B

ISOTOPE DATA FOR FAUNAL SAMPLES

431

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

BM 1 Burial Mounds Bahrain sheep/goat LM1 0.708263 5.4 -1.9 BM 2 Burial Mounds Bahrain sheep/goat LM2 0.708233 -1.2 -9.6 BM 3 Burial Mounds Bahrain sheep/goat LM3 0.708168 2.6 -9.4 BM 4 Burial Mounds Bahrain sheep/goat LM2 0.708209 11.8 -8.6 BM 5 Burial Mounds Bahrain sheep/goat LM2 0.708523 - -

Bar 6 Barbar Bahrain cattle LM1 0.708107 2.6 -3.2 Bar 7 Barbar Bahrain cattle LM2 0.708303 -1.3 -6.7

Bar 8 Barbar Bahrain cattle RM3 0.708184 3.9 -0.9 Bar 9 Barbar Bahrain cattle RM1 0.708108 3.3 -6.3

432 Bar 10 Barbar Bahrain cattle LM1 0.708082 1.0 -5.6 1 Bar 11 Barbar Bahrain cattle RM 0.708572 1.8 -6.1 Bar 12 Barbar Bahrain cattle LM2 0.708174 -2.1 -6.8 Bar 13 Barbar Bahrain cattle LM2 0.708170 -1.5 -7.8 Bar 14 Barbar Bahrain sheep/goat LM3 0.708472 3.6 -10.9 Bar 15 Barbar Bahrain sheep/goat LM3 0.708320 2.3 -3.4 Bar 16 Barbar Bahrain sheep/goat LM3 0.708256 -2.4 -7.1 Bar 17 Barbar Bahrain sheep/goat RM3 0.708349 2.8 -3.1 Bar 18 Barbar Bahrain sheep/goat RM1 0.708240 -1.5 -7.7 Bar 19 Barbar Bahrain sheep/goat RM1 0.708114 2.3 -6.0 Bar 20 Barbar Bahrain sheep/goat LM1 0.708518 5.3 -11.2

FAI 21 Failaka Kuwait cattle LM2 0.708861 0.4 -6.0

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

FAI 22 Failaka Kuwait cattle RM2 0.708580 1.5 -7.7 FAI 23 Failaka Kuwait cattle RM3 0.708669 0.3 -6.7 FAI 24 Failaka Kuwait cattle LM3 0.708750 1.5 1.3 FAI 25 Failaka Kuwait cattle LM2 0.708571 -2.3 -8.9 FAI 26 Failaka Kuwait cattle LM2 0.708669 2.8 -1.7 FAI 27 Failaka Kuwait cattle LM2 0.708417 2.3 4.1 FAI 28 Failaka Kuwait sheep/goat LM3 0.708423 1.8 -4.1 FAI 29 Failaka Kuwait sheep/goat LM2 0.709023 -0.2 -6.8 FAI 30 Failaka Kuwait sheep/goat RM2 0.709011 -0.2 -6.8

433 FAI 31 Failaka Kuwait sheep/goat RM3 0.709024 0.4 -7.0 FAI 32 Failaka Kuwait sheep/goat LM3 0.708427 5.3 -4.6 FAI 33 Failaka Kuwait sheep/goat LM3 0.708392 -1.4 -6.1 FAI 34 Failaka Kuwait sheep/goat LM1/2 0.708447 3.2 0.0 FAI 35 Failaka Kuwait sheep/goat RM2 0.708413 6.6 -6.7 UaN 36 Umm an-Nar Island UAE cattle LM1 0.708752 2.9 -4.5

UaN 37 Umm an-Nar Island UAE cattle RM3 0.708900 8.3 -1.7

UaN 38 Umm an-Nar Island UAE sheep/goat RM1 0.708826 4.8 -6.5

UaN 39 Umm an-Nar Island UAE sheep/goat LM1 0.708765 4.8 -9.0

UaN 40 Umm an-Nar Island UAE sheep/goat LM2 0.708838 2.7 -7.0

UaN 41 Umm an-Nar Island UAE sheep/goat RM2 0.708892 1.6 -8.0

UaN 42 Umm an-Nar Island UAE oryx RM3 0.708775 10.2 5.9

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

UaN 43 Umm an-Nar Island UAE sheep/goat LM2 0.708631 5.4 -7.3

UaN 44 Umm an-Nar Island UAE sheep/goat RM1 0.708779 4.9 -6.6

UaN 45 Umm an-Nar Island UAE sheep/goat LM1 0.708695 4.4 -5.7 UaN 46 Umm an-Nar Island UAE sheep/goat RM1 0.708792 3.0 -6.4 UaN 47 Umm an-Nar Island UAE sheep/goat LM2 0.708591 6.3 -6.9 UaN 48 Umm an-Nar Island UAE sheep/goat RM2 0.708867 3.4 -5.3 UaN 49 Umm an-Nar Island UAE sheep/goat RM2 0.708845 5.1 -5.3 UaN 50 Umm an-Nar Island UAE sheep/goat RM2 0.708867 3.8 -7.0

Dib 51 Dibba UAE sheep/goat LM1 0.708686 3.1 -1.3 Qid 52 Qidfa UAE sheep/goat RM1 0.708535 0.6 -7.0 434 TY 68 Tepe Yahya Iran cattle LM3 0.708301 1.8 -0.6 TY 69 Tepe Yahya Iran cattle RP3 0.708214 2.9 1.4 TY 70 Tepe Yahya Iran cattle RM1/2 0.708094 -0.2 -4.8 TY 71 Tepe Yahya Iran sheep/goat RM3 0.708652 4.8 -9.5 TY 72 Tepe Yahya Iran sheep/goat LM3 0.708184 6.1 -4.4 TY 73 Tepe Yahya Iran sheep/goat LM3 0.708267 3.8 -3.6 TY 74 Tepe Yahya Iran sheep/goat LM3 0.708221 -0.6 -2.4 TY 75 Tepe Yahya Iran sheep/goat LM3 0.708382 6.6 -7.4 TY 76 Tepe Yahya Iran pig RM3 0.708121 -2.6 -11.2 TY 77 Tepe Yahya Iran pig RM3 0.708132 -3.7 -10.7 Bal 78 Balakot Pakistan gazelle LM2 0.709113 3.1 -10.9 Bal 79 Balakot Pakistan pig RI1 0.708855 -3.6 ± 0.086* -1.2 ± 0.106* Bal 80 Balakot Pakistan cattle RM1 0.709668 -1.8 0.0

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

Bal 81 Balakot Pakistan cattle LM1/2 0.708886 -0.9 1.5 Bal 82 Balakot Pakistan sheep/goat RM1/2 0.708955 0.8 -4.8 Bal 83 Balakot Pakistan sheep/goat RM3 0.708901 3.6 -8.0 Bal 84 Balakot Pakistan cattle RM3 0.708745 2.7 -5.6 Bal 85 Balakot Pakistan cattle RM3 0.708887 -0.9 0.6 Bal 86 Balakot Pakistan cattle LM2 0.708911 0.2 1.7 Bal 87 Balakot Pakistan cattle RM3 0.708877 0.2 3.6

ALL 88 Allahdino Pakistan sheep/goat RM2 0.708645 4.9 -10.8 ALL 89 Allahdino Pakistan sheep/goat LM2/3 0.708759 -9.1 -13.8

435 ALL 90 Allahdino Pakistan goat RM2 0.708764 6.2 -9.3

ALL 91 Allahdino Pakistan goat LM2 0.708719 -0.2 -4.8 ALL 92 Allahdino Pakistan sheep/goat LM2 0.708757 3.6 -7.6 ALL 93 Allahdino Pakistan goat LM3 0.709133 3.4 -6.6 ALL 94 Allahdino Pakistan cattle RP4 0.708720 3.5 -10.8 ALL 95 Allahdino Pakistan goat LM1 0.710417 1.5 -3.4 ALL 96 Allahdino Pakistan goat LM2 0.708781 -0.4 -7.2 ALL 97 Allahdino Pakistan goat RM3 0.708446 3.1 -9.7

TA 98 Tell Abraq UAE goat RM3 0.708816 0.8 -6.5

TA 99 Tell Abraq UAE goat RM3 0.708701 3.1 0.4

TA 100 Tell Abraq UAE sheep RM1 0.708745 4.9 0.6

TA 101 Tell Abraq UAE sheep RM1 0.708843 2.4 -5.7

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

TA 102 Tell Abraq UAE sheep LM3 0.708746 1.9 0.5

TA 103 Tell Abraq UAE sheep RM1 0.708668 4.0 -0.7

TA 104 Tell Abraq UAE sheep LM3 0.708760 2.2 -0.2 TA 105 Tell Abraq UAE cattle RM1/2 0.708756 1.9 1.0

TA 106 Tell Abraq UAE cattle RM2 0.708788 4.5 -2.1 TA 107 Tell Abraq UAE cattle LM2 0.708797 3.8 4.6

TA 108 Tell Abraq UAE cattle RM2/3 0.708826 2.9 -1.8 TA 109 Tell Abraq UAE pig RM1 0.708862 -1.2 1.3 SH 110 Shimal UAE sheep/goat RM1/2 0.708890 -0.2 -6.7 1/2

436 SH 111 Shimal UAE sheep/goat RM 0.708646 0.3 -2.3 Shimal sheep/goat LM3

SH 112 UAE 0.708768 1.2 -4.5 SH 113 Shimal UAE sheep/goat RM1/2 0.708846 -0.6 -5.6 SH 114 Shimal UAE sheep/goat LM1 0.708973 4.1 -7.7 SH 115 Shimal UAE sheep/goat LM1/2 0.708686 2.8 -5.5 SH 116 Unar 1 UAE sheep/goat LP3/4 0.708711 3.7 -8.9 SH 117 Shimal UAE sheep/goat RM1 0.708856 -0.5 -7.1 SH 118 Shimal UAE sheep/goat LM3 0.708649 6.4 ± 0.003* -7.9 ± 0.006* SH 119 Shimal UAE sheep/goat LM1 0.708859 1.4 ± 0.170* -5.7 ± 0.158* *Duplicate measurement

APPENDIX C

ISOTOPE DATA FOR HUMAN SAMPLES

437

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C UaN 120 Umm an-Nar Island I UAE human RM1 0.708795 -2.0 -7.1 UaN 121 Umm an-Nar Island I UAE human RM1 0.708872 -1.9 -8.4 UaN 122 Umm an-Nar Island I UAE human RM1 0.709001 -1.8 0.9 UaN 123 Umm an-Nar Island I UAE human RM1 0.708905 -1.9 -7.5 UaN 124 Umm an-Nar Island I UAE human RM3 0.708926 -2.3 -2.6

UaN 125 Umm an-Nar Island II UAE human LM1 0.708896 -2.3 ± 0.184* -5.4 ± 0.080*

UaN 126 Umm an-Nar Island II UAE human LM2 0.708908 -3.2 ± 0.139* -4.6 ± 0.002*

UaN 127 Umm an-Nar Island II UAE human LM1 0.708916 -2.6 -2.6

UaN 128 Umm an-Nar Island II UAE human LM1 0.708906 -1.8 ± 0.161* -5.9 ± 0.022*

UaN 129 Umm an-Nar Island II UAE human LM3 0.709003 -2.5 -1.3

438 UaN 130 Umm an-Nar Island II UAE human LM1 0.708649 -1.1 -11.3

UaN 131 Umm an-Nar Island II UAE human LM3 0.708726 -2.0 -8.4

UaN 132 Umm an-Nar Island II UAE human LM1 0.708862 -2.5 -7.1

UaN 133 Umm an-Nar Island II UAE human LM1 0.708884 -2.5 -4.8

UaN 134 Umm an-Nar Island II UAE human LM1 0.708942 -2.7 -7.3

UaN 135 Umm an-Nar Island II UAE human LM1 0.708894 -2.7 -5.2

UaN 136 Umm an-Nar Island II UAE human LM1 0.708733 -1.7 -9.0

UaN 137 Umm an-Nar Island II UAE human LM1 0.708765 -2.1 -4.4

UaN 138 Umm an-Nar Island II UAE human LM1 0.708939 -2.5 -3.7

UaN 139 Umm an-Nar Island II UAE human LM1 0.708998 -2.4 -1.7

UaN 140 Umm an-Nar Island II UAE human LM1 0.708897 -2.5 -6.7 UaN 141 Umm an-Nar Island II UAE human LM1 0.708918 -2.4 -6.8

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

UaN 142 Umm an-Nar Island II UAE human LM1 0.708853 -2.1 -8.1

UaN 143 Umm an-Nar Island V UAE human LM1 0.709035 -1.1 -7.6

UaN 144 Umm an-Nar Island V UAE human LM1 0.708918 -2.3 -3.9

UaN 145 Umm an-Nar Island V UAE human LM3 0.708931 -1.4 -4.6

UaN 146 Umm an-Nar Island V UAE human LM1 0.708871 -2.6 -5.3

UaN 147 Umm an-Nar Island V UAE human LM1 0.708958 -1.7 -8.5

UaN 148 Umm an-Nar Island V UAE human LM1 0.708959 -2.7 -4.8

UaN 149 Umm an-Nar Island V UAE human LM1 0.708939 -2.3 ± 0.059* -2.8 ± 0.025*

UaN 150 Umm an-Nar Island V UAE human LM1 0.709029 -2.3 ± 0.078* -5.9 ± 0.037*

439 UaN 151 Umm an-Nar Island V UAE human LM1 0.708933 -2.1 -5.9

UaN 152 Umm an-Nar Island V UAE human LM1 0.708971 -1.7 -4.9

UaN 153 Umm an-Nar Island V UAE human LM1 0.708903 -2.7 -6.5

UaN 154 Umm an-Nar Island V UAE human LM1 0.708903 -3.1 ± 0.027* -5.9 ± 0.011*

UaN 155 Umm an-Nar Island V UAE human LM1 0.708881 -3.0 ± 0.148* -5.7 ± 0.004*

UaN 156 Umm an-Nar Island V UAE human LM1 0.708947 -2.5 ± 0.158* -6.7 ± 0.019*

UaN 157 Umm an-Nar Island V UAE human LM1 0.708908 -2.6 -8.4

TA 158 Tell Abraq UAE human LM1 0.708887 -1.9 -8.8

TA 159 Tell Abraq UAE human LM1 0.708899 -1.6 -10.8

TA 160 Tell Abraq UAE human LM1 0.708866 -2.1 -8.4

TA 161 Tell Abraq UAE human LM1 0.710661 -1.6 -7.5

TA 162 Tell Abraq UAE human LM1 0.708875 -1.4 -9.3

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

TA 163 Tell Abraq UAE human LM1 0.708839 -2.0 -9.2

TA 164 Tell Abraq UAE human LM1 0.708877 -1.3 -9.9

TA 165 Tell Abraq UAE human LM1 0.708179 -6.0 ± 0.095* -13.6 ± 0.034*

TA 166 Tell Abraq UAE human LM1 0.708906 -2.3 -9.2

TA 167 Tell Abraq UAE human LM1 0.708878 -1.9 -11.6

TA 168 Tell Abraq UAE human LM1 0.708869 -2.8 -8.9

TA 169 Tell Abraq UAE human LM1 0.708878 -2.3 ± 0.041* -9.8 ± 0.020*

TA 170 Tell Abraq UAE human LM1 0.708862 -1.2 -10.5

TA 171 Tell Abraq UAE human LM1 0.708884 -1.6 ± 0.030* -10.6 ± 0.029*

440 TA 172 Tell Abraq UAE human LM1 0.708886 -2.7 -5.9

TA 173 Tell Abraq UAE human LM1 0.708856 -2.0 -10.3

TA 174 Tell Abraq UAE human LM1 0.708863 -3.2 -5.7

TA 175 Tell Abraq UAE human LM1 0.708896 -2.5 -5.4

TA 176 Tell Abraq UAE human LM1 0.708901 -1.7 -8.0

TA 177 Tell Abraq UAE human LM1 0.708862 -1.5 -9.9

TA 178 Tell Abraq UAE human LM1 0.708871 -2.0 -7.4

TA 179 Tell Abraq UAE human LM1 0.708820 -2.5 -10.2

TA 180 Tell Abraq UAE human LM1 0.708888 -2.4 -8.6

TA 181 Tell Abraq UAE human LM1 0.708884 -2.4 ± 0.146* -8.1 ± 0.034*

TA 182 Tell Abraq UAE human LM1 0.708863 -2.4 -10.3

TA 183 Tell Abraq UAE human LM1 0.708878 -2.5 -6.6

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

TA 184 Tell Abraq UAE human LM1 0.708839 -2.3 ± 0.138* -6.6 ± 0.022*

TA 185 Tell Abraq UAE human LM1 0.708876 -1.9 ± 0.119* -9.2 ± 0.021*

TA 186 Tell Abraq UAE human LM1 0.708866 -2.5 ± 0.214* -6.6 ± 0.162*

TA 187 Tell Abraq UAE human LM2 0.708869 -2.4 -6.4

MW 188 Mowaihat UAE human LM1 0.708866 -2.1 -11.3

MW 189 Mowaihat UAE human LM1 0.708835 -1.8 -11.1

MW 190 Mowaihat UAE human LM1 0.708865 -2.0 -10.5

MW 191 Mowaihat UAE human LM3 0.708862 -2.7 -9.7 MW 192

441 Mowaihat UAE human LM1 0.708863 -2.0 -10.7

MW 193 Mowaihat UAE human LM3 0.708873 -2.3 -9.5

MW 194 Mowaihat UAE human LM1 0.708868 -2.4 -10.5

MW 195 Mowaihat UAE human LM1 0.708858 -2.0 -8.2

MW 196 Mowaihat UAE human LM2 0.708858 -2.4 -8.7

MW 197 Mowaihat UAE human LM1 0.708582 -0.7 -13.0

MW 198 Mowaihat UAE human LM1 0.708844 -2.5 -10.2

MW 199 Mowaihat UAE human LM1 0.708859 -2.4 -7.2

MW 200 Mowaihat UAE human LM1 0.708873 -3.1 ± 0.004* -6.2 ± 0.008*

MW 201 Mowaihat UAE human LM2 0.708866 -2.9 -5.1

MW 202 Mowaihat UAE human LM1 0.708879 -2.6 ± 0.009* -6.9 ± 0.069*

MW 203 Mowaihat UAE human LM3 0.708884 -3.0 -6.0

MW 204 Mowaihat UAE human LM1 0.708838 -1.8 -10.9

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

MW 205 Mowaihat UAE human LM3 0.708884 -2.6 -10.2

MW 206 Mowaihat UAE human LM1 0.708860 -2.1 ± 0.001* -11.0 ± 0.019*

SH 207 Shimal 95 UAE human RM1 0.708831 -3.0 -11.0

SH 208 Shimal 95 UAE human RM1 0.708797 -2.2 -10.7

SH 209 Shimal 103 UAE human LM1 0.708880 -2.2 -11.8

SH 210 Shimal 103 UAE human LM1 0.708764 -3.2 -10.3

SH 211 Shimal 103 UAE human LM1 0.708804 -2.4 -11.0

SH 212 Shimal 103 UAE human LM1 0.708822 -3.0 -9.9

SH 213 Shimal 103 UAE human LM1 0.708827 -2.6 -11.5

442 SH 214 Shimal 103 UAE human LM1 0.708828 -0.3 -12.6

SH 215 Shimal 103 UAE human LM1 0.708871 -2.6 ± 0.101* -11.8 ± 0.019*

RAK 216 Unar 1 UAE human LM1 0.708819 -3.6 -10.5

RAK 217 Unar 1 UAE human LM1 0.708757 -3.1 -11.8

RAK 218 Unar 1 UAE human LM1 0.708789 -4.0 -11.0

RAK 219 Unar 1 UAE human LM1 0.708770 -4.7 -12.3

RAK 220 Unar 1 UAE human LM1 0.708748 -12.2 -13.9

RAK 221 Unar 1 UAE human LM1 0.709012 -3.8 -9.8

RAK 222 Unar 1 UAE human LM1 0.708795 -2.9 -11.1

RAK 223 Unar 1 UAE human LM1 0.708767 -3.2 -11.1

RAK 224 Unar 1 UAE human LM1 0.708793 -10.6 -11.1

RAK 225 Unar 1 UAE human LM1 0.708788 -4.2 -11.1

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

RAK 226 Unar 1 UAE human LM1 0.708820 -3.5 -11.0

RAK 227 Unar 1 UAE human LM2 0.708814 -3.4 ± 0.001* -11.5 ± 0.006*

RAK 228 Unar 1 UAE human LM1 0.708777 -3.7 ± 0.003* -11.0 ± 0.036*

RAK 229 Unar 1 UAE human LM1 0.708737 -2.7 -10.6

RAK 230 Unar 1 UAE human LM1 0.708865 -4.1 ± 0.030* -11.1 ± 0.011*

RAK 231 Unar 1 UAE human LM1 0.708789 -2.5 ± 0.023* -8.9 ± 0.049*

RAK 232 Unar 1 UAE human LM1 0.708758 -3.5 ± 0.192* -11.6 ± 0.070*

RAK 233 Unar 1 UAE human LM1 0.708750 -2.8 -11.6

RAK 234 Unar 1 UAE human LM2 0.708763 -3.6 ± 0.057* -10.4 ± 0.018*

443 RAK 235 Unar 1 UAE human LM1 0.708754 -3.3 -11.9

RAK 236 Unar 1 UAE human LM1 0.708783 -3.2 -11.5

RAK 237 Unar 1 UAE human LM1 0.708852 -3.7 -10.8

RAK 238 Unar 1 UAE human LM1 0.708968 -3.5 ± 0.088* -11.9 ± 0.032*

RAK 239 Unar 1 UAE human LM1 0.708820 -3.0 ± 0.081* -11.7 ± 0.042*

RAK 240 Unar 1 UAE human LM2 0.708900 -4.1 -8.6

RAK 241 Unar 1 UAE human LM1 0.708784 -4.1 -11.1

RAK 242 Unar 1 UAE human LM1 0.708801 -4.5 -9.0

RAK 243 Unar 1 UAE human LM1 0.708832 -2.6 -11.3 Bid 244 Bidya 1 UAE human RM1 0.708693 -2.7 -11.1 Bid 245 Bidya 1 UAE human RM1 0.708640 -1.9 -11.1 Bid 246 Bidya 1 UAE human RM2 0.708227 -1.8 -11.7

ID Site Country Species Tooth 87Sr/86Sr δ18O δ13C

DH 247 Dadna UAE human RM2 0.708790 -2.2 -11.4

MR 248 Mereshed UAE human RM2 0.708646 -3.5 -11.0

Qid 250 Qidfa 4 UAE human RM1 0.708674 -2.4 ± 0.045* -11.8 ± 0.031*

Qid 251 Qidfa 4 UAE human RM2 0.708695 -2.4 -11.5

Dib 252 Dibba 76 UAE human LM2 0.708798 -2.7 -11.0

Dib 253 Dibba 76 UAE human LM3 0.708831 -2.5 -10.5

Dib 254 Dibba 76 UAE human LM1 0.709055 -2.6 -9.4

Dib 255 Dibba 76 UAE human LM2 0.708755 -1.8 -11.3 1

444 BM 257 Bahrain Burial Mounds Bahrain human LM 0.708222 -2.5 -10.9 BM 258 Bahrain Burial Mounds Bahrain human LM1 0.708317 -2.2 -12.1

BM 259 Bahrain Burial Mounds Bahrain human LM1 0.708175 -3.4 -12.5 BM 260 Bahrain Burial Mounds Bahrain human LM1 0.708351 -2.7 -12.5 BM 261 Bahrain Burial Mounds Bahrain human LM1 0.708226 -3.4 -12.4 BM 262 Bahrain Burial Mounds Bahrain human LM1 0.708220 -2.9 -11.4

OM 263 al-Khubayb Oman human LM1 0.708356 6.8 -10.1

OM 264 al-Khubayb Oman human LM2 0.708302 3.0 -10.5

OM 265 al-Khubayb Oman human LM3 0.708353 2.8 -11.3 *Duplicate measurement

APPENDIX D:

18O CONVERSION DATA

445 18 18 Appendix 4. Oxygen conversions from  Oc(VPDB) to  Odw(VSMOW). Outliers are 18 denoted in bold text. All  O values are measured in ‰.

18 18 18 18 ID Site δ Oc δ Oc δ Op δ Odw (VPDB) (VSMOW) (VSMOW) (VSMOW) UaN 120 Umm an-Nar Island I -2.0 28.8 19.8 -3.8 UaN 121 Umm an-Nar Island I -1.9 29.0 19.9 -3.6 UaN 122 Umm an-Nar Island I -1.8 29.1 20.0 -3.5 UaN 123 Umm an-Nar Island I -1.9 28.9 19.8 -3.7 UaN 124 Umm an-Nar Island I -2.3 28.6 19.5 -4.1 UaN 125 Umm an-Nar Island II -2.2 28.6 19.5 -4.1 UaN 126 Umm an-Nar Island II -3.2 27.6 18.6 -5.3 UaN 127 Umm an-Nar Island II -2.6 28.3 19.2 -4.5 UaN 128 Umm an-Nar Island II -1.8 29.1 20.0 -3.5 UaN 129 Umm an-Nar Island II -2.5 28.4 19.3 -4.4 UaN 130 Umm an-Nar Island II -1.1 29.8 20.7 -2.6 UaN 131 Umm an-Nar Island II -2.0 28.9 19.8 -3.7 UaN 132 Umm an-Nar Island II -2.5 28.3 19.2 -4.4 UaN 133 Umm an-Nar Island II -2.5 28.3 19.2 -4.5 UaN 134 Umm an-Nar Island II -2.7 28.1 19.1 -4.6 UaN 135 Umm an-Nar Island II -2.7 28.2 19.1 -4.6 UaN 136 Umm an-Nar Island II -1.7 29.2 20.1 -3.4 UaN 137 Umm an-Nar Island II -2.1 28.7 19.6 -3.9 UaN 138 Umm an-Nar Island II -2.5 28.3 19.2 -4.4 UaN 139 Umm an-Nar Island II -2.4 28.4 19.3 -4.3 UaN 140 Umm an-Nar Island II -2.5 28.3 19.3 -4.4 UaN 141 Umm an-Nar Island II -2.4 28.4 19.3 -4.3 UaN 142 Umm an-Nar Island II -2.1 28.7 19.7 -3.9 UaN 143 Umm an-Nar Island V -1.1 29.8 20.7 -2.6 UaN 144 Umm an-Nar Island V -2.3 28.5 19.5 -4.1 UaN 145 Umm an-Nar Island V -1.4 29.5 20.4 -2.9 UaN 146 Umm an-Nar Island V -2.6 28.3 19.2 -4.5 UaN 147 Umm an-Nar Island V -1.7 29.2 20.1 -3.3 UaN 148 Umm an-Nar Island V -2.7 28.1 19.1 -4.7 UaN 149 Umm an-Nar Island V -2.3 28.6 19.5 -4.1 UaN 150 Umm an-Nar Island V -2.3 28.5 19.4 -4.2 UaN 151 Umm an-Nar Island V -2.1 28.7 19.7 -3.9 UaN 152 Umm an-Nar Island V -1.7 29.2 20.1 -3.3 UaN 153 Umm an-Nar Island V -2.7 28.1 19.0 -4.7 UaN 154 Umm an-Nar Island V -3.1 27.7 18.7 -5.2 UaN 155 Umm an-Nar Island V -3.0 27.8 18.7 -5.1 446 UaN 156 Umm an-Nar Island V -2.5 28.3 19.2 -4.4 UaN 157 Umm an-Nar Island V -2.6 28.2 19.2 -4.5 TA 158 Tell Abraq -1.9 28.9 19.9 -3.6 TA 159 Tell Abraq -1.6 29.3 20.2 -3.2 TA 160 Tell Abraq -2.1 28.8 19.7 -3.9 TA 161 Tell Abraq -1.6 29.2 20.2 -3.3 TA 162 Tell Abraq -1.4 29.5 20.4 -2.9 TA 163 Tell Abraq -2.0 28.9 19.8 -3.7 TA 164 Tell Abraq -1.3 29.6 20.5 -2.9 TA 165 Tell Abraq -6.0 24.7 15.7 -9.0 TA 166 Tell Abraq -2.3 28.5 19.4 -4.2 TA 167 Tell Abraq -1.9 29.0 19.9 -3.6 TA 168 Tell Abraq -2.8 28.0 19.0 -4.8 TA 169 Tell Abraq -2.3 28.5 19.4 -4.2 TA 170 Tell Abraq -1.2 29.7 20.6 -2.7 TA 171 Tell Abraq -1.6 29.2 20.1 -3.3 TA 172 Tell Abraq -2.7 28.1 19.0 -4.7 TA 173 Tell Abraq -2.0 28.9 19.8 -3.7 TA 174 Tell Abraq -3.2 27.6 18.6 -5.3 TA 175 Tell Abraq -2.5 28.4 19.3 -4.4 TA 176 Tell Abraq -1.7 29.1 20.1 -3.4 TA 177 Tell Abraq -1.5 29.4 20.3 -3.1 TA 178 Tell Abraq -2.0 28.9 19.8 -3.7 TA 179 Tell Abraq -2.5 28.3 19.3 -4.4 TA 180 Tell Abraq -2.4 28.4 19.4 -4.3 TA 181 Tell Abraq -2.4 28.5 19.4 -4.2 TA 182 Tell Abraq -2.4 28.4 19.3 -4.3 TA 183 Tell Abraq -2.5 28.3 19.3 -4.4 TA 184 Tell Abraq -2.3 28.6 19.5 -4.1 TA 185 Tell Abraq -1.9 28.9 19.8 -3.7 TA 186 Tell Abraq -2.5 28.3 19.3 -4.4 TA 187 Tell Abraq -2.4 28.4 19.3 -4.3 MW 188 Mowaihat -2.1 28.8 19.7 -3.8 MW 189 Mowaihat -1.8 29.1 20.0 -3.5 MW 190 Mowaihat -2.0 28.8 19.8 -3.8 MW 191 Mowaihat -2.7 28.1 19.1 -4.7 MW 192 Mowaihat -2.0 28.8 19.8 -3.8 MW 193 Mowaihat -2.3 28.6 19.5 -4.1 MW 194 Mowaihat -2.4 28.4 19.3 -4.3 MW 195 Mowaihat -2.0 28.9 19.8 -3.7 MW 196 Mowaihat -2.4 28.4 19.4 -4.3 MW 197 Mowaihat -0.7 30.2 21.1 -2.1 447 MW 198 Mowaihat -2.5 28.3 19.2 -4.4 MW 199 Mowaihat -2.4 28.5 19.4 -4.2 MW 200 Mowaihat -3.1 27.7 18.6 -5.2 MW 201 Mowaihat -2.9 27.9 18.8 -5.0 MW 202 Mowaihat -2.6 28.3 19.2 -4.5 MW 203 Mowaihat -3.0 27.8 18.7 -5.1 MW 204 Mowaihat -1.8 29.1 20.0 -3.5 MW 205 Mowaihat -2.6 28.2 19.1 -4.6 MW 206 Mowaihat -2.0 28.8 19.7 -3.8 SH 207 Shimal 95 -3.0 27.8 18.8 -5.0 SH 208 Shimal 95 -2.2 28.6 19.5 -4.0 SH 209 Shimal 103 -2.2 28.7 19.6 -4.0 SH 210 Shimal 103 -3.2 27.6 18.5 -5.3 SH 211 Shimal 103 -2.4 28.5 19.4 -4.2 SH 212 Shimal 103 -3.0 27.8 18.7 -5.1 SH 213 Shimal 103 -2.6 28.2 19.1 -4.6 SH 214 Shimal 103 -0.3 30.6 21.5 -1.5 SH 215 Shimal 103 -2.6 28.2 19.2 -4.5 RAK 216 Unar 1 -3.6 27.2 18.2 -5.8 RAK 217 Unar 1 -3.1 27.7 18.6 -5.2 RAK 218 Unar 1 -4.0 26.7 17.7 -6.4 RAK 219 Unar 1 -4.7 26.1 17.1 -7.2 RAK 220 Unar 1 -12.2 18.3 9.5 -16.9 RAK 221 Unar 1 -3.8 27.0 18.0 -6.1 RAK 222 Unar 1 -2.9 27.9 18.8 -4.9 RAK 223 Unar 1 -3.2 27.6 18.6 -5.3 RAK 224 Unar 1 -10.6 20.0 11.1 -14.9 RAK 225 Unar 1 -4.2 26.5 17.5 -6.7 RAK 226 Unar 1 -3.5 27.3 18.2 -5.7 RAK 227 Unar 1 -3.4 27.5 18.4 -5.5 RAK 228 Unar 1 -3.7 27.1 18.1 -5.9 RAK 229 Unar 1 -2.7 28.1 19.0 -4.7 RAK 230 Unar 1 -4.1 26.7 17.7 -6.5 RAK 231 Unar 1 -2.5 28.3 19.2 -4.4 RAK 232 Unar 1 -3.5 27.3 18.3 -5.6 RAK 233 Unar 1 -2.8 28.1 19.0 -4.7 RAK 234 Unar 1 -3.6 27.2 18.2 -5.8 RAK 235 Unar 1 -3.3 27.5 18.4 -5.4 RAK 236 Unar 1 -3.2 27.6 18.5 -5.3 RAK 237 Unar 1 -3.7 27.1 18.1 -5.9 RAK 238 Unar 1 -3.5 27.3 18.2 -5.8 RAK 239 Unar 1 -3.0 27.8 18.7 -5.1 448 RAK 240 Unar 1 -4.1 26.7 17.6 -6.5 RAK 241 Unar 1 -4.1 26.7 17.6 -6.5 RAK 242 Unar 1 -4.5 26.3 17.3 -6.9 RAK 243 Unar 1 -2.6 28.3 19.2 -4.5 Bid 244 Bidya 1 -2.7 28.1 19.0 -4.7 Bid 245 Bidya 1 -1.9 28.9 19.8 -3.7 Bid 246 Bidya 1 -1.8 29.0 19.9 -3.5 DH 247 Dadna -2.2 28.6 19.5 -4.1 MR 248 Mereshid -3.5 27.3 18.3 -5.7 Qid 250 Qidfa 4 -2.3 28.5 19.4 -4.2 Qid 251 Qidfa 4 -2.4 28.4 19.4 -4.3 Dib 252 Dibba 76 -2.7 28.1 19.1 -4.6 Dib 253 Dibba 76 -2.5 28.3 19.2 -4.4 Dib 254 Dibba 76 -2.6 28.2 19.1 -4.6 Dib 255 Dibba 76 -1.8 29.1 20.0 -3.5

449